From 00fe1ae91e0d69e52e8212d23cd3ecc74a7259a0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Jul 2011 16:24:46 +0200 Subject: [PATCH 001/326] netfilter: xt_rateest: fix xt_rateest_mt_checkentry() commit 4a5a5c73b7cfee (slightly better error reporting) added some useless code in xt_rateest_mt_checkentry(). Fix this so that different error codes can really be returned. Signed-off-by: Eric Dumazet CC: Jan Engelhardt Signed-off-by: Patrick McHardy --- net/netfilter/xt_rateest.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/net/netfilter/xt_rateest.c b/net/netfilter/xt_rateest.c index 76a083184d8e..ed0db15ab00e 100644 --- a/net/netfilter/xt_rateest.c +++ b/net/netfilter/xt_rateest.c @@ -78,7 +78,7 @@ static int xt_rateest_mt_checkentry(const struct xt_mtchk_param *par) { struct xt_rateest_match_info *info = par->matchinfo; struct xt_rateest *est1, *est2; - int ret = false; + int ret = -EINVAL; if (hweight32(info->flags & (XT_RATEEST_MATCH_ABS | XT_RATEEST_MATCH_REL)) != 1) @@ -101,13 +101,12 @@ static int xt_rateest_mt_checkentry(const struct xt_mtchk_param *par) if (!est1) goto err1; + est2 = NULL; if (info->flags & XT_RATEEST_MATCH_REL) { est2 = xt_rateest_lookup(info->name2); if (!est2) goto err2; - } else - est2 = NULL; - + } info->est1 = est1; info->est2 = est2; @@ -116,7 +115,7 @@ static int xt_rateest_mt_checkentry(const struct xt_mtchk_param *par) err2: xt_rateest_put(est1); err1: - return -EINVAL; + return ret; } static void xt_rateest_mt_destroy(const struct xt_mtdtor_param *par) From 91c66c6893a3e2bb8a88a30cb76007d5d49d32c9 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Fri, 29 Jul 2011 16:38:49 +0200 Subject: [PATCH 002/326] netfilter: ip_queue: Fix small leak in ipq_build_packet_message() ipq_build_packet_message() in net/ipv4/netfilter/ip_queue.c and net/ipv6/netfilter/ip6_queue.c contain a small potential mem leak as far as I can tell. We allocate memory for 'skb' with alloc_skb() annd then call nlh = NLMSG_PUT(skb, 0, 0, IPQM_PACKET, size - sizeof(*nlh)); NLMSG_PUT is a macro NLMSG_PUT(skb, pid, seq, type, len) \ NLMSG_NEW(skb, pid, seq, type, len, 0) that expands to NLMSG_NEW, which is also a macro which expands to: NLMSG_NEW(skb, pid, seq, type, len, flags) \ ({ if (unlikely(skb_tailroom(skb) < (int)NLMSG_SPACE(len))) \ goto nlmsg_failure; \ __nlmsg_put(skb, pid, seq, type, len, flags); }) If we take the true branch of the 'if' statement and 'goto nlmsg_failure', then we'll, at that point, return from ipq_build_packet_message() without having assigned 'skb' to anything and we'll leak the memory we allocated for it when it goes out of scope. Fix this by placing a 'kfree(skb)' at 'nlmsg_failure'. I admit that I do not know how likely this to actually happen or even if there's something that guarantees that it will never happen - I'm not that familiar with this code, but if that is so, I've not been able to spot it. Signed-off-by: Jesper Juhl Signed-off-by: Patrick McHardy --- net/ipv4/netfilter/ip_queue.c | 1 + net/ipv6/netfilter/ip6_queue.c | 1 + 2 files changed, 2 insertions(+) diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 5c9b9d963918..48f7d5b4ff37 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -218,6 +218,7 @@ ipq_build_packet_message(struct nf_queue_entry *entry, int *errp) return skb; nlmsg_failure: + kfree_skb(skb); *errp = -EINVAL; printk(KERN_ERR "ip_queue: error creating packet message\n"); return NULL; diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 249394863284..87b243a25afa 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -218,6 +218,7 @@ ipq_build_packet_message(struct nf_queue_entry *entry, int *errp) return skb; nlmsg_failure: + kfree_skb(skb); *errp = -EINVAL; printk(KERN_ERR "ip6_queue: error creating packet message\n"); return NULL; From 9823d9ff483af4ce8804a9eb69600ca739cd1f58 Mon Sep 17 00:00:00 2001 From: Bart De Schuymer Date: Fri, 29 Jul 2011 16:40:30 +0200 Subject: [PATCH 003/326] netfilter: ebtables: fix ebtables build dependency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The configuration of ebtables shouldn't depend on CONFIG_BRIDGE_NETFILTER, only on CONFIG_NETFILTER. Reported-by: Sébastien Laveze Signed-off-by: Bart De Schuymer Signed-off-by: Patrick McHardy --- net/bridge/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index ba6f73eb06c6..a9aff9c7d027 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -4,7 +4,7 @@ menuconfig BRIDGE_NF_EBTABLES tristate "Ethernet Bridge tables (ebtables) support" - depends on BRIDGE && BRIDGE_NETFILTER + depends on BRIDGE && NETFILTER select NETFILTER_XTABLES help ebtables is a general, extensible frame/packet identification From 5185352c163a72cf969b2fbbfb89801b398896fd Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Tue, 9 Aug 2011 14:48:11 -0700 Subject: [PATCH 004/326] libceph: fix msgpool There were several problems here: 1- we weren't tagging allocations with the pool, so they were never returned to the pool. 2- msgpool_put didn't add back to the mempool, even it were called. 3- msgpool_release didn't clear the pool pointer, so it would have looped had #1 not been broken. These may or may not have been responsible for #1136 or #1381 (BUG due to non-empty mempool on umount). I can't seem to trigger the crash now using the method I was using before. Signed-off-by: Sage Weil --- net/ceph/msgpool.c | 40 +++++++++++++++++++++++++++++----------- 1 file changed, 29 insertions(+), 11 deletions(-) diff --git a/net/ceph/msgpool.c b/net/ceph/msgpool.c index d5f2d97ac05c..1f4cb30a42c5 100644 --- a/net/ceph/msgpool.c +++ b/net/ceph/msgpool.c @@ -7,27 +7,37 @@ #include -static void *alloc_fn(gfp_t gfp_mask, void *arg) +static void *msgpool_alloc(gfp_t gfp_mask, void *arg) { struct ceph_msgpool *pool = arg; - void *p; + struct ceph_msg *msg; - p = ceph_msg_new(0, pool->front_len, gfp_mask); - if (!p) - pr_err("msgpool %s alloc failed\n", pool->name); - return p; + msg = ceph_msg_new(0, pool->front_len, gfp_mask); + if (!msg) { + dout("msgpool_alloc %s failed\n", pool->name); + } else { + dout("msgpool_alloc %s %p\n", pool->name, msg); + msg->pool = pool; + } + return msg; } -static void free_fn(void *element, void *arg) +static void msgpool_free(void *element, void *arg) { - ceph_msg_put(element); + struct ceph_msgpool *pool = arg; + struct ceph_msg *msg = element; + + dout("msgpool_release %s %p\n", pool->name, msg); + msg->pool = NULL; + ceph_msg_put(msg); } int ceph_msgpool_init(struct ceph_msgpool *pool, int front_len, int size, bool blocking, const char *name) { + dout("msgpool %s init\n", name); pool->front_len = front_len; - pool->pool = mempool_create(size, alloc_fn, free_fn, pool); + pool->pool = mempool_create(size, msgpool_alloc, msgpool_free, pool); if (!pool->pool) return -ENOMEM; pool->name = name; @@ -36,14 +46,17 @@ int ceph_msgpool_init(struct ceph_msgpool *pool, void ceph_msgpool_destroy(struct ceph_msgpool *pool) { + dout("msgpool %s destroy\n", pool->name); mempool_destroy(pool->pool); } struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool, int front_len) { + struct ceph_msg *msg; + if (front_len > pool->front_len) { - pr_err("msgpool_get pool %s need front %d, pool size is %d\n", + dout("msgpool_get %s need front %d, pool size is %d\n", pool->name, front_len, pool->front_len); WARN_ON(1); @@ -51,14 +64,19 @@ struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool, return ceph_msg_new(0, front_len, GFP_NOFS); } - return mempool_alloc(pool->pool, GFP_NOFS); + msg = mempool_alloc(pool->pool, GFP_NOFS); + dout("msgpool_get %s %p\n", pool->name, msg); + return msg; } void ceph_msgpool_put(struct ceph_msgpool *pool, struct ceph_msg *msg) { + dout("msgpool_put %s %p\n", pool->name, msg); + /* reset msg front_len; user may have changed it */ msg->front.iov_len = pool->front_len; msg->hdr.front_len = cpu_to_le32(pool->front_len); kref_init(&msg->kref); /* retake single ref */ + mempool_free(msg, pool->pool); } From 4f6fdf08681cecd9f38499de7a02eb4f05f399a7 Mon Sep 17 00:00:00 2001 From: Chase Douglas Date: Fri, 5 Aug 2011 09:16:57 -0700 Subject: [PATCH 005/326] HID: magicmouse: Set resolution of touch surfaces Add touch surface resolution information. The size of the touch surfaces has been determined to the hundredth of a mm. Cc: Jiri Kosina Cc: Michael Poole Cc: linux-input@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Chase Douglas [jkosina@suse.cz: update comments and commit message] Signed-off-by: Jiri Kosina --- drivers/hid/hid-magicmouse.c | 56 +++++++++++++++++++++++++++++------- 1 file changed, 46 insertions(+), 10 deletions(-) diff --git a/drivers/hid/hid-magicmouse.c b/drivers/hid/hid-magicmouse.c index 0ec91c18a421..b5bdab3299bc 100644 --- a/drivers/hid/hid-magicmouse.c +++ b/drivers/hid/hid-magicmouse.c @@ -81,6 +81,28 @@ MODULE_PARM_DESC(report_undeciphered, "Report undeciphered multi-touch state fie #define NO_TOUCHES -1 #define SINGLE_TOUCH_UP -2 +/* Touch surface information. Dimension is in hundredths of a mm, min and max + * are in units. */ +#define MOUSE_DIMENSION_X (float)9056 +#define MOUSE_MIN_X -1100 +#define MOUSE_MAX_X 1258 +#define MOUSE_RES_X ((MOUSE_MAX_X - MOUSE_MIN_X) / (MOUSE_DIMENSION_X / 100)) +#define MOUSE_DIMENSION_Y (float)5152 +#define MOUSE_MIN_Y -1589 +#define MOUSE_MAX_Y 2047 +#define MOUSE_RES_Y ((MOUSE_MAX_Y - MOUSE_MIN_Y) / (MOUSE_DIMENSION_Y / 100)) + +#define TRACKPAD_DIMENSION_X (float)13000 +#define TRACKPAD_MIN_X -2909 +#define TRACKPAD_MAX_X 3167 +#define TRACKPAD_RES_X \ + ((TRACKPAD_MAX_X - TRACKPAD_MIN_X) / (TRACKPAD_DIMENSION_X / 100)) +#define TRACKPAD_DIMENSION_Y (float)11000 +#define TRACKPAD_MIN_Y -2456 +#define TRACKPAD_MAX_Y 2565 +#define TRACKPAD_RES_Y \ + ((TRACKPAD_MAX_Y - TRACKPAD_MIN_Y) / (TRACKPAD_DIMENSION_Y / 100)) + /** * struct magicmouse_sc - Tracks Magic Mouse-specific data. * @input: Input device through which we report events. @@ -406,17 +428,31 @@ static void magicmouse_setup_input(struct input_dev *input, struct hid_device *h * inverse of the reported Y. */ if (input->id.product == USB_DEVICE_ID_APPLE_MAGICMOUSE) { - input_set_abs_params(input, ABS_MT_POSITION_X, -1100, - 1358, 4, 0); - input_set_abs_params(input, ABS_MT_POSITION_Y, -1589, - 2047, 4, 0); + input_set_abs_params(input, ABS_MT_POSITION_X, + MOUSE_MIN_X, MOUSE_MAX_X, 4, 0); + input_set_abs_params(input, ABS_MT_POSITION_Y, + MOUSE_MIN_Y, MOUSE_MAX_Y, 4, 0); + + input_abs_set_res(input, ABS_MT_POSITION_X, + MOUSE_RES_X); + input_abs_set_res(input, ABS_MT_POSITION_Y, + MOUSE_RES_Y); } else { /* USB_DEVICE_ID_APPLE_MAGICTRACKPAD */ - input_set_abs_params(input, ABS_X, -2909, 3167, 4, 0); - input_set_abs_params(input, ABS_Y, -2456, 2565, 4, 0); - input_set_abs_params(input, ABS_MT_POSITION_X, -2909, - 3167, 4, 0); - input_set_abs_params(input, ABS_MT_POSITION_Y, -2456, - 2565, 4, 0); + input_set_abs_params(input, ABS_X, TRACKPAD_MIN_X, + TRACKPAD_MAX_X, 4, 0); + input_set_abs_params(input, ABS_Y, TRACKPAD_MIN_Y, + TRACKPAD_MAX_Y, 4, 0); + input_set_abs_params(input, ABS_MT_POSITION_X, + TRACKPAD_MIN_X, TRACKPAD_MAX_X, 4, 0); + input_set_abs_params(input, ABS_MT_POSITION_Y, + TRACKPAD_MIN_Y, TRACKPAD_MAX_Y, 4, 0); + + input_abs_set_res(input, ABS_X, TRACKPAD_RES_X); + input_abs_set_res(input, ABS_Y, TRACKPAD_RES_Y); + input_abs_set_res(input, ABS_MT_POSITION_X, + TRACKPAD_RES_X); + input_abs_set_res(input, ABS_MT_POSITION_Y, + TRACKPAD_RES_Y); } input_set_events_per_packet(input, 60); From 971c90bfa2f0b4fe52d6d9002178d547706f1343 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 4 Aug 2011 07:25:35 -0700 Subject: [PATCH 006/326] alarmtimers: Avoid possible null pointer traversal We don't check if old_setting is non null before assigning it, so correct this. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 59f369f98a04..1dee3f62a6a7 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -479,11 +479,8 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, if (!rtcdev) return -ENOTSUPP; - /* Save old values */ - old_setting->it_interval = - ktime_to_timespec(timr->it.alarmtimer.period); - old_setting->it_value = - ktime_to_timespec(timr->it.alarmtimer.node.expires); + if (old_setting) + alarm_timer_get(timr, old_setting); /* If the timer was already set, cancel it */ alarm_cancel(&timr->it.alarmtimer); From ea7802f630d356acaf66b3c0b28c00a945fc35dc Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 4 Aug 2011 07:51:56 -0700 Subject: [PATCH 007/326] alarmtimers: Memset itimerspec passed into alarm_timer_get Following common_timer_get, zero out the itimerspec passed in. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 1dee3f62a6a7..0e9263f6fd09 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -441,6 +441,8 @@ static int alarm_timer_create(struct k_itimer *new_timer) static void alarm_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting) { + memset(cur_setting, 0, sizeof(struct itimerspec)); + cur_setting->it_interval = ktime_to_timespec(timr->it.alarmtimer.period); cur_setting->it_value = From 6af7e471e5a7746b8024d70b4363d3dfe41d36b8 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 10:26:09 -0700 Subject: [PATCH 008/326] alarmtimers: Avoid possible denial of service with high freq periodic timers Its possible to jam up the alarm timers by setting very small interval timers, which will cause the alarmtimer subsystem to spend all of its time firing and restarting timers. This can effectivly lock up a box. A deeper fix is needed, closely mimicking the hrtimer code, but for now just cap the interval to 100us to avoid userland hanging the system. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 0e9263f6fd09..ea5e1a928d5b 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -481,6 +481,15 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, if (!rtcdev) return -ENOTSUPP; + /* + * XXX HACK! Currently we can DOS a system if the interval + * period on alarmtimers is too small. Cap the interval here + * to 100us and solve this properly in a future patch! -jstultz + */ + if ((new_setting->it_interval.tv_sec == 0) && + (new_setting->it_interval.tv_nsec < 100000)) + new_setting->it_interval.tv_nsec = 100000; + if (old_setting) alarm_timer_get(timr, old_setting); From bf6ed027bcc93f8d54d321fe87f0434b25699eb1 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 10 Aug 2011 21:11:26 +0800 Subject: [PATCH 009/326] rtc: ep93xx: Fix 'rtc' may be used uninitialized warning commit 92d921c5d "rtc: ep93xx: Initialize drvdata before registering device" ensures the drvdata is initialized prior to registering the rtc device. But it set the drvdata to an uninitialized pointer. Thus calling platform_get_drvdata in ep93xx_rtc_remove does not get correct address. This patch fixes below warning by adding struct rtc_device *rtc to struct ep93xx_rtc. Then set platform drvdata to ep93xx_rtc instead of rtc. CC drivers/rtc/rtc-ep93xx.o drivers/rtc/rtc-ep93xx.c: In function 'ep93xx_rtc_probe': drivers/rtc/rtc-ep93xx.c:154: warning: 'rtc' may be used uninitialized in this function Signed-off-by: Axel Lin Signed-off-by: John Stultz --- drivers/rtc/rtc-ep93xx.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/rtc/rtc-ep93xx.c b/drivers/rtc/rtc-ep93xx.c index 335551d333b2..14a42a1edc66 100644 --- a/drivers/rtc/rtc-ep93xx.c +++ b/drivers/rtc/rtc-ep93xx.c @@ -36,6 +36,7 @@ */ struct ep93xx_rtc { void __iomem *mmio_base; + struct rtc_device *rtc; }; static int ep93xx_rtc_get_swcomp(struct device *dev, unsigned short *preload, @@ -130,7 +131,6 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) { struct ep93xx_rtc *ep93xx_rtc; struct resource *res; - struct rtc_device *rtc; int err; ep93xx_rtc = devm_kzalloc(&pdev->dev, sizeof(*ep93xx_rtc), GFP_KERNEL); @@ -151,12 +151,12 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) return -ENXIO; pdev->dev.platform_data = ep93xx_rtc; - platform_set_drvdata(pdev, rtc); + platform_set_drvdata(pdev, ep93xx_rtc); - rtc = rtc_device_register(pdev->name, + ep93xx_rtc->rtc = rtc_device_register(pdev->name, &pdev->dev, &ep93xx_rtc_ops, THIS_MODULE); - if (IS_ERR(rtc)) { - err = PTR_ERR(rtc); + if (IS_ERR(ep93xx_rtc->rtc)) { + err = PTR_ERR(ep93xx_rtc->rtc); goto exit; } @@ -167,7 +167,7 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) return 0; fail: - rtc_device_unregister(rtc); + rtc_device_unregister(ep93xx_rtc->rtc); exit: platform_set_drvdata(pdev, NULL); pdev->dev.platform_data = NULL; @@ -176,11 +176,11 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) static int __exit ep93xx_rtc_remove(struct platform_device *pdev) { - struct rtc_device *rtc = platform_get_drvdata(pdev); + struct ep93xx_rtc *ep93xx_rtc = platform_get_drvdata(pdev); sysfs_remove_group(&pdev->dev.kobj, &ep93xx_rtc_sysfs_files); platform_set_drvdata(pdev, NULL); - rtc_device_unregister(rtc); + rtc_device_unregister(ep93xx_rtc->rtc); pdev->dev.platform_data = NULL; return 0; From dec35d19c4ec65b94df3b27b6e373f0d48c9cd32 Mon Sep 17 00:00:00 2001 From: Ilkka Koskinen Date: Wed, 16 Mar 2011 06:07:14 +0000 Subject: [PATCH 010/326] rtc: rtc-twl: Switch to using threaded irq The driver is accessing to i2c bus in interrupt handler. Therefore, it should use threaded irq. Signed-off-by: Ilkka Koskinen Acked-by: Balaji T K Signed-off-by: John Stultz --- drivers/rtc/rtc-twl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-twl.c b/drivers/rtc/rtc-twl.c index 9a81f778d6b2..1963cddbf214 100644 --- a/drivers/rtc/rtc-twl.c +++ b/drivers/rtc/rtc-twl.c @@ -462,7 +462,7 @@ static int __devinit twl_rtc_probe(struct platform_device *pdev) if (ret < 0) goto out1; - ret = request_irq(irq, twl_rtc_interrupt, + ret = request_threaded_irq(irq, NULL, twl_rtc_interrupt, IRQF_TRIGGER_RISING, dev_name(&rtc->dev), rtc); if (ret < 0) { From 34d623d11316cb69f9e8cc5eb50d3792b5c302b6 Mon Sep 17 00:00:00 2001 From: Sebastian Reichel Date: Tue, 31 May 2011 08:51:39 +0000 Subject: [PATCH 011/326] rtc: rtc-twl: Remove lockdep related local_irq_enable() Now that the irq is properly threaded (due to it needing i2c access) we should also remove the local_irq_enable() call in twl_rtc_interrupt. Testing this with Pandaboard, the RTC is still working. [Reworked commit message -jstultz] Signed-off-by: John Stultz --- drivers/rtc/rtc-twl.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/rtc/rtc-twl.c b/drivers/rtc/rtc-twl.c index 1963cddbf214..9677bbc433f9 100644 --- a/drivers/rtc/rtc-twl.c +++ b/drivers/rtc/rtc-twl.c @@ -362,14 +362,6 @@ static irqreturn_t twl_rtc_interrupt(int irq, void *rtc) int res; u8 rd_reg; -#ifdef CONFIG_LOCKDEP - /* WORKAROUND for lockdep forcing IRQF_DISABLED on us, which - * we don't want and can't tolerate. Although it might be - * friendlier not to borrow this thread context... - */ - local_irq_enable(); -#endif - res = twl_rtc_read_u8(&rd_reg, REG_RTC_STATUS_REG); if (res) goto out; From 938f97bcf1bdd1b681d5d14d1d7117a2e22d4434 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 22 Jul 2011 09:12:51 +0000 Subject: [PATCH 012/326] rtc: Fix RTC PIE frequency limit Thomas earlier submitted a fix to limit the RTC PIE freq, but picked 5000Hz out of the air. Willy noticed that we should instead use the 8192Hz max from the rtc man documentation. Cc: Willy Tarreau Cc: stable@kernel.org Cc: Thomas Gleixner Signed-off-by: John Stultz --- drivers/rtc/interface.c | 2 +- include/linux/rtc.h | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c index 3195dbd3ec34..eb4c88316a15 100644 --- a/drivers/rtc/interface.c +++ b/drivers/rtc/interface.c @@ -708,7 +708,7 @@ int rtc_irq_set_freq(struct rtc_device *rtc, struct rtc_task *task, int freq) int err = 0; unsigned long flags; - if (freq <= 0 || freq > 5000) + if (freq <= 0 || freq > RTC_MAX_FREQ) return -EINVAL; retry: spin_lock_irqsave(&rtc->irq_task_lock, flags); diff --git a/include/linux/rtc.h b/include/linux/rtc.h index b27ebea25660..93f4d035076b 100644 --- a/include/linux/rtc.h +++ b/include/linux/rtc.h @@ -97,6 +97,9 @@ struct rtc_pll_info { #define RTC_AF 0x20 /* Alarm interrupt */ #define RTC_UF 0x10 /* Update interrupt for 1Hz RTC */ + +#define RTC_MAX_FREQ 8192 + #ifdef __KERNEL__ #include From b3b46d76d0fcbb1f737107cec1a1ee87bc5e5fd3 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Thu, 11 Aug 2011 12:06:28 -0400 Subject: [PATCH 013/326] APEI: Fix WHEA _OSC call Bit 0 of the support parameter to the OSC call should be set in order to indicate that the OS supports the WHEA mechanism. Stuart Hayes tracked an APEI issue on some Dell platforms down to this. Reported-by: Stuart Hayes Signed-off-by: Matthew Garrett Signed-off-by: Len Brown --- drivers/acpi/apei/apei-base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/apei/apei-base.c b/drivers/acpi/apei/apei-base.c index 8041248fce9b..61540360d5ce 100644 --- a/drivers/acpi/apei/apei-base.c +++ b/drivers/acpi/apei/apei-base.c @@ -618,7 +618,7 @@ int apei_osc_setup(void) }; capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE; - capbuf[OSC_SUPPORT_TYPE] = 0; + capbuf[OSC_SUPPORT_TYPE] = 1; capbuf[OSC_CONTROL_TYPE] = 0; if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle)) From 96242116d483cd98ab55fb989ca096f6f9cc3738 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 11 Aug 2011 12:14:05 -0600 Subject: [PATCH 014/326] PNP: update pnp.debug usage (needs value on command line) Commit cdefba03e44 changed pnp.debug from a boot param to a module param, which means it needs a value when used on the command line. CC: Thomas Renninger Signed-off-by: Bjorn Helgaas Signed-off-by: Len Brown --- Documentation/kernel-parameters.txt | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e279b7242912..130713f71875 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2081,9 +2081,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. Override pmtimer IOPort with a hex value. e.g. pmtmr=0x508 - pnp.debug [PNP] - Enable PNP debug messages. This depends on the - CONFIG_PNP_DEBUG_MESSAGES option. + pnp.debug=1 [PNP] + Enable PNP debug messages (depends on the + CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time + via /sys/module/pnp/parameters/debug. We always show + current resource usage; turning this on also shows + possible settings and some assignment information. pnpacpi= [ACPI] { off } From 03ba176a29dae5b4849f45c0b5c89b9d78baa2c6 Mon Sep 17 00:00:00 2001 From: Chen Gong Date: Wed, 10 Aug 2011 10:46:22 +0800 Subject: [PATCH 015/326] ACPI APEI: Add Kconfig option IRQ_WORK for GHES IRQ_WORK is used by GHES, but it is selected by PERF_EVENT. For now PERF_EVENT is selected by x86 by default, but in concept, IRQ_WORK should be selected by GHES, not by others. Signed-off-by: Chen Gong Signed-off-by: Len Brown --- drivers/acpi/apei/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/acpi/apei/Kconfig b/drivers/acpi/apei/Kconfig index c34aa51af4ee..e3f47872ec22 100644 --- a/drivers/acpi/apei/Kconfig +++ b/drivers/acpi/apei/Kconfig @@ -13,6 +13,7 @@ config ACPI_APEI_GHES bool "APEI Generic Hardware Error Source" depends on ACPI_APEI && X86 select ACPI_HED + select IRQ_WORK select LLIST select GENERIC_ALLOCATOR help From 795858dbd253462a67e14272edeaae73c6074b17 Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Mon, 15 Aug 2011 13:02:37 -0700 Subject: [PATCH 016/326] ceph: fix encoding of ino only (not relative) paths A 'path' consists of a starting ino and relative component. Encode even when there is no relative component. This is primarily needed by the NFS reexport code. Signed-off-by: Sage Weil --- fs/ceph/mds_client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index fee028b5332e..86c59e16ba74 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -1595,7 +1595,7 @@ static int set_request_path_attr(struct inode *rinode, struct dentry *rdentry, r = build_dentry_path(rdentry, ppath, pathlen, ino, freepath); dout(" dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen, *ppath); - } else if (rpath) { + } else if (rpath || rino) { *ino = rino; *ppath = rpath; *pathlen = strlen(rpath); From 016f1c54408b1e92e2e8087bfc05ca0a9c258513 Mon Sep 17 00:00:00 2001 From: Michal Marek Date: Thu, 11 Aug 2011 12:29:46 +0200 Subject: [PATCH 017/326] UBIFS: not build debug messages with CONFIG_UBIFS_FS_DEBUG disabled With $ grep -e UBIFS_FS_DEBUG -e DYNAMIC_DEBUG .config # CONFIG_UBIFS_FS_DEBUG is not set CONFIG_DYNAMIC_DEBUG=y Debug messages are kept in the object files due to the dynamic_pr_debug() macro, even if they are never going to be printed: $ make fs/ubifs/super.o $ strings fs/ubifs/super.o | grep 'compiled on' compiled on: Aug 11 2011 at 12:21:38 Use plain printk to fix this. Signed-off-by: Michal Marek Signed-off-by: Artem Bityutskiy --- fs/ubifs/debug.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index 45174b534377..feb361e252ac 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -335,9 +335,9 @@ void dbg_debugfs_exit_fs(struct ubifs_info *c); #define DBGKEY(key) ((char *)(key)) #define DBGKEY1(key) ((char *)(key)) -#define ubifs_dbg_msg(fmt, ...) do { \ - if (0) \ - pr_debug(fmt "\n", ##__VA_ARGS__); \ +#define ubifs_dbg_msg(fmt, ...) do { \ + if (0) \ + printk(KERN_DEBUG fmt "\n", ##__VA_ARGS__); \ } while (0) #define dbg_dump_stack() From 9efabc84768ee8e79b50ad6ad6cff94d66da01f7 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 19 Aug 2011 19:02:27 +0300 Subject: [PATCH 018/326] UBI: do not link debug messages when debugging is disabled Michal Marek spotted the same issue in UBIFS and this patch fixes UBI, see "UBIFS: not build debug messages with CONFIG_UBIFS_FS_DEBUG disabled" When UBI debugging is disabled, we have debugging messages defined as: if (0) pr_debug() But pr_debug macro defines data structures with debugging data and makes the linux binary larger, even though we have "if (0)". Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/debug.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h index 65b5b76cc379..64fbb0021825 100644 --- a/drivers/mtd/ubi/debug.h +++ b/drivers/mtd/ubi/debug.h @@ -181,7 +181,7 @@ static inline int ubi_dbg_is_erase_failure(const struct ubi_device *ubi) #define ubi_dbg_msg(fmt, ...) do { \ if (0) \ - pr_debug(fmt "\n", ##__VA_ARGS__); \ + printk(KERN_DEBUG fmt "\n", ##__VA_ARGS__); \ } while (0) #define dbg_msg(fmt, ...) ubi_dbg_msg(fmt, ##__VA_ARGS__) From c956b753e706f24d18a026f8efa4df3b1919fcc9 Mon Sep 17 00:00:00 2001 From: Rajendra Nayak Date: Fri, 19 Aug 2011 16:59:39 -0600 Subject: [PATCH 019/326] OMAP: powerdomains: Make all powerdomain target states as ON at init Program all powerdomain target state as ON; this is to prevent domains from hitting low power states (if bootloader has target states set to something other than ON) and potentially even losing context while PM is not fully initialized, which can cause the system to crash. The PM late init code can then program the desired target state for all the power domains. Signed-off-by: Rajendra Nayak Signed-off-by: Santosh Shilimkar [paul@pwsan.com: dropped comment typo hunk; fixed comment indent and moved to kerneldoc; moved code to pwrdm_init(); changed pwrdm_init() argument name to prevent clash; cleaned up patch description] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/powerdomain.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/arch/arm/mach-omap2/powerdomain.c b/arch/arm/mach-omap2/powerdomain.c index 9af08473bf10..ef71fdd40fc4 100644 --- a/arch/arm/mach-omap2/powerdomain.c +++ b/arch/arm/mach-omap2/powerdomain.c @@ -195,28 +195,35 @@ static int _pwrdm_post_transition_cb(struct powerdomain *pwrdm, void *unused) /** * pwrdm_init - set up the powerdomain layer - * @pwrdm_list: array of struct powerdomain pointers to register + * @pwrdms: array of struct powerdomain pointers to register * @custom_funcs: func pointers for arch specific implementations * - * Loop through the array of powerdomains @pwrdm_list, registering all - * that are available on the current CPU. If pwrdm_list is supplied - * and not null, all of the referenced powerdomains will be - * registered. No return value. XXX pwrdm_list is not really a - * "list"; it is an array. Rename appropriately. + * Loop through the array of powerdomains @pwrdms, registering all + * that are available on the current CPU. Also, program all + * powerdomain target state as ON; this is to prevent domains from + * hitting low power states (if bootloader has target states set to + * something other than ON) and potentially even losing context while + * PM is not fully initialized. The PM late init code can then program + * the desired target state for all the power domains. No return + * value. */ -void pwrdm_init(struct powerdomain **pwrdm_list, struct pwrdm_ops *custom_funcs) +void pwrdm_init(struct powerdomain **pwrdms, struct pwrdm_ops *custom_funcs) { struct powerdomain **p = NULL; + struct powerdomain *temp_p; if (!custom_funcs) WARN(1, "powerdomain: No custom pwrdm functions registered\n"); else arch_pwrdm = custom_funcs; - if (pwrdm_list) { - for (p = pwrdm_list; *p; p++) + if (pwrdms) { + for (p = pwrdms; *p; p++) _pwrdm_register(*p); } + + list_for_each_entry(temp_p, &pwrdm_list, node) + pwrdm_set_next_pwrst(temp_p, PWRDM_POWER_ON); } /** From b1cbdb00da2ac00eb67fe277e563ff1f5093b4ba Mon Sep 17 00:00:00 2001 From: Santosh Shilimkar Date: Fri, 19 Aug 2011 16:59:39 -0600 Subject: [PATCH 020/326] OMAP: clockdomain: Wait for powerdomain to be ON when using clockdomain force wakeup While using clockdomain force wakeup method, not waiting for powerdomain to be effectively ON may end up locking the clockdomain FSM until a next wakeup event occurs. One such issue was seen on OMAP4430, where L4_PER was periodically getting stuck in in-transition state when transitioning from from OSWR to ON. This issue was reported and investigated by Patrick Titiano Signed-off-by: Santosh Shilimkar Signed-off-by: Rajendra Nayak Reported-by: Patrick Titiano Cc: Kevin Hilman Cc: Benoit Cousson Cc: Paul Walmsley [paul@pwsan.com: updated to apply; added transition wait on clkdm_deny_idle(); remove two superfluous pwrdm_wait_transition() calls] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/clockdomain.c | 2 ++ arch/arm/mach-omap2/pm.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/clockdomain.c b/arch/arm/mach-omap2/clockdomain.c index ab7db083f97f..8f0890685d7b 100644 --- a/arch/arm/mach-omap2/clockdomain.c +++ b/arch/arm/mach-omap2/clockdomain.c @@ -747,6 +747,7 @@ int clkdm_wakeup(struct clockdomain *clkdm) spin_lock_irqsave(&clkdm->lock, flags); clkdm->_flags &= ~_CLKDM_FLAG_HWSUP_ENABLED; ret = arch_clkdm->clkdm_wakeup(clkdm); + ret |= pwrdm_state_switch(clkdm->pwrdm.ptr); spin_unlock_irqrestore(&clkdm->lock, flags); return ret; } @@ -818,6 +819,7 @@ void clkdm_deny_idle(struct clockdomain *clkdm) spin_lock_irqsave(&clkdm->lock, flags); clkdm->_flags &= ~_CLKDM_FLAG_HWSUP_ENABLED; arch_clkdm->clkdm_deny_idle(clkdm); + pwrdm_state_switch(clkdm->pwrdm.ptr); spin_unlock_irqrestore(&clkdm->lock, flags); } diff --git a/arch/arm/mach-omap2/pm.c b/arch/arm/mach-omap2/pm.c index 3feb35911a32..472bf22d5e84 100644 --- a/arch/arm/mach-omap2/pm.c +++ b/arch/arm/mach-omap2/pm.c @@ -130,7 +130,6 @@ int omap_set_pwrdm_state(struct powerdomain *pwrdm, u32 state) } else { hwsup = clkdm_in_hwsup(pwrdm->pwrdm_clkdms[0]); clkdm_wakeup(pwrdm->pwrdm_clkdms[0]); - pwrdm_wait_transition(pwrdm); sleep_switch = FORCEWAKEUP_SWITCH; } } @@ -156,7 +155,6 @@ int omap_set_pwrdm_state(struct powerdomain *pwrdm, u32 state) return ret; } - pwrdm_wait_transition(pwrdm); pwrdm_state_switch(pwrdm); err: return ret; From 9c5f560173a466582d91bb06f4e3d2bafb0fee5c Mon Sep 17 00:00:00 2001 From: Paul Walmsley Date: Fri, 19 Aug 2011 16:59:56 -0600 Subject: [PATCH 021/326] OMAP4: clock: re-enable previous clockdomain enable/disable sequence MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After commit 665d001338b494d6d62810aa99b4c0fa1a0884b9 ("OMAP2+: hwmod: Follow the recommended PRCM module enable sequence"), device drivers for OMAP IP blocks that do not use runtime PM can cause oopses or kernel instability[1][2]. This is because those non-runtime PM drivers do not use the hwmod code, which implements the correct IP block enable and disable sequence. Several options for dealing with this problem have been proposed: 1. Add a new field to the OMAP struct clk to mark clocks that are currently used by non-runtime PM drivers. Modify the clock code to use the old clockdomain sequence for these marked clocks. As drivers are converted to use runtime PM, remove the annotation from the clocks. 2. Similar to #1, but associate the flag with the struct omap_clk instead. 3. Add IDLEST wait support to the OMAP4 clock code, similar to the way it is implemented for OMAP2/3, and enable it in each struct clk currently used by non-runtime PM drivers. As drivers are converted to use runtime PM, remove the annotation from the clocks. 4. Do nothing; leave the problem to those responsible for the unconverted drivers. 5. Re-enable clock-based clockdomain control in the OMAP4 clock code. This would revert back to the behavior of Linux 3.0, simply with a slightly longer module enable/disable latency. Unfortunately, no approach seemed particularly good. Options 1 through 3 seemed unwise due to the following reasons: A. The OMAP struct clks are intended primarily to describe hardware clock nodes, and the intention is that no driver-specific data should be stored there (applies to #1) B. The resulting patch would have been quite large for the -rc series (applies to #1, #2, #3) C. The patch would have been a new, yet temporary hack; and similar fixes have drawn negative comments in the recent past (see for example [3]) Option 4 is undesirable because commit 665d001338b494d6d62810aa99b4c0fa1a0884b9 ("OMAP2+: hwmod: Follow the recommended PRCM module enable sequence") has resulted in a less stable kernel; and kernel stability is more important than OMAP4 power management. Option 5 is the approach taken in this patch. This seemed to be the least intrusive approach for 3.1-rc. The approach in this patch was originally proposed by Ohad Ben-Cohen . I'm simply writing the commit message and passing it along. ... Thanks to Luciano Coelho for reporting the problem. Thanks to Ohad Ben-Cohen for tracking the problem down, generating a temporary workaround, and proposing a patch to deal with the problem. Thanks to Rajendra Nayak for proposing another patch to deal with the problem. Thanks to Felipe Balbi for comments. 1. Coelho, Luciano . _Re: Oops on ehci_hcd when booting 3.0.0-rc2 on panda_. Tue, 09 Aug 2011 14:26:08 +0300. Posted to the mailing list. Available from (among others) http://www.spinics.net/linux/lists/linux-omap/msg55213.html 2. Munegowda, Keshava . _Re: Oops on ehci_hcd when booting 3.0.0-rc2 on panda_. Thu, 11 Aug 2011 13:51:05 +0530. Posted to the mailing list. Available from (among others) http://www.spinics.net/linux/lists/linux-omap/msg55371.html 3. King, Russell . _Re: [PATCH 5/8] OMAP4: PM: TEMP: Prevent l3init from idling/force sleep_. Thu, 23 Jun 2011 16:22:49 +0100. Posted to the mailing list. Available from (among others) http://www.mail-archive.com/linux-omap@vger.kernel.org/msg51392.html Signed-off-by: Paul Walmsley Cc: Luciano Coelho Cc: Ohad Ben-Cohen Cc: Rajendra Nayak Cc: Benoît Cousson Acked-by: Santosh Shilimkar --- arch/arm/mach-omap2/clock44xx_data.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/clock44xx_data.c b/arch/arm/mach-omap2/clock44xx_data.c index 2af0e3f00ce1..a3a1827837e8 100644 --- a/arch/arm/mach-omap2/clock44xx_data.c +++ b/arch/arm/mach-omap2/clock44xx_data.c @@ -3379,7 +3379,13 @@ int __init omap4xxx_clk_init(void) } clk_init(&omap2_clk_functions); - omap2_clk_disable_clkdm_control(); + + /* + * Must stay commented until all OMAP SoC drivers are + * converted to runtime PM, or drivers may start crashing + * + * omap2_clk_disable_clkdm_control(); + */ for (c = omap44xx_clks; c < omap44xx_clks + ARRAY_SIZE(omap44xx_clks); c++) From 450a37d2eca6ddf6ea8186f57a7531318df6e796 Mon Sep 17 00:00:00 2001 From: Paul Walmsley Date: Sun, 21 Aug 2011 00:28:56 -0600 Subject: [PATCH 022/326] OMAP4: clock: fix compile warning Fix the following compile warning: arch/arm/mach-omap2/clock44xx_data.c: In function 'omap4xxx_clk_init': arch/arm/mach-omap2/clock44xx_data.c:3371:6: warning: 'cpu_clkflg' may be used uninitialized in this function The approach taken here is intended to work if omap4xxx_clk_init() is converted into an initcall. Thanks to Bjarne Steinsbo for proposing another approach. Signed-off-by: Paul Walmsley Cc: Bjarne Steinsbo --- arch/arm/mach-omap2/clock44xx_data.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mach-omap2/clock44xx_data.c b/arch/arm/mach-omap2/clock44xx_data.c index 2af0e3f00ce1..4304768da712 100644 --- a/arch/arm/mach-omap2/clock44xx_data.c +++ b/arch/arm/mach-omap2/clock44xx_data.c @@ -3376,6 +3376,8 @@ int __init omap4xxx_clk_init(void) } else if (cpu_is_omap446x()) { cpu_mask = RATE_IN_4460; cpu_clkflg = CK_446X; + } else { + return 0; } clk_init(&omap2_clk_functions); From 259a187ade45056fd44856654f78aa9e9f0f7c75 Mon Sep 17 00:00:00 2001 From: Noah Watkins Date: Mon, 22 Aug 2011 13:49:41 -0600 Subject: [PATCH 023/326] ceph: fix memory leak kfree does not clean up indirect allocations in ceph_fs_client and ceph_options (e.g. snapdir_name). Signed-off-by: Noah Watkins Signed-off-by: Sage Weil --- fs/ceph/super.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ceph/super.c b/fs/ceph/super.c index d47c5ec7fb1f..88bacaf385d9 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -813,8 +813,8 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type, fsc = create_fs_client(fsopt, opt); if (IS_ERR(fsc)) { res = ERR_CAST(fsc); - kfree(fsopt); - kfree(opt); + destroy_mount_options(fsopt); + ceph_destroy_options(opt); goto out_final; } From e21757a05730f03f18fbfc528a919e0205aa6a61 Mon Sep 17 00:00:00 2001 From: Paul Walmsley Date: Mon, 22 Aug 2011 16:13:00 -0600 Subject: [PATCH 024/326] OMAP3: clock: indicate that gpt12_fck and wdt1_fck are in the WKUP clockdomain The oscillator that supplies GPT12_FCLK and WDT1_FCLK exists in the WKUP powerdomain[1]. This resolves at least one boot-time warning: omap_hwmod: gpt12_fck: missing clockdomain for gpt12_fck. 1. _OMAP34xx Multimedia High Security (HS) Device Silicon Revision 3.1.x Security Addendum Version K (SWPU119K)_ Figure 3-29. August 2010. --- arch/arm/mach-omap2/clock3xxx_data.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mach-omap2/clock3xxx_data.c b/arch/arm/mach-omap2/clock3xxx_data.c index ffd55b1c4396..b9b844683147 100644 --- a/arch/arm/mach-omap2/clock3xxx_data.c +++ b/arch/arm/mach-omap2/clock3xxx_data.c @@ -3078,6 +3078,7 @@ static struct clk gpt12_fck = { .name = "gpt12_fck", .ops = &clkops_null, .parent = &secure_32k_fck, + .clkdm_name = "wkup_clkdm", .recalc = &followparent_recalc, }; @@ -3085,6 +3086,7 @@ static struct clk wdt1_fck = { .name = "wdt1_fck", .ops = &clkops_null, .parent = &secure_32k_fck, + .clkdm_name = "wkup_clkdm", .recalc = &followparent_recalc, }; From 1a878284473284f9577d44babf16d87152a05c33 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 29 Jul 2011 17:16:40 -0700 Subject: [PATCH 025/326] [SCSI] isci: fix sata response handling A bug (likely copy/paste) that has been carried from the original implementation. The unsolicited frame handling structure returns the d2h fis in the isci_request.stp.rsp buffer. Cc: Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/request.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/drivers/scsi/isci/request.c b/drivers/scsi/isci/request.c index a46e07ac789f..b4cf998385b3 100644 --- a/drivers/scsi/isci/request.c +++ b/drivers/scsi/isci/request.c @@ -2399,22 +2399,19 @@ static void isci_task_save_for_upper_layer_completion( } } -static void isci_request_process_stp_response(struct sas_task *task, - void *response_buffer) +static void isci_process_stp_response(struct sas_task *task, struct dev_to_host_fis *fis) { - struct dev_to_host_fis *d2h_reg_fis = response_buffer; struct task_status_struct *ts = &task->task_status; struct ata_task_resp *resp = (void *)&ts->buf[0]; - resp->frame_len = le16_to_cpu(*(__le16 *)(response_buffer + 6)); - memcpy(&resp->ending_fis[0], response_buffer + 16, 24); + resp->frame_len = sizeof(*fis); + memcpy(resp->ending_fis, fis, sizeof(*fis)); ts->buf_valid_size = sizeof(*resp); - /** - * If the device fault bit is set in the status register, then + /* If the device fault bit is set in the status register, then * set the sense data and return. */ - if (d2h_reg_fis->status & ATA_DF) + if (fis->status & ATA_DF) ts->stat = SAS_PROTO_RESPONSE; else ts->stat = SAM_STAT_GOOD; @@ -2428,7 +2425,6 @@ static void isci_request_io_request_complete(struct isci_host *ihost, { struct sas_task *task = isci_request_access_task(request); struct ssp_response_iu *resp_iu; - void *resp_buf; unsigned long task_flags; struct isci_remote_device *idev = isci_lookup_device(task->dev); enum service_response response = SAS_TASK_UNDELIVERED; @@ -2565,9 +2561,7 @@ static void isci_request_io_request_complete(struct isci_host *ihost, task); if (sas_protocol_ata(task->task_proto)) { - resp_buf = &request->stp.rsp; - isci_request_process_stp_response(task, - resp_buf); + isci_process_stp_response(task, &request->stp.rsp); } else if (SAS_PROTOCOL_SSP == task->task_proto) { /* crack the iu response buffer. */ From ee33e2b771f9e9e4aaba2bb2ace7b727fe451a8b Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 29 Jul 2011 17:16:45 -0700 Subject: [PATCH 026/326] [SCSI] isci: fix 32-bit operation when CONFIG_HIGHMEM64G=n The unsolicited frame control infrastructure requires a table of dma addresses for the hardware to lookup the frame buffer location by an index. The hardware expects the elements of this table to be 64-bit quantities, so we cannot reference these elements as dma_addr_t. All unsolicited frame protocols are affected, particularly SATA-PIO and SMP which prevented direct-attached SATA drives and expander-attached drives to not be discovered. Cc: Reported-by: Jacek Danecki Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/unsolicited_frame_control.c | 2 +- drivers/scsi/isci/unsolicited_frame_control.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/isci/unsolicited_frame_control.c b/drivers/scsi/isci/unsolicited_frame_control.c index e9e1e2abacb9..16f88ab939c8 100644 --- a/drivers/scsi/isci/unsolicited_frame_control.c +++ b/drivers/scsi/isci/unsolicited_frame_control.c @@ -72,7 +72,7 @@ int sci_unsolicited_frame_control_construct(struct isci_host *ihost) */ buf_len = SCU_MAX_UNSOLICITED_FRAMES * SCU_UNSOLICITED_FRAME_BUFFER_SIZE; header_len = SCU_MAX_UNSOLICITED_FRAMES * sizeof(struct scu_unsolicited_frame_header); - size = buf_len + header_len + SCU_MAX_UNSOLICITED_FRAMES * sizeof(dma_addr_t); + size = buf_len + header_len + SCU_MAX_UNSOLICITED_FRAMES * sizeof(uf_control->address_table.array[0]); /* * The Unsolicited Frame buffers are set at the start of the UF diff --git a/drivers/scsi/isci/unsolicited_frame_control.h b/drivers/scsi/isci/unsolicited_frame_control.h index 31cb9506f52d..75d896686f5a 100644 --- a/drivers/scsi/isci/unsolicited_frame_control.h +++ b/drivers/scsi/isci/unsolicited_frame_control.h @@ -214,7 +214,7 @@ struct sci_uf_address_table_array { * starting address of the UF address table. * 64-bit pointers are required by the hardware. */ - dma_addr_t *array; + u64 *array; /** * This field specifies the physical address location for the UF From 985af6f70dbb8a33b3af8a7c7df508d924650e37 Mon Sep 17 00:00:00 2001 From: Marcin Tomczak Date: Fri, 29 Jul 2011 17:16:50 -0700 Subject: [PATCH 027/326] [SCSI] isci: change sas phy timeouts from 54us to 59us Need the following workaround in the driver for interoperability with the older Intel SSD drives and any other SATA drive that may exhibit the same behavior. This is a corner case where SCU speed is limited to either 3G or 1.5G and the drive has a period of DC idle when it switches speed during SATA speed negotiation. Workaround :change PHYTOV[31:24] from 0x36 to 0x3B. Signed-off-by: Marcin Tomczak Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/phy.c | 13 +++++++++++++ drivers/scsi/isci/registers.h | 12 ++++++++++++ 2 files changed, 25 insertions(+) diff --git a/drivers/scsi/isci/phy.c b/drivers/scsi/isci/phy.c index 79313a7a2356..430fc8ff014a 100644 --- a/drivers/scsi/isci/phy.c +++ b/drivers/scsi/isci/phy.c @@ -104,6 +104,7 @@ sci_phy_link_layer_initialization(struct isci_phy *iphy, u32 parity_count = 0; u32 llctl, link_rate; u32 clksm_value = 0; + u32 sp_timeouts = 0; iphy->link_layer_registers = reg; @@ -211,6 +212,18 @@ sci_phy_link_layer_initialization(struct isci_phy *iphy, llctl |= SCU_SAS_LLCTL_GEN_VAL(MAX_LINK_RATE, link_rate); writel(llctl, &iphy->link_layer_registers->link_layer_control); + sp_timeouts = readl(&iphy->link_layer_registers->sas_phy_timeouts); + + /* Clear the default 0x36 (54us) RATE_CHANGE timeout value. */ + sp_timeouts &= ~SCU_SAS_PHYTOV_GEN_VAL(RATE_CHANGE, 0xFF); + + /* Set RATE_CHANGE timeout value to 0x3B (59us). This ensures SCU can + * lock with 3Gb drive when SCU max rate is set to 1.5Gb. + */ + sp_timeouts |= SCU_SAS_PHYTOV_GEN_VAL(RATE_CHANGE, 0x3B); + + writel(sp_timeouts, &iphy->link_layer_registers->sas_phy_timeouts); + if (is_a2(ihost->pdev)) { /* Program the max ARB time for the PHY to 700us so we inter-operate with * the PMC expander which shuts down PHYs if the expander PHY generates too diff --git a/drivers/scsi/isci/registers.h b/drivers/scsi/isci/registers.h index 9b266c7428e8..00afc738bbed 100644 --- a/drivers/scsi/isci/registers.h +++ b/drivers/scsi/isci/registers.h @@ -1299,6 +1299,18 @@ struct scu_transport_layer_registers { #define SCU_AFE_XCVRCR_OFFSET 0x00DC #define SCU_AFE_LUTCR_OFFSET 0x00E0 +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_ALIGN_DETECTION_SHIFT (0UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_ALIGN_DETECTION_MASK (0x000000FFUL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_HOT_PLUG_SHIFT (8UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_HOT_PLUG_MASK (0x0000FF00UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_COMSAS_DETECTION_SHIFT (16UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_COMSAS_DETECTION_MASK (0x00FF0000UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_RATE_CHANGE_SHIFT (24UL) +#define SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_RATE_CHANGE_MASK (0xFF000000UL) + +#define SCU_SAS_PHYTOV_GEN_VAL(name, value) \ + SCU_GEN_VALUE(SCU_SAS_PHY_TIMER_TIMEOUT_VALUES_##name, value) + #define SCU_SAS_LINK_LAYER_CONTROL_MAX_LINK_RATE_SHIFT (0) #define SCU_SAS_LINK_LAYER_CONTROL_MAX_LINK_RATE_MASK (0x00000003) #define SCU_SAS_LINK_LAYER_CONTROL_MAX_LINK_RATE_GEN1 (0) From 4ac13e177904280a2502c27029a72e3fd2957cde Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Fri, 29 Jul 2011 17:16:55 -0700 Subject: [PATCH 028/326] [SCSI] isci: Update MAINTAINERS entry for the isci driver Signed-off-by: Dave Jiang Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- MAINTAINERS | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 069ee3b5c651..4fe6854c57d9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3262,6 +3262,17 @@ F: Documentation/input/multi-touch-protocol.txt F: drivers/input/input-mt.c K: \b(ABS|SYN)_MT_ +INTEL C600 SERIES SAS CONTROLLER DRIVER +M: Intel SCU Linux support +M: Dan Williams +M: Dave Jiang +M: Ed Nadolski +L: linux-scsi@vger.kernel.org +T: git git://git.kernel.org/pub/scm/linux/kernel/git/djbw/isci.git +S: Maintained +F: drivers/scsi/isci/ +F: firmware/isci/ + INTEL IDLE DRIVER M: Len Brown L: linux-pm@lists.linux-foundation.org From 3a7bda830fad427768ed71c0ebf3448849c006b5 Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Fri, 29 Jul 2011 17:17:00 -0700 Subject: [PATCH 029/326] [SCSI] isci: Adding documentation to API change and fixup sysfs registration Adding API update for adding isci_id entry scsi_host sysfs entry. Also fixing up the sysfs registration to the scsi_host template Signed-off-by: Dave Jiang Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- .../ABI/testing/sysfs-class-scsi_host | 13 +++++++ drivers/scsi/isci/init.c | 36 +++++++++---------- 2 files changed, 31 insertions(+), 18 deletions(-) create mode 100644 Documentation/ABI/testing/sysfs-class-scsi_host diff --git a/Documentation/ABI/testing/sysfs-class-scsi_host b/Documentation/ABI/testing/sysfs-class-scsi_host new file mode 100644 index 000000000000..29a4f892e433 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-scsi_host @@ -0,0 +1,13 @@ +What: /sys/class/scsi_host/hostX/isci_id +Date: June 2011 +Contact: Dave Jiang +Description: + This file contains the enumerated host ID for the Intel + SCU controller. The Intel(R) C600 Series Chipset SATA/SAS + Storage Control Unit embeds up to two 4-port controllers in + a single PCI device. The controllers are enumerated in order + which usually means the lowest number scsi_host corresponds + with the first controller, but this association is not + guaranteed. The 'isci_id' attribute unambiguously identifies + the controller index: '0' for the first controller, + '1' for the second. diff --git a/drivers/scsi/isci/init.c b/drivers/scsi/isci/init.c index 61e0d09e2b57..e78320bbec4f 100644 --- a/drivers/scsi/isci/init.c +++ b/drivers/scsi/isci/init.c @@ -59,6 +59,7 @@ #include #include #include +#include #include "isci.h" #include "task.h" #include "probe_roms.h" @@ -113,6 +114,22 @@ unsigned char max_concurr_spinup = 1; module_param(max_concurr_spinup, byte, 0); MODULE_PARM_DESC(max_concurr_spinup, "Max concurrent device spinup"); +static ssize_t isci_show_id(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = container_of(dev, typeof(*shost), shost_dev); + struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost); + struct isci_host *ihost = container_of(sas_ha, typeof(*ihost), sas_ha); + + return snprintf(buf, PAGE_SIZE, "%d\n", ihost->id); +} + +static DEVICE_ATTR(isci_id, S_IRUGO, isci_show_id, NULL); + +struct device_attribute *isci_host_attrs[] = { + &dev_attr_isci_id, + NULL +}; + static struct scsi_host_template isci_sht = { .module = THIS_MODULE, @@ -138,6 +155,7 @@ static struct scsi_host_template isci_sht = { .slave_alloc = sas_slave_alloc, .target_destroy = sas_target_destroy, .ioctl = sas_ioctl, + .shost_attrs = isci_host_attrs, }; static struct sas_domain_function_template isci_transport_ops = { @@ -232,17 +250,6 @@ static int isci_register_sas_ha(struct isci_host *isci_host) return 0; } -static ssize_t isci_show_id(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct Scsi_Host *shost = container_of(dev, typeof(*shost), shost_dev); - struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost); - struct isci_host *ihost = container_of(sas_ha, typeof(*ihost), sas_ha); - - return snprintf(buf, PAGE_SIZE, "%d\n", ihost->id); -} - -static DEVICE_ATTR(isci_id, S_IRUGO, isci_show_id, NULL); - static void isci_unregister(struct isci_host *isci_host) { struct Scsi_Host *shost; @@ -251,7 +258,6 @@ static void isci_unregister(struct isci_host *isci_host) return; shost = isci_host->shost; - device_remove_file(&shost->shost_dev, &dev_attr_isci_id); sas_unregister_ha(&isci_host->sas_ha); @@ -415,14 +421,8 @@ static struct isci_host *isci_host_alloc(struct pci_dev *pdev, int id) if (err) goto err_shost_remove; - err = device_create_file(&shost->shost_dev, &dev_attr_isci_id); - if (err) - goto err_unregister_ha; - return isci_host; - err_unregister_ha: - sas_unregister_ha(&(isci_host->sas_ha)); err_shost_remove: scsi_remove_host(shost); err_shost: From 39ea2c5b5ffaa344467da53e885cfa4ac0105050 Mon Sep 17 00:00:00 2001 From: Jeff Skirvin Date: Fri, 29 Jul 2011 17:17:05 -0700 Subject: [PATCH 030/326] [SCSI] isci: Leave requests alone if already terminating. Instead of immediately completing any request that has a second termination call made on it, wait for the TC done/abort HW event. Signed-off-by: Jeff Skirvin Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/request.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/isci/request.c b/drivers/scsi/isci/request.c index b4cf998385b3..b5d3a8c4d329 100644 --- a/drivers/scsi/isci/request.c +++ b/drivers/scsi/isci/request.c @@ -732,12 +732,20 @@ sci_io_request_terminate(struct isci_request *ireq) sci_change_state(&ireq->sm, SCI_REQ_ABORTING); return SCI_SUCCESS; case SCI_REQ_TASK_WAIT_TC_RESP: + /* The task frame was already confirmed to have been + * sent by the SCU HW. Since the state machine is + * now only waiting for the task response itself, + * abort the request and complete it immediately + * and don't wait for the task response. + */ sci_change_state(&ireq->sm, SCI_REQ_ABORTING); sci_change_state(&ireq->sm, SCI_REQ_COMPLETED); return SCI_SUCCESS; case SCI_REQ_ABORTING: - sci_change_state(&ireq->sm, SCI_REQ_COMPLETED); - return SCI_SUCCESS; + /* If a request has a termination requested twice, return + * a failure indication, since HW confirmation of the first + * abort is still outstanding. + */ case SCI_REQ_COMPLETED: default: dev_warn(&ireq->owning_controller->pdev->dev, From 9b4be528999483d70a1ffc0accd102e477d5a503 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 29 Jul 2011 17:17:10 -0700 Subject: [PATCH 031/326] [SCSI] isci: dynamic interrupt coalescing Hardware allows both an outstanding number commands and a timeout value (whichever occurs first) as a gate to the next interrupt generation. This scheme at completion time looks at the remaining number of outstanding tasks and sets the timeout to maximize small transaction operation. If transactions are large (take more than a few 10s of microseconds to complete) then performance is not interrupt processing bound, so the small timeouts this scheme generates are overridden by the time it takes for a completion to arrive. Tested-by: Dave Jiang Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/host.c | 10 +++++++++- drivers/scsi/isci/host.h | 3 +++ 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/isci/host.c b/drivers/scsi/isci/host.c index 26072f1e9852..2328f98c7f1e 100644 --- a/drivers/scsi/isci/host.c +++ b/drivers/scsi/isci/host.c @@ -1091,6 +1091,7 @@ static void isci_host_completion_routine(unsigned long data) struct isci_request *request; struct isci_request *next_request; struct sas_task *task; + u16 active; INIT_LIST_HEAD(&completed_request_list); INIT_LIST_HEAD(&errored_request_list); @@ -1181,6 +1182,13 @@ static void isci_host_completion_routine(unsigned long data) } } + /* the coalesence timeout doubles at each encoding step, so + * update it based on the ilog2 value of the outstanding requests + */ + active = isci_tci_active(ihost); + writel(SMU_ICC_GEN_VAL(NUMBER, active) | + SMU_ICC_GEN_VAL(TIMER, ISCI_COALESCE_BASE + ilog2(active)), + &ihost->smu_registers->interrupt_coalesce_control); } /** @@ -1471,7 +1479,7 @@ static void sci_controller_ready_state_enter(struct sci_base_state_machine *sm) struct isci_host *ihost = container_of(sm, typeof(*ihost), sm); /* set the default interrupt coalescence number and timeout value. */ - sci_controller_set_interrupt_coalescence(ihost, 0x10, 250); + sci_controller_set_interrupt_coalescence(ihost, 0, 0); } static void sci_controller_ready_state_exit(struct sci_base_state_machine *sm) diff --git a/drivers/scsi/isci/host.h b/drivers/scsi/isci/host.h index 062101a39f79..9f33831a2f04 100644 --- a/drivers/scsi/isci/host.h +++ b/drivers/scsi/isci/host.h @@ -369,6 +369,9 @@ static inline struct isci_host *dev_to_ihost(struct domain_device *dev) #define ISCI_TAG_SEQ(tag) (((tag) >> 12) & (SCI_MAX_SEQ-1)) #define ISCI_TAG_TCI(tag) ((tag) & (SCI_MAX_IO_REQUESTS-1)) +/* interrupt coalescing baseline: 9 == 3 to 5us interrupt delay per command */ +#define ISCI_COALESCE_BASE 9 + /* expander attached sata devices require 3 rnc slots */ static inline int sci_remote_device_node_count(struct isci_remote_device *idev) { From 77cd72a53f6426f81b7f56a862402849ee903bda Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 29 Jul 2011 17:17:16 -0700 Subject: [PATCH 032/326] [SCSI] isci: fix event-get pointer increment Hardware only increments the put pointer on event types >= 4. Do not increment the get pointer for event type 3. Reported-by: Kapil Karkra Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/host.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/scsi/isci/host.c b/drivers/scsi/isci/host.c index 2328f98c7f1e..6981b773a88d 100644 --- a/drivers/scsi/isci/host.c +++ b/drivers/scsi/isci/host.c @@ -531,6 +531,9 @@ static void sci_controller_process_completions(struct isci_host *ihost) break; case SCU_COMPLETION_TYPE_EVENT: + sci_controller_event_completion(ihost, ent); + break; + case SCU_COMPLETION_TYPE_NOTIFY: { event_cycle ^= ((event_get+1) & SCU_MAX_EVENTS) << (SMU_COMPLETION_QUEUE_GET_EVENT_CYCLE_BIT_SHIFT - SCU_MAX_EVENTS_SHIFT); From 98e2a5a3a125608505783bdb95744997f76b3c30 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 29 Jul 2011 17:17:21 -0700 Subject: [PATCH 033/326] [SCSI] isci: add version number Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/isci/init.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/isci/init.c b/drivers/scsi/isci/init.c index e78320bbec4f..29aa34efb0f5 100644 --- a/drivers/scsi/isci/init.c +++ b/drivers/scsi/isci/init.c @@ -64,6 +64,14 @@ #include "task.h" #include "probe_roms.h" +#define MAJ 1 +#define MIN 0 +#define BUILD 0 +#define DRV_VERSION __stringify(MAJ) "." __stringify(MIN) "." \ + __stringify(BUILD) + +MODULE_VERSION(DRV_VERSION); + static struct scsi_transport_template *isci_transport_template; static DEFINE_PCI_DEVICE_TABLE(isci_id_table) = { @@ -540,7 +548,8 @@ static __init int isci_init(void) { int err; - pr_info("%s: Intel(R) C600 SAS Controller Driver\n", DRV_NAME); + pr_info("%s: Intel(R) C600 SAS Controller Driver - version %s\n", + DRV_NAME, DRV_VERSION); isci_transport_template = sas_domain_attach_transport(&isci_transport_ops); if (!isci_transport_template) From f1c39625d63c9f8eba8f036429c10a9cb9e32920 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 24 Aug 2011 09:54:24 -0700 Subject: [PATCH 034/326] xen: use non-tracing preempt in xen_clocksource_read() The tracing code used sched_clock() to get tracing timestamps, which ends up calling xen_clocksource_read(). xen_clocksource_read() must disable preemption, but if preemption tracing is enabled, this results in infinite recursion. I've only noticed this when boot-time tracing tests are enabled, but it seems like a generic bug. It looks like it would also affect kvm_clocksource_read(). Reported-by: Konrad Rzeszutek Wilk Signed-off-by: Jeremy Fitzhardinge Cc: Avi Kivity Cc: Marcelo Tosatti --- arch/x86/xen/time.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 5158c505bef9..163b4679556e 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -168,9 +168,10 @@ cycle_t xen_clocksource_read(void) struct pvclock_vcpu_time_info *src; cycle_t ret; - src = &get_cpu_var(xen_vcpu)->time; + preempt_disable_notrace(); + src = &__get_cpu_var(xen_vcpu)->time; ret = pvclock_clocksource_read(src); - put_cpu_var(xen_vcpu); + preempt_enable_notrace(); return ret; } From b569ad34926defcff998f214afeb260331165985 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Aug 2011 15:07:35 -0400 Subject: [PATCH 035/326] NFSv4: nfs4_proc_async_renew should use a GFP_NOFS allocation We shouldn't allow the renew daemon to do direct reclaim on the NFS partition. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8c77039e7a81..776b41a16469 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3397,7 +3397,7 @@ int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) if (!atomic_inc_not_zero(&clp->cl_count)) return -EIO; - data = kmalloc(sizeof(*data), GFP_KERNEL); + data = kmalloc(sizeof(*data), GFP_NOFS); if (data == NULL) return -ENOMEM; data->client = clp; From 8534d4ec055d854be6c94e8e5654fa87678ea5f7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Aug 2011 15:07:37 -0400 Subject: [PATCH 036/326] NFSv4: nfs4_proc_renew should be declared static Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 2 -- fs/nfs/nfs4proc.c | 4 ++-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 1ec1a85fa71c..58adfb6e3657 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -237,8 +237,6 @@ extern const struct inode_operations nfs4_dir_inode_operations; extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *); extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *); extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred); -extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *); -extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *); extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *); extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *); extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 776b41a16469..b358ec1d1711 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3386,7 +3386,7 @@ static const struct rpc_call_ops nfs4_renew_ops = { .rpc_release = nfs4_renew_release, }; -int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) +static int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) { struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], @@ -3406,7 +3406,7 @@ int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) &nfs4_renew_ops, data); } -int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred) +static int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred) { struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], From 2f60ea6b8ceda61ae08bef71a652eac36ec193b3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Aug 2011 15:07:37 -0400 Subject: [PATCH 037/326] NFSv4: The NFSv4.0 client must send RENEW calls if it holds a delegation RFC3530 states that if the client holds a delegation, then it is obliged to continue to send RENEW calls once every lease period in order to allow the server to return NFS4ERR_CB_PATH_DOWN if the callback path is unreachable. This is not required for NFSv4.1, since the server can at any time set the SEQ4_STATUS_CB_PATH_DOWN_SESSION in any SEQUENCE operation. Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 5 ++++- fs/nfs/nfs4proc.c | 8 ++++++-- fs/nfs/nfs4renewd.c | 12 +++++++++--- 3 files changed, 19 insertions(+), 6 deletions(-) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 58adfb6e3657..e1b660728675 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -56,6 +56,9 @@ enum nfs4_session_state { NFS4_SESSION_DRAINING, }; +#define NFS4_RENEW_TIMEOUT 0x01 +#define NFS4_RENEW_DELEGATION_CB 0x02 + struct nfs4_minor_version_ops { u32 minor_version; @@ -225,7 +228,7 @@ struct nfs4_state_recovery_ops { }; struct nfs4_state_maintenance_ops { - int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *); + int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *, unsigned); struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *); int (*renew_lease)(struct nfs_client *, struct rpc_cred *); }; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b358ec1d1711..e89940a2819d 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3386,7 +3386,7 @@ static const struct rpc_call_ops nfs4_renew_ops = { .rpc_release = nfs4_renew_release, }; -static int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) +static int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred, unsigned renew_flags) { struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], @@ -3395,6 +3395,8 @@ static int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) }; struct nfs4_renewdata *data; + if (renew_flags == 0) + return 0; if (!atomic_inc_not_zero(&clp->cl_count)) return -EIO; data = kmalloc(sizeof(*data), GFP_NOFS); @@ -5504,11 +5506,13 @@ static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, struct rpc_ return rpc_run_task(&task_setup_data); } -static int nfs41_proc_async_sequence(struct nfs_client *clp, struct rpc_cred *cred) +static int nfs41_proc_async_sequence(struct nfs_client *clp, struct rpc_cred *cred, unsigned renew_flags) { struct rpc_task *task; int ret = 0; + if ((renew_flags & NFS4_RENEW_TIMEOUT) == 0) + return 0; task = _nfs41_proc_sequence(clp, cred); if (IS_ERR(task)) ret = PTR_ERR(task); diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index df8e7f3ca56d..dc484c0eae7f 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -60,6 +60,7 @@ nfs4_renew_state(struct work_struct *work) struct rpc_cred *cred; long lease; unsigned long last, now; + unsigned renew_flags = 0; ops = clp->cl_mvops->state_renewal_ops; dprintk("%s: start\n", __func__); @@ -72,18 +73,23 @@ nfs4_renew_state(struct work_struct *work) last = clp->cl_last_renewal; now = jiffies; /* Are we close to a lease timeout? */ - if (time_after(now, last + lease/3)) { + if (time_after(now, last + lease/3)) + renew_flags |= NFS4_RENEW_TIMEOUT; + if (nfs_delegations_present(clp)) + renew_flags |= NFS4_RENEW_DELEGATION_CB; + + if (renew_flags != 0) { cred = ops->get_state_renewal_cred_locked(clp); spin_unlock(&clp->cl_lock); if (cred == NULL) { - if (!nfs_delegations_present(clp)) { + if (!(renew_flags & NFS4_RENEW_DELEGATION_CB)) { set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); goto out; } nfs_expire_all_delegations(clp); } else { /* Queue an asynchronous RENEW. */ - ops->sched_state_renewal(clp, cred); + ops->sched_state_renewal(clp, cred, renew_flags); put_rpccred(cred); goto out_exp; } From 042b60beb410caf68f576d63d6849d0f0a545eb0 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Aug 2011 15:07:37 -0400 Subject: [PATCH 038/326] NFSv4: renewd needs to be able to handle the NFS4ERR_CB_PATH_DOWN error The NFSv4 spec does not specify that the server must repeat that error, so in order to avoid having the delegations revoked, we should handle it immediately. Also note that NFS4ERR_CB_PATH_DOWN does in fact renew the lease... Signed-off-by: Trond Myklebust --- fs/nfs/nfs4_fs.h | 1 + fs/nfs/nfs4proc.c | 8 ++++++-- fs/nfs/nfs4state.c | 6 ++++++ 3 files changed, 13 insertions(+), 2 deletions(-) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index e1b660728675..3e93e9a1bee1 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -350,6 +350,7 @@ extern void nfs4_close_sync(struct nfs4_state *, fmode_t); extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t); extern void nfs4_schedule_lease_recovery(struct nfs_client *); extern void nfs4_schedule_state_manager(struct nfs_client *); +extern void nfs4_schedule_path_down_recovery(struct nfs_client *clp); extern void nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *); extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags); extern void nfs41_handle_recall_slot(struct nfs_client *clp); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index e89940a2819d..4700fae1ada0 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3374,9 +3374,13 @@ static void nfs4_renew_done(struct rpc_task *task, void *calldata) if (task->tk_status < 0) { /* Unless we're shutting down, schedule state recovery! */ - if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) != 0) + if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) == 0) + return; + if (task->tk_status != NFS4ERR_CB_PATH_DOWN) { nfs4_schedule_lease_recovery(clp); - return; + return; + } + nfs4_schedule_path_down_recovery(clp); } do_renew_lease(clp, timestamp); } diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 72ab97ef3d61..39914be40b03 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1038,6 +1038,12 @@ void nfs4_schedule_lease_recovery(struct nfs_client *clp) nfs4_schedule_state_manager(clp); } +void nfs4_schedule_path_down_recovery(struct nfs_client *clp) +{ + nfs_handle_cb_pathdown(clp); + nfs4_schedule_state_manager(clp); +} + static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state) { From 3bdf28feafc52864bd7f17b39deec64833a89d19 Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Tue, 23 Aug 2011 16:48:26 -0500 Subject: [PATCH 039/326] ASoC: MPC5200: replace of_device with platform_device 'struct of_device' no longer exists, and its functionality has been merged into platform_device. Update the MPC5200 audio DMA driver (mpc5200_dma) accordingly. This fixes a build break. Signed-off-by: Timur Tabi Acked-by: Liam Girdwood Signed-off-by: Mark Brown Cc: stable@kernel.org --- sound/soc/fsl/mpc5200_dma.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/fsl/mpc5200_dma.c b/sound/soc/fsl/mpc5200_dma.c index fd0dc46afc34..5c6c2457386e 100644 --- a/sound/soc/fsl/mpc5200_dma.c +++ b/sound/soc/fsl/mpc5200_dma.c @@ -369,7 +369,7 @@ static struct snd_soc_platform_driver mpc5200_audio_dma_platform = { .pcm_free = &psc_dma_free, }; -static int mpc5200_hpcd_probe(struct of_device *op) +static int mpc5200_hpcd_probe(struct platform_device *op) { phys_addr_t fifo; struct psc_dma *psc_dma; @@ -487,7 +487,7 @@ static int mpc5200_hpcd_probe(struct of_device *op) return ret; } -static int mpc5200_hpcd_remove(struct of_device *op) +static int mpc5200_hpcd_remove(struct platform_device *op) { struct psc_dma *psc_dma = dev_get_drvdata(&op->dev); @@ -519,7 +519,7 @@ MODULE_DEVICE_TABLE(of, mpc5200_hpcd_match); static struct platform_driver mpc5200_hpcd_of_driver = { .probe = mpc5200_hpcd_probe, .remove = mpc5200_hpcd_remove, - .dev = { + .driver = { .owner = THIS_MODULE, .name = "mpc5200-pcm-audio", .of_match_table = mpc5200_hpcd_match, From 5ef56c8fecedf403a346d02140e52a072d693d6b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 25 Aug 2011 14:42:51 +1000 Subject: [PATCH 040/326] md: report failure if a 'set faulty' request doesn't. Sometimes a device will refuse to be set faulty. e.g. RAID1 will never let the last working device become faulty. So check if "md_error()" did manage to set the faulty flag and fail with EBUSY if it didn't. Resolves-Debian-Bug: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=601198 Reported-by: Mike Hommey Signed-off-by: NeilBrown --- drivers/md/md.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 8e221a20f5d9..1cd9bfb45e9a 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2561,7 +2561,10 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len) int err = -EINVAL; if (cmd_match(buf, "faulty") && rdev->mddev->pers) { md_error(rdev->mddev, rdev); - err = 0; + if (test_bit(Faulty, &rdev->flags)) + err = 0; + else + err = -EBUSY; } else if (cmd_match(buf, "remove")) { if (rdev->raid_disk >= 0) err = -EBUSY; @@ -5983,6 +5986,8 @@ static int set_disk_faulty(mddev_t *mddev, dev_t dev) return -ENODEV; md_error(mddev, rdev); + if (!test_bit(Faulty, &rdev->flags)) + return -EBUSY; return 0; } From aeb9b211849621f592288ed5ad694de9eeaae87a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 25 Aug 2011 14:43:08 +1000 Subject: [PATCH 041/326] md: ensure changes to 'write-mostly' are reflected in metadata. The 'write-mostly' flag can be changed through sysfs. With 0.90 metadata, those changes are reflected in the metadata. For 1.x metadata, they aren't. So fix super_1_sync to record 'write-mostly' status. Signed-off-by: NeilBrown --- drivers/md/md.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index 1cd9bfb45e9a..9a880239219d 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1738,6 +1738,11 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) sb->level = cpu_to_le32(mddev->level); sb->layout = cpu_to_le32(mddev->layout); + if (test_bit(WriteMostly, &rdev->flags)) + sb->devflags |= WriteMostly1; + else + sb->devflags &= ~WriteMostly1; + if (mddev->bitmap && mddev->bitmap_info.file == NULL) { sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_info.offset); sb->feature_map = cpu_to_le32(MD_FEATURE_BITMAP_OFFSET); From a5bf4df0c88b88d34b6f0e3bc8a402dac7d14611 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Thu, 25 Aug 2011 14:43:34 +1000 Subject: [PATCH 042/326] md: use REQ_NOIDLE flag in md_super_write() Queue idling is used for the anticipation of immediate sequencial I/O's but md_super_write() is a kind of one- shot operation, coupled with md_super_wait(), so the idling in this case will be just a waste of time. Specifying REQ_NOIDLE prevents it. Instead of adding the flag to submit_bio() directly, use pre-defined macro WRITE_FLUSH_FUA. Signed-off-by: Namhyung Kim Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 9a880239219d..aca611711264 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -848,7 +848,7 @@ void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, bio->bi_end_io = super_written; atomic_inc(&mddev->pending_writes); - submit_bio(REQ_WRITE | REQ_SYNC | REQ_FLUSH | REQ_FUA, bio); + submit_bio(WRITE_FLUSH_FUA, bio); } void md_super_wait(mddev_t *mddev) From 1b6afa17581027218088a18a9ceda600e0ddba7a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 25 Aug 2011 14:43:53 +1000 Subject: [PATCH 043/326] md/linear: avoid corrupting structure while waiting for rcu_free to complete. I don't know what I was thinking putting 'rcu' after a dynamically sized array! The array could still be in use when we call rcu_free() (That is the point) so we mustn't corrupt it. Cc: stable@kernel.org Signed-off-by: NeilBrown --- drivers/md/linear.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/linear.h b/drivers/md/linear.h index 0ce29b61605a..2f2da05b2ce9 100644 --- a/drivers/md/linear.h +++ b/drivers/md/linear.h @@ -10,9 +10,9 @@ typedef struct dev_info dev_info_t; struct linear_private_data { + struct rcu_head rcu; sector_t array_sectors; dev_info_t disks[0]; - struct rcu_head rcu; }; From 35d851df23b093ee027f827fed2213ae5e88fc7a Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Thu, 25 Aug 2011 14:21:37 +0200 Subject: [PATCH 044/326] HID: magicmouse: ignore 'ivalid report id' while switching modes, v2 This is basically a more generic respin of 23746a6 ("HID: magicmouse: ignore 'ivalid report id' while switching modes") which got reverted later by c3a492. It turns out that on some configurations, this is actually still the case and we are not able to detect in runtime. The device reponds with 'invalid report id' when feature report switching it into multitouch mode is sent to it. This has been silently ignored before 0825411ade ("HID: bt: Wait for ACK on Sent Reports"), but since this commit, it propagates -EIO from the _raw callback . So let the driver ignore -EIO as response to 0xd7,0x01 report, as that's how the device reacts in normal mode. Sad, but following reality. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=35022 Reported-by: Chase Douglas Reported-by: Jaikumar Ganesh Tested-by: Chase Douglas Tested-by: Jaikumar Ganesh Signed-off-by: Jiri Kosina --- drivers/hid/hid-magicmouse.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/hid/hid-magicmouse.c b/drivers/hid/hid-magicmouse.c index b5bdab3299bc..f0fbd7bd239e 100644 --- a/drivers/hid/hid-magicmouse.c +++ b/drivers/hid/hid-magicmouse.c @@ -537,9 +537,17 @@ static int magicmouse_probe(struct hid_device *hdev, } report->size = 6; + /* + * Some devices repond with 'invalid report id' when feature + * report switching it into multitouch mode is sent to it. + * + * This results in -EIO from the _raw low-level transport callback, + * but there seems to be no other way of switching the mode. + * Thus the super-ugly hacky success check below. + */ ret = hdev->hid_output_raw_report(hdev, feature, sizeof(feature), HID_FEATURE_REPORT); - if (ret != sizeof(feature)) { + if (ret != -EIO && ret != sizeof(feature)) { hid_err(hdev, "unable to request touch data (%d)\n", ret); goto err_stop_hw; } From 6d1db0777981e1626ae71243984ac300b61789ff Mon Sep 17 00:00:00 2001 From: Clemens Werther Date: Thu, 25 Aug 2011 15:35:14 +0200 Subject: [PATCH 045/326] HID: add support for HuiJia USB Gamepad connector Create each gamepad as a separate joystick Signed-off-by: Clemens Werther Signed-off-by: Jiri Kosina --- drivers/hid/hid-ids.h | 1 + drivers/hid/usbhid/hid-quirks.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 7d27d2b0445a..7484e1b67249 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -277,6 +277,7 @@ #define USB_DEVICE_ID_PENPOWER 0x00f4 #define USB_VENDOR_ID_GREENASIA 0x0e8f +#define USB_DEVICE_ID_GREENASIA_DUAL_USB_JOYPAD 0x3013 #define USB_VENDOR_ID_GRETAGMACBETH 0x0971 #define USB_DEVICE_ID_GRETAGMACBETH_HUEY 0x2005 diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c index 4bdb5d46c52c..3146fdcda272 100644 --- a/drivers/hid/usbhid/hid-quirks.c +++ b/drivers/hid/usbhid/hid-quirks.c @@ -47,6 +47,7 @@ static const struct hid_blacklist { { USB_VENDOR_ID_AFATECH, USB_DEVICE_ID_AFATECH_AF9016, HID_QUIRK_FULLSPEED_INTERVAL }, { USB_VENDOR_ID_ETURBOTOUCH, USB_DEVICE_ID_ETURBOTOUCH, HID_QUIRK_MULTI_INPUT }, + { USB_VENDOR_ID_GREENASIA, USB_DEVICE_ID_GREENASIA_DUAL_USB_JOYPAD, HID_QUIRK_MULTI_INPUT }, { USB_VENDOR_ID_PANTHERLORD, USB_DEVICE_ID_PANTHERLORD_TWIN_USB_JOYSTICK, HID_QUIRK_MULTI_INPUT | HID_QUIRK_SKIP_OUTPUT_REPORTS }, { USB_VENDOR_ID_PLAYDOTCOM, USB_DEVICE_ID_PLAYDOTCOM_EMS_USBII, HID_QUIRK_MULTI_INPUT }, { USB_VENDOR_ID_TOUCHPACK, USB_DEVICE_ID_TOUCHPACK_RTS, HID_QUIRK_MULTI_INPUT }, From 242d621964dd8641df53f7d51d4c6ead655cc5a6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 24 Aug 2011 05:57:51 +0000 Subject: [PATCH 046/326] xfs: deprecate the nodelaylog mount option Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Signed-off-by: Alex Elder --- Documentation/feature-removal-schedule.txt | 8 ++++++++ fs/xfs/xfs_super.c | 2 ++ 2 files changed, 10 insertions(+) diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index c4a6e148732a..4dc465477665 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -592,3 +592,11 @@ Why: In 3.0, we can now autodetect internal 3G device and already have interface that was used by acer-wmi driver. It will replaced by information log when acer-wmi initial. Who: Lee, Chun-Yi + +---------------------------- +What: The XFS nodelaylog mount option +When: 3.3 +Why: The delaylog mode that has been the default since 2.6.39 has proven + stable, and the old code is in the way of additional improvements in + the log code. +Who: Christoph Hellwig diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 9a72dda58bd0..c1b022f20d35 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -356,6 +356,8 @@ xfs_parseargs( mp->m_flags |= XFS_MOUNT_DELAYLOG; } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { mp->m_flags &= ~XFS_MOUNT_DELAYLOG; + xfs_warn(mp, + "nodelaylog is deprecated and will be removed in Linux 3.3"); } else if (!strcmp(this_char, MNTOPT_DISCARD)) { mp->m_flags |= XFS_MOUNT_DISCARD; } else if (!strcmp(this_char, MNTOPT_NODISCARD)) { From 0b0e1d6cbcc8627970e0399df8f06edd690ec7d9 Mon Sep 17 00:00:00 2001 From: "Stephen M. Cameron" Date: Tue, 9 Aug 2011 08:17:30 -0500 Subject: [PATCH 047/326] [SCSI] hpsa: fix problem that OBDR devices are not detected The test to detect OBDR ("One Button Disaster Recovery") cd-rom devices was comparing against uninitialized data. Fixed by moving the test for the device to where the inquiry data is collected, and uninitialized variable altogether as it wasn't really being used. Signed-off-by: Stephen M. Cameron Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/hpsa.c | 47 ++++++++++++++++++++++++++------------------- 1 file changed, 27 insertions(+), 20 deletions(-) diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index ec61bdb833ac..1f32f0610bc0 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -1548,10 +1548,17 @@ static inline void hpsa_set_bus_target_lun(struct hpsa_scsi_dev_t *device, } static int hpsa_update_device_info(struct ctlr_info *h, - unsigned char scsi3addr[], struct hpsa_scsi_dev_t *this_device) + unsigned char scsi3addr[], struct hpsa_scsi_dev_t *this_device, + unsigned char *is_OBDR_device) { -#define OBDR_TAPE_INQ_SIZE 49 + +#define OBDR_SIG_OFFSET 43 +#define OBDR_TAPE_SIG "$DR-10" +#define OBDR_SIG_LEN (sizeof(OBDR_TAPE_SIG) - 1) +#define OBDR_TAPE_INQ_SIZE (OBDR_SIG_OFFSET + OBDR_SIG_LEN) + unsigned char *inq_buff; + unsigned char *obdr_sig; inq_buff = kzalloc(OBDR_TAPE_INQ_SIZE, GFP_KERNEL); if (!inq_buff) @@ -1583,6 +1590,16 @@ static int hpsa_update_device_info(struct ctlr_info *h, else this_device->raid_level = RAID_UNKNOWN; + if (is_OBDR_device) { + /* See if this is a One-Button-Disaster-Recovery device + * by looking for "$DR-10" at offset 43 in inquiry data. + */ + obdr_sig = &inq_buff[OBDR_SIG_OFFSET]; + *is_OBDR_device = (this_device->devtype == TYPE_ROM && + strncmp(obdr_sig, OBDR_TAPE_SIG, + OBDR_SIG_LEN) == 0); + } + kfree(inq_buff); return 0; @@ -1716,7 +1733,7 @@ static int add_msa2xxx_enclosure_device(struct ctlr_info *h, return 0; } - if (hpsa_update_device_info(h, scsi3addr, this_device)) + if (hpsa_update_device_info(h, scsi3addr, this_device, NULL)) return 0; (*nmsa2xxx_enclosures)++; hpsa_set_bus_target_lun(this_device, bus, target, 0); @@ -1808,7 +1825,6 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) */ struct ReportLUNdata *physdev_list = NULL; struct ReportLUNdata *logdev_list = NULL; - unsigned char *inq_buff = NULL; u32 nphysicals = 0; u32 nlogicals = 0; u32 ndev_allocated = 0; @@ -1824,11 +1840,9 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) GFP_KERNEL); physdev_list = kzalloc(reportlunsize, GFP_KERNEL); logdev_list = kzalloc(reportlunsize, GFP_KERNEL); - inq_buff = kmalloc(OBDR_TAPE_INQ_SIZE, GFP_KERNEL); tmpdevice = kzalloc(sizeof(*tmpdevice), GFP_KERNEL); - if (!currentsd || !physdev_list || !logdev_list || - !inq_buff || !tmpdevice) { + if (!currentsd || !physdev_list || !logdev_list || !tmpdevice) { dev_err(&h->pdev->dev, "out of memory\n"); goto out; } @@ -1863,7 +1877,7 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) /* adjust our table of devices */ nmsa2xxx_enclosures = 0; for (i = 0; i < nphysicals + nlogicals + 1; i++) { - u8 *lunaddrbytes; + u8 *lunaddrbytes, is_OBDR = 0; /* Figure out where the LUN ID info is coming from */ lunaddrbytes = figure_lunaddrbytes(h, raid_ctlr_position, @@ -1874,7 +1888,8 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) continue; /* Get device type, vendor, model, device id */ - if (hpsa_update_device_info(h, lunaddrbytes, tmpdevice)) + if (hpsa_update_device_info(h, lunaddrbytes, tmpdevice, + &is_OBDR)) continue; /* skip it if we can't talk to it. */ figure_bus_target_lun(h, lunaddrbytes, &bus, &target, &lun, tmpdevice); @@ -1898,7 +1913,7 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) hpsa_set_bus_target_lun(this_device, bus, target, lun); switch (this_device->devtype) { - case TYPE_ROM: { + case TYPE_ROM: /* We don't *really* support actual CD-ROM devices, * just "One Button Disaster Recovery" tape drive * which temporarily pretends to be a CD-ROM drive. @@ -1906,15 +1921,8 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) * device by checking for "$DR-10" in bytes 43-48 of * the inquiry data. */ - char obdr_sig[7]; -#define OBDR_TAPE_SIG "$DR-10" - strncpy(obdr_sig, &inq_buff[43], 6); - obdr_sig[6] = '\0'; - if (strncmp(obdr_sig, OBDR_TAPE_SIG, 6) != 0) - /* Not OBDR device, ignore it. */ - break; - } - ncurrent++; + if (is_OBDR) + ncurrent++; break; case TYPE_DISK: if (i < nphysicals) @@ -1947,7 +1955,6 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) for (i = 0; i < ndev_allocated; i++) kfree(currentsd[i]); kfree(currentsd); - kfree(inq_buff); kfree(physdev_list); kfree(logdev_list); } From 01350d05539d1c95ef3568d062d864ab76ae7670 Mon Sep 17 00:00:00 2001 From: "Stephen M. Cameron" Date: Tue, 9 Aug 2011 08:18:01 -0500 Subject: [PATCH 048/326] [SCSI] hpsa: fix physical device lun and target numbering problem If a physical device exposed to the OS by hpsa is replaced (e.g. one hot plug tape drive is replaced by another, or a tape drive is placed into "OBDR" mode in which it acts like a CD-ROM device) and a rescan is initiated, the replaced device will be added to the SCSI midlayer with target and lun numbers set to -1. After that, a panic is likely to ensue. When a physical device is replaced, the lun and target number should be preserved. Signed-off-by: Stephen M. Cameron Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/hpsa.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index 1f32f0610bc0..b200b736b000 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -676,6 +676,16 @@ static void hpsa_scsi_replace_entry(struct ctlr_info *h, int hostno, BUG_ON(entry < 0 || entry >= HPSA_MAX_SCSI_DEVS_PER_HBA); removed[*nremoved] = h->dev[entry]; (*nremoved)++; + + /* + * New physical devices won't have target/lun assigned yet + * so we need to preserve the values in the slot we are replacing. + */ + if (new_entry->target == -1) { + new_entry->target = h->dev[entry]->target; + new_entry->lun = h->dev[entry]->lun; + } + h->dev[entry] = new_entry; added[*nadded] = new_entry; (*nadded)++; From a7402deb324f62106566f5a95199a54c41e200ef Mon Sep 17 00:00:00 2001 From: Mike Waychison Date: Fri, 12 Aug 2011 21:04:30 +0000 Subject: [PATCH 049/326] rtc: Initialized rtc_time->tm_isdst Even though the Linux kernel does not use the tm_isdst field, it is exposed as part of the ABI. This field can accidentally be left initialized, which is why we currently memset buffers returned to userland in rtc_read_time. There is a case however where the field can return garbage from the stack though when using the RTC_ALM_READ ioctl on the rtc device. This ioctl invokes rtc_read_alarm, which is careful to memset the rtc_wkalrm buffer that is copied to userland, but it then uses a struct copy to assign to alarm->time given the return value from rtc_ktime_to_tm(). rtc_ktime_to_tm() is implemented by calling rtc_time_to_tm using a derivative seconds counds from ktime, but rtc_time_to_tm does not assign a value to ->tm_isdst. This results in garbage from rtc_ktime_to_tm()'s frame ending up being copied out to userland as part of the returned rtc_wkalrm. Fix this by initializing rtc_time->tm_isdst to 0 in rtc_time_to_tm. Signed-off-by: Mike Waychison Signed-off-by: John Stultz --- drivers/rtc/rtc-lib.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/rtc/rtc-lib.c b/drivers/rtc/rtc-lib.c index 075f1708deae..c4cf05731118 100644 --- a/drivers/rtc/rtc-lib.c +++ b/drivers/rtc/rtc-lib.c @@ -85,6 +85,8 @@ void rtc_time_to_tm(unsigned long time, struct rtc_time *tm) time -= tm->tm_hour * 3600; tm->tm_min = time / 60; tm->tm_sec = time - tm->tm_min * 60; + + tm->tm_isdst = 0; } EXPORT_SYMBOL(rtc_time_to_tm); From 7e72c686347562b4a275c97b4bdd7a79c1f23c65 Mon Sep 17 00:00:00 2001 From: Todd Poynor Date: Wed, 10 Aug 2011 20:20:36 -0700 Subject: [PATCH 050/326] rtc: twl: Fix registration vs. init order Only register as an RTC device after the hardware has been successfully initialized. The RTC class driver will call back to this driver to read a pending alarm, and other drivers watching for new devices on the RTC class may read the RTC time upon registration. Such access might occur while the RTC is stopped, prior to clearing pending alarms, etc. The new ordering also avoids leaving the platform device drvdata set to an unregistered struct rtc_device * on probe errors. Signed-off-by: Todd Poynor Signed-off-by: John Stultz --- drivers/rtc/rtc-twl.c | 54 ++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 29 deletions(-) diff --git a/drivers/rtc/rtc-twl.c b/drivers/rtc/rtc-twl.c index 9677bbc433f9..20687d55e7a7 100644 --- a/drivers/rtc/rtc-twl.c +++ b/drivers/rtc/rtc-twl.c @@ -420,24 +420,12 @@ static struct rtc_class_ops twl_rtc_ops = { static int __devinit twl_rtc_probe(struct platform_device *pdev) { struct rtc_device *rtc; - int ret = 0; + int ret = -EINVAL; int irq = platform_get_irq(pdev, 0); u8 rd_reg; if (irq <= 0) - return -EINVAL; - - rtc = rtc_device_register(pdev->name, - &pdev->dev, &twl_rtc_ops, THIS_MODULE); - if (IS_ERR(rtc)) { - ret = PTR_ERR(rtc); - dev_err(&pdev->dev, "can't register RTC device, err %ld\n", - PTR_ERR(rtc)); - goto out0; - - } - - platform_set_drvdata(pdev, rtc); + goto out1; ret = twl_rtc_read_u8(&rd_reg, REG_RTC_STATUS_REG); if (ret < 0) @@ -454,14 +442,6 @@ static int __devinit twl_rtc_probe(struct platform_device *pdev) if (ret < 0) goto out1; - ret = request_threaded_irq(irq, NULL, twl_rtc_interrupt, - IRQF_TRIGGER_RISING, - dev_name(&rtc->dev), rtc); - if (ret < 0) { - dev_err(&pdev->dev, "IRQ is not free.\n"); - goto out1; - } - if (twl_class_is_6030()) { twl6030_interrupt_unmask(TWL6030_RTC_INT_MASK, REG_INT_MSK_LINE_A); @@ -472,28 +452,44 @@ static int __devinit twl_rtc_probe(struct platform_device *pdev) /* Check RTC module status, Enable if it is off */ ret = twl_rtc_read_u8(&rd_reg, REG_RTC_CTRL_REG); if (ret < 0) - goto out2; + goto out1; if (!(rd_reg & BIT_RTC_CTRL_REG_STOP_RTC_M)) { dev_info(&pdev->dev, "Enabling TWL-RTC.\n"); rd_reg = BIT_RTC_CTRL_REG_STOP_RTC_M; ret = twl_rtc_write_u8(rd_reg, REG_RTC_CTRL_REG); if (ret < 0) - goto out2; + goto out1; } /* init cached IRQ enable bits */ ret = twl_rtc_read_u8(&rtc_irq_bits, REG_RTC_INTERRUPTS_REG); if (ret < 0) - goto out2; + goto out1; - return ret; + rtc = rtc_device_register(pdev->name, + &pdev->dev, &twl_rtc_ops, THIS_MODULE); + if (IS_ERR(rtc)) { + ret = PTR_ERR(rtc); + dev_err(&pdev->dev, "can't register RTC device, err %ld\n", + PTR_ERR(rtc)); + goto out1; + } + + ret = request_threaded_irq(irq, NULL, twl_rtc_interrupt, + IRQF_TRIGGER_RISING, + dev_name(&rtc->dev), rtc); + if (ret < 0) { + dev_err(&pdev->dev, "IRQ is not free.\n"); + goto out2; + } + + platform_set_drvdata(pdev, rtc); + return 0; out2: - free_irq(irq, rtc); -out1: rtc_device_unregister(rtc); -out0: +out1: return ret; } From cfb7d557242783bc3bfe77683ced20b4909258ec Mon Sep 17 00:00:00 2001 From: Ping Cheng Date: Fri, 26 Aug 2011 23:10:02 -0700 Subject: [PATCH 051/326] Input: wacom - remove pressure for touch devices Touch devices do not report valid pressure or capacitance. Signed-off-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- drivers/input/tablet/wacom_sys.c | 14 -------------- drivers/input/tablet/wacom_wac.c | 6 +++--- 2 files changed, 3 insertions(+), 17 deletions(-) diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c index d27c9d91630b..958b4eb6369d 100644 --- a/drivers/input/tablet/wacom_sys.c +++ b/drivers/input/tablet/wacom_sys.c @@ -229,13 +229,6 @@ static int wacom_parse_hid(struct usb_interface *intf, struct hid_descriptor *hi get_unaligned_le16(&report[i + 3]); i += 4; } - } else if (usage == WCM_DIGITIZER) { - /* max pressure isn't reported - features->pressure_max = (unsigned short) - (report[i+4] << 8 | report[i + 3]); - */ - features->pressure_max = 255; - i += 4; } break; @@ -291,13 +284,6 @@ static int wacom_parse_hid(struct usb_interface *intf, struct hid_descriptor *hi pen = 1; i++; break; - - case HID_USAGE_UNDEFINED: - if (usage == WCM_DESKTOP && finger) /* capacity */ - features->pressure_max = - get_unaligned_le16(&report[i + 3]); - i += 4; - break; } break; diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index c1c2f7b28d89..3eccf212d5b2 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -818,7 +818,6 @@ static int wacom_bpt_touch(struct wacom_wac *wacom) x <<= 5; y <<= 5; } - input_report_abs(input, ABS_MT_PRESSURE, p); input_report_abs(input, ABS_MT_POSITION_X, x); input_report_abs(input, ABS_MT_POSITION_Y, y); } @@ -1056,10 +1055,11 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, features->x_fuzz, 0); input_set_abs_params(input_dev, ABS_Y, 0, features->y_max, features->y_fuzz, 0); - input_set_abs_params(input_dev, ABS_PRESSURE, 0, features->pressure_max, - features->pressure_fuzz, 0); if (features->device_type == BTN_TOOL_PEN) { + input_set_abs_params(input_dev, ABS_PRESSURE, 0, features->pressure_max, + features->pressure_fuzz, 0); + /* penabled devices have fixed resolution for each model */ input_abs_set_res(input_dev, ABS_X, features->x_resolution); input_abs_set_res(input_dev, ABS_Y, features->y_resolution); From 1fab84aa635572fbd74df8fd4fd25ea0a24c76e5 Mon Sep 17 00:00:00 2001 From: Jason Gerecke Date: Fri, 26 Aug 2011 23:18:22 -0700 Subject: [PATCH 052/326] Input: wacom - advertise BTN_TOOL_PEN and BTN_STYLUS for PenPartner The Wacom PenPartner should advertise its stylus tip and button in addition to the eraser tool. These are both physically present on the hardware, and emitted from 'wacom_penpartner_irq'. Signed-off-by: Jason Gerecke Reviewed-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- drivers/input/tablet/wacom_wac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index 3eccf212d5b2..2d88316d0e54 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -1191,13 +1191,13 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, case PL: case PTU: case DTU: - __set_bit(BTN_TOOL_PEN, input_dev->keybit); - __set_bit(BTN_STYLUS, input_dev->keybit); __set_bit(BTN_STYLUS2, input_dev->keybit); /* fall through */ case PENPARTNER: + __set_bit(BTN_TOOL_PEN, input_dev->keybit); __set_bit(BTN_TOOL_RUBBER, input_dev->keybit); + __set_bit(BTN_STYLUS, input_dev->keybit); break; case BAMBOO_PT: From 7b727acc412c9320dc56a0fd7312febf8710ac0e Mon Sep 17 00:00:00 2001 From: axel lin Date: Thu, 25 Aug 2011 09:42:09 -0700 Subject: [PATCH 053/326] Input: cm109 - fix checking return value of usb_control_msg If successful, usb_control_msg returns the number of bytes transferred, otherwise a negative error number. Signed-off-by: Axel Lin Signed-off-by: Dmitry Torokhov --- drivers/input/misc/cm109.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/cm109.c b/drivers/input/misc/cm109.c index b09c7d127219..ab860511f016 100644 --- a/drivers/input/misc/cm109.c +++ b/drivers/input/misc/cm109.c @@ -475,7 +475,7 @@ static void cm109_toggle_buzzer_sync(struct cm109_dev *dev, int on) le16_to_cpu(dev->ctl_req->wIndex), dev->ctl_data, USB_PKT_LEN, USB_CTRL_SET_TIMEOUT); - if (error && error != EINTR) + if (error < 0 && error != -EINTR) err("%s: usb_control_msg() failed %d", __func__, error); } From 8c6756603976e9d21bba9913cd80c38ec529a1fb Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Fri, 26 Aug 2011 23:37:33 -0700 Subject: [PATCH 054/326] Input: adp5588-keys - remove incorrect modalias For i2c drivers, we should use "i2c:" prefix for modalias. MODULE_DEVICE_TABLE will setup the modulalias for us, thus adding a MODULE_ALIAS is redundant (in addition to being incorrect). Signed-off-by: Axel Lin Acked-by: Michael Hennerich Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/adp5588-keys.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/input/keyboard/adp5588-keys.c b/drivers/input/keyboard/adp5588-keys.c index 7b404e5443ed..e34eeb8ae371 100644 --- a/drivers/input/keyboard/adp5588-keys.c +++ b/drivers/input/keyboard/adp5588-keys.c @@ -668,4 +668,3 @@ module_exit(adp5588_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michael Hennerich "); MODULE_DESCRIPTION("ADP5588/87 Keypad driver"); -MODULE_ALIAS("platform:adp5588-keys"); From 130655ef097940b627e8e04fa7c6f3b51cf24f85 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 23 Aug 2011 08:36:59 +0800 Subject: [PATCH 055/326] slub: add slab with one free object to partial list tail The slab has just one free object, adding it to partial list head doesn't make sense. And it can cause lock contentation. For example, 1. CPU takes the slab from partial list 2. fetch an object 3. switch to another slab 4. free an object, then the slab is added to partial list again In this way n->list_lock will be heavily contended. In fact, Alex had a hackbench regression. 3.1-rc1 performance drops about 70% against 3.0. This patch fixes it. Acked-by: Christoph Lameter Reported-by: Alex Shi Signed-off-by: Shaohua Li Signed-off-by: Shaohua Li Signed-off-by: Pekka Enberg --- mm/slub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/slub.c b/mm/slub.c index 9f662d70eb47..7c54fe83a90c 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -2377,7 +2377,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, */ if (unlikely(!prior)) { remove_full(s, page); - add_partial(n, page, 0); + add_partial(n, page, 1); stat(s, FREE_ADD_PARTIAL); } } From 8cb2049c744809193ed3707a37c09676a24599ee Mon Sep 17 00:00:00 2001 From: Arun Easi Date: Tue, 16 Aug 2011 11:29:22 -0700 Subject: [PATCH 056/326] [SCSI] qla2xxx: T10 DIF - Handle uninitalized sectors. Driver needs to update protection bytes for uninitialized sectors as they are not DMA-d. Signed-off-by: Arun Easi Reviewed-by: Andrew Vasquez Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_attr.c | 5 +- drivers/scsi/qla2xxx/qla_fw.h | 5 + drivers/scsi/qla2xxx/qla_inline.h | 21 +++ drivers/scsi/qla2xxx/qla_iocb.c | 233 ++++++++++++++++++++++++++++-- drivers/scsi/qla2xxx/qla_isr.c | 90 ++++++++++-- drivers/scsi/qla2xxx/qla_os.c | 19 ++- 6 files changed, 335 insertions(+), 38 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 7836eb01c7fc..810067099801 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1788,11 +1788,14 @@ qla24xx_vport_create(struct fc_vport *fc_vport, bool disable) if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { + int prot = 0; vha->flags.difdix_supported = 1; ql_dbg(ql_dbg_user, vha, 0x7082, "Registered for DIF/DIX type 1 and 3 protection.\n"); + if (ql2xenabledif == 1) + prot = SHOST_DIX_TYPE0_PROTECTION; scsi_host_set_prot(vha->host, - SHOST_DIF_TYPE1_PROTECTION + prot | SHOST_DIF_TYPE1_PROTECTION | SHOST_DIF_TYPE2_PROTECTION | SHOST_DIF_TYPE3_PROTECTION | SHOST_DIX_TYPE1_PROTECTION diff --git a/drivers/scsi/qla2xxx/qla_fw.h b/drivers/scsi/qla2xxx/qla_fw.h index 691783abfb69..aa69486dc064 100644 --- a/drivers/scsi/qla2xxx/qla_fw.h +++ b/drivers/scsi/qla2xxx/qla_fw.h @@ -537,6 +537,11 @@ struct sts_entry_24xx { /* * If DIF Error is set in comp_status, these additional fields are * defined: + * + * !!! NOTE: Firmware sends expected/actual DIF data in big endian + * format; but all of the "data" field gets swab32-d in the beginning + * of qla2x00_status_entry(). + * * &data[10] : uint8_t report_runt_bg[2]; - computed guard * &data[12] : uint8_t actual_dif[8]; - DIF Data received * &data[20] : uint8_t expected_dif[8]; - DIF Data computed diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h index d2e904bc21c0..c06e5f9b431e 100644 --- a/drivers/scsi/qla2xxx/qla_inline.h +++ b/drivers/scsi/qla2xxx/qla_inline.h @@ -102,3 +102,24 @@ qla2x00_set_fcport_state(fc_port_t *fcport, int state) fcport->d_id.b.al_pa); } } + +static inline int +qla2x00_hba_err_chk_enabled(unsigned char op) +{ + switch (op) { + case SCSI_PROT_READ_STRIP: + case SCSI_PROT_WRITE_INSERT: + if (ql2xenablehba_err_chk >= 1) + return 1; + break; + case SCSI_PROT_READ_PASS: + case SCSI_PROT_WRITE_PASS: + if (ql2xenablehba_err_chk >= 2) + return 1; + break; + case SCSI_PROT_READ_INSERT: + case SCSI_PROT_WRITE_STRIP: + return 1; + } + return 0; +} diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c index 49d6906af886..09ad3ce60064 100644 --- a/drivers/scsi/qla2xxx/qla_iocb.c +++ b/drivers/scsi/qla2xxx/qla_iocb.c @@ -717,12 +717,17 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, unsigned char op = scsi_get_prot_op(cmd); switch (scsi_get_prot_type(cmd)) { - /* For TYPE 0 protection: no checking */ case SCSI_PROT_DIF_TYPE0: - pkt->ref_tag_mask[0] = 0x00; - pkt->ref_tag_mask[1] = 0x00; - pkt->ref_tag_mask[2] = 0x00; - pkt->ref_tag_mask[3] = 0x00; + /* + * No check for ql2xenablehba_err_chk, as it would be an + * I/O error if hba tag generation is not done. + */ + pkt->ref_tag = cpu_to_le32((uint32_t) + (0xffffffff & scsi_get_lba(cmd))); + pkt->ref_tag_mask[0] = 0xff; + pkt->ref_tag_mask[1] = 0xff; + pkt->ref_tag_mask[2] = 0xff; + pkt->ref_tag_mask[3] = 0xff; break; /* @@ -730,7 +735,7 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, * match LBA in CDB + N */ case SCSI_PROT_DIF_TYPE2: - if (!ql2xenablehba_err_chk) + if (!qla2x00_hba_err_chk_enabled(op)) break; if (scsi_prot_sg_count(cmd)) { @@ -763,7 +768,7 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, * 16 bit app tag. */ case SCSI_PROT_DIF_TYPE1: - if (!ql2xenablehba_err_chk) + if (!qla2x00_hba_err_chk_enabled(op)) break; if (protcnt && (op == SCSI_PROT_WRITE_STRIP || @@ -798,7 +803,161 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, scsi_get_prot_type(cmd), cmd); } +struct qla2_sgx { + dma_addr_t dma_addr; /* OUT */ + uint32_t dma_len; /* OUT */ + uint32_t tot_bytes; /* IN */ + struct scatterlist *cur_sg; /* IN */ + + /* for book keeping, bzero on initial invocation */ + uint32_t bytes_consumed; + uint32_t num_bytes; + uint32_t tot_partial; + + /* for debugging */ + uint32_t num_sg; + srb_t *sp; +}; + +static int +qla24xx_get_one_block_sg(uint32_t blk_sz, struct qla2_sgx *sgx, + uint32_t *partial) +{ + struct scatterlist *sg; + uint32_t cumulative_partial, sg_len; + dma_addr_t sg_dma_addr; + + if (sgx->num_bytes == sgx->tot_bytes) + return 0; + + sg = sgx->cur_sg; + cumulative_partial = sgx->tot_partial; + + sg_dma_addr = sg_dma_address(sg); + sg_len = sg_dma_len(sg); + + sgx->dma_addr = sg_dma_addr + sgx->bytes_consumed; + + if ((cumulative_partial + (sg_len - sgx->bytes_consumed)) >= blk_sz) { + sgx->dma_len = (blk_sz - cumulative_partial); + sgx->tot_partial = 0; + sgx->num_bytes += blk_sz; + *partial = 0; + } else { + sgx->dma_len = sg_len - sgx->bytes_consumed; + sgx->tot_partial += sgx->dma_len; + *partial = 1; + } + + sgx->bytes_consumed += sgx->dma_len; + + if (sg_len == sgx->bytes_consumed) { + sg = sg_next(sg); + sgx->num_sg++; + sgx->cur_sg = sg; + sgx->bytes_consumed = 0; + } + + return 1; +} + +static int +qla24xx_walk_and_build_sglist_no_difb(struct qla_hw_data *ha, srb_t *sp, + uint32_t *dsd, uint16_t tot_dsds) +{ + void *next_dsd; + uint8_t avail_dsds = 0; + uint32_t dsd_list_len; + struct dsd_dma *dsd_ptr; + struct scatterlist *sg_prot; + uint32_t *cur_dsd = dsd; + uint16_t used_dsds = tot_dsds; + + uint32_t prot_int; + uint32_t partial; + struct qla2_sgx sgx; + dma_addr_t sle_dma; + uint32_t sle_dma_len, tot_prot_dma_len = 0; + struct scsi_cmnd *cmd = sp->cmd; + + prot_int = cmd->device->sector_size; + + memset(&sgx, 0, sizeof(struct qla2_sgx)); + sgx.tot_bytes = scsi_bufflen(sp->cmd); + sgx.cur_sg = scsi_sglist(sp->cmd); + sgx.sp = sp; + + sg_prot = scsi_prot_sglist(sp->cmd); + + while (qla24xx_get_one_block_sg(prot_int, &sgx, &partial)) { + + sle_dma = sgx.dma_addr; + sle_dma_len = sgx.dma_len; +alloc_and_fill: + /* Allocate additional continuation packets? */ + if (avail_dsds == 0) { + avail_dsds = (used_dsds > QLA_DSDS_PER_IOCB) ? + QLA_DSDS_PER_IOCB : used_dsds; + dsd_list_len = (avail_dsds + 1) * 12; + used_dsds -= avail_dsds; + + /* allocate tracking DS */ + dsd_ptr = kzalloc(sizeof(struct dsd_dma), GFP_ATOMIC); + if (!dsd_ptr) + return 1; + + /* allocate new list */ + dsd_ptr->dsd_addr = next_dsd = + dma_pool_alloc(ha->dl_dma_pool, GFP_ATOMIC, + &dsd_ptr->dsd_list_dma); + + if (!next_dsd) { + /* + * Need to cleanup only this dsd_ptr, rest + * will be done by sp_free_dma() + */ + kfree(dsd_ptr); + return 1; + } + + list_add_tail(&dsd_ptr->list, + &((struct crc_context *)sp->ctx)->dsd_list); + + sp->flags |= SRB_CRC_CTX_DSD_VALID; + + /* add new list to cmd iocb or last list */ + *cur_dsd++ = cpu_to_le32(LSD(dsd_ptr->dsd_list_dma)); + *cur_dsd++ = cpu_to_le32(MSD(dsd_ptr->dsd_list_dma)); + *cur_dsd++ = dsd_list_len; + cur_dsd = (uint32_t *)next_dsd; + } + *cur_dsd++ = cpu_to_le32(LSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(MSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(sle_dma_len); + avail_dsds--; + + if (partial == 0) { + /* Got a full protection interval */ + sle_dma = sg_dma_address(sg_prot) + tot_prot_dma_len; + sle_dma_len = 8; + + tot_prot_dma_len += sle_dma_len; + if (tot_prot_dma_len == sg_dma_len(sg_prot)) { + tot_prot_dma_len = 0; + sg_prot = sg_next(sg_prot); + } + + partial = 1; /* So as to not re-enter this block */ + goto alloc_and_fill; + } + } + /* Null termination */ + *cur_dsd++ = 0; + *cur_dsd++ = 0; + *cur_dsd++ = 0; + return 0; +} static int qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd, uint16_t tot_dsds) @@ -981,7 +1140,7 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, struct scsi_cmnd *cmd; struct scatterlist *cur_seg; int sgc; - uint32_t total_bytes; + uint32_t total_bytes = 0; uint32_t data_bytes; uint32_t dif_bytes; uint8_t bundling = 1; @@ -1023,8 +1182,10 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, __constant_cpu_to_le16(CF_READ_DATA); } - tot_prot_dsds = scsi_prot_sg_count(cmd); - if (!tot_prot_dsds) + if ((scsi_get_prot_op(sp->cmd) == SCSI_PROT_READ_INSERT) || + (scsi_get_prot_op(sp->cmd) == SCSI_PROT_WRITE_STRIP) || + (scsi_get_prot_op(sp->cmd) == SCSI_PROT_READ_STRIP) || + (scsi_get_prot_op(sp->cmd) == SCSI_PROT_WRITE_INSERT)) bundling = 0; /* Allocate CRC context from global pool */ @@ -1107,15 +1268,28 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, cmd_pkt->fcp_rsp_dseg_len = 0; /* Let response come in status iocb */ /* Compute dif len and adjust data len to incude protection */ - total_bytes = data_bytes; dif_bytes = 0; blk_size = cmd->device->sector_size; - if (scsi_get_prot_op(cmd) != SCSI_PROT_NORMAL) { - dif_bytes = (data_bytes / blk_size) * 8; - total_bytes += dif_bytes; + dif_bytes = (data_bytes / blk_size) * 8; + + switch (scsi_get_prot_op(sp->cmd)) { + case SCSI_PROT_READ_INSERT: + case SCSI_PROT_WRITE_STRIP: + total_bytes = data_bytes; + data_bytes += dif_bytes; + break; + + case SCSI_PROT_READ_STRIP: + case SCSI_PROT_WRITE_INSERT: + case SCSI_PROT_READ_PASS: + case SCSI_PROT_WRITE_PASS: + total_bytes = data_bytes + dif_bytes; + break; + default: + BUG(); } - if (!ql2xenablehba_err_chk) + if (!qla2x00_hba_err_chk_enabled(scsi_get_prot_op(cmd))) fw_prot_opts |= 0x10; /* Disable Guard tag checking */ if (!bundling) { @@ -1151,7 +1325,12 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, cmd_pkt->control_flags |= __constant_cpu_to_le16(CF_DATA_SEG_DESCR_ENABLE); - if (qla24xx_walk_and_build_sglist(ha, sp, cur_dsd, + + if (!bundling && tot_prot_dsds) { + if (qla24xx_walk_and_build_sglist_no_difb(ha, sp, + cur_dsd, tot_dsds)) + goto crc_queuing_error; + } else if (qla24xx_walk_and_build_sglist(ha, sp, cur_dsd, (tot_dsds - tot_prot_dsds))) goto crc_queuing_error; @@ -1414,6 +1593,22 @@ qla24xx_dif_start_scsi(srb_t *sp) goto queuing_error; else sp->flags |= SRB_DMA_VALID; + + if ((scsi_get_prot_op(cmd) == SCSI_PROT_READ_INSERT) || + (scsi_get_prot_op(cmd) == SCSI_PROT_WRITE_STRIP)) { + struct qla2_sgx sgx; + uint32_t partial; + + memset(&sgx, 0, sizeof(struct qla2_sgx)); + sgx.tot_bytes = scsi_bufflen(cmd); + sgx.cur_sg = scsi_sglist(cmd); + sgx.sp = sp; + + nseg = 0; + while (qla24xx_get_one_block_sg( + cmd->device->sector_size, &sgx, &partial)) + nseg++; + } } else nseg = 0; @@ -1428,6 +1623,11 @@ qla24xx_dif_start_scsi(srb_t *sp) goto queuing_error; else sp->flags |= SRB_CRC_PROT_DMA_VALID; + + if ((scsi_get_prot_op(cmd) == SCSI_PROT_READ_INSERT) || + (scsi_get_prot_op(cmd) == SCSI_PROT_WRITE_STRIP)) { + nseg = scsi_bufflen(cmd) / cmd->device->sector_size; + } } else { nseg = 0; } @@ -1454,6 +1654,7 @@ qla24xx_dif_start_scsi(srb_t *sp) /* Build header part of command packet (excluding the OPCODE). */ req->current_outstanding_cmd = handle; req->outstanding_cmds[handle] = sp; + sp->handle = handle; sp->cmd->host_scribble = (unsigned char *)(unsigned long)handle; req->cnt -= req_cnt; diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index b16b7725dee0..53339f10a598 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -1435,25 +1435,27 @@ struct scsi_dif_tuple { * ASC/ASCQ fields in the sense buffer with ILLEGAL_REQUEST * to indicate to the kernel that the HBA detected error. */ -static inline void +static inline int qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) { struct scsi_qla_host *vha = sp->fcport->vha; struct scsi_cmnd *cmd = sp->cmd; - struct scsi_dif_tuple *ep = - (struct scsi_dif_tuple *)&sts24->data[20]; - struct scsi_dif_tuple *ap = - (struct scsi_dif_tuple *)&sts24->data[12]; + uint8_t *ap = &sts24->data[12]; + uint8_t *ep = &sts24->data[20]; uint32_t e_ref_tag, a_ref_tag; uint16_t e_app_tag, a_app_tag; uint16_t e_guard, a_guard; - e_ref_tag = be32_to_cpu(ep->ref_tag); - a_ref_tag = be32_to_cpu(ap->ref_tag); - e_app_tag = be16_to_cpu(ep->app_tag); - a_app_tag = be16_to_cpu(ap->app_tag); - e_guard = be16_to_cpu(ep->guard); - a_guard = be16_to_cpu(ap->guard); + /* + * swab32 of the "data" field in the beginning of qla2x00_status_entry() + * would make guard field appear at offset 2 + */ + a_guard = le16_to_cpu(*(uint16_t *)(ap + 2)); + a_app_tag = le16_to_cpu(*(uint16_t *)(ap + 0)); + a_ref_tag = le32_to_cpu(*(uint32_t *)(ap + 4)); + e_guard = le16_to_cpu(*(uint16_t *)(ep + 2)); + e_app_tag = le16_to_cpu(*(uint16_t *)(ep + 0)); + e_ref_tag = le32_to_cpu(*(uint32_t *)(ep + 4)); ql_dbg(ql_dbg_io, vha, 0x3023, "iocb(s) %p Returned STATUS.\n", sts24); @@ -1465,6 +1467,63 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) cmd->cmnd[0], (u64)scsi_get_lba(cmd), a_ref_tag, e_ref_tag, a_app_tag, e_app_tag, a_guard, e_guard); + /* + * Ignore sector if: + * For type 3: ref & app tag is all 'f's + * For type 0,1,2: app tag is all 'f's + */ + if ((a_app_tag == 0xffff) && + ((scsi_get_prot_type(cmd) != SCSI_PROT_DIF_TYPE3) || + (a_ref_tag == 0xffffffff))) { + uint32_t blocks_done, resid; + sector_t lba_s = scsi_get_lba(cmd); + + /* 2TB boundary case covered automatically with this */ + blocks_done = e_ref_tag - (uint32_t)lba_s + 1; + + resid = scsi_bufflen(cmd) - (blocks_done * + cmd->device->sector_size); + + scsi_set_resid(cmd, resid); + cmd->result = DID_OK << 16; + + /* Update protection tag */ + if (scsi_prot_sg_count(cmd)) { + uint32_t i, j = 0, k = 0, num_ent; + struct scatterlist *sg; + struct sd_dif_tuple *spt; + + /* Patch the corresponding protection tags */ + scsi_for_each_prot_sg(cmd, sg, + scsi_prot_sg_count(cmd), i) { + num_ent = sg_dma_len(sg) / 8; + if (k + num_ent < blocks_done) { + k += num_ent; + continue; + } + j = blocks_done - k - 1; + k = blocks_done; + break; + } + + if (k != blocks_done) { + qla_printk(KERN_WARNING, sp->fcport->vha->hw, + "unexpected tag values tag:lba=%x:%lx)\n", + e_ref_tag, lba_s); + return 1; + } + + spt = page_address(sg_page(sg)) + sg->offset; + spt += j; + + spt->app_tag = 0xffff; + if (scsi_get_prot_type(cmd) == SCSI_PROT_DIF_TYPE3) + spt->ref_tag = 0xffffffff; + } + + return 0; + } + /* check guard */ if (e_guard != a_guard) { scsi_build_sense_buffer(1, cmd->sense_buffer, ILLEGAL_REQUEST, @@ -1472,7 +1531,7 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) set_driver_byte(cmd, DRIVER_SENSE); set_host_byte(cmd, DID_ABORT); cmd->result |= SAM_STAT_CHECK_CONDITION << 1; - return; + return 1; } /* check appl tag */ @@ -1482,7 +1541,7 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) set_driver_byte(cmd, DRIVER_SENSE); set_host_byte(cmd, DID_ABORT); cmd->result |= SAM_STAT_CHECK_CONDITION << 1; - return; + return 1; } /* check ref tag */ @@ -1492,8 +1551,9 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) set_driver_byte(cmd, DRIVER_SENSE); set_host_byte(cmd, DID_ABORT); cmd->result |= SAM_STAT_CHECK_CONDITION << 1; - return; + return 1; } + return 1; } /** @@ -1767,7 +1827,7 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt) break; case CS_DIF_ERROR: - qla2x00_handle_dif_error(sp, sts24); + logit = qla2x00_handle_dif_error(sp, sts24); break; default: cp->result = DID_ERROR << 16; diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index e02df276804e..d65a3005b439 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -106,17 +106,21 @@ MODULE_PARM_DESC(ql2xmaxqdepth, "Maximum queue depth to report for target devices."); /* Do not change the value of this after module load */ -int ql2xenabledif = 1; +int ql2xenabledif = 0; module_param(ql2xenabledif, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(ql2xenabledif, " Enable T10-CRC-DIF " - " Default is 0 - No DIF Support. 1 - Enable it"); + " Default is 0 - No DIF Support. 1 - Enable it" + ", 2 - Enable DIF for all types, except Type 0."); -int ql2xenablehba_err_chk; +int ql2xenablehba_err_chk = 2; module_param(ql2xenablehba_err_chk, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(ql2xenablehba_err_chk, - " Enable T10-CRC-DIF Error isolation by HBA" - " Default is 0 - Error isolation disabled, 1 - Enable it"); + " Enable T10-CRC-DIF Error isolation by HBA:\n" + " Default is 1.\n" + " 0 -- Error isolation disabled\n" + " 1 -- Error isolation enabled only for DIX Type 0\n" + " 2 -- Error isolation enabled for all Types\n"); int ql2xiidmaenable=1; module_param(ql2xiidmaenable, int, S_IRUGO); @@ -2380,11 +2384,14 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { + int prot = 0; base_vha->flags.difdix_supported = 1; ql_dbg(ql_dbg_init, base_vha, 0x00f1, "Registering for DIF/DIX type 1 and 3 protection.\n"); + if (ql2xenabledif == 1) + prot = SHOST_DIX_TYPE0_PROTECTION; scsi_host_set_prot(host, - SHOST_DIF_TYPE1_PROTECTION + prot | SHOST_DIF_TYPE1_PROTECTION | SHOST_DIF_TYPE2_PROTECTION | SHOST_DIF_TYPE3_PROTECTION | SHOST_DIX_TYPE1_PROTECTION From e02587d777bfb398f70709fd3a92fa0154959003 Mon Sep 17 00:00:00 2001 From: Arun Easi Date: Tue, 16 Aug 2011 11:29:23 -0700 Subject: [PATCH 057/326] [SCSI] qla2xxx: T10 DIF - Fix incorrect error reporting. This fix: - Disables app tag peeking; correct tag check will be added when the SCSI API is available. - Always derive ref_tag from scsi_get_lba() - Removes incorrect swap of FCP_LUN in FCP_CMND - Moves app-tag error check before ref-tag check. The reason being, currently there is no interface in SCSI to retrieve the app-tag for protection I/Os, so driver puts zero for app-tag in the firmware interface, but requests not to validate it, but when a ref-tag error is detected by firmware, it would put expected/actual tags for all the protection tags (guard/app/ref). As driver checks for app tag error first, a ref-tag error is incorrectly flagged as app-tag error. - Convert HBA specific checks to capability based. Signed-off-by: Arun Easi Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_attr.c | 2 +- drivers/scsi/qla2xxx/qla_dbg.c | 36 ++++++++++---------- drivers/scsi/qla2xxx/qla_def.h | 2 ++ drivers/scsi/qla2xxx/qla_inline.h | 12 +++++-- drivers/scsi/qla2xxx/qla_iocb.c | 55 ++++++++++++------------------- drivers/scsi/qla2xxx/qla_isr.c | 19 ++++++----- drivers/scsi/qla2xxx/qla_mid.c | 2 +- drivers/scsi/qla2xxx/qla_os.c | 4 +-- 8 files changed, 65 insertions(+), 67 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 810067099801..a31e05f3bfd4 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1786,7 +1786,7 @@ qla24xx_vport_create(struct fc_vport *fc_vport, bool disable) fc_vport_set_state(fc_vport, FC_VPORT_LINKDOWN); } - if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) { + if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { int prot = 0; vha->flags.difdix_supported = 1; diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index 2155071f3100..d79cd8a5f831 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -8,24 +8,24 @@ /* * Table for showing the current message id in use for particular level * Change this table for addition of log/debug messages. - * ----------------------------------------------------- - * | Level | Last Value Used | - * ----------------------------------------------------- - * | Module Init and Probe | 0x0116 | - * | Mailbox commands | 0x111e | - * | Device Discovery | 0x2083 | - * | Queue Command and IO tracing | 0x302e | - * | DPC Thread | 0x401c | - * | Async Events | 0x5059 | - * | Timer Routines | 0x600d | - * | User Space Interactions | 0x709c | - * | Task Management | 0x8043 | - * | AER/EEH | 0x900f | - * | Virtual Port | 0xa007 | - * | ISP82XX Specific | 0xb027 | - * | MultiQ | 0xc00b | - * | Misc | 0xd00b | - * ----------------------------------------------------- + * ---------------------------------------------------------------------- + * | Level | Last Value Used | Holes | + * ---------------------------------------------------------------------- + * | Module Init and Probe | 0x0116 | | + * | Mailbox commands | 0x1126 | | + * | Device Discovery | 0x2083 | | + * | Queue Command and IO tracing | 0x302e | 0x3008 | + * | DPC Thread | 0x401c | | + * | Async Events | 0x5059 | | + * | Timer Routines | 0x600d | | + * | User Space Interactions | 0x709d | | + * | Task Management | 0x8041 | | + * | AER/EEH | 0x900f | | + * | Virtual Port | 0xa007 | | + * | ISP82XX Specific | 0xb04f | | + * | MultiQ | 0xc00b | | + * | Misc | 0xd00b | | + * ---------------------------------------------------------------------- */ #include "qla_def.h" diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index cc5a79259d33..a03eaf40f377 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -2529,6 +2529,7 @@ struct qla_hw_data { #define DT_ISP8021 BIT_14 #define DT_ISP_LAST (DT_ISP8021 << 1) +#define DT_T10_PI BIT_25 #define DT_IIDMA BIT_26 #define DT_FWI2 BIT_27 #define DT_ZIO_SUPPORTED BIT_28 @@ -2572,6 +2573,7 @@ struct qla_hw_data { #define IS_NOCACHE_VPD_TYPE(ha) (IS_QLA81XX(ha)) #define IS_ALOGIO_CAPABLE(ha) (IS_QLA23XX(ha) || IS_FWI2_CAPABLE(ha)) +#define IS_T10_PI_CAPABLE(ha) ((ha)->device_type & DT_T10_PI) #define IS_IIDMA_CAPABLE(ha) ((ha)->device_type & DT_IIDMA) #define IS_FWI2_CAPABLE(ha) ((ha)->device_type & DT_FWI2) #define IS_ZIO_SUPPORTED(ha) ((ha)->device_type & DT_ZIO_SUPPORTED) diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h index c06e5f9b431e..9902834e0b74 100644 --- a/drivers/scsi/qla2xxx/qla_inline.h +++ b/drivers/scsi/qla2xxx/qla_inline.h @@ -104,9 +104,17 @@ qla2x00_set_fcport_state(fc_port_t *fcport, int state) } static inline int -qla2x00_hba_err_chk_enabled(unsigned char op) +qla2x00_hba_err_chk_enabled(srb_t *sp) { - switch (op) { + /* + * Uncomment when corresponding SCSI changes are done. + * + if (!sp->cmd->prot_chk) + return 0; + * + */ + + switch (scsi_get_prot_op(sp->cmd)) { case SCSI_PROT_READ_STRIP: case SCSI_PROT_WRITE_INSERT: if (ql2xenablehba_err_chk >= 1) diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c index 09ad3ce60064..dbec89622a0f 100644 --- a/drivers/scsi/qla2xxx/qla_iocb.c +++ b/drivers/scsi/qla2xxx/qla_iocb.c @@ -709,12 +709,11 @@ struct fw_dif_context { * */ static inline void -qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, +qla24xx_set_t10dif_tags(srb_t *sp, struct fw_dif_context *pkt, unsigned int protcnt) { - struct sd_dif_tuple *spt; + struct scsi_cmnd *cmd = sp->cmd; scsi_qla_host_t *vha = shost_priv(cmd->device->host); - unsigned char op = scsi_get_prot_op(cmd); switch (scsi_get_prot_type(cmd)) { case SCSI_PROT_DIF_TYPE0: @@ -724,6 +723,10 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, */ pkt->ref_tag = cpu_to_le32((uint32_t) (0xffffffff & scsi_get_lba(cmd))); + + if (!qla2x00_hba_err_chk_enabled(sp)) + break; + pkt->ref_tag_mask[0] = 0xff; pkt->ref_tag_mask[1] = 0xff; pkt->ref_tag_mask[2] = 0xff; @@ -735,20 +738,16 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, * match LBA in CDB + N */ case SCSI_PROT_DIF_TYPE2: - if (!qla2x00_hba_err_chk_enabled(op)) - break; - - if (scsi_prot_sg_count(cmd)) { - spt = page_address(sg_page(scsi_prot_sglist(cmd))) + - scsi_prot_sglist(cmd)[0].offset; - pkt->app_tag = swab32(spt->app_tag); - pkt->app_tag_mask[0] = 0xff; - pkt->app_tag_mask[1] = 0xff; - } + pkt->app_tag = __constant_cpu_to_le16(0); + pkt->app_tag_mask[0] = 0x0; + pkt->app_tag_mask[1] = 0x0; pkt->ref_tag = cpu_to_le32((uint32_t) (0xffffffff & scsi_get_lba(cmd))); + if (!qla2x00_hba_err_chk_enabled(sp)) + break; + /* enable ALL bytes of the ref tag */ pkt->ref_tag_mask[0] = 0xff; pkt->ref_tag_mask[1] = 0xff; @@ -768,26 +767,15 @@ qla24xx_set_t10dif_tags(struct scsi_cmnd *cmd, struct fw_dif_context *pkt, * 16 bit app tag. */ case SCSI_PROT_DIF_TYPE1: - if (!qla2x00_hba_err_chk_enabled(op)) + pkt->ref_tag = cpu_to_le32((uint32_t) + (0xffffffff & scsi_get_lba(cmd))); + pkt->app_tag = __constant_cpu_to_le16(0); + pkt->app_tag_mask[0] = 0x0; + pkt->app_tag_mask[1] = 0x0; + + if (!qla2x00_hba_err_chk_enabled(sp)) break; - if (protcnt && (op == SCSI_PROT_WRITE_STRIP || - op == SCSI_PROT_WRITE_PASS)) { - spt = page_address(sg_page(scsi_prot_sglist(cmd))) + - scsi_prot_sglist(cmd)[0].offset; - ql_dbg(ql_dbg_io, vha, 0x3008, - "LBA from user %p, lba = 0x%x for cmd=%p.\n", - spt, (int)spt->ref_tag, cmd); - pkt->ref_tag = swab32(spt->ref_tag); - pkt->app_tag_mask[0] = 0x0; - pkt->app_tag_mask[1] = 0x0; - } else { - pkt->ref_tag = cpu_to_le32((uint32_t) - (0xffffffff & scsi_get_lba(cmd))); - pkt->app_tag = __constant_cpu_to_le16(0); - pkt->app_tag_mask[0] = 0x0; - pkt->app_tag_mask[1] = 0x0; - } /* enable ALL bytes of the ref tag */ pkt->ref_tag_mask[0] = 0xff; pkt->ref_tag_mask[1] = 0xff; @@ -1208,7 +1196,7 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, INIT_LIST_HEAD(&crc_ctx_pkt->dsd_list); - qla24xx_set_t10dif_tags(cmd, (struct fw_dif_context *) + qla24xx_set_t10dif_tags(sp, (struct fw_dif_context *) &crc_ctx_pkt->ref_tag, tot_prot_dsds); cmd_pkt->crc_context_address[0] = cpu_to_le32(LSD(crc_ctx_dma)); @@ -1237,7 +1225,6 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, fcp_cmnd->additional_cdb_len |= 2; int_to_scsilun(sp->cmd->device->lun, &fcp_cmnd->lun); - host_to_fcp_swap((uint8_t *)&fcp_cmnd->lun, sizeof(fcp_cmnd->lun)); memcpy(fcp_cmnd->cdb, cmd->cmnd, cmd->cmd_len); cmd_pkt->fcp_cmnd_dseg_len = cpu_to_le16(fcp_cmnd_len); cmd_pkt->fcp_cmnd_dseg_address[0] = cpu_to_le32( @@ -1289,7 +1276,7 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, BUG(); } - if (!qla2x00_hba_err_chk_enabled(scsi_get_prot_op(cmd))) + if (!qla2x00_hba_err_chk_enabled(sp)) fw_prot_opts |= 0x10; /* Disable Guard tag checking */ if (!bundling) { diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 53339f10a598..ec53e87781a5 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -1534,6 +1534,16 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) return 1; } + /* check ref tag */ + if (e_ref_tag != a_ref_tag) { + scsi_build_sense_buffer(1, cmd->sense_buffer, ILLEGAL_REQUEST, + 0x10, 0x3); + set_driver_byte(cmd, DRIVER_SENSE); + set_host_byte(cmd, DID_ABORT); + cmd->result |= SAM_STAT_CHECK_CONDITION << 1; + return 1; + } + /* check appl tag */ if (e_app_tag != a_app_tag) { scsi_build_sense_buffer(1, cmd->sense_buffer, ILLEGAL_REQUEST, @@ -1544,15 +1554,6 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) return 1; } - /* check ref tag */ - if (e_ref_tag != a_ref_tag) { - scsi_build_sense_buffer(1, cmd->sense_buffer, ILLEGAL_REQUEST, - 0x10, 0x3); - set_driver_byte(cmd, DRIVER_SENSE); - set_host_byte(cmd, DID_ABORT); - cmd->result |= SAM_STAT_CHECK_CONDITION << 1; - return 1; - } return 1; } diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c index c706ed370000..f488cc69fc79 100644 --- a/drivers/scsi/qla2xxx/qla_mid.c +++ b/drivers/scsi/qla2xxx/qla_mid.c @@ -472,7 +472,7 @@ qla24xx_create_vhost(struct fc_vport *fc_vport) host->can_queue = base_vha->req->length + 128; host->this_id = 255; host->cmd_per_lun = 3; - if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) + if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) host->max_cmd_len = 32; else host->max_cmd_len = MAX_CMDSZ; diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index d65a3005b439..f57c292845a5 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -2255,7 +2255,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) host->this_id = 255; host->cmd_per_lun = 3; host->unique_id = host->host_no; - if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) + if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) host->max_cmd_len = 32; else host->max_cmd_len = MAX_CMDSZ; @@ -2382,7 +2382,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) "Detected hba at address=%p.\n", ha); - if ((IS_QLA25XX(ha) || IS_QLA81XX(ha)) && ql2xenabledif) { + if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { int prot = 0; base_vha->flags.difdix_supported = 1; From 42cd4f5dc2a3de31bfd24642ab4e8b21834a6b78 Mon Sep 17 00:00:00 2001 From: Chad Dupuis Date: Tue, 16 Aug 2011 11:29:24 -0700 Subject: [PATCH 058/326] [SCSI] qla2xxx: Fix qla24xx revision check while enabling interrupts. Since we enable interrupts before initializing the firmware, use the chip revision from PCI config space directly to perform the chip revision check. Also remove the unnecessary firmware attributes test. Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_isr.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index ec53e87781a5..477767fcfd1e 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -2529,11 +2529,10 @@ qla2x00_request_irqs(struct qla_hw_data *ha, struct rsp_que *rsp) goto skip_msi; } - if (IS_QLA2432(ha) && (ha->pdev->revision < QLA_MSIX_CHIP_REV_24XX || - !QLA_MSIX_FW_MODE_1(ha->fw_attributes))) { + if (IS_QLA2432(ha) && (ha->pdev->revision < QLA_MSIX_CHIP_REV_24XX)) { ql_log(ql_log_warn, vha, 0x0035, "MSI-X; Unsupported ISP2432 (0x%X, 0x%X).\n", - ha->pdev->revision, ha->fw_attributes); + ha->pdev->revision, QLA_MSIX_CHIP_REV_24XX); goto skip_msix; } From 7594206493880007fd68a18d6e9f380a1afe20d4 Mon Sep 17 00:00:00 2001 From: Saurav Kashyap Date: Tue, 16 Aug 2011 11:29:25 -0700 Subject: [PATCH 059/326] [SCSI] qla2xxx: Acquire hardware lock while manipulating dsd list. The dsd list shouldn't be manipulated without taking the per host hardware lock to prevent multiple callers from trampling upon one another. Signed-off-by: Saurav Kashyap Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_os.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index f57c292845a5..2caab83c4c9f 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -913,7 +913,10 @@ qla2xxx_eh_abort(struct scsi_cmnd *cmd) "Abort command mbx success.\n"); wait = 1; } + + spin_lock_irqsave(&ha->hardware_lock, flags); qla2x00_sp_compl(ha, sp); + spin_unlock_irqrestore(&ha->hardware_lock, flags); /* Wait for the command to be returned. */ if (wait) { From bc91ade9b7bc274d625c9b24c04d365a2daf481e Mon Sep 17 00:00:00 2001 From: Chad Dupuis Date: Tue, 16 Aug 2011 11:29:26 -0700 Subject: [PATCH 060/326] [SCSI] qla2xxx: Double check for command completion if abort mailbox command fails. Close a small window where we could falsely fail an abort request if the mailbox command fails but the command was returned during interrupt context. Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_os.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 2caab83c4c9f..4cace3f20c04 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -918,6 +918,10 @@ qla2xxx_eh_abort(struct scsi_cmnd *cmd) qla2x00_sp_compl(ha, sp); spin_unlock_irqrestore(&ha->hardware_lock, flags); + /* Did the command return during mailbox execution? */ + if (ret == FAILED && !CMD_SP(cmd)) + ret = SUCCESS; + /* Wait for the command to be returned. */ if (wait) { if (qla2x00_eh_wait_on_command(cmd) != QLA_SUCCESS) { From 3553d343e7acc418988cb8f22cd5b4976e7b484a Mon Sep 17 00:00:00 2001 From: Saurav Kashyap Date: Tue, 16 Aug 2011 11:29:27 -0700 Subject: [PATCH 061/326] [SCSI] qla2xxx: Save and restore irq in the response queue interrupt handler. Signed-off-by: Saurav Kashyap Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_nx.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index 5cbf33a50b14..02704fe8afab 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -2208,6 +2208,7 @@ qla82xx_msix_rsp_q(int irq, void *dev_id) struct qla_hw_data *ha; struct rsp_que *rsp; struct device_reg_82xx __iomem *reg; + unsigned long flags; rsp = (struct rsp_que *) dev_id; if (!rsp) { @@ -2218,11 +2219,11 @@ qla82xx_msix_rsp_q(int irq, void *dev_id) ha = rsp->hw; reg = &ha->iobase->isp82; - spin_lock_irq(&ha->hardware_lock); + spin_lock_irqsave(&ha->hardware_lock, flags); vha = pci_get_drvdata(ha->pdev); qla24xx_process_response_queue(vha, rsp); WRT_REG_DWORD(®->host_int, 0); - spin_unlock_irq(&ha->hardware_lock); + spin_unlock_irqrestore(&ha->hardware_lock, flags); return IRQ_HANDLED; } From 58b48576966ed0afd3f63ef17480ec12748a7119 Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Tue, 16 Aug 2011 11:29:28 -0700 Subject: [PATCH 062/326] [SCSI] qla2xxx: Correct inadvertent loop state transitions during port-update handling. Transitioning to a LOOP_UPDATE loop-state could cause the driver to miss normal link/target processing. LOOP_UPDATE is a crufty artifact leftover from at time the driver performed it's own internal command-queuing. Safely remove this state. Signed-off-by: Andrew Vasquez Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_init.c | 3 --- drivers/scsi/qla2xxx/qla_isr.c | 1 - 2 files changed, 4 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index def694271bf7..37da04d3db26 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -3838,15 +3838,12 @@ qla2x00_loop_resync(scsi_qla_host_t *vha) req = vha->req; rsp = req->rsp; - atomic_set(&vha->loop_state, LOOP_UPDATE); clear_bit(ISP_ABORT_RETRY, &vha->dpc_flags); if (vha->flags.online) { if (!(rval = qla2x00_fw_ready(vha))) { /* Wait at most MAX_TARGET RSCNs for a stable link. */ wait_time = 256; do { - atomic_set(&vha->loop_state, LOOP_UPDATE); - /* Issue a marker after FW becomes ready. */ qla2x00_marker(vha, req, rsp, 0, 0, MK_SYNC_ALL); diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 477767fcfd1e..646fc5263d50 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -719,7 +719,6 @@ qla2x00_async_event(scsi_qla_host_t *vha, struct rsp_que *rsp, uint16_t *mb) vha->flags.rscn_queue_overflow = 1; } - atomic_set(&vha->loop_state, LOOP_UPDATE); atomic_set(&vha->loop_down_timer, 0); vha->flags.management_server_logged_in = 0; From 51cc9a8e5f610a0d0881b45410c37890e02a2f76 Mon Sep 17 00:00:00 2001 From: Saurav Kashyap Date: Tue, 16 Aug 2011 11:29:29 -0700 Subject: [PATCH 063/326] [SCSI] qla2xxx: Set the task attributes after memsetting fcp cmnd. The memset of the fcp_cmnd struct needs to be moved so that it will not zero-out valid data. Signed-off-by: Saurav Kashyap Signed-off-by: Chad Dupuis Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_nx.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index 02704fe8afab..049807cda419 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -2839,6 +2839,16 @@ qla82xx_start_scsi(srb_t *sp) int_to_scsilun(sp->cmd->device->lun, &cmd_pkt->lun); host_to_fcp_swap((uint8_t *)&cmd_pkt->lun, sizeof(cmd_pkt->lun)); + /* build FCP_CMND IU */ + memset(ctx->fcp_cmnd, 0, sizeof(struct fcp_cmnd)); + int_to_scsilun(sp->cmd->device->lun, &ctx->fcp_cmnd->lun); + ctx->fcp_cmnd->additional_cdb_len = additional_cdb_len; + + if (cmd->sc_data_direction == DMA_TO_DEVICE) + ctx->fcp_cmnd->additional_cdb_len |= 1; + else if (cmd->sc_data_direction == DMA_FROM_DEVICE) + ctx->fcp_cmnd->additional_cdb_len |= 2; + /* * Update tagged queuing modifier -- default is TSK_SIMPLE (0). */ @@ -2855,16 +2865,6 @@ qla82xx_start_scsi(srb_t *sp) } } - /* build FCP_CMND IU */ - memset(ctx->fcp_cmnd, 0, sizeof(struct fcp_cmnd)); - int_to_scsilun(sp->cmd->device->lun, &ctx->fcp_cmnd->lun); - ctx->fcp_cmnd->additional_cdb_len = additional_cdb_len; - - if (cmd->sc_data_direction == DMA_TO_DEVICE) - ctx->fcp_cmnd->additional_cdb_len |= 1; - else if (cmd->sc_data_direction == DMA_FROM_DEVICE) - ctx->fcp_cmnd->additional_cdb_len |= 2; - memcpy(ctx->fcp_cmnd->cdb, cmd->cmnd, cmd->cmd_len); fcp_dl = (uint32_t *)(ctx->fcp_cmnd->cdb + 16 + From 7ca3c803e85080afdff4097e60fefec865027809 Mon Sep 17 00:00:00 2001 From: Chad Dupuis Date: Tue, 16 Aug 2011 11:29:30 -0700 Subject: [PATCH 064/326] [SCSI] qla2xxx: Update version number to 8.03.07.07-k. Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h index 062c97bf62f5..13b6357c1fa2 100644 --- a/drivers/scsi/qla2xxx/qla_version.h +++ b/drivers/scsi/qla2xxx/qla_version.h @@ -7,7 +7,7 @@ /* * Driver version */ -#define QLA2XXX_VERSION "8.03.07.03-k" +#define QLA2XXX_VERSION "8.03.07.07-k" #define QLA_DRIVER_MAJOR_VER 8 #define QLA_DRIVER_MINOR_VER 3 From 66506f761772c87fd4ff31b94b298888d5d58d77 Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Mon, 15 Aug 2011 10:28:18 +0800 Subject: [PATCH 065/326] mmc: sdhci-esdhc-imx: add missing inclusion of linux/module.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are the following warnings and errorx when compiling the driver. The patch adds the missing inclusion of linux/module.h to fix them. drivers/mmc/host/sdhci-esdhc-imx.c:563:12: error: ‘THIS_MODULE’ undeclared here (not in a function) [..] Signed-off-by: Shawn Guo Signed-off-by: Chris Ball --- drivers/mmc/host/sdhci-esdhc-imx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index 0e9780f5a4a9..4dc0028086a3 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include From 848e7d5b46b9b0ee613a106bc460acf6a09a8546 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Thu, 25 Aug 2011 12:40:47 -0700 Subject: [PATCH 066/326] [SCSI] fcoe: Fix deadlock between fip's recv_work and rtnl The rtnl cannot be held durrng the fcoe_interface_put. If it is the last reference on the fcoe_interface the fcoe_ctlr_destroy will be called as a part of the cleanup, ultimately calling cancel_work_sync(&fip->recv_work); If we are processing a flogi response we will be in the recv_work context and we will lock the rtnl to add a new unicast MAC address. This is how the deadlock can occur. The fix is simply to move the rtnl_lock/unlock into fcoe_interface_cleanup so that it can be unlocked before fcoe_interface_put is called. Here is the lockdep report: Jul 21 11:26:35 bubba [ 223.870702] ul 21 11:26:35 bubba [ 223.870704] ======================================================= Jul 21 11:26:35 bubba [ 223.871255] [ INFO: possible circular locking dependency detected ] Jul 21 11:26:35 bubba [ 223.871530] 3.0.0-rc7+ #1 Jul 21 11:26:35 bubba [ 223.871797] ------------------------------------------------------- Jul 21 11:26:35 bubba [ 223.872072] lockdeptest.sh/3464 is trying to acquire lock: Jul 21 11:26:35 bubba [ 223.872345] ((&fip->recv_work) Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] wait_on_work+0x0/0xbd Jul 21 11:26:35 bubba [ 223.873022] Jul 21 11:26:35 bubba [ 223.873023] but task is already holding lock: Jul 21 11:26:35 bubba [ 223.873555] (rtnl_mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] rtnl_lock+0x12/0x14 Jul 21 11:26:35 bubba [ 223.874229] Jul 21 11:26:35 bubba [ 223.874230] which lock already depends on the new lock. Jul 21 11:26:35 bubba [ 223.874231] Jul 21 11:26:35 bubba [ 223.875032] Jul 21 11:26:35 bubba [ 223.875033] the existing dependency chain (in reverse order) is: Jul 21 11:26:35 bubba [ 223.875573] Jul 21 11:26:35 bubba [ 223.875573] -> #1 Jul 21 11:26:35 bubba (rtnl_mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba : Jul 21 11:26:35 bubba [ 223.876301] Jul 21 11:26:35 bubba [] lock_acquire+0xd2/0xf7 Jul 21 11:26:35 bubba [ 223.876645] Jul 21 11:26:35 bubba [] __mutex_lock_common+0x47/0x30d Jul 21 11:26:35 bubba [ 223.876991] Jul 21 11:26:35 bubba [] mutex_lock_nested+0x3b/0x40 Jul 21 11:26:35 bubba [ 223.877334] Jul 21 11:26:35 bubba [] rtnl_lock+0x12/0x14 Jul 21 11:26:35 bubba [ 223.877675] Jul 21 11:26:35 bubba [] fcoe_update_src_mac+0x2b/0x80 [fcoe] Jul 21 11:26:35 bubba [ 223.878022] Jul 21 11:26:35 bubba [] fcoe_flogi_resp+0x5e/0x79 [fcoe] Jul 21 11:26:35 bubba [ 223.878366] Jul 21 11:26:35 bubba [] fc_exch_recv+0x7f5/0x9da [libfc] Jul 21 11:26:35 bubba [ 223.878713] Jul 21 11:26:35 bubba [] fcoe_ctlr_recv_work+0x71f/0x10dc [libfcoe] Jul 21 11:26:35 bubba [ 223.879258] Jul 21 11:26:35 bubba [] process_one_work+0x1d7/0x347 Jul 21 11:26:35 bubba [ 223.879601] Jul 21 11:26:35 bubba [] worker_thread+0xf8/0x17c Jul 21 11:26:35 bubba [ 223.879944] Jul 21 11:26:35 bubba [] kthread+0x7d/0x85 Jul 21 11:26:35 bubba [ 223.880287] Jul 21 11:26:35 bubba [] kernel_thread_helper+0x4/0x10 Jul 21 11:26:35 bubba [ 223.880634] Jul 21 11:26:35 bubba [ 223.880635] -> #0 Jul 21 11:26:35 bubba ((&fip->recv_work) Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba : Jul 21 11:26:35 bubba [ 223.881357] Jul 21 11:26:35 bubba [] __lock_acquire+0xb1d/0xe2c Jul 21 11:26:35 bubba [ 223.881695] Jul 21 11:26:35 bubba [] lock_acquire+0xd2/0xf7 Jul 21 11:26:35 bubba [ 223.882033] Jul 21 11:26:35 bubba [] wait_on_work+0x50/0xbd Jul 21 11:26:35 bubba [ 223.882378] Jul 21 11:26:35 bubba [] __cancel_work_timer+0xb6/0xf4 Jul 21 11:26:35 bubba [ 223.882718] Jul 21 11:26:35 bubba [] cancel_work_sync+0xb/0xd Jul 21 11:26:35 bubba [ 223.883057] Jul 21 11:26:35 bubba [] fcoe_ctlr_destroy+0x1d/0x67 [libfcoe] Jul 21 11:26:35 bubba [ 223.883399] Jul 21 11:26:35 bubba [] fcoe_interface_release+0x21/0x45 [fcoe] Jul 21 11:26:35 bubba [ 223.883940] Jul 21 11:26:35 bubba [] kref_put+0x43/0x4d Jul 21 11:26:35 bubba [ 223.884280] Jul 21 11:26:35 bubba [] fcoe_interface_put+0x17/0x19 [fcoe] Jul 21 11:26:35 bubba [ 223.884624] Jul 21 11:26:35 bubba [] fcoe_interface_cleanup+0x188/0x193 [fcoe] Jul 21 11:26:35 bubba [ 223.885163] Jul 21 11:26:35 bubba [] fcoe_destroy+0x52/0x72 [fcoe] Jul 21 11:26:35 bubba [ 223.885502] Jul 21 11:26:35 bubba [] fcoe_transport_destroy+0xab/0x110 [libfcoe] Jul 21 11:26:35 bubba [ 223.886045] Jul 21 11:26:35 bubba [] param_attr_store+0x43/0x62 Jul 21 11:26:35 bubba [ 223.886385] Jul 21 11:26:35 bubba [] module_attr_store+0x21/0x25 Jul 21 11:26:35 bubba [ 223.886728] Jul 21 11:26:35 bubba [] sysfs_write_file+0x103/0x13f Jul 21 11:26:35 bubba [ 223.887068] Jul 21 11:26:35 bubba [] vfs_write+0xa7/0xfa Jul 21 11:26:35 bubba [ 223.887406] Jul 21 11:26:35 bubba [] sys_write+0x45/0x69 Jul 21 11:26:35 bubba [ 223.887742] Jul 21 11:26:35 bubba [] system_call_fastpath+0x16/0x1b Jul 21 11:26:35 bubba [ 223.888083] Jul 21 11:26:35 bubba [ 223.888084] other info that might help us debug this: Jul 21 11:26:35 bubba [ 223.888085] Jul 21 11:26:35 bubba [ 223.888879] Possible unsafe locking scenario: Jul 21 11:26:35 bubba [ 223.888881] Jul 21 11:26:35 bubba [ 223.889411] CPU0 CPU1 Jul 21 11:26:35 bubba [ 223.889683] ---- ---- Jul 21 11:26:35 bubba [ 223.889955] lock( Jul 21 11:26:35 bubba rtnl_mutex Jul 21 11:26:35 bubba ); Jul 21 11:26:35 bubba [ 223.890349] lock( Jul 21 11:26:35 bubba (&fip->recv_work) Jul 21 11:26:35 bubba ); Jul 21 11:26:35 bubba [ 223.890751] lock( Jul 21 11:26:35 bubba rtnl_mutex Jul 21 11:26:35 bubba ); Jul 21 11:26:35 bubba [ 223.891154] lock( Jul 21 11:26:35 bubba (&fip->recv_work) Jul 21 11:26:35 bubba ); Jul 21 11:26:35 bubba [ 223.891549] Jul 21 11:26:35 bubba [ 223.891550] *** DEADLOCK *** Jul 21 11:26:35 bubba [ 223.891551] Jul 21 11:26:35 bubba [ 223.892347] 6 locks held by lockdeptest.sh/3464: Jul 21 11:26:35 bubba [ 223.892621] #0: Jul 21 11:26:35 bubba (&buffer->mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] sysfs_write_file+0x37/0x13f Jul 21 11:26:35 bubba [ 223.893359] #1: Jul 21 11:26:35 bubba (s_active Jul 21 11:26:35 bubba ){++++.+} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] sysfs_write_file+0xe2/0x13f Jul 21 11:26:35 bubba [ 223.894094] #2: Jul 21 11:26:35 bubba (param_lock Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] param_attr_store+0x36/0x62 Jul 21 11:26:35 bubba [ 223.894835] #3: Jul 21 11:26:35 bubba (ft_mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] fcoe_transport_destroy+0x1e/0x110 [libfcoe] Jul 21 11:26:35 bubba [ 223.895574] #4: Jul 21 11:26:35 bubba (fcoe_config_mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] fcoe_destroy+0x18/0x72 [fcoe] Jul 21 11:26:35 bubba [ 223.896314] #5: Jul 21 11:26:35 bubba (rtnl_mutex Jul 21 11:26:35 bubba ){+.+.+.} Jul 21 11:26:35 bubba , at: Jul 21 11:26:35 bubba [] rtnl_lock+0x12/0x14 Jul 21 11:26:35 bubba [ 223.897047] Jul 21 11:26:35 bubba [ 223.897048] stack backtrace: Jul 21 11:26:35 bubba [ 223.897578] Pid: 3464, comm: lockdeptest.sh Not tainted 3.0.0-rc7+ #1 Jul 21 11:26:35 bubba [ 223.897853] Call Trace: Jul 21 11:26:35 bubba [ 223.898128] [] print_circular_bug+0x1f8/0x209 Jul 21 11:26:35 bubba [ 223.898416] [] __lock_acquire+0xb1d/0xe2c Jul 21 11:26:35 bubba [ 223.898699] [] ? wait_on_cpu_work+0xe6/0xe6 Jul 21 11:26:35 bubba [ 223.898982] [] lock_acquire+0xd2/0xf7 Jul 21 11:26:35 bubba [ 223.899263] [] ? wait_on_cpu_work+0xe6/0xe6 Jul 21 11:26:35 bubba [ 223.899547] [] ? mod_timer+0x8f/0x98 Jul 21 11:26:35 bubba [ 223.899827] [] wait_on_work+0x50/0xbd Jul 21 11:26:35 bubba [ 223.900108] [] ? wait_on_cpu_work+0xe6/0xe6 Jul 21 11:26:35 bubba [ 223.900390] [] __cancel_work_timer+0xb6/0xf4 Jul 21 11:26:35 bubba [ 223.900671] [] cancel_work_sync+0xb/0xd Jul 21 11:26:35 bubba [ 223.900953] [] fcoe_ctlr_destroy+0x1d/0x67 [libfcoe] Jul 21 11:26:35 bubba [ 223.901237] [] fcoe_interface_release+0x21/0x45 [fcoe] Jul 21 11:26:35 bubba [ 223.901522] [] ? fcoe_enable+0x6b/0x6b [fcoe] Jul 21 11:26:35 bubba [ 223.901803] [] kref_put+0x43/0x4d Jul 21 11:26:35 bubba [ 223.902083] [] fcoe_interface_put+0x17/0x19 [fcoe] Jul 21 11:26:35 bubba [ 223.902367] [] fcoe_interface_cleanup+0x188/0x193 [fcoe] Jul 21 11:26:35 bubba [ 223.902653] [] ? mutex_lock_nested+0x3b/0x40 Jul 21 11:26:35 bubba [ 223.902939] [] fcoe_destroy+0x52/0x72 [fcoe] Jul 21 11:26:35 bubba [ 223.903223] [] fcoe_transport_destroy+0xab/0x110 [libfcoe] Jul 21 11:26:35 bubba [ 223.903508] [] param_attr_store+0x43/0x62 Jul 21 11:26:35 bubba [ 223.903792] [] module_attr_store+0x21/0x25 Jul 21 11:26:35 bubba [ 223.904075] [] sysfs_write_file+0x103/0x13f Jul 21 11:26:35 bubba [ 223.904357] [] vfs_write+0xa7/0xfa Jul 21 11:26:35 bubba [ 223.904642] [] ? fget_light+0x35/0x96 Jul 21 11:26:35 bubba [ 223.904923] [] sys_write+0x45/0x69 Jul 21 11:26:35 bubba [ 223.905204] [] system_call_fastpath+0x16/0x1b Jul 21 11:26:36 bubba [ 223.964438] ixgbe 0000:05:00.0: eth3: detected SFP+: 5 Jul 21 11:26:37 bubba [ 225.196702] ixgbe 0000:05:00.0: eth3: NIC Link is Up 10 Gbps, Flow Control: None Signed-off-by: Robert Love Tested-by: Ross Brattain Reviewed-by: Yi Zou Signed-off-by: James Bottomley --- drivers/scsi/fcoe/fcoe.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c index ba710e350ac5..5d0e9a24ae94 100644 --- a/drivers/scsi/fcoe/fcoe.c +++ b/drivers/scsi/fcoe/fcoe.c @@ -432,6 +432,8 @@ void fcoe_interface_cleanup(struct fcoe_interface *fcoe) u8 flogi_maddr[ETH_ALEN]; const struct net_device_ops *ops; + rtnl_lock(); + /* * Don't listen for Ethernet packets anymore. * synchronize_net() ensures that the packet handlers are not running @@ -461,6 +463,8 @@ void fcoe_interface_cleanup(struct fcoe_interface *fcoe) " specific feature for LLD.\n"); } + rtnl_unlock(); + /* Release the self-reference taken during fcoe_interface_create() */ fcoe_interface_put(fcoe); } @@ -1951,11 +1955,8 @@ static void fcoe_destroy_work(struct work_struct *work) fcoe_if_destroy(port->lport); /* Do not tear down the fcoe interface for NPIV port */ - if (!npiv) { - rtnl_lock(); + if (!npiv) fcoe_interface_cleanup(fcoe); - rtnl_unlock(); - } mutex_unlock(&fcoe_config_mutex); } @@ -2009,8 +2010,9 @@ static int fcoe_create(struct net_device *netdev, enum fip_state fip_mode) printk(KERN_ERR "fcoe: Failed to create interface (%s)\n", netdev->name); rc = -EIO; + rtnl_unlock(); fcoe_interface_cleanup(fcoe); - goto out_nodev; + goto out_nortnl; } /* Make this the "master" N_Port */ @@ -2027,6 +2029,7 @@ static int fcoe_create(struct net_device *netdev, enum fip_state fip_mode) out_nodev: rtnl_unlock(); +out_nortnl: mutex_unlock(&fcoe_config_mutex); return rc; } From 77a2b73a7805a3c6a473b6741aa514ef40295d26 Mon Sep 17 00:00:00 2001 From: Vasu Dev Date: Thu, 25 Aug 2011 12:40:52 -0700 Subject: [PATCH 067/326] [SCSI] libfc: fix fc_eh_host_reset Current fc_eh_host_reset leaves lport offline permanently due to FLOGI response getting handled by LOGO response from last reset as both had same exchange id. So fix this by having end to end exches clean-up using exchange abort along exches reset done from fc_eh_host_reset. This would avoid exchanges collision between the sessions across the reset. In this case implicit login should have done that but no aborting support for FIP frames, so just wait till lport->r_a_tov before restarting next flogi to ensure all exchanges are good to use again for next session. Below is the trace of LOGO from older session coming ahead of FLOGI response with same exche id 0x203:- 617 86.435165 4e.00.0b -> ff.ff.fc FC ELS LOGO 0x203 618 86.435195 4e.00.0b -> b6.02.00 FC ELS LOGO 0x213 619 86.435220 4e.00.0b -> 18.03.00 FC ELS LOGO 0x223 620 86.435244 4e.00.0b -> 18.02.00 FC ELS LOGO 0x233 621 86.435267 4e.00.0b -> 18.01.00 FC ELS LOGO 0x243 622 86.435349 00.00.00 -> ff.ff.fe FC ELS FLOGI 0x203 623 86.435549 ff.ff.fc -> 4e.00.0b FC ELS ACC (LOGO) 0x203 624 86.438721 ff.ff.fe -> 4e.00.0b FC ELS ACC (FLOGI) 0x203 625 86.442059 18.03.00 -> 4e.00.0b FC ELS ACC (LOGO) 0x223 626 86.443683 b6.02.00 -> 4e.00.0b FC ELS ACC (LOGO) 0x213 627 86.447693 18.01.00 -> 4e.00.0b FC ELS ACC (LOGO) 0x243 628 86.453499 18.02.00 -> 4e.00.0b FC ELS ACC (LOGO) 0x233 Signed-off-by: Vasu Dev Tested-by: Ross Brattain Reviewed-by: Yi Zou Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_exch.c | 51 ++++++++++++++++++++++++----------- drivers/scsi/libfc/fc_lport.c | 11 +++++++- 2 files changed, 45 insertions(+), 17 deletions(-) diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c index 01ff082dc34c..744fefe81341 100644 --- a/drivers/scsi/libfc/fc_exch.c +++ b/drivers/scsi/libfc/fc_exch.c @@ -494,6 +494,9 @@ static int fc_seq_send(struct fc_lport *lport, struct fc_seq *sp, */ error = lport->tt.frame_send(lport, fp); + if (fh->fh_type == FC_TYPE_BLS) + return error; + /* * Update the exchange and sequence flags, * assuming all frames for the sequence have been sent. @@ -575,42 +578,35 @@ static void fc_seq_set_resp(struct fc_seq *sp, } /** - * fc_seq_exch_abort() - Abort an exchange and sequence - * @req_sp: The sequence to be aborted + * fc_exch_abort_locked() - Abort an exchange + * @ep: The exchange to be aborted * @timer_msec: The period of time to wait before aborting * - * Generally called because of a timeout or an abort from the upper layer. + * Locking notes: Called with exch lock held + * + * Return value: 0 on success else error code */ -static int fc_seq_exch_abort(const struct fc_seq *req_sp, - unsigned int timer_msec) +static int fc_exch_abort_locked(struct fc_exch *ep, + unsigned int timer_msec) { struct fc_seq *sp; - struct fc_exch *ep; struct fc_frame *fp; int error; - ep = fc_seq_exch(req_sp); - - spin_lock_bh(&ep->ex_lock); if (ep->esb_stat & (ESB_ST_COMPLETE | ESB_ST_ABNORMAL) || - ep->state & (FC_EX_DONE | FC_EX_RST_CLEANUP)) { - spin_unlock_bh(&ep->ex_lock); + ep->state & (FC_EX_DONE | FC_EX_RST_CLEANUP)) return -ENXIO; - } /* * Send the abort on a new sequence if possible. */ sp = fc_seq_start_next_locked(&ep->seq); - if (!sp) { - spin_unlock_bh(&ep->ex_lock); + if (!sp) return -ENOMEM; - } ep->esb_stat |= ESB_ST_SEQ_INIT | ESB_ST_ABNORMAL; if (timer_msec) fc_exch_timer_set_locked(ep, timer_msec); - spin_unlock_bh(&ep->ex_lock); /* * If not logged into the fabric, don't send ABTS but leave @@ -632,6 +628,28 @@ static int fc_seq_exch_abort(const struct fc_seq *req_sp, return error; } +/** + * fc_seq_exch_abort() - Abort an exchange and sequence + * @req_sp: The sequence to be aborted + * @timer_msec: The period of time to wait before aborting + * + * Generally called because of a timeout or an abort from the upper layer. + * + * Return value: 0 on success else error code + */ +static int fc_seq_exch_abort(const struct fc_seq *req_sp, + unsigned int timer_msec) +{ + struct fc_exch *ep; + int error; + + ep = fc_seq_exch(req_sp); + spin_lock_bh(&ep->ex_lock); + error = fc_exch_abort_locked(ep, timer_msec); + spin_unlock_bh(&ep->ex_lock); + return error; +} + /** * fc_exch_timeout() - Handle exchange timer expiration * @work: The work_struct identifying the exchange that timed out @@ -1715,6 +1733,7 @@ static void fc_exch_reset(struct fc_exch *ep) int rc = 1; spin_lock_bh(&ep->ex_lock); + fc_exch_abort_locked(ep, 0); ep->state |= FC_EX_RST_CLEANUP; if (cancel_delayed_work(&ep->timeout_work)) atomic_dec(&ep->ex_refcnt); /* drop hold for timer */ diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c index e55ed9cf23fb..628f347404f9 100644 --- a/drivers/scsi/libfc/fc_lport.c +++ b/drivers/scsi/libfc/fc_lport.c @@ -88,6 +88,7 @@ */ #include +#include #include #include @@ -1029,8 +1030,16 @@ static void fc_lport_enter_reset(struct fc_lport *lport) FCH_EVT_LIPRESET, 0); fc_vports_linkchange(lport); fc_lport_reset_locked(lport); - if (lport->link_up) + if (lport->link_up) { + /* + * Wait upto resource allocation time out before + * doing re-login since incomplete FIP exchanged + * from last session may collide with exchanges + * in new session. + */ + msleep(lport->r_a_tov); fc_lport_enter_flogi(lport); + } } /** From 21cc0bd3a9e524b44a4f0ff05ac612aa0ff1a26e Mon Sep 17 00:00:00 2001 From: Vasu Dev Date: Thu, 25 Aug 2011 12:40:57 -0700 Subject: [PATCH 068/326] [SCSI] libfc: block SCSI eh thread for blocked rports Call fc_block_scsi_eh() in all fcoe eh to blocks the scsi_eh thread for blocked rports. Signed-off-by: Vasu Dev Tested-by: Ross Brattain Reviewed-by: Yi Zou Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_fcp.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c index afb63c843144..4c41ee816f0b 100644 --- a/drivers/scsi/libfc/fc_fcp.c +++ b/drivers/scsi/libfc/fc_fcp.c @@ -2019,6 +2019,11 @@ int fc_eh_abort(struct scsi_cmnd *sc_cmd) struct fc_fcp_internal *si; int rc = FAILED; unsigned long flags; + int rval; + + rval = fc_block_scsi_eh(sc_cmd); + if (rval) + return rval; lport = shost_priv(sc_cmd->device->host); if (lport->state != LPORT_ST_READY) @@ -2068,9 +2073,9 @@ int fc_eh_device_reset(struct scsi_cmnd *sc_cmd) int rc = FAILED; int rval; - rval = fc_remote_port_chkready(rport); + rval = fc_block_scsi_eh(sc_cmd); if (rval) - goto out; + return rval; lport = shost_priv(sc_cmd->device->host); @@ -2116,6 +2121,8 @@ int fc_eh_host_reset(struct scsi_cmnd *sc_cmd) FC_SCSI_DBG(lport, "Resetting host\n"); + fc_block_scsi_eh(sc_cmd); + lport->tt.lport_reset(lport); wait_tmo = jiffies + FC_HOST_RESET_TIMEOUT; while (!fc_fcp_lport_queue_ready(lport) && time_before(jiffies, From 3ee17f59c5378af8d245f82498e3919b7de2ab40 Mon Sep 17 00:00:00 2001 From: Yi Zou Date: Thu, 25 Aug 2011 12:41:03 -0700 Subject: [PATCH 069/326] [SCSI] libfc: fix referencing to fc_fcp_pkt from the frame pointer via fr_fsp() In commit 6a716a8, while releasing the DDP context in case frame_send() failed, the frame may already be freed, so we should store the pointer to fc_fcp_pkt and release the DDP context using the locally stored fsp instead of getting fsp from the fr_fsp(fp) on a frame. Signed-off-by: Yi Zou Reported-by: Bhanu Prakash Gollapudi Tested-by: Ross Brattain Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_exch.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c index 744fefe81341..d261e982a2fa 100644 --- a/drivers/scsi/libfc/fc_exch.c +++ b/drivers/scsi/libfc/fc_exch.c @@ -1981,6 +1981,7 @@ static struct fc_seq *fc_exch_seq_send(struct fc_lport *lport, struct fc_exch *ep; struct fc_seq *sp = NULL; struct fc_frame_header *fh; + struct fc_fcp_pkt *fsp = NULL; int rc = 1; ep = fc_exch_alloc(lport, fp); @@ -2003,8 +2004,10 @@ static struct fc_seq *fc_exch_seq_send(struct fc_lport *lport, fc_exch_setup_hdr(ep, fp, ep->f_ctl); sp->cnt++; - if (ep->xid <= lport->lro_xid && fh->fh_r_ctl == FC_RCTL_DD_UNSOL_CMD) + if (ep->xid <= lport->lro_xid && fh->fh_r_ctl == FC_RCTL_DD_UNSOL_CMD) { + fsp = fr_fsp(fp); fc_fcp_ddp_setup(fr_fsp(fp), ep->xid); + } if (unlikely(lport->tt.frame_send(lport, fp))) goto err; @@ -2018,7 +2021,8 @@ static struct fc_seq *fc_exch_seq_send(struct fc_lport *lport, spin_unlock_bh(&ep->ex_lock); return sp; err: - fc_fcp_ddp_done(fr_fsp(fp)); + if (fsp) + fc_fcp_ddp_done(fsp); rc = fc_exch_done_locked(ep); spin_unlock_bh(&ep->ex_lock); if (!rc) From 610602f369b4c810c9df05e431abd38f38cb8e0d Mon Sep 17 00:00:00 2001 From: Eddie Wai Date: Fri, 26 Aug 2011 11:16:47 -0700 Subject: [PATCH 070/326] [SCSI] bnx2i: Fixed the endian on TTT for NOP out transmission The iscsi_nopout task's TTT is defined as __be32 while the DMA memory to the chip is CPU specific. This creates a problem for unsolicited NOP-In responses where the TTT is not the RESERVED tag of 0xFFs. This patch adds a call to be32_to_cpu for the TTT specified. Signed-off-by: Eddie Wai Signed-off-by: James Bottomley --- drivers/scsi/bnx2i/bnx2i_hwi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/bnx2i/bnx2i_hwi.c b/drivers/scsi/bnx2i/bnx2i_hwi.c index 9ae80cd5953b..dba72a4e6a1c 100644 --- a/drivers/scsi/bnx2i/bnx2i_hwi.c +++ b/drivers/scsi/bnx2i/bnx2i_hwi.c @@ -563,7 +563,7 @@ int bnx2i_send_iscsi_nopout(struct bnx2i_conn *bnx2i_conn, nopout_wqe->itt = ((u16)task->itt | (ISCSI_TASK_TYPE_MPATH << ISCSI_TMF_REQUEST_TYPE_SHIFT)); - nopout_wqe->ttt = nopout_hdr->ttt; + nopout_wqe->ttt = be32_to_cpu(nopout_hdr->ttt); nopout_wqe->flags = 0; if (!unsol) nopout_wqe->flags = ISCSI_NOP_OUT_REQUEST_LOCAL_COMPLETION; From c259e01a1ec90063042f758e409cd26b2a0963c8 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 22 Jun 2011 19:47:00 +0200 Subject: [PATCH 071/326] sched: Separate the scheduler entry for preemption Block-IO and workqueues call into notifier functions from the scheduler core code with interrupts and preemption disabled. These calls should be made before entering the scheduler core. To simplify this, separate the scheduler core code into __schedule(). __schedule() is directly called from the places which set PREEMPT_ACTIVE and from schedule(). This allows us to add the work checks into schedule(), so they are only called when a task voluntary goes to sleep. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Cc: Tejun Heo Cc: Jens Axboe Cc: Linus Torvalds Cc: stable@kernel.org # 2.6.39+ Link: http://lkml.kernel.org/r/20110622174918.813258321@linutronix.de Signed-off-by: Ingo Molnar --- kernel/sched.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..ec15e8129cf7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4279,9 +4279,9 @@ pick_next_task(struct rq *rq) } /* - * schedule() is the main scheduler function. + * __schedule() is the main scheduler function. */ -asmlinkage void __sched schedule(void) +static void __sched __schedule(void) { struct task_struct *prev, *next; unsigned long *switch_count; @@ -4369,6 +4369,11 @@ asmlinkage void __sched schedule(void) if (need_resched()) goto need_resched; } + +asmlinkage void schedule(void) +{ + __schedule(); +} EXPORT_SYMBOL(schedule); #ifdef CONFIG_MUTEX_SPIN_ON_OWNER @@ -4435,7 +4440,7 @@ asmlinkage void __sched notrace preempt_schedule(void) do { add_preempt_count_notrace(PREEMPT_ACTIVE); - schedule(); + __schedule(); sub_preempt_count_notrace(PREEMPT_ACTIVE); /* @@ -4463,7 +4468,7 @@ asmlinkage void __sched preempt_schedule_irq(void) do { add_preempt_count(PREEMPT_ACTIVE); local_irq_enable(); - schedule(); + __schedule(); local_irq_disable(); sub_preempt_count(PREEMPT_ACTIVE); @@ -5588,7 +5593,7 @@ static inline int should_resched(void) static void __cond_resched(void) { add_preempt_count(PREEMPT_ACTIVE); - schedule(); + __schedule(); sub_preempt_count(PREEMPT_ACTIVE); } From 9c40cef2b799f9b5e7fa5de4d2ad3a0168ba118c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 22 Jun 2011 19:47:01 +0200 Subject: [PATCH 072/326] sched: Move blk_schedule_flush_plug() out of __schedule() There is no real reason to run blk_schedule_flush_plug() with interrupts and preemption disabled. Move it into schedule() and call it when the task is going voluntarily to sleep. There might be false positives when the task is woken between that call and actually scheduling, but that's not really different from being woken immediately after switching away. This fixes a deadlock in the scheduler where the blk_schedule_flush_plug() callchain enables interrupts and thereby allows a wakeup to happen of the task that's going to sleep. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Cc: Tejun Heo Cc: Jens Axboe Cc: Linus Torvalds Cc: stable@kernel.org # 2.6.39+ Link: http://lkml.kernel.org/n/tip-dwfxtra7yg1b5r65m32ywtct@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index ec15e8129cf7..511732c39b6e 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4322,16 +4322,6 @@ static void __sched __schedule(void) if (to_wakeup) try_to_wake_up_local(to_wakeup); } - - /* - * If we are going to sleep and we have plugged IO - * queued, make sure to submit it to avoid deadlocks. - */ - if (blk_needs_flush_plug(prev)) { - raw_spin_unlock(&rq->lock); - blk_schedule_flush_plug(prev); - raw_spin_lock(&rq->lock); - } } switch_count = &prev->nvcsw; } @@ -4370,8 +4360,23 @@ static void __sched __schedule(void) goto need_resched; } +static inline void sched_submit_work(struct task_struct *tsk) +{ + if (!tsk->state) + return; + /* + * If we are going to sleep and we have plugged IO queued, + * make sure to submit it to avoid deadlocks. + */ + if (blk_needs_flush_plug(tsk)) + blk_schedule_flush_plug(tsk); +} + asmlinkage void schedule(void) { + struct task_struct *tsk = current; + + sched_submit_work(tsk); __schedule(); } EXPORT_SYMBOL(schedule); From feff8fa0075bdfd43c841e9d689ed81adda988d6 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Thu, 18 Aug 2011 20:36:57 +0800 Subject: [PATCH 073/326] sched: Fix a memory leak in __sdt_free() This patch fixes the following memory leak: unreferenced object 0xffff880107266800 (size 512): comm "sched-powersave", pid 3718, jiffies 4323097853 (age 27495.450s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] create_object+0x187/0x28b [] kmemleak_alloc+0x73/0x98 [] __kmalloc_node+0x104/0x159 [] kzalloc_node.clone.97+0x15/0x17 [] build_sched_domains+0xb7/0x7f3 [] partition_sched_domains+0x1db/0x24a [] do_rebuild_sched_domains+0x3b/0x47 [] rebuild_sched_domains+0x10/0x12 [] sched_power_savings_store+0x6c/0x7b [] sched_mc_power_savings_store+0x16/0x18 [] sysdev_class_store+0x20/0x22 [] sysfs_write_file+0x108/0x144 [] vfs_write+0xaf/0x102 [] sys_write+0x4d/0x74 [] system_call_fastpath+0x16/0x1b [] 0xffffffffffffffff Signed-off-by: WANG Cong Signed-off-by: Peter Zijlstra Cc: stable@kernel.org # 3.0 Link: http://lkml.kernel.org/r/1313671017-4112-1-git-send-email-amwang@redhat.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/sched.c b/kernel/sched.c index 511732c39b6e..c79e7c63a4aa 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7453,6 +7453,7 @@ static void __sdt_free(const struct cpumask *cpu_map) struct sched_domain *sd = *per_cpu_ptr(sdd->sd, j); if (sd && (sd->flags & SD_OVERLAP)) free_sched_groups(sd->groups, 0); + kfree(*per_cpu_ptr(sdd->sd, j)); kfree(*per_cpu_ptr(sdd->sg, j)); kfree(*per_cpu_ptr(sdd->sgp, j)); } From a8d757ef076f0f95f13a918808824058de25b3eb Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 25 Aug 2011 15:58:03 +0200 Subject: [PATCH 074/326] perf events: Fix slow and broken cgroup context switch code The current cgroup context switch code was incorrect leading to bogus counts. Furthermore, as soon as there was an active cgroup event on a CPU, the context switch cost on that CPU would increase by a significant amount as demonstrated by a simple ping/pong example: $ ./pong Both processes pinned to CPU1, running for 10s 10684.51 ctxsw/s Now start a cgroup perf stat: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 100 $ ./pong Both processes pinned to CPU1, running for 10s 6674.61 ctxsw/s That's a 37% penalty. Note that pong is not even in the monitored cgroup. The results shown by perf stat are bogus: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 100 Performance counter stats for 'sleep 100': CPU1 cycles test CPU1 16,984,189,138 cycles # 0.000 GHz The second 'cycles' event should report a count @ CPU clock (here 2.4GHz) as it is counting across all cgroups. The patch below fixes the bogus accounting and bypasses any cgroup switches in case the outgoing and incoming tasks are in the same cgroup. With this patch the same test now yields: $ ./pong Both processes pinned to CPU1, running for 10s 10775.30 ctxsw/s Start perf stat with cgroup: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Run pong outside the cgroup: $ /pong Both processes pinned to CPU1, running for 10s 10687.80 ctxsw/s The penalty is now less than 2%. And the results for perf stat are correct: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Performance counter stats for 'sleep 10': CPU1 cycles test # 0.000 GHz CPU1 23,933,981,448 cycles # 0.000 GHz Now perf stat reports the correct counts for for the non cgroup event. If we run pong inside the cgroup, then we also get the correct counts: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Performance counter stats for 'sleep 10': CPU1 22,297,726,205 cycles test # 0.000 GHz CPU1 23,933,981,448 cycles # 0.000 GHz 10.001457237 seconds time elapsed Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110825135803.GA4697@quad Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 24 +++++++++------ kernel/events/core.c | 63 ++++++++++++++++++++++++++++++++------ kernel/sched.c | 2 +- 3 files changed, 69 insertions(+), 20 deletions(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 245bafdafd5e..c816075c01ce 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -944,8 +944,10 @@ extern void perf_pmu_unregister(struct pmu *pmu); extern int perf_num_counters(void); extern const char *perf_pmu_name(void); -extern void __perf_event_task_sched_in(struct task_struct *task); -extern void __perf_event_task_sched_out(struct task_struct *task, struct task_struct *next); +extern void __perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task); +extern void __perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next); extern int perf_event_init_task(struct task_struct *child); extern void perf_event_exit_task(struct task_struct *child); extern void perf_event_free_task(struct task_struct *task); @@ -1059,17 +1061,20 @@ perf_sw_event(u32 event_id, u64 nr, struct pt_regs *regs, u64 addr) extern struct jump_label_key perf_sched_events; -static inline void perf_event_task_sched_in(struct task_struct *task) +static inline void perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { if (static_branch(&perf_sched_events)) - __perf_event_task_sched_in(task); + __perf_event_task_sched_in(prev, task); } -static inline void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next) +static inline void perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next) { perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, NULL, 0); - __perf_event_task_sched_out(task, next); + if (static_branch(&perf_sched_events)) + __perf_event_task_sched_out(prev, next); } extern void perf_event_mmap(struct vm_area_struct *vma); @@ -1139,10 +1144,11 @@ extern void perf_event_disable(struct perf_event *event); extern void perf_event_task_tick(void); #else static inline void -perf_event_task_sched_in(struct task_struct *task) { } +perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { } static inline void -perf_event_task_sched_out(struct task_struct *task, - struct task_struct *next) { } +perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next) { } static inline int perf_event_init_task(struct task_struct *child) { return 0; } static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } diff --git a/kernel/events/core.c b/kernel/events/core.c index b8785e26ee1c..45847fbb599a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -399,14 +399,54 @@ void perf_cgroup_switch(struct task_struct *task, int mode) local_irq_restore(flags); } -static inline void perf_cgroup_sched_out(struct task_struct *task) +static inline void perf_cgroup_sched_out(struct task_struct *task, + struct task_struct *next) { - perf_cgroup_switch(task, PERF_CGROUP_SWOUT); + struct perf_cgroup *cgrp1; + struct perf_cgroup *cgrp2 = NULL; + + /* + * we come here when we know perf_cgroup_events > 0 + */ + cgrp1 = perf_cgroup_from_task(task); + + /* + * next is NULL when called from perf_event_enable_on_exec() + * that will systematically cause a cgroup_switch() + */ + if (next) + cgrp2 = perf_cgroup_from_task(next); + + /* + * only schedule out current cgroup events if we know + * that we are switching to a different cgroup. Otherwise, + * do no touch the cgroup events. + */ + if (cgrp1 != cgrp2) + perf_cgroup_switch(task, PERF_CGROUP_SWOUT); } -static inline void perf_cgroup_sched_in(struct task_struct *task) +static inline void perf_cgroup_sched_in(struct task_struct *prev, + struct task_struct *task) { - perf_cgroup_switch(task, PERF_CGROUP_SWIN); + struct perf_cgroup *cgrp1; + struct perf_cgroup *cgrp2 = NULL; + + /* + * we come here when we know perf_cgroup_events > 0 + */ + cgrp1 = perf_cgroup_from_task(task); + + /* prev can never be NULL */ + cgrp2 = perf_cgroup_from_task(prev); + + /* + * only need to schedule in cgroup events if we are changing + * cgroup during ctxsw. Cgroup events were not scheduled + * out of ctxsw out if that was not the case. + */ + if (cgrp1 != cgrp2) + perf_cgroup_switch(task, PERF_CGROUP_SWIN); } static inline int perf_cgroup_connect(int fd, struct perf_event *event, @@ -518,11 +558,13 @@ static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx) { } -static inline void perf_cgroup_sched_out(struct task_struct *task) +static inline void perf_cgroup_sched_out(struct task_struct *task, + struct task_struct *next) { } -static inline void perf_cgroup_sched_in(struct task_struct *task) +static inline void perf_cgroup_sched_in(struct task_struct *prev, + struct task_struct *task) { } @@ -1988,7 +2030,7 @@ void __perf_event_task_sched_out(struct task_struct *task, * cgroup event are system-wide mode only */ if (atomic_read(&__get_cpu_var(perf_cgroup_events))) - perf_cgroup_sched_out(task); + perf_cgroup_sched_out(task, next); } static void task_ctx_sched_out(struct perf_event_context *ctx) @@ -2153,7 +2195,8 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx, * accessing the event control register. If a NMI hits, then it will * keep the event running. */ -void __perf_event_task_sched_in(struct task_struct *task) +void __perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { struct perf_event_context *ctx; int ctxn; @@ -2171,7 +2214,7 @@ void __perf_event_task_sched_in(struct task_struct *task) * cgroup event are system-wide mode only */ if (atomic_read(&__get_cpu_var(perf_cgroup_events))) - perf_cgroup_sched_in(task); + perf_cgroup_sched_in(prev, task); } static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count) @@ -2427,7 +2470,7 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx) * ctxswin cgroup events which are already scheduled * in. */ - perf_cgroup_sched_out(current); + perf_cgroup_sched_out(current, NULL); raw_spin_lock(&ctx->lock); task_ctx_sched_out(ctx); diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..0408cdc6d572 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3065,7 +3065,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_disable(); #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ - perf_event_task_sched_in(current); + perf_event_task_sched_in(prev, current); #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_enable(); #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ From 59ec6da2e3e53a3e4f3cb4d35b6449e05f1bcc18 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 19 Aug 2011 17:53:12 +0900 Subject: [PATCH 075/326] MAINTAINERS: Add some missed Wolfson files Mostly input related. Signed-off-by: Mark Brown --- MAINTAINERS | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 4f555d8e5346..0b4ccdd35bbb 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7200,6 +7200,9 @@ W: http://opensource.wolfsonmicro.com/content/linux-drivers-wolfson-devices S: Supported F: Documentation/hwmon/wm83?? F: drivers/leds/leds-wm83*.c +F: drivers/input/misc/wm831x-on.c +F: drivers/input/touchscreen/wm831x-ts.c +F: drivers/input/touchscreen/wm97*.c F: drivers/mfd/wm8*.c F: drivers/power/wm83*.c F: drivers/rtc/rtc-wm83*.c @@ -7209,6 +7212,7 @@ F: drivers/watchdog/wm83*_wdt.c F: include/linux/mfd/wm831x/ F: include/linux/mfd/wm8350/ F: include/linux/mfd/wm8400* +F: include/linux/wm97xx.h F: include/sound/wm????.h F: sound/soc/codecs/wm* From 1c1bdd324cd50ac55f7ebf95ef249d946c6e4361 Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 26 Aug 2011 12:42:11 +0530 Subject: [PATCH 076/326] ath9k_hw: Fix init mode register regression The commit 172805ad46b78717a738ca5c7908c68f0326d3a9 overwirtes additional clock settings of AR9330 to all AR9300 chips. Cc: stable@kernel.org Signed-off-by: Rajkumar Manoharan Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/ar9003_phy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath9k/ar9003_phy.c b/drivers/net/wireless/ath/ath9k/ar9003_phy.c index 1baca8e4715d..fcafec0605f4 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_phy.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_phy.c @@ -671,7 +671,7 @@ static int ar9003_hw_process_ini(struct ath_hw *ah, REG_WRITE_ARRAY(&ah->iniModesAdditional, modesIndex, regWrites); - if (AR_SREV_9300(ah)) + if (AR_SREV_9330(ah)) REG_WRITE_ARRAY(&ah->iniModesAdditional, 1, regWrites); if (AR_SREV_9340(ah) && !ah->is_clk_25mhz) From 7c2510120e9b43b0caf32c3786a6ab831f7d9e87 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Fri, 26 Aug 2011 17:24:59 +0200 Subject: [PATCH 077/326] iwlegacy: fix BUG_ON(info->control.rates[0].idx < 0) When trying to connect to 5GHz we can provide negative index to mac80211 what trigger BUG_ON. Reason of iwl-3945-rs malfunction on 5GHz is unknown and needs further investigation. For now, to do not trigger a bug, correct value and just print WARNING. Address bug: https://bugzilla.redhat.com/show_bug.cgi?id=730653 Reported-and-tested-by: Jan Teichmann Cc: stable@kernel.org Signed-off-by: Stanislaw Gruszka Signed-off-by: John W. Linville --- drivers/net/wireless/iwlegacy/iwl-3945-rs.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/iwlegacy/iwl-3945-rs.c b/drivers/net/wireless/iwlegacy/iwl-3945-rs.c index 977bd2477c6a..164bcae821f8 100644 --- a/drivers/net/wireless/iwlegacy/iwl-3945-rs.c +++ b/drivers/net/wireless/iwlegacy/iwl-3945-rs.c @@ -822,12 +822,15 @@ static void iwl3945_rs_get_rate(void *priv_r, struct ieee80211_sta *sta, out: - rs_sta->last_txrate_idx = index; - if (sband->band == IEEE80211_BAND_5GHZ) - info->control.rates[0].idx = rs_sta->last_txrate_idx - - IWL_FIRST_OFDM_RATE; - else + if (sband->band == IEEE80211_BAND_5GHZ) { + if (WARN_ON_ONCE(index < IWL_FIRST_OFDM_RATE)) + index = IWL_FIRST_OFDM_RATE; + rs_sta->last_txrate_idx = index; + info->control.rates[0].idx = index - IWL_FIRST_OFDM_RATE; + } else { + rs_sta->last_txrate_idx = index; info->control.rates[0].idx = rs_sta->last_txrate_idx; + } IWL_DEBUG_RATE(priv, "leave: %d\n", index); } From b33c25d6a62ac253caabda2b5f43258abff451c0 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Mon, 29 Aug 2011 23:01:58 -0400 Subject: [PATCH 078/326] acpica: ACPI_MAX_SLEEP should be 2 sec, not 20 This limit is a workaround for AML that sleeps too long, but the workaround didn't work b/c of a typo. https://bugzilla.kernel.org/show_bug.cgi?id=13195 Signed-off-by: Len Brown cc: stable@kernel.org # 2.6.35..3.0 --- drivers/acpi/acpica/acconfig.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/acpica/acconfig.h b/drivers/acpi/acpica/acconfig.h index bc533dde16c4..f895a244ca7e 100644 --- a/drivers/acpi/acpica/acconfig.h +++ b/drivers/acpi/acpica/acconfig.h @@ -121,7 +121,7 @@ /* Maximum sleep allowed via Sleep() operator */ -#define ACPI_MAX_SLEEP 20000 /* Two seconds */ +#define ACPI_MAX_SLEEP 2000 /* Two seconds */ /****************************************************************************** * From 7da64a0abc3b2c6cbd3521672e9bb74dd560bb89 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 30 Aug 2011 16:20:17 +1000 Subject: [PATCH 079/326] md: fix clearing of 'blocked' flag in the presence of bad blocks. When the 'blocked' flag on a device is cleared while there are unacknowledged bad blocks we must fail the device. This is needed for backwards compatability of the interface. The code currently uses the wrong test for "unacknowledged bad blocks exist". Change it to the right test. Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index aca611711264..3742ce8b0acf 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2592,7 +2592,7 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len) err = 0; } else if (cmd_match(buf, "-blocked")) { if (!test_bit(Faulty, &rdev->flags) && - test_bit(BlockedBadBlocks, &rdev->flags)) { + rdev->badblocks.unacked_exist) { /* metadata handler doesn't understand badblocks, * so we need to fail the device */ From 3b217116edaac634bf31e85c35708298059a8171 Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Tue, 30 Aug 2011 10:58:22 +0200 Subject: [PATCH 080/326] KVM: Fix instruction size issue in pvclock scaling Commit de2d1a524e94 ("KVM: Fix register corruption in pvclock_scale_delta") introduced a mul instruction that may have only a memory operand; the assembler therefore cannot select the correct size: pvclock.s:229: Error: no instruction mnemonic suffix given and no register operands; can't size instruction In this example the assembler is: #APP mul -48(%rbp) ; shrd $32, %rdx, %rax #NO_APP A simple solution is to use mulq. Signed-off-by: Duncan Sands Signed-off-by: Avi Kivity --- arch/x86/include/asm/pvclock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h index a518c0a45044..c59cc97fe6c1 100644 --- a/arch/x86/include/asm/pvclock.h +++ b/arch/x86/include/asm/pvclock.h @@ -44,7 +44,7 @@ static inline u64 pvclock_scale_delta(u64 delta, u32 mul_frac, int shift) : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) ); #elif defined(__x86_64__) __asm__ ( - "mul %[mul_frac] ; shrd $32, %[hi], %[lo]" + "mulq %[mul_frac] ; shrd $32, %[hi], %[lo]" : [lo]"=a"(product), [hi]"=d"(tmp) : "0"(delta), From c6675233f9015d3c0460c8aab53ed9b99d915c64 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 30 Aug 2011 15:01:20 +0200 Subject: [PATCH 081/326] netfilter: nf_queue: reject NF_STOLEN verdicts from userspace A userspace listener may send (bogus) NF_STOLEN verdict, which causes skb leak. This problem was previously fixed via 64507fdbc29c3a622180378210ecea8659b14e40 (netfilter: nf_queue: fix NF_STOLEN skb leak) but this had to be reverted because NF_STOLEN can also be returned by a netfilter hook when iterating the rules in nf_reinject. Reject userspace NF_STOLEN verdict, as suggested by Michal Miroslaw. This is complementary to commit fad54440438a7c231a6ae347738423cbabc936d9 (netfilter: avoid double free in nf_reinject). Cc: Julian Anastasov Cc: Eric Dumazet Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/ipv4/netfilter/ip_queue.c | 11 ++++------- net/ipv6/netfilter/ip6_queue.c | 11 ++++------- net/netfilter/nfnetlink_queue.c | 4 ++-- 3 files changed, 10 insertions(+), 16 deletions(-) diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 48f7d5b4ff37..e59aabd0eae4 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -314,7 +314,7 @@ ipq_set_verdict(struct ipq_verdict_msg *vmsg, unsigned int len) { struct nf_queue_entry *entry; - if (vmsg->value > NF_MAX_VERDICT) + if (vmsg->value > NF_MAX_VERDICT || vmsg->value == NF_STOLEN) return -EINVAL; entry = ipq_find_dequeue_entry(vmsg->id); @@ -359,12 +359,9 @@ ipq_receive_peer(struct ipq_peer_msg *pmsg, break; case IPQM_VERDICT: - if (pmsg->msg.verdict.value > NF_MAX_VERDICT) - status = -EINVAL; - else - status = ipq_set_verdict(&pmsg->msg.verdict, - len - sizeof(*pmsg)); - break; + status = ipq_set_verdict(&pmsg->msg.verdict, + len - sizeof(*pmsg)); + break; default: status = -EINVAL; } diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 87b243a25afa..e63c3972a739 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -314,7 +314,7 @@ ipq_set_verdict(struct ipq_verdict_msg *vmsg, unsigned int len) { struct nf_queue_entry *entry; - if (vmsg->value > NF_MAX_VERDICT) + if (vmsg->value > NF_MAX_VERDICT || vmsg->value == NF_STOLEN) return -EINVAL; entry = ipq_find_dequeue_entry(vmsg->id); @@ -359,12 +359,9 @@ ipq_receive_peer(struct ipq_peer_msg *pmsg, break; case IPQM_VERDICT: - if (pmsg->msg.verdict.value > NF_MAX_VERDICT) - status = -EINVAL; - else - status = ipq_set_verdict(&pmsg->msg.verdict, - len - sizeof(*pmsg)); - break; + status = ipq_set_verdict(&pmsg->msg.verdict, + len - sizeof(*pmsg)); + break; default: status = -EINVAL; } diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 00bd475eab4b..a80b0cb03f17 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -646,8 +646,8 @@ verdicthdr_get(const struct nlattr * const nfqa[]) return NULL; vhdr = nla_data(nfqa[NFQA_VERDICT_HDR]); - verdict = ntohl(vhdr->verdict); - if ((verdict & NF_VERDICT_MASK) > NF_MAX_VERDICT) + verdict = ntohl(vhdr->verdict) & NF_VERDICT_MASK; + if (verdict > NF_MAX_VERDICT || verdict == NF_STOLEN) return NULL; return vhdr; } From 4c6e4209662b2a4147cde16c2144a253a7430a49 Mon Sep 17 00:00:00 2001 From: Sanket Shah Date: Tue, 30 Aug 2011 15:23:03 +0200 Subject: [PATCH 082/326] netfilter: nf_ct_pptp: fix DNATed PPTP connection address translation When both the server and the client are NATed, the set-link-info control packet containing the peer's call-id field is not properly translated. I have verified that it was working in 2.6.16.13 kernel previously but due to rewrite, this scenario stopped working (Not knowing exact version when it stopped working). Signed-off-by: Sanket Shah Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_pptp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 2fd4565144de..31d56b23b9e9 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -364,6 +364,7 @@ pptp_inbound_pkt(struct sk_buff *skb, break; case PPTP_WAN_ERROR_NOTIFY: + case PPTP_SET_LINK_INFO: case PPTP_ECHO_REQUEST: case PPTP_ECHO_REPLY: /* I don't have to explain these ;) */ From 4a5cc84ae7e19fb7a72a30332ba67af43e0ad1ad Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 30 Aug 2011 15:45:10 +0200 Subject: [PATCH 083/326] netfilter: nf_ct_tcp: fix incorrect handling of invalid TCP option Michael M. Builov reported that in the tcp_options and tcp_sack functions of netfilter TCP conntrack the incorrect handling of invalid TCP option with too big opsize may lead to read access beyond tcp-packet or buffer allocated on stack (netfilter bugzilla #738). The fix is to stop parsing the options at detecting the broken option. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_proto_tcp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 37bf94394be0..afc4ab7cfe01 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -409,7 +409,7 @@ static void tcp_options(const struct sk_buff *skb, if (opsize < 2) /* "silly options" */ return; if (opsize > length) - break; /* don't parse partial options */ + return; /* don't parse partial options */ if (opcode == TCPOPT_SACK_PERM && opsize == TCPOLEN_SACK_PERM) @@ -469,7 +469,7 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, if (opsize < 2) /* "silly options" */ return; if (opsize > length) - break; /* don't parse partial options */ + return; /* don't parse partial options */ if (opcode == TCPOPT_SACK && opsize >= (TCPOLEN_SACK_BASE From bb9fc37358ffa9de1cc2b2b6f1a559b926ef50d9 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 30 Aug 2011 15:46:13 +0200 Subject: [PATCH 084/326] netfilter: nf_ct_tcp: wrong multiplication of TCPOLEN_TSTAMP_ALIGNED in tcp_sack skips fastpath The wrong multiplication of TCPOLEN_TSTAMP_ALIGNED by 4 skips the fast path for the timestamp-only option. Bug reported by Michael M. Builov (netfilter bugzilla #738). Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_proto_tcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index afc4ab7cfe01..8235b86b4e87 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -447,7 +447,7 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, BUG_ON(ptr == NULL); /* Fast path for timestamp-only option */ - if (length == TCPOLEN_TSTAMP_ALIGNED*4 + if (length == TCPOLEN_TSTAMP_ALIGNED && *(__be32 *)ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | (TCPOPT_TIMESTAMP << 8) From 768b1031dc9f3eabe3e9f603e33617a0215c12a8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 30 Aug 2011 16:33:04 +0200 Subject: [PATCH 085/326] netfilter: update netfilter git URL Netfilter git trees are moving to a directory shared by Pablo and myself, update git URLs. Signed-off-by: Patrick McHardy --- MAINTAINERS | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 1d2e79db0f58..a6669b2a77e0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4386,7 +4386,8 @@ L: netfilter@vger.kernel.org L: coreteam@netfilter.org W: http://www.netfilter.org/ W: http://www.iptables.org/ -T: git git://git.kernel.org/pub/scm/linux/kernel/git/kaber/nf-2.6.git +T: git git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-2.6.git +T: git git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next-2.6.git S: Supported F: include/linux/netfilter* F: include/linux/netfilter/ From 0e4660cbe51276e86dbdab17228733dbcdb49249 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 29 Aug 2011 10:06:14 +0200 Subject: [PATCH 086/326] ath9k_hw: fix calibration on 5 ghz ADC calibrations cannot run on 5 GHz with fast clock enabled. They need to be disabled, otherwise they'll hang and IQ mismatch calibration will not be run either. Signed-off-by: Felix Fietkau Reported-by: Adrian Chadd Cc: stable@kernel.org Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/ar9002_calib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath9k/ar9002_calib.c b/drivers/net/wireless/ath/ath9k/ar9002_calib.c index 2d4c0910295b..2d394af82171 100644 --- a/drivers/net/wireless/ath/ath9k/ar9002_calib.c +++ b/drivers/net/wireless/ath/ath9k/ar9002_calib.c @@ -41,7 +41,8 @@ static bool ar9002_hw_is_cal_supported(struct ath_hw *ah, case ADC_DC_CAL: /* Run ADC Gain Cal for non-CCK & non 2GHz-HT20 only */ if (!IS_CHAN_B(chan) && - !(IS_CHAN_2GHZ(chan) && IS_CHAN_HT20(chan))) + !((IS_CHAN_2GHZ(chan) || IS_CHAN_A_FAST_CLOCK(ah, chan)) && + IS_CHAN_HT20(chan))) supported = true; break; } From ec0506dbe4e240ecd4c32bf74c84a88ce1ddb414 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maciej=20=C5=BBenczykowski?= Date: Sun, 28 Aug 2011 12:35:31 +0000 Subject: [PATCH 087/326] net: relax PKTINFO non local ipv6 udp xmit check MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allow transparent sockets to be less restrictive about the source ip of ipv6 udp packets being sent. Google-Bug-Id: 5018138 Signed-off-by: Maciej Å»enczykowski CC: "Erik Kline" CC: "Lorenzo Colitti" Signed-off-by: David S. Miller --- include/net/transp_v6.h | 1 + net/ipv6/datagram.c | 5 +++-- net/ipv6/ip6_flowlabel.c | 8 ++++---- net/ipv6/ipv6_sockglue.c | 2 +- net/ipv6/raw.c | 4 ++-- net/ipv6/udp.c | 4 ++-- 6 files changed, 13 insertions(+), 11 deletions(-) diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 5271a741c3a3..498433dd067d 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -39,6 +39,7 @@ extern int datagram_recv_ctl(struct sock *sk, struct sk_buff *skb); extern int datagram_send_ctl(struct net *net, + struct sock *sk, struct msghdr *msg, struct flowi6 *fl6, struct ipv6_txoptions *opt, diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 9ef1831746ef..b46e9f88ce37 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -599,7 +599,7 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) return 0; } -int datagram_send_ctl(struct net *net, +int datagram_send_ctl(struct net *net, struct sock *sk, struct msghdr *msg, struct flowi6 *fl6, struct ipv6_txoptions *opt, int *hlimit, int *tclass, int *dontfrag) @@ -658,7 +658,8 @@ int datagram_send_ctl(struct net *net, if (addr_type != IPV6_ADDR_ANY) { int strict = __ipv6_addr_src_scope(addr_type) <= IPV6_ADDR_SCOPE_LINKLOCAL; - if (!ipv6_chk_addr(net, &src_info->ipi6_addr, + if (!inet_sk(sk)->transparent && + !ipv6_chk_addr(net, &src_info->ipi6_addr, strict ? dev : NULL, 0)) err = -EINVAL; else diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index f3caf1b8d572..543039450193 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -322,8 +322,8 @@ static int fl6_renew(struct ip6_flowlabel *fl, unsigned long linger, unsigned lo } static struct ip6_flowlabel * -fl_create(struct net *net, struct in6_flowlabel_req *freq, char __user *optval, - int optlen, int *err_p) +fl_create(struct net *net, struct sock *sk, struct in6_flowlabel_req *freq, + char __user *optval, int optlen, int *err_p) { struct ip6_flowlabel *fl = NULL; int olen; @@ -360,7 +360,7 @@ fl_create(struct net *net, struct in6_flowlabel_req *freq, char __user *optval, msg.msg_control = (void*)(fl->opt+1); memset(&flowi6, 0, sizeof(flowi6)); - err = datagram_send_ctl(net, &msg, &flowi6, fl->opt, &junk, + err = datagram_send_ctl(net, sk, &msg, &flowi6, fl->opt, &junk, &junk, &junk); if (err) goto done; @@ -528,7 +528,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen) if (freq.flr_label & ~IPV6_FLOWLABEL_MASK) return -EINVAL; - fl = fl_create(net, &freq, optval, optlen, &err); + fl = fl_create(net, sk, &freq, optval, optlen, &err); if (fl == NULL) return err; sfl1 = kmalloc(sizeof(*sfl1), GFP_KERNEL); diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 147ede38ab48..2fbda5fc4cc4 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -475,7 +475,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, msg.msg_controllen = optlen; msg.msg_control = (void*)(opt+1); - retv = datagram_send_ctl(net, &msg, &fl6, opt, &junk, &junk, + retv = datagram_send_ctl(net, sk, &msg, &fl6, opt, &junk, &junk, &junk); if (retv) goto done; diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 6a79f3081bdb..343852e5c703 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -817,8 +817,8 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, memset(opt, 0, sizeof(struct ipv6_txoptions)); opt->tot_len = sizeof(struct ipv6_txoptions); - err = datagram_send_ctl(sock_net(sk), msg, &fl6, opt, &hlimit, - &tclass, &dontfrag); + err = datagram_send_ctl(sock_net(sk), sk, msg, &fl6, opt, + &hlimit, &tclass, &dontfrag); if (err < 0) { fl6_sock_release(flowlabel); return err; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 29213b51c499..bb95e8e1c6f9 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1090,8 +1090,8 @@ int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, memset(opt, 0, sizeof(struct ipv6_txoptions)); opt->tot_len = sizeof(*opt); - err = datagram_send_ctl(sock_net(sk), msg, &fl6, opt, &hlimit, - &tclass, &dontfrag); + err = datagram_send_ctl(sock_net(sk), sk, msg, &fl6, opt, + &hlimit, &tclass, &dontfrag); if (err < 0) { fl6_sock_release(flowlabel); return err; From 29c486df6a208432b370bd4be99ae1369ede28d8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 30 Aug 2011 18:57:00 -0400 Subject: [PATCH 088/326] net: ipv4: relax AF_INET check in bind() commit d0733d2e29b65 (Check for mistakenly passed in non-IPv4 address) added regression on legacy apps that use bind() with AF_UNSPEC family. Relax the check, but make sure the bind() is done on INADDR_ANY addresses, as AF_UNSPEC has probably no sane meaning for other addresses. Bugzilla reference : https://bugzilla.kernel.org/show_bug.cgi?id=42012 Signed-off-by: Eric Dumazet Reported-and-bisected-by: Rene Meier CC: Marcus Meissner Signed-off-by: David S. Miller --- net/ipv4/af_inet.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 1b745d412cf6..dd2b9478ddd1 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -466,8 +466,13 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) goto out; if (addr->sin_family != AF_INET) { + /* Compatibility games : accept AF_UNSPEC (mapped to AF_INET) + * only if s_addr is INADDR_ANY. + */ err = -EAFNOSUPPORT; - goto out; + if (addr->sin_family != AF_UNSPEC || + addr->sin_addr.s_addr != htonl(INADDR_ANY)) + goto out; } chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr); From bfbed02ff54b3f97c8a00ef1657049448f1f172a Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Mon, 29 Aug 2011 23:34:47 +0000 Subject: [PATCH 089/326] MAINTAINERS: Update ATLX driver maintainers jie.yang@atheros.com bounces and I get a 550 "Unknown address error". Perhaps they have moved on? Signed-off-by: Ian Campbell Cc: Jay Cliburn Cc: Chris Snook Cc: Jie Yang Cc: Andrew Morton Cc: Joe Perches Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 261232d8adf9..2829f69484ef 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1278,7 +1278,6 @@ F: drivers/input/misc/ati_remote2.c ATLX ETHERNET DRIVERS M: Jay Cliburn M: Chris Snook -M: Jie Yang L: netdev@vger.kernel.org W: http://sourceforge.net/projects/atl1 W: http://atl1.sourceforge.net From 40a9f52e584936267228740bf7c16c5343166d11 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Mon, 29 Aug 2011 23:41:09 +0000 Subject: [PATCH 090/326] MAINTAINERS: Update Cisco VIC driver maintainers vkolluri@cisco.com bounces and I get "Unknown address error 550". Signed-off-by: Ian Campbell Cc: Christian Benvenuti Cc: Roopa Prabhu Cc: David Wang Cc: Andrew Morton Cc: Joe Perches Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 2829f69484ef..735bdf90d15c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1757,7 +1757,6 @@ F: Documentation/zh_CN/ CISCO VIC ETHERNET NIC DRIVER M: Christian Benvenuti -M: Vasanthy Kolluri M: Roopa Prabhu M: David Wang S: Supported From 31a0479546a7f91ec959560be01a0b1243a2b20c Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Mon, 29 Aug 2011 23:42:58 +0000 Subject: [PATCH 091/326] MAINTAINERS: Update BNA 10G Maintainer ddutt@brocade.com bounces with 550 "RESOLVER.ADR.RecipNotFound" Signed-off-by: Ian Campbell Cc: Rasesh Mody Cc: Andrew Morton Cc: Joe Perches Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 735bdf90d15c..959ff41a14c4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1573,7 +1573,6 @@ F: drivers/scsi/bfa/ BROCADE BNA 10 GIGABIT ETHERNET DRIVER M: Rasesh Mody -M: Debashis Dutt L: netdev@vger.kernel.org S: Supported F: drivers/net/bna/ From e2faeec2de9e2c73958e6ea6065dde1e8cd6f3a2 Mon Sep 17 00:00:00 2001 From: Jeff Kirsher Date: Tue, 30 Aug 2011 20:58:56 -0400 Subject: [PATCH 092/326] e1000: Fix driver to be used on PA RISC C8000 workstations The checksum field in the EEPROM on HPPA is really not a checksum but a signature (0x16d6). So allow 0x16d6 as the matching checksum on HPPA systems. This issue is present on longterm/stable kernels, I have verified that this patch is applicable back to at least 2.6.32.y kernels. v2- changed ifdef to use CONFIG_PARISC instead of __hppa__ CC: Guy Martin CC: Rolf Eike Beer CC: Matt Turner Reported-by: Mikulas Patocka Signed-off-by: Jeff Kirsher Acked-by: Jesse Brandeburg Signed-off-by: David S. Miller --- drivers/net/e1000/e1000_hw.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index 8545c7aa93eb..a5a89ecb6f36 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -4026,6 +4026,12 @@ s32 e1000_validate_eeprom_checksum(struct e1000_hw *hw) checksum += eeprom_data; } +#ifdef CONFIG_PARISC + /* This is a signature and not a checksum on HP c8000 */ + if ((hw->subsystem_vendor_id == 0x103C) && (eeprom_data == 0x16d6)) + return E1000_SUCCESS; + +#endif if (checksum == (u16) EEPROM_SUM) return E1000_SUCCESS; else { From 39c428f753fc9514e5042d4ec736f80065e6c3a7 Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Tue, 16 Aug 2011 18:44:39 -0400 Subject: [PATCH 093/326] powerpc/85xx: enable the audio drivers in the defconfigs Enable the audio drivers in the non-corenet 85xx defconfigs so that audio is enabled on the Freescale P1022DS reference board. Signed-off-by: Timur Tabi Signed-off-by: Kumar Gala --- arch/powerpc/configs/mpc85xx_defconfig | 1 + arch/powerpc/configs/mpc85xx_smp_defconfig | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/powerpc/configs/mpc85xx_defconfig b/arch/powerpc/configs/mpc85xx_defconfig index fcd85d2c72dc..a3467bfb7671 100644 --- a/arch/powerpc/configs/mpc85xx_defconfig +++ b/arch/powerpc/configs/mpc85xx_defconfig @@ -139,6 +139,7 @@ CONFIG_SND=y CONFIG_SND_INTEL8X0=y # CONFIG_SND_PPC is not set # CONFIG_SND_USB is not set +CONFIG_SND_SOC=y CONFIG_HID_A4TECH=y CONFIG_HID_APPLE=y CONFIG_HID_BELKIN=y diff --git a/arch/powerpc/configs/mpc85xx_smp_defconfig b/arch/powerpc/configs/mpc85xx_smp_defconfig index 908c941fc24c..9693f6ed3da0 100644 --- a/arch/powerpc/configs/mpc85xx_smp_defconfig +++ b/arch/powerpc/configs/mpc85xx_smp_defconfig @@ -140,6 +140,7 @@ CONFIG_SND=y CONFIG_SND_INTEL8X0=y # CONFIG_SND_PPC is not set # CONFIG_SND_USB is not set +CONFIG_SND_SOC=y CONFIG_HID_A4TECH=y CONFIG_HID_APPLE=y CONFIG_HID_BELKIN=y From e09e2fb51385fe6f9beca0a27ae6f1299097bbcd Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Fri, 22 Jul 2011 15:48:08 -0500 Subject: [PATCH 094/326] powerpc/85xx: enable caam crypto driver by default corenet based SoCs have SEC4 h/w, so enable the SEC4 driver, caam, and the algorithms it supports, and disable the SEC2/3 driver, talitos. Signed-off-by: Kim Phillips Signed-off-by: Kumar Gala --- arch/powerpc/configs/85xx/p1023rds_defconfig | 1 + arch/powerpc/configs/corenet32_smp_defconfig | 1 + arch/powerpc/configs/corenet64_smp_defconfig | 5 ++++- 3 files changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/configs/85xx/p1023rds_defconfig b/arch/powerpc/configs/85xx/p1023rds_defconfig index 980ff8f61fd4..3ff5a81c709f 100644 --- a/arch/powerpc/configs/85xx/p1023rds_defconfig +++ b/arch/powerpc/configs/85xx/p1023rds_defconfig @@ -171,3 +171,4 @@ CONFIG_CRYPTO_SHA256=y CONFIG_CRYPTO_SHA512=y CONFIG_CRYPTO_AES=y # CONFIG_CRYPTO_ANSI_CPRNG is not set +CONFIG_CRYPTO_DEV_FSL_CAAM=y diff --git a/arch/powerpc/configs/corenet32_smp_defconfig b/arch/powerpc/configs/corenet32_smp_defconfig index 10562a5c65b9..4311d02a3bfd 100644 --- a/arch/powerpc/configs/corenet32_smp_defconfig +++ b/arch/powerpc/configs/corenet32_smp_defconfig @@ -185,3 +185,4 @@ CONFIG_CRYPTO_SHA256=y CONFIG_CRYPTO_SHA512=y CONFIG_CRYPTO_AES=y # CONFIG_CRYPTO_ANSI_CPRNG is not set +CONFIG_CRYPTO_DEV_FSL_CAAM=y diff --git a/arch/powerpc/configs/corenet64_smp_defconfig b/arch/powerpc/configs/corenet64_smp_defconfig index d32283555b53..c92c204a204b 100644 --- a/arch/powerpc/configs/corenet64_smp_defconfig +++ b/arch/powerpc/configs/corenet64_smp_defconfig @@ -100,5 +100,8 @@ CONFIG_DEBUG_INFO=y CONFIG_SYSCTL_SYSCALL_CHECK=y CONFIG_VIRQ_DEBUG=y CONFIG_CRYPTO_PCBC=m +CONFIG_CRYPTO_SHA256=y +CONFIG_CRYPTO_SHA512=y +CONFIG_CRYPTO_AES=y # CONFIG_CRYPTO_ANSI_CPRNG is not set -CONFIG_CRYPTO_DEV_TALITOS=y +CONFIG_CRYPTO_DEV_FSL_CAAM=y From 0c81e4b426a17785ebd4f41d1e14a818706bdc90 Mon Sep 17 00:00:00 2001 From: Chunhe Lan Date: Fri, 12 Aug 2011 19:00:09 +0800 Subject: [PATCH 095/326] powerpc/p1023rds: Fix the error of bank-width of nor flash In the p1023rds, a physical bus of nor flash is 16 bits width. The bank-width is width (in bytes) of the bus width. So, the value of bank-width of nor flash is not one, and it should be two. Signed-off-by: Chunhe Lan Signed-off-by: Kumar Gala --- arch/powerpc/boot/dts/p1023rds.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/boot/dts/p1023rds.dts b/arch/powerpc/boot/dts/p1023rds.dts index bfa96aa8f2ca..d9b776740a67 100644 --- a/arch/powerpc/boot/dts/p1023rds.dts +++ b/arch/powerpc/boot/dts/p1023rds.dts @@ -387,7 +387,7 @@ nor@0,0 { #size-cells = <1>; compatible = "cfi-flash"; reg = <0x0 0x0 0x02000000>; - bank-width = <1>; + bank-width = <2>; device-width = <1>; partition@0 { label = "ramdisk"; From 43220aa0f22cd3ce5b30246d50ccd696d119edea Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 31 Aug 2011 12:49:14 +1000 Subject: [PATCH 096/326] md/raid5: fix a hang on device failure. Waiting for a 'blocked' rdev to become unblocked in the raid5d thread cannot work with internal metadata as it is the raid5d thread which will clear the blocked flag. This wasn't a problem in 3.0 and earlier as we only set the blocked flag when external metadata was used then. However we now set it always, so we need to be more careful. Signed-off-by: NeilBrown --- drivers/md/raid5.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index dbae459fb02d..43709fa6b6df 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3336,7 +3336,7 @@ static void handle_stripe(struct stripe_head *sh) finish: /* wait for this device to become unblocked */ - if (unlikely(s.blocked_rdev)) + if (conf->mddev->external && unlikely(s.blocked_rdev)) md_wait_for_blocked_rdev(s.blocked_rdev, conf->mddev); if (s.handle_bad_blocks) From 728a52222461a8cf0d5c375da1ef514a72d5194b Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Fri, 26 Aug 2011 16:33:52 +0300 Subject: [PATCH 097/326] ASoC: soc-dapm: Fix parameter comment for snd_soc_dapm_free We have dapm_context instead of codec parameter. Signed-off-by: Peter Ujfalusi Signed-off-by: Mark Brown --- sound/soc/soc-dapm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 7e15914b3633..d67c637557a7 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -2763,7 +2763,7 @@ EXPORT_SYMBOL_GPL(snd_soc_dapm_ignore_suspend); /** * snd_soc_dapm_free - free dapm resources - * @card: SoC device + * @dapm: DAPM context * * Free all dapm widgets and resources. */ From 63fa0a288cfedca681175fe13cf15677e944cdb2 Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Sat, 27 Aug 2011 18:24:12 +0200 Subject: [PATCH 098/326] ASoC: snd_soc_codec_{readable,writable}_register change default to true Change the default return value of snd_soc_codec_{readable,writable}_register to true when no codec specific callback for this function is given. Otherwise all registers of that codec will neither be readable nor writable, which is most certainly not what we want. Signed-off-by: Lars-Peter Clausen Acked-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/soc-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index b085d8e87574..d2ef014af215 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -1633,7 +1633,7 @@ int snd_soc_codec_readable_register(struct snd_soc_codec *codec, if (codec->readable_register) return codec->readable_register(codec, reg); else - return 0; + return 1; } EXPORT_SYMBOL_GPL(snd_soc_codec_readable_register); @@ -1651,7 +1651,7 @@ int snd_soc_codec_writable_register(struct snd_soc_codec *codec, if (codec->writable_register) return codec->writable_register(codec, reg); else - return 0; + return 1; } EXPORT_SYMBOL_GPL(snd_soc_codec_writable_register); From 6c5b756aaa3e4a20c8d4b47b9dc4799b9cfdfcb8 Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Sat, 27 Aug 2011 18:24:13 +0200 Subject: [PATCH 099/326] ASoC: Fix register cache sync register_writable WARN_ONs Currently the condition for these WARN_ONs is reversed and they are placed before the actual check whether we are going to write to that register. So if the codec implements the register_writable callback we'll get a warning for each writable register when syncing the register cache. While we are at it change the check to use snd_soc_codec_writable_register instead of open-coding it. Signed-off-by: Lars-Peter Clausen Acked-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/soc-cache.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/sound/soc/soc-cache.c b/sound/soc/soc-cache.c index d9f8aded51f3..20b7f3b003a3 100644 --- a/sound/soc/soc-cache.c +++ b/sound/soc/soc-cache.c @@ -203,14 +203,14 @@ static int snd_soc_rbtree_cache_sync(struct snd_soc_codec *codec) rbnode = rb_entry(node, struct snd_soc_rbtree_node, node); for (i = 0; i < rbnode->blklen; ++i) { regtmp = rbnode->base_reg + i; - WARN_ON(codec->writable_register && - codec->writable_register(codec, regtmp)); val = snd_soc_rbtree_get_register(rbnode, i); def = snd_soc_get_cache_val(codec->reg_def_copy, i, rbnode->word_size); if (val == def) continue; + WARN_ON(!snd_soc_codec_writable_register(codec, regtmp)); + codec->cache_bypass = 1; ret = snd_soc_write(codec, regtmp, val); codec->cache_bypass = 0; @@ -563,8 +563,7 @@ static int snd_soc_lzo_cache_sync(struct snd_soc_codec *codec) lzo_blocks = codec->reg_cache; for_each_set_bit(i, lzo_blocks[0]->sync_bmp, lzo_blocks[0]->sync_bmp_nbits) { - WARN_ON(codec->writable_register && - codec->writable_register(codec, i)); + WARN_ON(!snd_soc_codec_writable_register(codec, i)); ret = snd_soc_cache_read(codec, i, &val); if (ret) return ret; @@ -823,8 +822,6 @@ static int snd_soc_flat_cache_sync(struct snd_soc_codec *codec) codec_drv = codec->driver; for (i = 0; i < codec_drv->reg_cache_size; ++i) { - WARN_ON(codec->writable_register && - codec->writable_register(codec, i)); ret = snd_soc_cache_read(codec, i, &val); if (ret) return ret; @@ -832,6 +829,9 @@ static int snd_soc_flat_cache_sync(struct snd_soc_codec *codec) if (snd_soc_get_cache_val(codec->reg_def_copy, i, codec_drv->reg_word_size) == val) continue; + + WARN_ON(!snd_soc_codec_writable_register(codec, i)); + ret = snd_soc_write(codec, i, val); if (ret) return ret; From 117ef9570bf3a332eca443231f41dd3a52fcfb2c Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Tue, 30 Aug 2011 00:28:42 -0300 Subject: [PATCH 100/326] ASoC: imx: Fix build warning of unused 'card' variable Fixes the following warning: CC sound/soc/imx/imx-pcm-fiq.o sound/soc/imx/imx-pcm-fiq.c: In function 'imx_pcm_fiq_new': sound/soc/imx/imx-pcm-fiq.c:243: warning: unused variable 'card' CC sound/soc/imx/imx-pcm-dma-mx2.o Signed-off-by: Fabio Estevam Signed-off-by: Mark Brown --- sound/soc/imx/imx-pcm-fiq.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/soc/imx/imx-pcm-fiq.c b/sound/soc/imx/imx-pcm-fiq.c index 309c59e6fb6c..7945625e0e08 100644 --- a/sound/soc/imx/imx-pcm-fiq.c +++ b/sound/soc/imx/imx-pcm-fiq.c @@ -240,7 +240,6 @@ static int ssi_irq = 0; static int imx_pcm_fiq_new(struct snd_soc_pcm_runtime *rtd) { - struct snd_card *card = rtd->card->snd_card; struct snd_soc_dai *dai = rtd->cpu_dai; struct snd_pcm *pcm = rtd->pcm; int ret; From 9adceaa5b3d2480e2252c4a7f9c4bd7d66b8c4a2 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Tue, 30 Aug 2011 20:22:04 +0100 Subject: [PATCH 101/326] drm/radeon/kms: set a default max_pixel_clock On some Power rv100 cards, we have no ATY OF table, but we have no combios table either, and hence we refuse all modes on VGA-0 since we end up with a 0 max pixel clock. Signed-off-by: Dave Airlie Cc: stable@kernel.org Reviewed-by: Alex Deucher Reviewed-by: Jerome Glisse --- drivers/gpu/drm/radeon/radeon_clocks.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c index dcd0863e31ae..b6e18c8db9f5 100644 --- a/drivers/gpu/drm/radeon/radeon_clocks.c +++ b/drivers/gpu/drm/radeon/radeon_clocks.c @@ -219,6 +219,9 @@ void radeon_get_clock_info(struct drm_device *dev) } else { DRM_INFO("Using generic clock info\n"); + /* may need to be per card */ + rdev->clock.max_pixel_clock = 35000; + if (rdev->flags & RADEON_IS_IGP) { p1pll->reference_freq = 1432; p2pll->reference_freq = 1432; From 7f310a5d4e8525ac0cc2f58c973d2100ce034410 Mon Sep 17 00:00:00 2001 From: Eric B Munson Date: Thu, 23 Jun 2011 16:34:38 -0400 Subject: [PATCH 102/326] perf_event: Fix broken calc_timer_values() We detected a serious issue with PERF_SAMPLE_READ and timing information when events were being multiplexing. Samples would have time_running > time_enabled. That was easy to reproduce with a libpfm4 example (ran 3 times to cause multiplexing on Core 2): $ syst_smpl -e uops_retired:freq=1 & $ syst_smpl -e uops_retired:freq=1 & $ syst_smpl -e uops_retired:freq=1 & IIP:0x0000000040062d ... PERIOD:2355332948 ENA=40144625315 RUN=60014875184 syst_smpl: WARNING: time_running > time_enabled 63277537998 uops_retired:freq=1 , scaled The bug was not present in kernel up to (and including) 3.0. It turns out the bug was introduced by the following commit: commit c4794295917ebeda8013b6cb9c8d71ab4f74a1fa events: Move lockless timer calculation into helper function The parameters of the function got reversed yet the call sites were not updated to reflect the change. That lead to time_running and time_enabled being swapped. That had no effect when there was no multiplexing because in that case time_running = time_enabled but it would show up in any other scenario. Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110829124112.GA4828@quad Signed-off-by: Ingo Molnar --- kernel/events/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 45847fbb599a..0f857782d06f 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3396,8 +3396,8 @@ static int perf_event_index(struct perf_event *event) } static void calc_timer_values(struct perf_event *event, - u64 *running, - u64 *enabled) + u64 *enabled, + u64 *running) { u64 now, ctx_time; From 20afc60f892d285fde179ead4b24e6a7938c2f1b Mon Sep 17 00:00:00 2001 From: Andrey Vagin Date: Tue, 30 Aug 2011 12:32:36 +0400 Subject: [PATCH 103/326] x86, perf: Check that current->mm is alive before getting user callchain An event may occur when an mm is already released. I added an event in dequeue_entity() and caught a panic with the following backtrace: [ 434.421110] BUG: unable to handle kernel NULL pointer dereference at 0000000000000050 [ 434.421258] IP: [] __get_user_pages_fast+0x9c/0x120 ... [ 434.421258] Call Trace: [ 434.421258] [] copy_from_user_nmi+0x51/0xf0 [ 434.421258] [] ? sched_clock_local+0x25/0x90 [ 434.421258] [] perf_callchain_user+0x128/0x170 [ 434.421258] [] ? __perf_event_header__init_id+0xed/0x100 [ 434.421258] [] perf_prepare_sample+0x200/0x280 [ 434.421258] [] __perf_event_overflow+0x1b8/0x290 [ 434.421258] [] ? tg_shares_up+0x0/0x670 [ 434.421258] [] ? walk_tg_tree+0x6a/0xb0 [ 434.421258] [] perf_swevent_overflow+0xc4/0xf0 [ 434.421258] [] do_perf_sw_event+0x1e0/0x250 [ 434.421258] [] perf_tp_event+0x44/0x70 [ 434.421258] [] ftrace_profile_sched_block+0xdf/0x110 [ 434.421258] [] dequeue_entity+0x2ad/0x2d0 [ 434.421258] [] dequeue_task_fair+0x1c/0x60 [ 434.421258] [] dequeue_task+0x9a/0xb0 [ 434.421258] [] deactivate_task+0x42/0xe0 [ 434.421258] [] thread_return+0x191/0x808 [ 434.421258] [] ? switch_task_namespaces+0x24/0x60 [ 434.421258] [] do_exit+0x464/0x910 [ 434.421258] [] do_group_exit+0x58/0xd0 [ 434.421258] [] sys_exit_group+0x17/0x20 [ 434.421258] [] system_call_fastpath+0x16/0x1b Signed-off-by: Andrey Vagin Signed-off-by: Peter Zijlstra Cc: stable@kernel.org Link: http://lkml.kernel.org/r/1314693156-24131-1-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 4ee3abf20ed6..cfa62ec090ec 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1900,6 +1900,9 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) perf_callchain_store(entry, regs->ip); + if (!current->mm) + return; + if (perf_callchain_user32(regs, entry)) return; From 8c0bec2151a47906bf779c6715a10ce04453ab77 Mon Sep 17 00:00:00 2001 From: Jiaying Zhang Date: Wed, 31 Aug 2011 11:50:51 -0400 Subject: [PATCH 104/326] ext4: remove i_mutex lock in ext4_evict_inode to fix lockdep complaining The i_mutex lock and flush_completed_IO() added by commit 2581fdc810 in ext4_evict_inode() causes lockdep complaining about potential deadlock in several places. In most/all of these LOCKDEP complaints it looks like it's a false positive, since many of the potential circular locking cases can't take place by the time the ext4_evict_inode() is called; but since at the very least it may mask real problems, we need to address this. This change removes the flush_completed_IO() and i_mutex lock in ext4_evict_inode(). Instead, we take a different approach to resolve the software lockup that commit 2581fdc810 intends to fix. Rather than having ext4-dio-unwritten thread wait for grabing the i_mutex lock of an inode, we use mutex_trylock() instead, and simply requeue the work item if we fail to grab the inode's i_mutex lock. This should speed up work queue processing in general and also prevents the following deadlock scenario: During page fault, shrink_icache_memory is called that in turn evicts another inode B. Inode B has some pending io_end work so it calls ext4_ioend_wait() that waits for inode B's i_ioend_count to become zero. However, inode B's ioend work was queued behind some of inode A's ioend work on the same cpu's ext4-dio-unwritten workqueue. As the ext4-dio-unwritten thread on that cpu is processing inode A's ioend work, it tries to grab inode A's i_mutex lock. Since the i_mutex lock of inode A is still hold before the page fault happened, we enter a deadlock. Signed-off-by: Jiaying Zhang Signed-off-by: "Theodore Ts'o" --- fs/ext4/ext4.h | 1 + fs/ext4/inode.c | 3 --- fs/ext4/page-io.c | 18 +++++++++++++++++- 3 files changed, 18 insertions(+), 4 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index e717dfd2f2b4..b7d7bd0f066e 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -175,6 +175,7 @@ struct mpage_da_data { */ #define EXT4_IO_END_UNWRITTEN 0x0001 #define EXT4_IO_END_ERROR 0x0002 +#define EXT4_IO_END_QUEUED 0x0004 struct ext4_io_page { struct page *p_page; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c4da98a959ae..18d2558b7624 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -121,9 +121,6 @@ void ext4_evict_inode(struct inode *inode) trace_ext4_evict_inode(inode); - mutex_lock(&inode->i_mutex); - ext4_flush_completed_IO(inode); - mutex_unlock(&inode->i_mutex); ext4_ioend_wait(inode); if (inode->i_nlink) { diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 78839af7ce29..92f38ee13f8a 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -142,7 +142,23 @@ static void ext4_end_io_work(struct work_struct *work) unsigned long flags; int ret; - mutex_lock(&inode->i_mutex); + if (!mutex_trylock(&inode->i_mutex)) { + /* + * Requeue the work instead of waiting so that the work + * items queued after this can be processed. + */ + queue_work(EXT4_SB(inode->i_sb)->dio_unwritten_wq, &io->work); + /* + * To prevent the ext4-dio-unwritten thread from keeping + * requeueing end_io requests and occupying cpu for too long, + * yield the cpu if it sees an end_io request that has already + * been requeued. + */ + if (io->flag & EXT4_IO_END_QUEUED) + yield(); + io->flag |= EXT4_IO_END_QUEUED; + return; + } ret = ext4_end_io_nolock(io); if (ret < 0) { mutex_unlock(&inode->i_mutex); From 08c14071fda4e69abb9d5b1566651cd092b158d3 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 18 Aug 2011 15:23:47 +0300 Subject: [PATCH 105/326] mmc: rename mmc_host_clk_{ungate|gate} to mmc_host_clk_{hold|release} As per suggestion by Linus Walleij: > If you think the names of the functions are confusing then > you may rename them, say like this: > > mmc_host_clk_ungate() -> mmc_host_clk_hold() > mmc_host_clk_gate() -> mmc_host_clk_release() > > Which would make the usecases more clear (This is CC'd to stable@ because the next two patches, which fix observable races, depend on it.) Signed-off-by: Mika Westerberg Reviewed-by: Linus Walleij Cc: Signed-off-by: Chris Ball --- drivers/mmc/core/core.c | 4 ++-- drivers/mmc/core/host.c | 10 +++++----- drivers/mmc/core/host.h | 8 ++++---- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 91a0a7460ebb..63ffc65f84af 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -133,7 +133,7 @@ void mmc_request_done(struct mmc_host *host, struct mmc_request *mrq) if (mrq->done) mrq->done(mrq); - mmc_host_clk_gate(host); + mmc_host_clk_release(host); } } @@ -192,7 +192,7 @@ mmc_start_request(struct mmc_host *host, struct mmc_request *mrq) mrq->stop->mrq = mrq; } } - mmc_host_clk_ungate(host); + mmc_host_clk_hold(host); led_trigger_event(host->led, LED_FULL); host->ops->request(host, mrq); } diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c index b29d3e8fd3a2..96a26b2bf5f0 100644 --- a/drivers/mmc/core/host.c +++ b/drivers/mmc/core/host.c @@ -119,14 +119,14 @@ static void mmc_host_clk_gate_work(struct work_struct *work) } /** - * mmc_host_clk_ungate - ungate hardware MCI clocks + * mmc_host_clk_hold - ungate hardware MCI clocks * @host: host to ungate. * * Makes sure the host ios.clock is restored to a non-zero value * past this call. Increase clock reference count and ungate clock * if we're the first user. */ -void mmc_host_clk_ungate(struct mmc_host *host) +void mmc_host_clk_hold(struct mmc_host *host) { unsigned long flags; @@ -164,14 +164,14 @@ static bool mmc_host_may_gate_card(struct mmc_card *card) } /** - * mmc_host_clk_gate - gate off hardware MCI clocks + * mmc_host_clk_release - gate off hardware MCI clocks * @host: host to gate. * * Calls the host driver with ios.clock set to zero as often as possible * in order to gate off hardware MCI clocks. Decrease clock reference * count and schedule disabling of clock. */ -void mmc_host_clk_gate(struct mmc_host *host) +void mmc_host_clk_release(struct mmc_host *host) { unsigned long flags; @@ -231,7 +231,7 @@ static inline void mmc_host_clk_exit(struct mmc_host *host) if (cancel_work_sync(&host->clk_gate_work)) mmc_host_clk_gate_delayed(host); if (host->clk_gated) - mmc_host_clk_ungate(host); + mmc_host_clk_hold(host); /* There should be only one user now */ WARN_ON(host->clk_requests > 1); } diff --git a/drivers/mmc/core/host.h b/drivers/mmc/core/host.h index de199f911928..fb8a5cd2e4a1 100644 --- a/drivers/mmc/core/host.h +++ b/drivers/mmc/core/host.h @@ -16,16 +16,16 @@ int mmc_register_host_class(void); void mmc_unregister_host_class(void); #ifdef CONFIG_MMC_CLKGATE -void mmc_host_clk_ungate(struct mmc_host *host); -void mmc_host_clk_gate(struct mmc_host *host); +void mmc_host_clk_hold(struct mmc_host *host); +void mmc_host_clk_release(struct mmc_host *host); unsigned int mmc_host_clk_rate(struct mmc_host *host); #else -static inline void mmc_host_clk_ungate(struct mmc_host *host) +static inline void mmc_host_clk_hold(struct mmc_host *host) { } -static inline void mmc_host_clk_gate(struct mmc_host *host) +static inline void mmc_host_clk_release(struct mmc_host *host) { } From 778e277cb82411c9002ca28ccbd216c4d9eb9158 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 18 Aug 2011 15:23:48 +0300 Subject: [PATCH 106/326] mmc: core: prevent aggressive clock gating racing with ios updates We have seen at least two different races when clock gating kicks in in a middle of ios structure update. First one happens when ios->clock is changed outside of aggressive clock gating framework, for example via mmc_set_clock(). The race might happen when we run following code: mmc_set_ios(): ... if (ios->clock > 0) mmc_set_ungated(host); Now if gating kicks in right after the condition check we end up setting host->clk_gated to false even though we have just gated the clock. Next time a request is started we try to ungate and restore the clock in mmc_host_clk_hold(). However since we have host->clk_gated set to false the original clock is not restored. This eventually will cause the host controller to hang since its clock is disabled while we are trying to issue a request. For example on Intel Medfield platform we see: [ 13.818610] mmc2: Timeout waiting for hardware interrupt. [ 13.818698] sdhci: =========== REGISTER DUMP (mmc2)=========== [ 13.818753] sdhci: Sys addr: 0x00000000 | Version: 0x00008901 [ 13.818804] sdhci: Blk size: 0x00000000 | Blk cnt: 0x00000000 [ 13.818853] sdhci: Argument: 0x00000000 | Trn mode: 0x00000000 [ 13.818903] sdhci: Present: 0x1fff0000 | Host ctl: 0x00000001 [ 13.818951] sdhci: Power: 0x0000000d | Blk gap: 0x00000000 [ 13.819000] sdhci: Wake-up: 0x00000000 | Clock: 0x00000000 [ 13.819049] sdhci: Timeout: 0x00000000 | Int stat: 0x00000000 [ 13.819098] sdhci: Int enab: 0x00ff00c3 | Sig enab: 0x00ff00c3 [ 13.819147] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 [ 13.819196] sdhci: Caps: 0x6bee32b2 | Caps_1: 0x00000000 [ 13.819245] sdhci: Cmd: 0x00000000 | Max curr: 0x00000000 [ 13.819292] sdhci: Host ctl2: 0x00000000 [ 13.819331] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x00000000 [ 13.819377] sdhci: =========================================== [ 13.919605] mmc2: Reset 0x2 never completed. and it never recovers. Second race might happen while running mmc_power_off(): static void mmc_power_off(struct mmc_host *host) { host->ios.clock = 0; host->ios.vdd = 0; [ clock gating kicks in here ] /* * Reset ocr mask to be the highest possible voltage supported for * this mmc host. This value will be used at next power up. */ host->ocr = 1 << (fls(host->ocr_avail) - 1); if (!mmc_host_is_spi(host)) { host->ios.bus_mode = MMC_BUSMODE_OPENDRAIN; host->ios.chip_select = MMC_CS_DONTCARE; } host->ios.power_mode = MMC_POWER_OFF; host->ios.bus_width = MMC_BUS_WIDTH_1; host->ios.timing = MMC_TIMING_LEGACY; mmc_set_ios(host); } If the clock gating worker kicks in while we are only partially updated the ios structure the host controller gets incomplete ios and might not work as supposed. Again on Intel Medfield platform we get: [ 4.185349] kernel BUG at drivers/mmc/host/sdhci.c:1155! [ 4.185422] invalid opcode: 0000 [#1] PREEMPT SMP [ 4.185509] Modules linked in: [ 4.185565] [ 4.185608] Pid: 4, comm: kworker/0:0 Not tainted 3.0.0+ #240 Intel Corporation Medfield/iCDKA [ 4.185742] EIP: 0060:[] EFLAGS: 00010083 CPU: 0 [ 4.185827] EIP is at sdhci_set_power+0x3e/0xd0 [ 4.185891] EAX: f5ff98e0 EBX: f5ff98e0 ECX: 00000000 EDX: 00000001 [ 4.185970] ESI: f5ff977c EDI: f5ff9904 EBP: f644fe98 ESP: f644fe94 [ 4.186049] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 [ 4.186125] Process kworker/0:0 (pid: 4, ti=f644e000 task=f644c0e0 task.ti=f644e000) [ 4.186219] Stack: [ 4.186257] f5ff98e0 f644feb0 c1365173 00000282 f5ff9460 f5ff96e0 f5ff96e0 f644feec [ 4.186418] c1355bd8 f644c0e0 c1499c3d f5ff96e0 f644fed4 00000006 f5ff96e0 00000286 [ 4.186579] f644fedc c107922b f644feec 00000286 f5ff9460 f5ff9700 f644ff10 c135839e [ 4.186739] Call Trace: [ 4.186802] [] sdhci_set_ios+0x1c3/0x340 [ 4.186883] [] mmc_gate_clock+0x68/0x120 [ 4.186963] [] ? _raw_spin_unlock_irqrestore+0x4d/0x60 [ 4.187052] [] ? trace_hardirqs_on+0xb/0x10 [ 4.187134] [] mmc_host_clk_gate_delayed+0xbe/0x130 [ 4.187219] [] ? process_one_work+0xf9/0x5b0 [ 4.187300] [] mmc_host_clk_gate_work+0xd/0x10 [ 4.187379] [] process_one_work+0x172/0x5b0 [ 4.187457] [] ? process_one_work+0xf9/0x5b0 [ 4.187538] [] ? mmc_host_clk_gate_delayed+0x130/0x130 [ 4.187625] [] worker_thread+0x118/0x330 [ 4.187700] [] ? preempt_schedule+0x2e/0x50 [ 4.187779] [] ? rescuer_thread+0x1f0/0x1f0 [ 4.187857] [] kthread+0x74/0x80 [ 4.187931] [] ? __init_kthread_worker+0x60/0x60 [ 4.188015] [] kernel_thread_helper+0x6/0xd [ 4.188079] Code: 81 fa 00 00 04 00 0f 84 a7 00 00 00 7f 21 81 fa 80 00 00 00 0f 84 92 00 00 00 81 fa 00 00 0 [ 4.188780] EIP: [] sdhci_set_power+0x3e/0xd0 SS:ESP 0068:f644fe94 [ 4.188898] ---[ end trace a7b23eecc71777e4 ]--- This BUG() comes from the fact that ios.power_mode was still in previous value (MMC_POWER_ON) and ios.vdd was set to zero. We prevent these by inhibiting the clock gating while we update the ios structure. Both problems can be reproduced by simply running the device in a reboot loop. Signed-off-by: Mika Westerberg Reviewed-by: Linus Walleij Tested-by: Chris Ball Cc: Signed-off-by: Chris Ball --- drivers/mmc/core/core.c | 31 +++++++++++++++++++++++++++++-- 1 file changed, 29 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 63ffc65f84af..b27b94078c21 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -728,15 +728,17 @@ static inline void mmc_set_ios(struct mmc_host *host) */ void mmc_set_chip_select(struct mmc_host *host, int mode) { + mmc_host_clk_hold(host); host->ios.chip_select = mode; mmc_set_ios(host); + mmc_host_clk_release(host); } /* * Sets the host clock to the highest possible frequency that * is below "hz". */ -void mmc_set_clock(struct mmc_host *host, unsigned int hz) +static void __mmc_set_clock(struct mmc_host *host, unsigned int hz) { WARN_ON(hz < host->f_min); @@ -747,6 +749,13 @@ void mmc_set_clock(struct mmc_host *host, unsigned int hz) mmc_set_ios(host); } +void mmc_set_clock(struct mmc_host *host, unsigned int hz) +{ + mmc_host_clk_hold(host); + __mmc_set_clock(host, hz); + mmc_host_clk_release(host); +} + #ifdef CONFIG_MMC_CLKGATE /* * This gates the clock by setting it to 0 Hz. @@ -779,7 +788,7 @@ void mmc_ungate_clock(struct mmc_host *host) if (host->clk_old) { BUG_ON(host->ios.clock); /* This call will also set host->clk_gated to false */ - mmc_set_clock(host, host->clk_old); + __mmc_set_clock(host, host->clk_old); } } @@ -807,8 +816,10 @@ void mmc_set_ungated(struct mmc_host *host) */ void mmc_set_bus_mode(struct mmc_host *host, unsigned int mode) { + mmc_host_clk_hold(host); host->ios.bus_mode = mode; mmc_set_ios(host); + mmc_host_clk_release(host); } /* @@ -816,8 +827,10 @@ void mmc_set_bus_mode(struct mmc_host *host, unsigned int mode) */ void mmc_set_bus_width(struct mmc_host *host, unsigned int width) { + mmc_host_clk_hold(host); host->ios.bus_width = width; mmc_set_ios(host); + mmc_host_clk_release(host); } /** @@ -1015,8 +1028,10 @@ u32 mmc_select_voltage(struct mmc_host *host, u32 ocr) ocr &= 3 << bit; + mmc_host_clk_hold(host); host->ios.vdd = bit; mmc_set_ios(host); + mmc_host_clk_release(host); } else { pr_warning("%s: host doesn't support card's voltages\n", mmc_hostname(host)); @@ -1063,8 +1078,10 @@ int mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage, bool cmd11 */ void mmc_set_timing(struct mmc_host *host, unsigned int timing) { + mmc_host_clk_hold(host); host->ios.timing = timing; mmc_set_ios(host); + mmc_host_clk_release(host); } /* @@ -1072,8 +1089,10 @@ void mmc_set_timing(struct mmc_host *host, unsigned int timing) */ void mmc_set_driver_type(struct mmc_host *host, unsigned int drv_type) { + mmc_host_clk_hold(host); host->ios.drv_type = drv_type; mmc_set_ios(host); + mmc_host_clk_release(host); } /* @@ -1091,6 +1110,8 @@ static void mmc_power_up(struct mmc_host *host) { int bit; + mmc_host_clk_hold(host); + /* If ocr is set, we use it */ if (host->ocr) bit = ffs(host->ocr) - 1; @@ -1126,10 +1147,14 @@ static void mmc_power_up(struct mmc_host *host) * time required to reach a stable voltage. */ mmc_delay(10); + + mmc_host_clk_release(host); } static void mmc_power_off(struct mmc_host *host) { + mmc_host_clk_hold(host); + host->ios.clock = 0; host->ios.vdd = 0; @@ -1147,6 +1172,8 @@ static void mmc_power_off(struct mmc_host *host) host->ios.bus_width = MMC_BUS_WIDTH_1; host->ios.timing = MMC_TIMING_LEGACY; mmc_set_ios(host); + + mmc_host_clk_release(host); } /* From 50a50f9248497484c678631a9c1a719f1aaeab79 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 18 Aug 2011 15:23:49 +0300 Subject: [PATCH 107/326] mmc: core: use non-reentrant workqueue for clock gating The default multithread workqueue can cause the same work to be executed concurrently on a different CPUs. This isn't really suitable for clock gating as it might already gated the clock and gating it twice results both host->clk_old and host->ios.clock to be set to 0. To prevent this from happening we use system_nrt_wq instead. Signed-off-by: Mika Westerberg Reviewed-by: Linus Walleij Tested-by: Chris Ball Cc: Signed-off-by: Chris Ball --- drivers/mmc/core/host.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c index 96a26b2bf5f0..793d0a0dad8d 100644 --- a/drivers/mmc/core/host.c +++ b/drivers/mmc/core/host.c @@ -179,7 +179,7 @@ void mmc_host_clk_release(struct mmc_host *host) host->clk_requests--; if (mmc_host_may_gate_card(host->card) && !host->clk_requests) - schedule_work(&host->clk_gate_work); + queue_work(system_nrt_wq, &host->clk_gate_work); spin_unlock_irqrestore(&host->clk_lock, flags); } From b91df1593e361109f1fe665ce17c5e87ca60582b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 19 Aug 2011 10:07:07 +0900 Subject: [PATCH 108/326] mmc: sdhi: initialise mmc_data->flags before use This corrects a logic error that I introduced in "mmc: sdhi: Add write16_hook" Reported-by: Magnus Damm Signed-off-by: Simon Horman Signed-off-by: Chris Ball --- drivers/mmc/host/sh_mobile_sdhi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/host/sh_mobile_sdhi.c b/drivers/mmc/host/sh_mobile_sdhi.c index 774f6439d7ce..0c4a672f5db6 100644 --- a/drivers/mmc/host/sh_mobile_sdhi.c +++ b/drivers/mmc/host/sh_mobile_sdhi.c @@ -120,11 +120,11 @@ static int __devinit sh_mobile_sdhi_probe(struct platform_device *pdev) mmc_data->hclk = clk_get_rate(priv->clk); mmc_data->set_pwr = sh_mobile_sdhi_set_pwr; mmc_data->get_cd = sh_mobile_sdhi_get_cd; - if (mmc_data->flags & TMIO_MMC_HAS_IDLE_WAIT) - mmc_data->write16_hook = sh_mobile_sdhi_write16_hook; mmc_data->capabilities = MMC_CAP_MMC_HIGHSPEED; if (p) { mmc_data->flags = p->tmio_flags; + if (mmc_data->flags & TMIO_MMC_HAS_IDLE_WAIT) + mmc_data->write16_hook = sh_mobile_sdhi_write16_hook; mmc_data->ocr_mask = p->tmio_ocr_mask; mmc_data->capabilities |= p->tmio_caps; From 93c712f99d8e412b2d297edfe9f59b90636897c1 Mon Sep 17 00:00:00 2001 From: Subhash Jadavani Date: Tue, 9 Aug 2011 12:19:31 +0530 Subject: [PATCH 109/326] mmc: sd: UHS-I bus speed should be set last in UHS initialization mmc_sd_init_uhs_card function sets the driver type, current limit and bus speed mode on card as well as on host controller side. Currently bus speed mode is set by sending CMD6 to card and immediately setting the timing mode in host controller. But then before initiating tuning sequence, it also tries to set current limit by sending CMD6 to card which results in data timeout errors in controller if bus speed mode is SDR50/SDR104 mode. So basically bus speed mode should be set only after current limit is set in the card and immediately after setting the bus speed mode, tuning sequence should be initiated. Signed-off-by: Subhash Jadavani Reviewed-by: Arindam Nath Signed-off-by: Chris Ball --- drivers/mmc/core/sd.c | 81 ++++++++++++++++++++++++++++--------------- 1 file changed, 53 insertions(+), 28 deletions(-) diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c index 633975ff2bb3..0370e03e3142 100644 --- a/drivers/mmc/core/sd.c +++ b/drivers/mmc/core/sd.c @@ -469,56 +469,75 @@ static int sd_select_driver_type(struct mmc_card *card, u8 *status) return 0; } -static int sd_set_bus_speed_mode(struct mmc_card *card, u8 *status) +static void sd_update_bus_speed_mode(struct mmc_card *card) { - unsigned int bus_speed = 0, timing = 0; - int err; - /* * If the host doesn't support any of the UHS-I modes, fallback on * default speed. */ if (!(card->host->caps & (MMC_CAP_UHS_SDR12 | MMC_CAP_UHS_SDR25 | - MMC_CAP_UHS_SDR50 | MMC_CAP_UHS_SDR104 | MMC_CAP_UHS_DDR50))) - return 0; + MMC_CAP_UHS_SDR50 | MMC_CAP_UHS_SDR104 | MMC_CAP_UHS_DDR50))) { + card->sd_bus_speed = 0; + return; + } if ((card->host->caps & MMC_CAP_UHS_SDR104) && (card->sw_caps.sd3_bus_mode & SD_MODE_UHS_SDR104)) { - bus_speed = UHS_SDR104_BUS_SPEED; - timing = MMC_TIMING_UHS_SDR104; - card->sw_caps.uhs_max_dtr = UHS_SDR104_MAX_DTR; + card->sd_bus_speed = UHS_SDR104_BUS_SPEED; } else if ((card->host->caps & MMC_CAP_UHS_DDR50) && (card->sw_caps.sd3_bus_mode & SD_MODE_UHS_DDR50)) { - bus_speed = UHS_DDR50_BUS_SPEED; - timing = MMC_TIMING_UHS_DDR50; - card->sw_caps.uhs_max_dtr = UHS_DDR50_MAX_DTR; + card->sd_bus_speed = UHS_DDR50_BUS_SPEED; } else if ((card->host->caps & (MMC_CAP_UHS_SDR104 | MMC_CAP_UHS_SDR50)) && (card->sw_caps.sd3_bus_mode & SD_MODE_UHS_SDR50)) { - bus_speed = UHS_SDR50_BUS_SPEED; - timing = MMC_TIMING_UHS_SDR50; - card->sw_caps.uhs_max_dtr = UHS_SDR50_MAX_DTR; + card->sd_bus_speed = UHS_SDR50_BUS_SPEED; } else if ((card->host->caps & (MMC_CAP_UHS_SDR104 | MMC_CAP_UHS_SDR50 | MMC_CAP_UHS_SDR25)) && (card->sw_caps.sd3_bus_mode & SD_MODE_UHS_SDR25)) { - bus_speed = UHS_SDR25_BUS_SPEED; - timing = MMC_TIMING_UHS_SDR25; - card->sw_caps.uhs_max_dtr = UHS_SDR25_MAX_DTR; + card->sd_bus_speed = UHS_SDR25_BUS_SPEED; } else if ((card->host->caps & (MMC_CAP_UHS_SDR104 | MMC_CAP_UHS_SDR50 | MMC_CAP_UHS_SDR25 | MMC_CAP_UHS_SDR12)) && (card->sw_caps.sd3_bus_mode & SD_MODE_UHS_SDR12)) { - bus_speed = UHS_SDR12_BUS_SPEED; - timing = MMC_TIMING_UHS_SDR12; - card->sw_caps.uhs_max_dtr = UHS_SDR12_MAX_DTR; + card->sd_bus_speed = UHS_SDR12_BUS_SPEED; + } +} + +static int sd_set_bus_speed_mode(struct mmc_card *card, u8 *status) +{ + int err; + unsigned int timing = 0; + + switch (card->sd_bus_speed) { + case UHS_SDR104_BUS_SPEED: + timing = MMC_TIMING_UHS_SDR104; + card->sw_caps.uhs_max_dtr = UHS_SDR104_MAX_DTR; + break; + case UHS_DDR50_BUS_SPEED: + timing = MMC_TIMING_UHS_DDR50; + card->sw_caps.uhs_max_dtr = UHS_DDR50_MAX_DTR; + break; + case UHS_SDR50_BUS_SPEED: + timing = MMC_TIMING_UHS_SDR50; + card->sw_caps.uhs_max_dtr = UHS_SDR50_MAX_DTR; + break; + case UHS_SDR25_BUS_SPEED: + timing = MMC_TIMING_UHS_SDR25; + card->sw_caps.uhs_max_dtr = UHS_SDR25_MAX_DTR; + break; + case UHS_SDR12_BUS_SPEED: + timing = MMC_TIMING_UHS_SDR12; + card->sw_caps.uhs_max_dtr = UHS_SDR12_MAX_DTR; + break; + default: + return 0; } - card->sd_bus_speed = bus_speed; - err = mmc_sd_switch(card, 1, 0, bus_speed, status); + err = mmc_sd_switch(card, 1, 0, card->sd_bus_speed, status); if (err) return err; - if ((status[16] & 0xF) != bus_speed) + if ((status[16] & 0xF) != card->sd_bus_speed) printk(KERN_WARNING "%s: Problem setting bus speed mode!\n", mmc_hostname(card->host)); else { @@ -618,18 +637,24 @@ static int mmc_sd_init_uhs_card(struct mmc_card *card) mmc_set_bus_width(card->host, MMC_BUS_WIDTH_4); } + /* + * Select the bus speed mode depending on host + * and card capability. + */ + sd_update_bus_speed_mode(card); + /* Set the driver strength for the card */ err = sd_select_driver_type(card, status); if (err) goto out; - /* Set bus speed mode of the card */ - err = sd_set_bus_speed_mode(card, status); + /* Set current limit for the card */ + err = sd_set_current_limit(card, status); if (err) goto out; - /* Set current limit for the card */ - err = sd_set_current_limit(card, status); + /* Set bus speed mode of the card */ + err = sd_set_bus_speed_mode(card, status); if (err) goto out; From 49bb1e619568ec84785ceb366f07db2a6f0b64cc Mon Sep 17 00:00:00 2001 From: Girish K S Date: Fri, 26 Aug 2011 14:58:18 +0530 Subject: [PATCH 110/326] mmc: sdhci-s3c: Fix mmc card I/O problem This patch fixes the problem in sdhci-s3c host driver for Samsung Soc's. During the card identification stage the mmc core driver enumerates for the best bus width in combination with the highest available data rate. It starts enumerating from the highest bus width (8) to lowest width (1). In case of few MMC cards the 4-bit bus enumeration fails and tries the 1-bit bus enumeration. When switched to 1-bit bus mode the host driver has to clear the previous bus width setting and apply the new setting. The current patch will clear the previous bus mode and apply the new mode setting. Signed-off-by: Girish K S Acked-by: Jaehoon Chung Cc: Signed-off-by: Chris Ball --- drivers/mmc/host/sdhci-s3c.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c index 2bd7bf4fece7..fe886d6c474a 100644 --- a/drivers/mmc/host/sdhci-s3c.c +++ b/drivers/mmc/host/sdhci-s3c.c @@ -302,6 +302,8 @@ static int sdhci_s3c_platform_8bit_width(struct sdhci_host *host, int width) ctrl &= ~SDHCI_CTRL_8BITBUS; break; default: + ctrl &= ~SDHCI_CTRL_4BITBUS; + ctrl &= ~SDHCI_CTRL_8BITBUS; break; } From aca420bc51f48b0701963ba3a6234442a0cabebd Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Wed, 31 Aug 2011 14:45:53 -0700 Subject: [PATCH 111/326] libceph: fix leak of osd structs during shutdown We want to remove all OSDs, not just those on the idle LRU. Signed-off-by: Sage Weil --- net/ceph/osd_client.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index ce310eee708d..16836a7df7a6 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -685,6 +685,18 @@ static void __remove_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd) put_osd(osd); } +static void remove_all_osds(struct ceph_osd_client *osdc) +{ + dout("__remove_old_osds %p\n", osdc); + mutex_lock(&osdc->request_mutex); + while (!RB_EMPTY_ROOT(&osdc->osds)) { + struct ceph_osd *osd = rb_entry(rb_first(&osdc->osds), + struct ceph_osd, o_node); + __remove_osd(osdc, osd); + } + mutex_unlock(&osdc->request_mutex); +} + static void __move_osd_to_lru(struct ceph_osd_client *osdc, struct ceph_osd *osd) { @@ -701,14 +713,14 @@ static void __remove_osd_from_lru(struct ceph_osd *osd) list_del_init(&osd->o_osd_lru); } -static void remove_old_osds(struct ceph_osd_client *osdc, int remove_all) +static void remove_old_osds(struct ceph_osd_client *osdc) { struct ceph_osd *osd, *nosd; dout("__remove_old_osds %p\n", osdc); mutex_lock(&osdc->request_mutex); list_for_each_entry_safe(osd, nosd, &osdc->osd_lru, o_osd_lru) { - if (!remove_all && time_before(jiffies, osd->lru_ttl)) + if (time_before(jiffies, osd->lru_ttl)) break; __remove_osd(osdc, osd); } @@ -751,6 +763,7 @@ static void __insert_osd(struct ceph_osd_client *osdc, struct ceph_osd *new) struct rb_node *parent = NULL; struct ceph_osd *osd = NULL; + dout("__insert_osd %p osd%d\n", new, new->o_osd); while (*p) { parent = *p; osd = rb_entry(parent, struct ceph_osd, o_node); @@ -1144,7 +1157,7 @@ static void handle_osds_timeout(struct work_struct *work) dout("osds timeout\n"); down_read(&osdc->map_sem); - remove_old_osds(osdc, 0); + remove_old_osds(osdc); up_read(&osdc->map_sem); schedule_delayed_work(&osdc->osds_timeout_work, @@ -1862,8 +1875,7 @@ void ceph_osdc_stop(struct ceph_osd_client *osdc) ceph_osdmap_destroy(osdc->osdmap); osdc->osdmap = NULL; } - remove_old_osds(osdc, 1); - WARN_ON(!RB_EMPTY_ROOT(&osdc->osds)); + remove_all_osds(osdc); mempool_destroy(osdc->req_mempool); ceph_msgpool_destroy(&osdc->msgpool_op); ceph_msgpool_destroy(&osdc->msgpool_op_reply); From 866e4ed77448a0c311e1b055eb72ea05423fd799 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 27 Aug 2011 05:57:44 +0000 Subject: [PATCH 112/326] xfs: fix xfs_mark_inode_dirty during umount During umount we do not add a dirty inode to the lru and wait for it to become clean first, but force writeback of data and metadata with I_WILL_FREE set. Currently there is no way for XFS to detect that the inode has been redirtied for metadata operations, as we skip the mark_inode_dirty call during teardown. Fix this by setting i_update_core nanually in that case, so that the inode gets flushed during inode reclaim. Alternatively we could enable calling mark_inode_dirty for inodes in I_WILL_FREE state, and let the VFS dirty tracking handle this. I decided against this as we will get better I/O patterns from reclaim compared to the synchronous writeout in write_inode_now, and always marking the inode dirty in some way from xfs_mark_inode_dirty is a better safetly net in either case. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Signed-off-by: Alex Elder (cherry picked from commit da6742a5a4cc844a9982fdd936ddb537c0747856) Signed-off-by: Alex Elder --- fs/xfs/xfs_iops.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index b9c172b3fbbe..673704fab748 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -70,9 +70,8 @@ xfs_synchronize_times( } /* - * If the linux inode is valid, mark it dirty. - * Used when committing a dirty inode into a transaction so that - * the inode will get written back by the linux code + * If the linux inode is valid, mark it dirty, else mark the dirty state + * in the XFS inode to make sure we pick it up when reclaiming the inode. */ void xfs_mark_inode_dirty_sync( @@ -82,6 +81,10 @@ xfs_mark_inode_dirty_sync( if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) mark_inode_dirty_sync(inode); + else { + barrier(); + ip->i_update_core = 1; + } } void @@ -92,6 +95,11 @@ xfs_mark_inode_dirty( if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) mark_inode_dirty(inode); + else { + barrier(); + ip->i_update_core = 1; + } + } /* From d312ae878b6aed3912e1acaaf5d0b2a9d08a4f11 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Fri, 19 Aug 2011 15:57:16 +0100 Subject: [PATCH 113/326] xen: use maximum reservation to limit amount of usable RAM Use the domain's maximum reservation to limit the amount of extra RAM for the memory balloon. This reduces the size of the pages tables and the amount of reserved low memory (which defaults to about 1/32 of the total RAM). On a system with 8 GiB of RAM with the domain limited to 1 GiB the kernel reports: Before: Memory: 627792k/4472000k available After: Memory: 549740k/11132224k available A increase of about 76 MiB (~1.5% of the unused 7 GiB). The reserved low memory is also reduced from 253 MiB to 32 MiB. The total additional usable RAM is 329 MiB. For dom0, this requires at patch to Xen ('x86: use 'dom0_mem' to limit the number of pages for dom0') (c/s 23790) CC: stable@kernel.org Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/setup.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 02ffd9e48c9f..ff3dfa176814 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -183,6 +183,19 @@ static unsigned long __init xen_set_identity(const struct e820entry *list, PFN_UP(start_pci), PFN_DOWN(last)); return identity; } + +static unsigned long __init xen_get_max_pages(void) +{ + unsigned long max_pages = MAX_DOMAIN_PAGES; + domid_t domid = DOMID_SELF; + int ret; + + ret = HYPERVISOR_memory_op(XENMEM_maximum_reservation, &domid); + if (ret > 0) + max_pages = ret; + return min(max_pages, MAX_DOMAIN_PAGES); +} + /** * machine_specific_memory_setup - Hook for machine specific memory setup. **/ @@ -291,6 +304,12 @@ char * __init xen_memory_setup(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); + extra_limit = xen_get_max_pages(); + if (extra_limit >= max_pfn) + extra_pages = extra_limit - max_pfn; + else + extra_pages = 0; + extra_pages += xen_return_unused_memory(xen_start_info->nr_pages, &e820); /* From 58d84c4ee0389ddeb86238d5d8359a982c9f7a5b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sat, 27 Aug 2011 05:57:55 +0000 Subject: [PATCH 114/326] xfs: fix ->write_inode return values Currently we always redirty an inode that was attempted to be written out synchronously but has been cleaned by an AIL pushed internall, which is rather bogus. Fix that by doing the i_update_core check early on and return 0 for it. Also include async calls for it, as doing any work for those is just as pointless. While we're at it also fix the sign for the EIO return in case of a filesystem shutdown, and fix the completely non-sensical locking around xfs_log_inode. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Signed-off-by: Alex Elder (cherry picked from commit 297db93bb74cf687510313eb235a7aec14d67e97) Signed-off-by: Alex Elder --- fs/xfs/xfs_super.c | 34 +++++++++------------------------- 1 file changed, 9 insertions(+), 25 deletions(-) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index c1b022f20d35..2366c54cc4fa 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -879,33 +879,17 @@ xfs_log_inode( struct xfs_trans *tp; int error; - xfs_iunlock(ip, XFS_ILOCK_SHARED); tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); - if (error) { xfs_trans_cancel(tp, 0); - /* we need to return with the lock hold shared */ - xfs_ilock(ip, XFS_ILOCK_SHARED); return error; } xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * Note - it's possible that we might have pushed ourselves out of the - * way during trans_reserve which would flush the inode. But there's - * no guarantee that the inode buffer has actually gone out yet (it's - * delwri). Plus the buffer could be pinned anyway if it's part of - * an inode in another recent transaction. So we play it safe and - * fire off the transaction anyway. - */ - xfs_trans_ijoin(tp, ip); + xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - error = xfs_trans_commit(tp, 0); - xfs_ilock_demote(ip, XFS_ILOCK_EXCL); - - return error; + return xfs_trans_commit(tp, 0); } STATIC int @@ -920,7 +904,9 @@ xfs_fs_write_inode( trace_xfs_write_inode(ip); if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); + return -XFS_ERROR(EIO); + if (!ip->i_update_core) + return 0; if (wbc->sync_mode == WB_SYNC_ALL) { /* @@ -931,12 +917,10 @@ xfs_fs_write_inode( * of synchronous log foces dramatically. */ xfs_ioend_wait(ip); - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (ip->i_update_core) { - error = xfs_log_inode(ip); - if (error) - goto out_unlock; - } + error = xfs_log_inode(ip); + if (error) + goto out; + return 0; } else { /* * We make this non-blocking if the inode is contended, return From d198d499148a0c64a41b3aba9e7dd43772832b91 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Thu, 1 Sep 2011 13:46:55 +0200 Subject: [PATCH 115/326] xen: x86_32: do not enable iterrupts when returning from exception in interrupt context If vmalloc page_fault happens inside of interrupt handler with interrupts disabled then on exit path from exception handler when there is no pending interrupts, the following code (arch/x86/xen/xen-asm_32.S:112): cmpw $0x0001, XEN_vcpu_info_pending(%eax) sete XEN_vcpu_info_mask(%eax) will enable interrupts even if they has been previously disabled according to eflags from the bounce frame (arch/x86/xen/xen-asm_32.S:99) testb $X86_EFLAGS_IF>>8, 8+1+ESP_OFFSET(%esp) setz XEN_vcpu_info_mask(%eax) Solution is in setting XEN_vcpu_info_mask only when it should be set according to cmpw $0x0001, XEN_vcpu_info_pending(%eax) but not clearing it if there isn't any pending events. Reproducer for bug is attached to RHBZ 707552 CC: stable@kernel.org Signed-off-by: Igor Mammedov Acked-by: Jeremy Fitzhardinge Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/xen-asm_32.S | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S index 22a2093b5862..b040b0e518ca 100644 --- a/arch/x86/xen/xen-asm_32.S +++ b/arch/x86/xen/xen-asm_32.S @@ -113,11 +113,13 @@ xen_iret_start_crit: /* * If there's something pending, mask events again so we can - * jump back into xen_hypervisor_callback + * jump back into xen_hypervisor_callback. Otherwise do not + * touch XEN_vcpu_info_mask. */ - sete XEN_vcpu_info_mask(%eax) + jne 1f + movb $1, XEN_vcpu_info_mask(%eax) - popl %eax +1: popl %eax /* * From this point on the registers are restored and the stack From ed467e69f16e6b480e2face7bc5963834d025f91 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 1 Sep 2011 09:48:27 -0400 Subject: [PATCH 116/326] xen/smp: Warn user why they keel over - nosmp or noapic and what to use instead. We have hit a couple of customer bugs where they would like to use those parameters to run an UP kernel - but both of those options turn of important sources of interrupt information so we end up not being able to boot. The correct way is to pass in 'dom0_max_vcpus=1' on the Xen hypervisor line and the kernel will patch itself to be a UP kernel. Fixes bug: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=637308 CC: stable@kernel.org Acked-by: Ian Campbell Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/smp.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index e79dbb95482b..d4fc6d454f8d 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -32,6 +32,7 @@ #include #include +#include #include "xen-ops.h" #include "mmu.h" @@ -207,6 +208,15 @@ static void __init xen_smp_prepare_cpus(unsigned int max_cpus) unsigned cpu; unsigned int i; + if (skip_ioapic_setup) { + char *m = (max_cpus == 0) ? + "The nosmp parameter is incompatible with Xen; " \ + "use Xen dom0_max_vcpus=1 parameter" : + "The noapic parameter is incompatible with Xen"; + + xen_raw_printk(m); + panic(m); + } xen_init_lock_cpu(0); smp_store_cpu_info(0); From d054ac16eeb658bccadb06b12c39cee22243b10f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 1 Sep 2011 17:46:15 +0000 Subject: [PATCH 117/326] drm/radeon/kms: make sure pci max read request size is valid on evergreen+ (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If the bios or OS sets the pci max read request size to 0 or an invalid value (6,7), it can result in a hang or slowdown. Check and set it to something sane if it's invalid. Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=42162 v2: use pci reg defines from include/linux/pci_regs.h Signed-off-by: Alex Deucher Cc: stable@kernel.org Reviewed-by: Michel Dänzer Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen.c | 27 +++++++++++++++++++++++++++ drivers/gpu/drm/radeon/ni.c | 3 +++ 2 files changed, 30 insertions(+) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index d8d71a399f52..dc0a5b56c81a 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -41,6 +41,31 @@ static void evergreen_gpu_init(struct radeon_device *rdev); void evergreen_fini(struct radeon_device *rdev); static void evergreen_pcie_gen2_enable(struct radeon_device *rdev); +void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev) +{ + u16 ctl, v; + int cap, err; + + cap = pci_pcie_cap(rdev->pdev); + if (!cap) + return; + + err = pci_read_config_word(rdev->pdev, cap + PCI_EXP_DEVCTL, &ctl); + if (err) + return; + + v = (ctl & PCI_EXP_DEVCTL_READRQ) >> 12; + + /* if bios or OS sets MAX_READ_REQUEST_SIZE to an invalid value, fix it + * to avoid hangs or perfomance issues + */ + if ((v == 0) || (v == 6) || (v == 7)) { + ctl &= ~PCI_EXP_DEVCTL_READRQ; + ctl |= (2 << 12); + pci_write_config_word(rdev->pdev, cap + PCI_EXP_DEVCTL, ctl); + } +} + void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc) { /* enable the pflip int */ @@ -1863,6 +1888,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev) WREG32(GRBM_CNTL, GRBM_READ_TIMEOUT(0xff)); + evergreen_fix_pci_max_read_req_size(rdev); + cc_gc_shader_pipe_config = RREG32(CC_GC_SHADER_PIPE_CONFIG) & ~2; cc_gc_shader_pipe_config |= diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index a2e00fa9c618..cbf57d75d925 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -39,6 +39,7 @@ extern int evergreen_mc_wait_for_idle(struct radeon_device *rdev); extern void evergreen_mc_program(struct radeon_device *rdev); extern void evergreen_irq_suspend(struct radeon_device *rdev); extern int evergreen_mc_init(struct radeon_device *rdev); +extern void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev); #define EVERGREEN_PFP_UCODE_SIZE 1120 #define EVERGREEN_PM4_UCODE_SIZE 1376 @@ -669,6 +670,8 @@ static void cayman_gpu_init(struct radeon_device *rdev) WREG32(GRBM_CNTL, GRBM_READ_TIMEOUT(0xff)); + evergreen_fix_pci_max_read_req_size(rdev); + mc_shared_chmap = RREG32(MC_SHARED_CHMAP); mc_arb_ramcfg = RREG32(MC_ARB_RAMCFG); From f1ca1512e765337a7c09eb875eedef8ea4e07654 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Fri, 2 Sep 2011 14:10:32 +0200 Subject: [PATCH 118/326] iommu/amd: Make sure iommu->need_sync contains correct value The value is only set to true but never set back to false, which causes to many completion-wait commands to be sent to hardware. Fix it with this patch. Cc: stable@kernel.org Signed-off-by: Joerg Roedel --- drivers/iommu/amd_iommu.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index a14f8dc23462..45652231dae8 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -605,7 +605,9 @@ static void build_inv_all(struct iommu_cmd *cmd) * Writes the command to the IOMMUs command buffer and informs the * hardware about the new command. */ -static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) +static int iommu_queue_command_sync(struct amd_iommu *iommu, + struct iommu_cmd *cmd, + bool sync) { u32 left, tail, head, next_tail; unsigned long flags; @@ -639,13 +641,18 @@ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) copy_cmd_to_buffer(iommu, cmd, tail); /* We need to sync now to make sure all commands are processed */ - iommu->need_sync = true; + iommu->need_sync = sync; spin_unlock_irqrestore(&iommu->lock, flags); return 0; } +static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) +{ + return iommu_queue_command_sync(iommu, cmd, true); +} + /* * This function queues a completion wait command into the command * buffer of an IOMMU @@ -661,7 +668,7 @@ static int iommu_completion_wait(struct amd_iommu *iommu) build_completion_wait(&cmd, (u64)&sem); - ret = iommu_queue_command(iommu, &cmd); + ret = iommu_queue_command_sync(iommu, &cmd, false); if (ret) return ret; From e33acde91140f1809952d1c135c36feb66a51887 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Fri, 2 Sep 2011 14:19:50 +0200 Subject: [PATCH 119/326] iommu/amd: Don't take domain->lock recursivly The domain_flush_devices() function takes the domain->lock. But this function is only called from update_domain() which itself is already called unter the domain->lock. This causes a deadlock situation when the dma-address-space of a domain grows larger than 1GB. Cc: stable@kernel.org Signed-off-by: Joerg Roedel --- drivers/iommu/amd_iommu.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 45652231dae8..0e4227f457af 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -847,14 +847,9 @@ static void domain_flush_complete(struct protection_domain *domain) static void domain_flush_devices(struct protection_domain *domain) { struct iommu_dev_data *dev_data; - unsigned long flags; - - spin_lock_irqsave(&domain->lock, flags); list_for_each_entry(dev_data, &domain->dev_list, list) device_flush_dte(dev_data); - - spin_unlock_irqrestore(&domain->lock, flags); } /**************************************************************************** From fb492c9160f3d40d09456a79cc669fba74d7d9cc Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Tue, 30 Aug 2011 17:45:10 +0100 Subject: [PATCH 120/326] ARM: 7067/1: mm: keep significant bits in pfn_valid When ARCH_HAS_HOLES_MEMORYMODEL is selected, pfn_valid calls memblock_is_memory to test validity of a pfn: > memblock_is_memory(pfn << PAGE_SHIFT); On LPAE systems this cuts off the top bits, as the shift occurs before the value is promoted to a phys_addr_t. This patch replaces the shift with a call to __pfn_to_phys (which casts pfn to phys_addr_t before shifting), preventing the loss of significant bits. Signed-off-by: Mark Rutland Acked-by: Will Deacon Signed-off-by: Russell King --- arch/arm/mm/init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index 91bca355cd31..cc7e2d8be9aa 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -298,7 +298,7 @@ static void __init arm_bootmem_free(unsigned long min, unsigned long max_low, #ifdef CONFIG_HAVE_ARCH_PFN_VALID int pfn_valid(unsigned long pfn) { - return memblock_is_memory(pfn << PAGE_SHIFT); + return memblock_is_memory(__pfn_to_phys(pfn)); } EXPORT_SYMBOL(pfn_valid); #endif From ddf28352b80c86754a6424e3a61e8bdf9213b3c7 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 4 Sep 2011 15:45:10 -0700 Subject: [PATCH 121/326] Linux 3.1-rc5 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index c3e90c530a65..03d97aa8c73e 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = -rc5 NAME = "Divemaster Edition" # *DOCUMENTATION* From 10167873a415ba642aa2eee0c310ebd5a4633573 Mon Sep 17 00:00:00 2001 From: Paul Walmsley Date: Sun, 4 Sep 2011 20:20:53 -0600 Subject: [PATCH 122/326] OMAP2430: hwmod: musb: add missing terminator to omap2430_usbhsotg_addrs[] Add a missing array terminator to omap2430_usbhsotg_addrs[]. Without this terminator, the omap_hwmod resource building code runs off the end of the array, resulting in at least this error -- if not worse behavior: [ 0.578002] musb-omap2430: failed to claim resource 4 [ 0.583465] omap_device: musb-omap2430: build failed (-16) [ 0.589294] Could not build omap_device for musb-omap2430 usb_otg_hs This should have been part of commit 78183f3fdf76f422431a81852468be01b36db325 ("omap_hwmod: use a null structure record to terminate omap_hwmod_addr_space arrays") but was evidently missed. Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_2430_data.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-omap2/omap_hwmod_2430_data.c b/arch/arm/mach-omap2/omap_hwmod_2430_data.c index 16743c7d6e8e..408193d8e044 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2430_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2430_data.c @@ -192,6 +192,7 @@ static struct omap_hwmod_addr_space omap2430_usbhsotg_addrs[] = { .pa_end = OMAP243X_HS_BASE + SZ_4K - 1, .flags = ADDR_TYPE_RT }, + { } }; /* l4_core ->usbhsotg interface */ From 1df726ef0a700587a712a3660b2caa8e533c7de9 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 5 Sep 2011 08:58:29 +0100 Subject: [PATCH 123/326] NET: am79c961: fix race in link status code The link status code operates from a timer, and writes the index register without first taking a lock. A well-placed interrupt between writing the index register and reading the data register could change the index register on us, which will return wrong data. Add the necessary lock. Signed-off-by: Russell King --- drivers/net/arm/am79c961a.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/arm/am79c961a.c b/drivers/net/arm/am79c961a.c index 52fe21e1e2cd..3b1416e3d217 100644 --- a/drivers/net/arm/am79c961a.c +++ b/drivers/net/arm/am79c961a.c @@ -308,8 +308,11 @@ static void am79c961_timer(unsigned long data) struct net_device *dev = (struct net_device *)data; struct dev_priv *priv = netdev_priv(dev); unsigned int lnkstat, carrier; + unsigned long flags; + spin_lock_irqsave(&priv->chip_lock, flags); lnkstat = read_ireg(dev->base_addr, ISALED0) & ISALED0_LNKST; + spin_unlock_irqrestore(&priv->chip_lock, flags); carrier = netif_carrier_ok(dev); if (lnkstat && !carrier) { From da063d260969c4e5e5f91d911ba87f7f6b48ead0 Mon Sep 17 00:00:00 2001 From: Per Forlin Date: Mon, 29 Aug 2011 13:33:32 +0200 Subject: [PATCH 124/326] dmaengine/ste_dma40: add missing kernel doc for pending_queue Signed-off-by: Per Forlin Acked-by: Linus Walleij Signed-off-by: Vinod Koul --- drivers/dma/ste_dma40.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c index cd3a7c726bf8..486b6c0b44e3 100644 --- a/drivers/dma/ste_dma40.c +++ b/drivers/dma/ste_dma40.c @@ -174,6 +174,7 @@ struct d40_base; * @tasklet: Tasklet that gets scheduled from interrupt context to complete a * transfer and call client callback. * @client: Cliented owned descriptor list. + * @pending_queue: Submitted jobs, to be issued by issue_pending() * @active: Active descriptor. * @queue: Queued jobs. * @dma_cfg: The client configuration of this dma channel. From 3b3d5b0f855b3eec45a02832e97c3c1890ff8823 Mon Sep 17 00:00:00 2001 From: Per Forlin Date: Mon, 29 Aug 2011 13:33:33 +0200 Subject: [PATCH 125/326] dmaengine/ste_dma40: remove duplicate call to d40_pool_lli_free(). d40_desc_free() already calls d40_pool_lli_free(). Signed-off-by: Per Forlin Acked-by: Linus Walleij Signed-off-by: Vinod Koul --- drivers/dma/ste_dma40.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c index 486b6c0b44e3..37388d10497a 100644 --- a/drivers/dma/ste_dma40.c +++ b/drivers/dma/ste_dma40.c @@ -478,7 +478,6 @@ static struct d40_desc *d40_desc_get(struct d40_chan *d40c) list_for_each_entry_safe(d, _d, &d40c->client, node) if (async_tx_test_ack(&d->txd)) { - d40_pool_lli_free(d40c, d); d40_desc_remove(d); desc = d; memset(desc, 0, sizeof(*desc)); @@ -1209,7 +1208,6 @@ static void dma_tasklet(unsigned long data) if (!d40d->cyclic) { if (async_tx_test_ack(&d40d->txd)) { - d40_pool_lli_free(d40c, d40d); d40_desc_remove(d40d); d40_desc_free(d40c, d40d); } else { @@ -1606,7 +1604,6 @@ static int d40_free_dma(struct d40_chan *d40c) /* Release client owned descriptors */ if (!list_empty(&d40c->client)) list_for_each_entry_safe(d, _d, &d40c->client, node) { - d40_pool_lli_free(d40c, d); d40_desc_remove(d); d40_desc_free(d40c, d); } From 7404368c22b4910ab839238e48d96be45180f6fc Mon Sep 17 00:00:00 2001 From: Per Forlin Date: Mon, 29 Aug 2011 13:33:34 +0200 Subject: [PATCH 126/326] dmaengine/ste_dma40: fix Oops due to double free of client descriptor The client list may exist in two lists at the same time. This makes free fail since the same desc is freed multiple times. Remove desc from client list when adding it to the pending queue. Move free of client owned descriptors from free_dma() to terminate_all(). Unable to handle kernel paging request at virtual address 00100104 pgd = dea8c000 [00100104] *pgd=1ea62831, *pte=00000000, *ppte=00000000 Internal error: Oops: 817 [#1] PREEMPT SMP Modules linked in: CPU: 0 Not tainted (3.1.0-rc3+ #58) PC is at d40_free_chan_resources+0x64/0x330 Signed-off-by: Per Forlin Acked-by: Linus Walleij Signed-off-by: Vinod Koul --- drivers/dma/ste_dma40.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c index 37388d10497a..92ec0a26401a 100644 --- a/drivers/dma/ste_dma40.c +++ b/drivers/dma/ste_dma40.c @@ -644,8 +644,11 @@ static struct d40_desc *d40_first_active_get(struct d40_chan *d40c) return d; } +/* remove desc from current queue and add it to the pending_queue */ static void d40_desc_queue(struct d40_chan *d40c, struct d40_desc *desc) { + d40_desc_remove(desc); + desc->is_in_client_list = false; list_add_tail(&desc->node, &d40c->pending_queue); } @@ -803,6 +806,7 @@ static int d40_channel_execute_command(struct d40_chan *d40c, static void d40_term_all(struct d40_chan *d40c) { struct d40_desc *d40d; + struct d40_desc *_d; /* Release active descriptors */ while ((d40d = d40_first_active_get(d40c))) { @@ -822,6 +826,14 @@ static void d40_term_all(struct d40_chan *d40c) d40_desc_free(d40c, d40d); } + /* Release client owned descriptors */ + if (!list_empty(&d40c->client)) + list_for_each_entry_safe(d40d, _d, &d40c->client, node) { + d40_desc_remove(d40d); + d40_desc_free(d40c, d40d); + } + + d40c->pending_tx = 0; d40c->busy = false; } @@ -1594,20 +1606,10 @@ static int d40_free_dma(struct d40_chan *d40c) u32 event; struct d40_phy_res *phy = d40c->phy_chan; bool is_src; - struct d40_desc *d; - struct d40_desc *_d; - /* Terminate all queued and active transfers */ d40_term_all(d40c); - /* Release client owned descriptors */ - if (!list_empty(&d40c->client)) - list_for_each_entry_safe(d, _d, &d40c->client, node) { - d40_desc_remove(d); - d40_desc_free(d40c, d); - } - if (phy == NULL) { chan_err(d40c, "phy == null\n"); return -EINVAL; From 82babbb361f207a80cffa8ac34c2b6a0b62acc88 Mon Sep 17 00:00:00 2001 From: Per Forlin Date: Mon, 29 Aug 2011 13:33:35 +0200 Subject: [PATCH 127/326] dmaengine/ste_dma40: fix memory leak due to prepared descriptors Prepared descriptors that are not submitted will not be freed. Add prepared descriptor to a list to be able to release them upon dmaengine_terminate_all(). Signed-off-by: Per Forlin Acked-by: Linus Walleij Signed-off-by: Vinod Koul --- drivers/dma/ste_dma40.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c index 92ec0a26401a..467e4dcb20a0 100644 --- a/drivers/dma/ste_dma40.c +++ b/drivers/dma/ste_dma40.c @@ -177,6 +177,7 @@ struct d40_base; * @pending_queue: Submitted jobs, to be issued by issue_pending() * @active: Active descriptor. * @queue: Queued jobs. + * @prepare_queue: Prepared jobs. * @dma_cfg: The client configuration of this dma channel. * @configured: whether the dma_cfg configuration is valid * @base: Pointer to the device instance struct. @@ -204,6 +205,7 @@ struct d40_chan { struct list_head pending_queue; struct list_head active; struct list_head queue; + struct list_head prepare_queue; struct stedma40_chan_cfg dma_cfg; bool configured; struct d40_base *base; @@ -833,6 +835,13 @@ static void d40_term_all(struct d40_chan *d40c) d40_desc_free(d40c, d40d); } + /* Release descriptors in prepare queue */ + if (!list_empty(&d40c->prepare_queue)) + list_for_each_entry_safe(d40d, _d, + &d40c->prepare_queue, node) { + d40_desc_remove(d40d); + d40_desc_free(d40c, d40d); + } d40c->pending_tx = 0; d40c->busy = false; @@ -1911,6 +1920,12 @@ d40_prep_sg(struct dma_chan *dchan, struct scatterlist *sg_src, goto err; } + /* + * add descriptor to the prepare queue in order to be able + * to free them later in terminate_all + */ + list_add_tail(&desc->node, &chan->prepare_queue); + spin_unlock_irqrestore(&chan->lock, flags); return &desc->txd; @@ -2400,6 +2415,7 @@ static void __init d40_chan_init(struct d40_base *base, struct dma_device *dma, INIT_LIST_HEAD(&d40c->queue); INIT_LIST_HEAD(&d40c->pending_queue); INIT_LIST_HEAD(&d40c->client); + INIT_LIST_HEAD(&d40c->prepare_queue); tasklet_init(&d40c->tasklet, dma_tasklet, (unsigned long) d40c); From d7cb6667090511755fc8bb294982783b087baef7 Mon Sep 17 00:00:00 2001 From: Jonas Bonn Date: Sat, 30 Jul 2011 16:15:42 +0200 Subject: [PATCH 128/326] openrisc: don't use pt_regs in struct sigcontext As it was decided not to export struct pt_regs to userspace, struct sigcontext shouldn't be using it either. The pt_regs struct for OpenRISC is kernel internal and the layout of the registers may change in the future. The struct user_regs_struct is what is guaranteed to remain stable, so struct sigcontext may use that instead. This patch removes the usage of struct pt_regs in struct sigcontext and makes according changes in signal.c to get the register layout right. The usp field is removed from the sigcontext structure as this information is already contained in the user_regs_struct. Signed-off-by: Jonas Bonn Reviewed-by: Emilio Cota --- arch/openrisc/include/asm/sigcontext.h | 7 +------ arch/openrisc/kernel/signal.c | 29 ++++++++++---------------- 2 files changed, 12 insertions(+), 24 deletions(-) diff --git a/arch/openrisc/include/asm/sigcontext.h b/arch/openrisc/include/asm/sigcontext.h index 54a5c50132e3..b79c2b19afbe 100644 --- a/arch/openrisc/include/asm/sigcontext.h +++ b/arch/openrisc/include/asm/sigcontext.h @@ -23,16 +23,11 @@ /* This struct is saved by setup_frame in signal.c, to keep the current context while a signal handler is executed. It's restored by sys_sigreturn. - - To keep things simple, we use pt_regs here even though normally you just - specify the list of regs to save. Then we can use copy_from_user on the - entire regs instead of a bunch of get_user's as well... */ struct sigcontext { - struct pt_regs regs; /* needs to be first */ + struct user_regs_struct regs; /* needs to be first */ unsigned long oldmask; - unsigned long usp; /* usp before stacking this gunk on it */ }; #endif /* __ASM_OPENRISC_SIGCONTEXT_H */ diff --git a/arch/openrisc/kernel/signal.c b/arch/openrisc/kernel/signal.c index 5f759c76834e..95207ab0c99e 100644 --- a/arch/openrisc/kernel/signal.c +++ b/arch/openrisc/kernel/signal.c @@ -52,31 +52,25 @@ struct rt_sigframe { static int restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc) { unsigned int err = 0; - unsigned long old_usp; /* Alwys make any pending restarted system call return -EINTR */ current_thread_info()->restart_block.fn = do_no_restart_syscall; - /* restore the regs from &sc->regs (same as sc, since regs is first) + /* + * Restore the regs from &sc->regs. * (sc is already checked for VERIFY_READ since the sigframe was * checked in sys_sigreturn previously) */ - - if (__copy_from_user(regs, sc, sizeof(struct pt_regs))) + if (__copy_from_user(regs, sc->regs.gpr, 32 * sizeof(unsigned long))) + goto badframe; + if (__copy_from_user(®s->pc, &sc->regs.pc, sizeof(unsigned long))) + goto badframe; + if (__copy_from_user(®s->sr, &sc->regs.sr, sizeof(unsigned long))) goto badframe; /* make sure the SM-bit is cleared so user-mode cannot fool us */ regs->sr &= ~SPR_SR_SM; - /* restore the old USP as it was before we stacked the sc etc. - * (we cannot just pop the sigcontext since we aligned the sp and - * stuff after pushing it) - */ - - err |= __get_user(old_usp, &sc->usp); - - regs->sp = old_usp; - /* TODO: the other ports use regs->orig_XX to disable syscall checks * after this completes, but we don't use that mechanism. maybe we can * use it now ? @@ -137,18 +131,17 @@ static int setup_sigcontext(struct sigcontext *sc, struct pt_regs *regs, unsigned long mask) { int err = 0; - unsigned long usp = regs->sp; - /* copy the regs. they are first in sc so we can use sc directly */ + /* copy the regs */ - err |= __copy_to_user(sc, regs, sizeof(struct pt_regs)); + err |= __copy_to_user(sc->regs.gpr, regs, 32 * sizeof(unsigned long)); + err |= __copy_to_user(&sc->regs.pc, ®s->pc, sizeof(unsigned long)); + err |= __copy_to_user(&sc->regs.sr, ®s->sr, sizeof(unsigned long)); /* then some other stuff */ err |= __put_user(mask, &sc->oldmask); - err |= __put_user(usp, &sc->usp); - return err; } From 5204f5e3f5b3c706e52682590de5974a82ea54f9 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Mon, 5 Sep 2011 08:07:47 -0700 Subject: [PATCH 129/326] regmap: Remove bitrotted module_put()s The conversion to per bus type registration functions means we don't need to do module_get()s to hold the bus types in memory (their users will link to them) so we removed all those calls. This left module_put() calls in the cleanup paths which aren't needed and which cause unbalanced puts if we ever try to unload anything. Reported-by: Jonathan Cameron Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 0eef4da1ac61..20663f8dae45 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -168,13 +168,11 @@ struct regmap *regmap_init(struct device *dev, map->work_buf = kmalloc(map->format.buf_size, GFP_KERNEL); if (map->work_buf == NULL) { ret = -ENOMEM; - goto err_bus; + goto err_map; } return map; -err_bus: - module_put(map->bus->owner); err_map: kfree(map); err: @@ -188,7 +186,6 @@ EXPORT_SYMBOL_GPL(regmap_init); void regmap_exit(struct regmap *map) { kfree(map->work_buf); - module_put(map->bus->owner); kfree(map); } EXPORT_SYMBOL_GPL(regmap_exit); From 747da0f80e566500421bd7760b2e050fea3fde5e Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Sun, 4 Sep 2011 08:18:18 -0700 Subject: [PATCH 130/326] ASoC: Fix reporting of partial jack updates We need to report the entire jack state to the core jack code, not just the bits that were being updated by the caller, otherwise the status reported by other detection methods will be omitted from the state seen by userspace. Signed-off-by: Mark Brown Acked-by: Liam Girdwood Cc: stable@kernel.org --- sound/soc/soc-jack.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/soc-jack.c b/sound/soc/soc-jack.c index 38b00131b2fe..fa31d9c2abd8 100644 --- a/sound/soc/soc-jack.c +++ b/sound/soc/soc-jack.c @@ -105,7 +105,7 @@ void snd_soc_jack_report(struct snd_soc_jack *jack, int status, int mask) snd_soc_dapm_sync(dapm); - snd_jack_report(jack->jack, status); + snd_jack_report(jack->jack, jack->status); out: mutex_unlock(&codec->mutex); From c5d2e650bd805a00ff9af537d5b5dede598a198c Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Mon, 5 Sep 2011 13:49:57 +0200 Subject: [PATCH 131/326] ASoC: Blackfin: bf5xx-ad193x: Fix codec device name Fix the codec_name field of the dai_link to match the actual device name of the codec. Otherwise the card won't be instantiated. Signed-off-by: Lars-Peter Clausen Acked-by: Liam Girdwood Signed-off-by: Mark Brown Cc: stable@kernel.org --- sound/soc/blackfin/bf5xx-ad193x.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/blackfin/bf5xx-ad193x.c b/sound/soc/blackfin/bf5xx-ad193x.c index a118a0fb9d81..5956584ea3a4 100644 --- a/sound/soc/blackfin/bf5xx-ad193x.c +++ b/sound/soc/blackfin/bf5xx-ad193x.c @@ -103,7 +103,7 @@ static struct snd_soc_dai_link bf5xx_ad193x_dai[] = { .cpu_dai_name = "bfin-tdm.0", .codec_dai_name ="ad193x-hifi", .platform_name = "bfin-tdm-pcm-audio", - .codec_name = "ad193x.5", + .codec_name = "spi0.5", .ops = &bf5xx_ad193x_ops, }, { @@ -112,7 +112,7 @@ static struct snd_soc_dai_link bf5xx_ad193x_dai[] = { .cpu_dai_name = "bfin-tdm.1", .codec_dai_name ="ad193x-hifi", .platform_name = "bfin-tdm-pcm-audio", - .codec_name = "ad193x.5", + .codec_name = "spi0.5", .ops = &bf5xx_ad193x_ops, }, }; From b06947b50053f2d21ad8ddf218cdb64fc8026896 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 2 Sep 2011 14:23:09 +0000 Subject: [PATCH 132/326] drm/radeon/kms: fix DP detect and EDID fetch for DP bridges Sink type is always DP for DP bridges and EDID fetch on DP bridges is always i2c over aux rather than plain i2c. Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_connectors.c | 37 ++++++++++++++-------- drivers/gpu/drm/radeon/radeon_display.c | 19 +++++++---- 2 files changed, 36 insertions(+), 20 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 4f0c1ecac72e..c4b8741dbf58 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1297,12 +1297,33 @@ radeon_dp_detect(struct drm_connector *connector, bool force) if (!radeon_dig_connector->edp_on) atombios_set_edp_panel_power(connector, ATOM_TRANSMITTER_ACTION_POWER_OFF); - } else { - /* need to setup ddc on the bridge */ - if (radeon_connector_encoder_is_dp_bridge(connector)) { + } else if (radeon_connector_encoder_is_dp_bridge(connector)) { + /* DP bridges are always DP */ + radeon_dig_connector->dp_sink_type = CONNECTOR_OBJECT_ID_DISPLAYPORT; + /* get the DPCD from the bridge */ + radeon_dp_getdpcd(radeon_connector); + + if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) + ret = connector_status_connected; + else { + /* need to setup ddc on the bridge */ if (encoder) radeon_atom_ext_encoder_setup_ddc(encoder); + if (radeon_ddc_probe(radeon_connector, + radeon_connector->requires_extended_probe)) + ret = connector_status_connected; } + + if ((ret == connector_status_disconnected) && + radeon_connector->dac_load_detect) { + struct drm_encoder *encoder = radeon_best_single_encoder(connector); + struct drm_encoder_helper_funcs *encoder_funcs; + if (encoder) { + encoder_funcs = encoder->helper_private; + ret = encoder_funcs->detect(encoder, connector); + } + } + } else { radeon_dig_connector->dp_sink_type = radeon_dp_getsinktype(radeon_connector); if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) { ret = connector_status_connected; @@ -1318,16 +1339,6 @@ radeon_dp_detect(struct drm_connector *connector, bool force) ret = connector_status_connected; } } - - if ((ret == connector_status_disconnected) && - radeon_connector->dac_load_detect) { - struct drm_encoder *encoder = radeon_best_single_encoder(connector); - struct drm_encoder_helper_funcs *encoder_funcs; - if (encoder) { - encoder_funcs = encoder->helper_private; - ret = encoder_funcs->detect(encoder, connector); - } - } } radeon_connector_update_scratch_regs(connector, ret); diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 1a858944e4f3..6cc17fb96a57 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -707,16 +707,21 @@ int radeon_ddc_get_modes(struct radeon_connector *radeon_connector) radeon_router_select_ddc_port(radeon_connector); if ((radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_DisplayPort) || - (radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_eDP)) { + (radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_eDP) || + radeon_connector_encoder_is_dp_bridge(&radeon_connector->base)) { struct radeon_connector_atom_dig *dig = radeon_connector->con_priv; + if ((dig->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT || dig->dp_sink_type == CONNECTOR_OBJECT_ID_eDP) && dig->dp_i2c_bus) - radeon_connector->edid = drm_get_edid(&radeon_connector->base, &dig->dp_i2c_bus->adapter); - } - if (!radeon_connector->ddc_bus) - return -1; - if (!radeon_connector->edid) { - radeon_connector->edid = drm_get_edid(&radeon_connector->base, &radeon_connector->ddc_bus->adapter); + radeon_connector->edid = drm_get_edid(&radeon_connector->base, + &dig->dp_i2c_bus->adapter); + else if (radeon_connector->ddc_bus && !radeon_connector->edid) + radeon_connector->edid = drm_get_edid(&radeon_connector->base, + &radeon_connector->ddc_bus->adapter); + } else { + if (radeon_connector->ddc_bus && !radeon_connector->edid) + radeon_connector->edid = drm_get_edid(&radeon_connector->base, + &radeon_connector->ddc_bus->adapter); } if (!radeon_connector->edid) { From 5441ae5eb3614d3c28f77073370738a2820c88e4 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 25 Jul 2011 18:06:32 +0000 Subject: [PATCH 133/326] fs/9p: Add fid before dentry instantiation d_instantiate marks the dentry positive. So a parallel lookup and mkdir of the directory can find dentry that doesn't have fid attached. This can result in both the code path doing v9fs_fid_add which results in v9fs_dentry leak. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Eric Van Hensbergen --- fs/9p/vfs_inode.c | 4 +--- fs/9p/vfs_inode_dotl.c | 8 ++++---- 2 files changed, 5 insertions(+), 7 deletions(-) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 8bb5507e822f..43dd540663af 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -645,13 +645,11 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err); goto error; } - d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) goto error; - + d_instantiate(dentry, inode); return ofid; - error: if (ofid) p9_client_clunk(ofid); diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index b6c8ed205192..0ca224c8bb60 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -281,10 +281,10 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, P9_DPRINTK(P9_DEBUG_VFS, "inode creation failed %d\n", err); goto error; } - d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) goto error; + d_instantiate(dentry, inode); /* Now set the ACL based on the default value */ v9fs_set_create_acl(dentry, &dacl, &pacl); @@ -403,10 +403,10 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, err); goto error; } - d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) goto error; + d_instantiate(dentry, inode); fid = NULL; } else { /* @@ -657,10 +657,10 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, err); goto error; } - d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) goto error; + d_instantiate(dentry, inode); fid = NULL; } else { /* Not in cached mode. No need to populate inode with stat */ @@ -810,10 +810,10 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, err); goto error; } - d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) goto error; + d_instantiate(dentry, inode); fid = NULL; } else { /* From 45089142b1497dab2327d60f6c71c40766fc3ea4 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 25 Jul 2011 18:06:33 +0000 Subject: [PATCH 134/326] fs/9p: Don't update file type when updating file attributes We should only update attributes that we can change on stat2inode. Also do file type initialization in v9fs_init_inode. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Eric Van Hensbergen --- fs/9p/v9fs_vfs.h | 4 +- fs/9p/vfs_inode.c | 91 +++++++++++++++++++++++------------------- fs/9p/vfs_inode_dotl.c | 23 +++++++---- fs/9p/vfs_super.c | 2 +- 4 files changed, 68 insertions(+), 52 deletions(-) diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index 46ce357ca1ab..7ac1faec2bde 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -54,9 +54,9 @@ extern struct kmem_cache *v9fs_inode_cache; struct inode *v9fs_alloc_inode(struct super_block *sb); void v9fs_destroy_inode(struct inode *inode); -struct inode *v9fs_get_inode(struct super_block *sb, int mode); +struct inode *v9fs_get_inode(struct super_block *sb, int mode, dev_t); int v9fs_init_inode(struct v9fs_session_info *v9ses, - struct inode *inode, int mode); + struct inode *inode, int mode, dev_t); void v9fs_evict_inode(struct inode *inode); ino_t v9fs_qid2ino(struct p9_qid *qid); void v9fs_stat2inode(struct p9_wstat *, struct inode *, struct super_block *); diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 43dd540663af..3563cace0a2e 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -95,15 +95,18 @@ static int unixmode2p9mode(struct v9fs_session_info *v9ses, int mode) /** * p9mode2unixmode- convert plan9 mode bits to unix mode bits * @v9ses: v9fs session information - * @mode: mode to convert + * @stat: p9_wstat from which mode need to be derived + * @rdev: major number, minor number in case of device files. * */ - -static int p9mode2unixmode(struct v9fs_session_info *v9ses, int mode) +static int p9mode2unixmode(struct v9fs_session_info *v9ses, + struct p9_wstat *stat, dev_t *rdev) { int res; + int mode = stat->mode; - res = mode & 0777; + res = mode & S_IALLUGO; + *rdev = 0; if ((mode & P9_DMDIR) == P9_DMDIR) res |= S_IFDIR; @@ -116,9 +119,26 @@ static int p9mode2unixmode(struct v9fs_session_info *v9ses, int mode) && (v9ses->nodev == 0)) res |= S_IFIFO; else if ((mode & P9_DMDEVICE) && (v9fs_proto_dotu(v9ses)) - && (v9ses->nodev == 0)) - res |= S_IFBLK; - else + && (v9ses->nodev == 0)) { + char type = 0, ext[32]; + int major = -1, minor = -1; + + strncpy(ext, stat->extension, sizeof(ext)); + sscanf(ext, "%c %u %u", &type, &major, &minor); + switch (type) { + case 'c': + res |= S_IFCHR; + break; + case 'b': + res |= S_IFBLK; + break; + default: + P9_DPRINTK(P9_DEBUG_ERROR, + "Unknown special type %c %s\n", type, + stat->extension); + }; + *rdev = MKDEV(major, minor); + } else res |= S_IFREG; if (v9fs_proto_dotu(v9ses)) { @@ -131,7 +151,6 @@ static int p9mode2unixmode(struct v9fs_session_info *v9ses, int mode) if ((mode & P9_DMSETVTX) == P9_DMSETVTX) res |= S_ISVTX; } - return res; } @@ -242,13 +261,13 @@ void v9fs_destroy_inode(struct inode *inode) } int v9fs_init_inode(struct v9fs_session_info *v9ses, - struct inode *inode, int mode) + struct inode *inode, int mode, dev_t rdev) { int err = 0; inode_init_owner(inode, NULL, mode); inode->i_blocks = 0; - inode->i_rdev = 0; + inode->i_rdev = rdev; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_mapping->a_ops = &v9fs_addr_operations; @@ -335,7 +354,7 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses, * */ -struct inode *v9fs_get_inode(struct super_block *sb, int mode) +struct inode *v9fs_get_inode(struct super_block *sb, int mode, dev_t rdev) { int err; struct inode *inode; @@ -348,7 +367,7 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode) P9_EPRINTK(KERN_WARNING, "Problem allocating inode\n"); return ERR_PTR(-ENOMEM); } - err = v9fs_init_inode(v9ses, inode, mode); + err = v9fs_init_inode(v9ses, inode, mode, rdev); if (err) { iput(inode); return ERR_PTR(err); @@ -435,11 +454,12 @@ void v9fs_evict_inode(struct inode *inode) static int v9fs_test_inode(struct inode *inode, void *data) { int umode; + dev_t rdev; struct v9fs_inode *v9inode = V9FS_I(inode); struct p9_wstat *st = (struct p9_wstat *)data; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(inode); - umode = p9mode2unixmode(v9ses, st->mode); + umode = p9mode2unixmode(v9ses, st, &rdev); /* don't match inode of different type */ if ((inode->i_mode & S_IFMT) != (umode & S_IFMT)) return 0; @@ -473,6 +493,7 @@ static struct inode *v9fs_qid_iget(struct super_block *sb, struct p9_wstat *st, int new) { + dev_t rdev; int retval, umode; unsigned long i_ino; struct inode *inode; @@ -496,8 +517,8 @@ static struct inode *v9fs_qid_iget(struct super_block *sb, * later. */ inode->i_ino = i_ino; - umode = p9mode2unixmode(v9ses, st->mode); - retval = v9fs_init_inode(v9ses, inode, umode); + umode = p9mode2unixmode(v9ses, st, &rdev); + retval = v9fs_init_inode(v9ses, inode, umode, rdev); if (retval) goto error; @@ -1000,7 +1021,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, return PTR_ERR(st); v9fs_stat2inode(st, dentry->d_inode, dentry->d_inode->i_sb); - generic_fillattr(dentry->d_inode, stat); + generic_fillattr(dentry->d_inode, stat); p9stat_free(st); kfree(st); @@ -1084,6 +1105,7 @@ void v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, struct super_block *sb) { + mode_t mode; char ext[32]; char tag_name[14]; unsigned int i_nlink; @@ -1119,31 +1141,9 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, inode->i_nlink = i_nlink; } } - inode->i_mode = p9mode2unixmode(v9ses, stat->mode); - if ((S_ISBLK(inode->i_mode)) || (S_ISCHR(inode->i_mode))) { - char type = 0; - int major = -1; - int minor = -1; - - strncpy(ext, stat->extension, sizeof(ext)); - sscanf(ext, "%c %u %u", &type, &major, &minor); - switch (type) { - case 'c': - inode->i_mode &= ~S_IFBLK; - inode->i_mode |= S_IFCHR; - break; - case 'b': - break; - default: - P9_DPRINTK(P9_DEBUG_ERROR, - "Unknown special type %c %s\n", type, - stat->extension); - }; - inode->i_rdev = MKDEV(major, minor); - init_special_inode(inode, inode->i_mode, inode->i_rdev); - } else - inode->i_rdev = 0; - + mode = stat->mode & S_IALLUGO; + mode |= inode->i_mode & ~S_IALLUGO; + inode->i_mode = mode; i_size_write(inode, stat->length); /* not real number of blocks, but 512 byte ones ... */ @@ -1409,6 +1409,8 @@ v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode) { + int umode; + dev_t rdev; loff_t i_size; struct p9_wstat *st; struct v9fs_session_info *v9ses; @@ -1417,6 +1419,12 @@ int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode) st = p9_client_stat(fid); if (IS_ERR(st)) return PTR_ERR(st); + /* + * Don't update inode if the file type is different + */ + umode = p9mode2unixmode(v9ses, st, &rdev); + if ((inode->i_mode & S_IFMT) != (umode & S_IFMT)) + goto out; spin_lock(&inode->i_lock); /* @@ -1428,6 +1436,7 @@ int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode) if (v9ses->cache) inode->i_size = i_size; spin_unlock(&inode->i_lock); +out: p9stat_free(st); kfree(st); return 0; diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 0ca224c8bb60..a3f2540cc4b2 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -153,7 +153,8 @@ static struct inode *v9fs_qid_iget_dotl(struct super_block *sb, * later. */ inode->i_ino = i_ino; - retval = v9fs_init_inode(v9ses, inode, st->st_mode); + retval = v9fs_init_inode(v9ses, inode, + st->st_mode, new_decode_dev(st->st_rdev)); if (retval) goto error; @@ -414,7 +415,7 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, * inode with stat. We need to get an inode * so that we can set the acl with dentry */ - inode = v9fs_get_inode(dir->i_sb, mode); + inode = v9fs_get_inode(dir->i_sb, mode, 0); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto error; @@ -540,6 +541,7 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) void v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode) { + mode_t mode; struct v9fs_inode *v9inode = V9FS_I(inode); if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) { @@ -552,11 +554,10 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode) inode->i_uid = stat->st_uid; inode->i_gid = stat->st_gid; inode->i_nlink = stat->st_nlink; - inode->i_mode = stat->st_mode; - inode->i_rdev = new_decode_dev(stat->st_rdev); - if ((S_ISBLK(inode->i_mode)) || (S_ISCHR(inode->i_mode))) - init_special_inode(inode, inode->i_mode, inode->i_rdev); + mode = stat->st_mode & S_IALLUGO; + mode |= inode->i_mode & ~S_IALLUGO; + inode->i_mode = mode; i_size_write(inode, stat->st_size); inode->i_blocks = stat->st_blocks; @@ -664,7 +665,7 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, fid = NULL; } else { /* Not in cached mode. No need to populate inode with stat */ - inode = v9fs_get_inode(dir->i_sb, S_IFLNK); + inode = v9fs_get_inode(dir->i_sb, S_IFLNK, 0); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto error; @@ -820,7 +821,7 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, * Not in cached mode. No need to populate inode with stat. * socket syscall returns a fd, so we need instantiate */ - inode = v9fs_get_inode(dir->i_sb, mode); + inode = v9fs_get_inode(dir->i_sb, mode, rdev); if (IS_ERR(inode)) { err = PTR_ERR(inode); goto error; @@ -886,6 +887,11 @@ int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode) st = p9_client_getattr_dotl(fid, P9_STATS_ALL); if (IS_ERR(st)) return PTR_ERR(st); + /* + * Don't update inode if the file type is different + */ + if ((inode->i_mode & S_IFMT) != (st->st_mode & S_IFMT)) + goto out; spin_lock(&inode->i_lock); /* @@ -897,6 +903,7 @@ int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode) if (v9ses->cache) inode->i_size = i_size; spin_unlock(&inode->i_lock); +out: kfree(st); return 0; } diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index feef6cdc1fd2..c70251d47ed1 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -149,7 +149,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, else sb->s_d_op = &v9fs_dentry_operations; - inode = v9fs_get_inode(sb, S_IFDIR | mode); + inode = v9fs_get_inode(sb, S_IFDIR | mode, 0); if (IS_ERR(inode)) { retval = PTR_ERR(inode); goto release_sb; From b49d8b5d7007a673796f3f99688b46931293873e Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Wed, 17 Aug 2011 16:56:04 +0000 Subject: [PATCH 135/326] net/9p: Fix kernel crash with msize 512K With msize equal to 512K (PAGE_SIZE * VIRTQUEUE_NUM), we hit multiple crashes. This patch fix those. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Eric Van Hensbergen --- net/9p/trans_virtio.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 175b5135bdcf..e317583fcc73 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -263,7 +263,6 @@ p9_virtio_request(struct p9_client *client, struct p9_req_t *req) { int in, out, inp, outp; struct virtio_chan *chan = client->trans; - char *rdata = (char *)req->rc+sizeof(struct p9_fcall); unsigned long flags; size_t pdata_off = 0; struct trans_rpage_info *rpinfo = NULL; @@ -346,7 +345,8 @@ p9_virtio_request(struct p9_client *client, struct p9_req_t *req) * Arrange in such a way that server places header in the * alloced memory and payload onto the user buffer. */ - inp = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM, rdata, 11); + inp = pack_sg_list(chan->sg, out, + VIRTQUEUE_NUM, req->rc->sdata, 11); /* * Running executables in the filesystem may result in * a read request with kernel buffer as opposed to user buffer. @@ -366,8 +366,8 @@ p9_virtio_request(struct p9_client *client, struct p9_req_t *req) } in += inp; } else { - in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM, rdata, - req->rc->capacity); + in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM, + req->rc->sdata, req->rc->capacity); } err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc); @@ -592,7 +592,14 @@ static struct p9_trans_module p9_virtio_trans = { .close = p9_virtio_close, .request = p9_virtio_request, .cancel = p9_virtio_cancel, - .maxsize = PAGE_SIZE*VIRTQUEUE_NUM, + + /* + * We leave one entry for input and one entry for response + * headers. We also skip one more entry to accomodate, address + * that are not at page boundary, that can result in an extra + * page in zero copy. + */ + .maxsize = PAGE_SIZE * (VIRTQUEUE_NUM - 3), .pref = P9_TRANS_PREF_PAYLOAD_SEP, .def = 0, .owner = THIS_MODULE, From f88657ce3f9713a0c62101dffb0e972a979e77b9 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Wed, 3 Aug 2011 19:55:32 +0530 Subject: [PATCH 136/326] fs/9p: Add OS dependent open flags in 9p protocol Some of the flags are OS/arch dependent we add a 9p protocol value which maps to asm-generic/fcntl.h values in Linux Based on the original patch from Venkateswararao Jujjuri Signed-off-by: Aneesh Kumar K.V --- fs/9p/v9fs_vfs.h | 2 ++ fs/9p/vfs_file.c | 2 +- fs/9p/vfs_inode.c | 16 +++++++++++- fs/9p/vfs_inode_dotl.c | 55 +++++++++++++++++++++++++++++++++++++++++- include/net/9p/9p.h | 24 ++++++++++++++++++ 5 files changed, 96 insertions(+), 3 deletions(-) diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index 7ac1faec2bde..410ffd6ceb5f 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -83,4 +83,6 @@ static inline void v9fs_invalidate_inode_attr(struct inode *inode) v9inode->cache_validity |= V9FS_INO_INVALID_ATTR; return; } + +int v9fs_open_to_dotl_flags(int flags); #endif diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index 3c173fcc2c5a..c2f107583125 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -65,7 +65,7 @@ int v9fs_file_open(struct inode *inode, struct file *file) v9inode = V9FS_I(inode); v9ses = v9fs_inode2v9ses(inode); if (v9fs_proto_dotl(v9ses)) - omode = file->f_flags; + omode = v9fs_open_to_dotl_flags(file->f_flags); else omode = v9fs_uflags2omode(file->f_flags, v9fs_proto_dotu(v9ses)); diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 3563cace0a2e..9e3ea6ce6951 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -552,6 +552,19 @@ v9fs_inode_from_fid(struct v9fs_session_info *v9ses, struct p9_fid *fid, return inode; } +/** + * v9fs_at_to_dotl_flags- convert Linux specific AT flags to + * plan 9 AT flag. + * @flags: flags to convert + */ +static int v9fs_at_to_dotl_flags(int flags) +{ + int rflags = 0; + if (flags & AT_REMOVEDIR) + rflags |= P9_DOTL_AT_REMOVEDIR; + return rflags; +} + /** * v9fs_remove - helper function to remove files and directories * @dir: directory inode that is being deleted @@ -579,7 +592,8 @@ static int v9fs_remove(struct inode *dir, struct dentry *dentry, int flags) return retval; } if (v9fs_proto_dotl(v9ses)) - retval = p9_client_unlinkat(dfid, dentry->d_name.name, flags); + retval = p9_client_unlinkat(dfid, dentry->d_name.name, + v9fs_at_to_dotl_flags(flags)); if (retval == -EOPNOTSUPP) { /* Try the one based on path */ v9fid = v9fs_fid_clone(dentry); diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index a3f2540cc4b2..aded79fcd5cf 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -191,6 +191,58 @@ v9fs_inode_from_fid_dotl(struct v9fs_session_info *v9ses, struct p9_fid *fid, return inode; } +struct dotl_openflag_map { + int open_flag; + int dotl_flag; +}; + +static int v9fs_mapped_dotl_flags(int flags) +{ + int i; + int rflags = 0; + struct dotl_openflag_map dotl_oflag_map[] = { + { O_CREAT, P9_DOTL_CREATE }, + { O_EXCL, P9_DOTL_EXCL }, + { O_NOCTTY, P9_DOTL_NOCTTY }, + { O_TRUNC, P9_DOTL_TRUNC }, + { O_APPEND, P9_DOTL_APPEND }, + { O_NONBLOCK, P9_DOTL_NONBLOCK }, + { O_DSYNC, P9_DOTL_DSYNC }, + { FASYNC, P9_DOTL_FASYNC }, + { O_DIRECT, P9_DOTL_DIRECT }, + { O_LARGEFILE, P9_DOTL_LARGEFILE }, + { O_DIRECTORY, P9_DOTL_DIRECTORY }, + { O_NOFOLLOW, P9_DOTL_NOFOLLOW }, + { O_NOATIME, P9_DOTL_NOATIME }, + { O_CLOEXEC, P9_DOTL_CLOEXEC }, + { O_SYNC, P9_DOTL_SYNC}, + }; + for (i = 0; i < ARRAY_SIZE(dotl_oflag_map); i++) { + if (flags & dotl_oflag_map[i].open_flag) + rflags |= dotl_oflag_map[i].dotl_flag; + } + return rflags; +} + +/** + * v9fs_open_to_dotl_flags- convert Linux specific open flags to + * plan 9 open flag. + * @flags: flags to convert + */ +int v9fs_open_to_dotl_flags(int flags) +{ + int rflags = 0; + + /* + * We have same bits for P9_DOTL_READONLY, P9_DOTL_WRONLY + * and P9_DOTL_NOACCESS + */ + rflags |= flags & O_ACCMODE; + rflags |= v9fs_mapped_dotl_flags(flags); + + return rflags; +} + /** * v9fs_vfs_create_dotl - VFS hook to create files for 9P2000.L protocol. * @dir: directory inode that is being created @@ -259,7 +311,8 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, "Failed to get acl values in creat %d\n", err); goto error; } - err = p9_client_create_dotl(ofid, name, flags, mode, gid, &qid); + err = p9_client_create_dotl(ofid, name, v9fs_open_to_dotl_flags(flags), + mode, gid, &qid); if (err < 0) { P9_DPRINTK(P9_DEBUG_VFS, "p9_client_open_dotl failed in creat %d\n", diff --git a/include/net/9p/9p.h b/include/net/9p/9p.h index 342dcf13d039..957ab99897a1 100644 --- a/include/net/9p/9p.h +++ b/include/net/9p/9p.h @@ -288,6 +288,30 @@ enum p9_perm_t { P9_DMSETVTX = 0x00010000, }; +/* 9p2000.L open flags */ +#define P9_DOTL_RDONLY 00000000 +#define P9_DOTL_WRONLY 00000001 +#define P9_DOTL_RDWR 00000002 +#define P9_DOTL_NOACCESS 00000003 +#define P9_DOTL_CREATE 00000100 +#define P9_DOTL_EXCL 00000200 +#define P9_DOTL_NOCTTY 00000400 +#define P9_DOTL_TRUNC 00001000 +#define P9_DOTL_APPEND 00002000 +#define P9_DOTL_NONBLOCK 00004000 +#define P9_DOTL_DSYNC 00010000 +#define P9_DOTL_FASYNC 00020000 +#define P9_DOTL_DIRECT 00040000 +#define P9_DOTL_LARGEFILE 00100000 +#define P9_DOTL_DIRECTORY 00200000 +#define P9_DOTL_NOFOLLOW 00400000 +#define P9_DOTL_NOATIME 01000000 +#define P9_DOTL_CLOEXEC 02000000 +#define P9_DOTL_SYNC 04000000 + +/* 9p2000.L at flags */ +#define P9_DOTL_AT_REMOVEDIR 0x200 + /** * enum p9_qid_t - QID types * @P9_QTDIR: directory From 73f507171cfa407b19f254aef95cbb058c8180cf Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 16 Aug 2011 22:19:28 +0530 Subject: [PATCH 137/326] fs/9p: Always ask new inode in lookup for cache mode disabled This make sure we don't end up reusing the unlinked inode object. The ideal way is to use inode i_generation. But i_generation is not available in userspace always. Signed-off-by: Aneesh Kumar K.V --- fs/9p/vfs_inode.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 9e3ea6ce6951..e3c03db3c788 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -825,6 +825,7 @@ static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nameidata) { + struct dentry *res; struct super_block *sb; struct v9fs_session_info *v9ses; struct p9_fid *dfid, *fid; @@ -856,22 +857,35 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(result); } - - inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb); + /* + * Make sure we don't use a wrong inode due to parallel + * unlink. For cached mode create calls request for new + * inode. But with cache disabled, lookup should do this. + */ + if (v9ses->cache) + inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb); + else + inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); if (IS_ERR(inode)) { result = PTR_ERR(inode); inode = NULL; goto error; } - result = v9fs_fid_add(dentry, fid); if (result < 0) goto error_iput; - inst_out: - d_add(dentry, inode); - return NULL; - + /* + * If we had a rename on the server and a parallel lookup + * for the new name, then make sure we instantiate with + * the new name. ie look up for a/b, while on server somebody + * moved b under k and client parallely did a lookup for + * k/b. + */ + res = d_materialise_unique(dentry, inode); + if (!IS_ERR(res)) + return res; + result = PTR_ERR(res); error_iput: iput(inode); error: From 51b8b4fb32271d39fbdd760397406177b2b0fd36 Mon Sep 17 00:00:00 2001 From: Jim Garlick Date: Sun, 21 Aug 2011 00:21:18 +0530 Subject: [PATCH 138/326] fs/9p: Use protocol-defined value for lock/getlock 'type' field. Signed-off-by: Jim Garlick Signed-off-by: Aneesh Kumar K.V --- fs/9p/vfs_file.c | 34 +++++++++++++++++++++++++++------- include/net/9p/9p.h | 5 +++++ 2 files changed, 32 insertions(+), 7 deletions(-) diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index c2f107583125..62857a810a79 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -169,7 +169,18 @@ static int v9fs_file_do_lock(struct file *filp, int cmd, struct file_lock *fl) /* convert posix lock to p9 tlock args */ memset(&flock, 0, sizeof(flock)); - flock.type = fl->fl_type; + /* map the lock type */ + switch (fl->fl_type) { + case F_RDLCK: + flock.type = P9_LOCK_TYPE_RDLCK; + break; + case F_WRLCK: + flock.type = P9_LOCK_TYPE_WRLCK; + break; + case F_UNLCK: + flock.type = P9_LOCK_TYPE_UNLCK; + break; + } flock.start = fl->fl_start; if (fl->fl_end == OFFSET_MAX) flock.length = 0; @@ -245,7 +256,7 @@ static int v9fs_file_getlock(struct file *filp, struct file_lock *fl) /* convert posix lock to p9 tgetlock args */ memset(&glock, 0, sizeof(glock)); - glock.type = fl->fl_type; + glock.type = P9_LOCK_TYPE_UNLCK; glock.start = fl->fl_start; if (fl->fl_end == OFFSET_MAX) glock.length = 0; @@ -257,17 +268,26 @@ static int v9fs_file_getlock(struct file *filp, struct file_lock *fl) res = p9_client_getlock_dotl(fid, &glock); if (res < 0) return res; - if (glock.type != F_UNLCK) { - fl->fl_type = glock.type; + /* map 9p lock type to os lock type */ + switch (glock.type) { + case P9_LOCK_TYPE_RDLCK: + fl->fl_type = F_RDLCK; + break; + case P9_LOCK_TYPE_WRLCK: + fl->fl_type = F_WRLCK; + break; + case P9_LOCK_TYPE_UNLCK: + fl->fl_type = F_UNLCK; + break; + } + if (glock.type != P9_LOCK_TYPE_UNLCK) { fl->fl_start = glock.start; if (glock.length == 0) fl->fl_end = OFFSET_MAX; else fl->fl_end = glock.start + glock.length - 1; fl->fl_pid = glock.proc_id; - } else - fl->fl_type = F_UNLCK; - + } return res; } diff --git a/include/net/9p/9p.h b/include/net/9p/9p.h index 957ab99897a1..a6326ef8ade6 100644 --- a/include/net/9p/9p.h +++ b/include/net/9p/9p.h @@ -312,6 +312,11 @@ enum p9_perm_t { /* 9p2000.L at flags */ #define P9_DOTL_AT_REMOVEDIR 0x200 +/* 9p2000.L lock type */ +#define P9_LOCK_TYPE_RDLCK 0 +#define P9_LOCK_TYPE_WRLCK 1 +#define P9_LOCK_TYPE_UNLCK 2 + /** * enum p9_qid_t - QID types * @P9_QTDIR: directory From 8efcc57dedfebc99c3cd39564e3fc47cd1a24b75 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Wed, 3 Aug 2011 18:04:29 +0900 Subject: [PATCH 139/326] mfd: Fix value of WM8994_CONFIGURE_GPIO This needs to be an out of band value for the register and on this device registers are 16 bit so we must shift left one to the 17th bit. Signed-off-by: Mark Brown Cc: stable@kernel.org Signed-off-by: Samuel Ortiz --- include/linux/mfd/wm8994/pdata.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/mfd/wm8994/pdata.h b/include/linux/mfd/wm8994/pdata.h index d12f8d635a81..97cf4f27d647 100644 --- a/include/linux/mfd/wm8994/pdata.h +++ b/include/linux/mfd/wm8994/pdata.h @@ -26,7 +26,7 @@ struct wm8994_ldo_pdata { struct regulator_init_data *init_data; }; -#define WM8994_CONFIGURE_GPIO 0x8000 +#define WM8994_CONFIGURE_GPIO 0x10000 #define WM8994_DRC_REGS 5 #define WM8994_EQ_REGS 20 From aa9d842c5f2da6cdef2777f2f062f61898be89d3 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Wed, 10 Aug 2011 10:05:49 +0200 Subject: [PATCH 140/326] mfd: Rename wm8350 static gpio_set_debounce() The kernel already has a function with this name declared in asm-generic/gpio.h. So if this header leaks into wm8350/gpio.c we get drivers/mfd/wm8350-gpio.c:40:12: error: conflicting types for 'gpio_set_debounce' include/asm-generic/gpio.h:156:12: note: previous declaration of 'gpio_set_debounce' was here Fix this by adding a wm8350_ prefix to the function. Signed-off-by: Sascha Hauer Acked-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/mfd/wm8350-gpio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/mfd/wm8350-gpio.c b/drivers/mfd/wm8350-gpio.c index ebf99bef392f..d584f6b4d6e2 100644 --- a/drivers/mfd/wm8350-gpio.c +++ b/drivers/mfd/wm8350-gpio.c @@ -37,7 +37,7 @@ static int gpio_set_dir(struct wm8350 *wm8350, int gpio, int dir) return ret; } -static int gpio_set_debounce(struct wm8350 *wm8350, int gpio, int db) +static int wm8350_gpio_set_debounce(struct wm8350 *wm8350, int gpio, int db) { if (db == WM8350_GPIO_DEBOUNCE_ON) return wm8350_set_bits(wm8350, WM8350_GPIO_DEBOUNCE, @@ -210,7 +210,7 @@ int wm8350_gpio_config(struct wm8350 *wm8350, int gpio, int dir, int func, goto err; if (gpio_set_polarity(wm8350, gpio, pol)) goto err; - if (gpio_set_debounce(wm8350, gpio, debounce)) + if (wm8350_gpio_set_debounce(wm8350, gpio, debounce)) goto err; if (gpio_set_dir(wm8350, gpio, dir)) goto err; From 66cc5b8e50af87b0bbd0f179d76d2826f4549c13 Mon Sep 17 00:00:00 2001 From: Kyle Manna Date: Thu, 11 Aug 2011 22:33:12 -0500 Subject: [PATCH 141/326] mfd: Copy the device pointer to the twl4030-madc structure Worst case this fixes the following error: [ 72.086212] (NULL device *): conversion timeout! Best case it prevents a crash Signed-off-by: Kyle Manna Signed-off-by: Samuel Ortiz --- drivers/mfd/twl4030-madc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/mfd/twl4030-madc.c b/drivers/mfd/twl4030-madc.c index b5d598c3aa71..cb44b53dee35 100644 --- a/drivers/mfd/twl4030-madc.c +++ b/drivers/mfd/twl4030-madc.c @@ -706,6 +706,8 @@ static int __devinit twl4030_madc_probe(struct platform_device *pdev) if (!madc) return -ENOMEM; + madc->dev = &pdev->dev; + /* * Phoenix provides 2 interrupt lines. The first one is connected to * the OMAP. The other one can be connected to the other processor such From d0e84caeb4cd535923884735906e5730329505b4 Mon Sep 17 00:00:00 2001 From: Kyle Manna Date: Thu, 11 Aug 2011 22:33:14 -0500 Subject: [PATCH 142/326] mfd: Check for twl4030-madc NULL pointer If the twl4030-madc device wasn't registered, and another device, such as twl4030-madc-hwmon, calls twl4030_madc_conversion() a NULL pointer is dereferenced. Signed-off-by: Kyle Manna Signed-off-by: Samuel Ortiz --- drivers/mfd/twl4030-madc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mfd/twl4030-madc.c b/drivers/mfd/twl4030-madc.c index cb44b53dee35..7cbf2aa9e64f 100644 --- a/drivers/mfd/twl4030-madc.c +++ b/drivers/mfd/twl4030-madc.c @@ -510,8 +510,9 @@ int twl4030_madc_conversion(struct twl4030_madc_request *req) u8 ch_msb, ch_lsb; int ret; - if (!req) + if (!req || !twl4030_madc) return -EINVAL; + mutex_lock(&twl4030_madc->lock); if (req->method < TWL4030_MADC_RT || req->method > TWL4030_MADC_SW2) { ret = -EINVAL; From fa948761e685fb03823b3029e5b6bdb52229d6ce Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Mon, 15 Aug 2011 12:42:03 +0200 Subject: [PATCH 143/326] mfd: Fix initialisation of tps65910 interrupts Fix regression introduced by commit a2974732ca7614aaf0baf9d6dd3ad893d50ce1c5 (TPS65911: Add new irq definitions) which caused irq_num to be incorrectly set for tps65910. Cc: stable@kernel.org Signed-off-by: Johan Hovold Acked-by: Graeme Gregory Signed-off-by: Samuel Ortiz --- drivers/mfd/tps65910-irq.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/mfd/tps65910-irq.c b/drivers/mfd/tps65910-irq.c index 2bfad5c86cc7..a56be931551c 100644 --- a/drivers/mfd/tps65910-irq.c +++ b/drivers/mfd/tps65910-irq.c @@ -178,8 +178,10 @@ int tps65910_irq_init(struct tps65910 *tps65910, int irq, switch (tps65910_chip_id(tps65910)) { case TPS65910: tps65910->irq_num = TPS65910_NUM_IRQ; + break; case TPS65911: tps65910->irq_num = TPS65911_NUM_IRQ; + break; } /* Register with genirq */ From 7eb3154e6caf7945ce60c196637b7ac06213befd Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Thu, 18 Aug 2011 16:37:35 +0900 Subject: [PATCH 144/326] mfd: Set MAX8997 irq pointer Required platform information is not handed to max8997-irq.c properly. This patch enables to hand over such information to max8997-irq.c so that max8997-irq functions properly. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Samuel Ortiz --- drivers/mfd/max8997.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/mfd/max8997.c b/drivers/mfd/max8997.c index 5d1fca0277ef..f83103b8970d 100644 --- a/drivers/mfd/max8997.c +++ b/drivers/mfd/max8997.c @@ -135,10 +135,13 @@ static int max8997_i2c_probe(struct i2c_client *i2c, max8997->dev = &i2c->dev; max8997->i2c = i2c; max8997->type = id->driver_data; + max8997->irq = i2c->irq; if (!pdata) goto err; + max8997->irq_base = pdata->irq_base; + max8997->ono = pdata->ono; max8997->wakeup = pdata->wakeup; mutex_init(&max8997->iolock); @@ -152,6 +155,8 @@ static int max8997_i2c_probe(struct i2c_client *i2c, pm_runtime_set_active(max8997->dev); + max8997_irq_init(max8997); + mfd_add_devices(max8997->dev, -1, max8997_devs, ARRAY_SIZE(max8997_devs), NULL, 0); From e600cffe618ff0da29ae1f8b8d3824ce0e2409fc Mon Sep 17 00:00:00 2001 From: Anand Gadiyar Date: Thu, 18 Aug 2011 16:14:31 +0530 Subject: [PATCH 145/326] mfd: Make omap-usb-host TLL mode work again This code section seems to have been accidentally copy pasted. It causes incorrect bits to be set up in the TLL_CHANNEL_CONF register and prevents the TLL mode from working correctly. Cc: stable@kernel.org Signed-off-by: Anand Gadiyar Cc: Keshava Munegowda Acked-by: Felipe Balbi Signed-off-by: Samuel Ortiz --- drivers/mfd/omap-usb-host.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/mfd/omap-usb-host.c b/drivers/mfd/omap-usb-host.c index 29601e7d606d..0f19ab14de88 100644 --- a/drivers/mfd/omap-usb-host.c +++ b/drivers/mfd/omap-usb-host.c @@ -676,7 +676,6 @@ static void usbhs_omap_tll_init(struct device *dev, u8 tll_channel_count) | OMAP_TLL_CHANNEL_CONF_ULPINOBITSTUFF | OMAP_TLL_CHANNEL_CONF_ULPIDDRMODE); - reg |= (1 << (i + 1)); } else continue; From 417e206b16e18bc729346b6db668031498975b8e Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 19 Aug 2011 16:57:54 +0800 Subject: [PATCH 146/326] mfd: Fix omap-usb-host build failure The patch fixes the build failure: drivers/mfd/omap-usb-host.c:1034:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1034:1: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' drivers/mfd/omap-usb-host.c:1034:1: warning: parameter names (without types) in function declaration drivers/mfd/omap-usb-host.c:1040:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1040:1: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' drivers/mfd/omap-usb-host.c:1040:1: warning: parameter names (without types) in function declaration drivers/mfd/omap-usb-host.c:1045:13: error: 'THIS_MODULE' undeclared here (not in a function) drivers/mfd/omap-usb-host.c:1050:15: error: expected declaration specifiers or '...' before string constant drivers/mfd/omap-usb-host.c:1050:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1050:1: warning: type defaults to 'int' in declaration of 'MODULE_AUTHOR' drivers/mfd/omap-usb-host.c:1050:15: warning: function declaration isn't a prototype drivers/mfd/omap-usb-host.c:1051:14: error: expected declaration specifiers or '...' before string constant drivers/mfd/omap-usb-host.c:1051:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1051:1: warning: type defaults to 'int' in declaration of 'MODULE_ALIAS' drivers/mfd/omap-usb-host.c:1051:14: warning: function declaration isn't a prototype drivers/mfd/omap-usb-host.c:1052:16: error: expected declaration specifiers or '...' before string constant drivers/mfd/omap-usb-host.c:1052:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1052:1: warning: type defaults to 'int' in declaration of 'MODULE_LICENSE' drivers/mfd/omap-usb-host.c:1052:16: warning: function declaration isn't a prototype drivers/mfd/omap-usb-host.c:1053:20: error: expected declaration specifiers or '...' before string constant drivers/mfd/omap-usb-host.c:1053:1: warning: data definition has no type or storage class drivers/mfd/omap-usb-host.c:1053:1: warning: type defaults to 'int' in declaration of 'MODULE_DESCRIPTION' drivers/mfd/omap-usb-host.c:1053:20: warning: function declaration isn't a prototype make[2]: *** [drivers/mfd/omap-usb-host.o] Error 1 CC fs/proc/namespaces.o make[1]: *** [drivers/mfd] Error 2 make: *** [drivers] Error 2 make: *** Waiting for unfinished jobs.... Signed-off-by: Ming Lei Signed-off-by: Samuel Ortiz --- drivers/mfd/omap-usb-host.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mfd/omap-usb-host.c b/drivers/mfd/omap-usb-host.c index 0f19ab14de88..86e14583a082 100644 --- a/drivers/mfd/omap-usb-host.c +++ b/drivers/mfd/omap-usb-host.c @@ -17,6 +17,7 @@ * along with this program. If not, see . */ #include +#include #include #include #include From ff71c182f461da5ae9d2d65f8a63f5a9193b9be1 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sun, 28 Aug 2011 13:01:49 -0700 Subject: [PATCH 147/326] hwmon: (max16065) Fix current calculation Current calculation is completely wrong. Add missing brackets to fix it. Signed-off-by: Guenter Roeck Acked-by: Jean Delvare Cc: stable@kernel.org # 3.0+ --- drivers/hwmon/max16065.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hwmon/max16065.c b/drivers/hwmon/max16065.c index d94a24fdf4ba..dd2d7b9620c2 100644 --- a/drivers/hwmon/max16065.c +++ b/drivers/hwmon/max16065.c @@ -124,7 +124,7 @@ static inline int MV_TO_LIMIT(int mv, int range) static inline int ADC_TO_CURR(int adc, int gain) { - return adc * 1400000 / gain * 255; + return adc * 1400000 / (gain * 255); } /* From d91aae1e52e5289a94f4ddff968decfc8d37271e Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Mon, 29 Aug 2011 22:53:20 -0700 Subject: [PATCH 148/326] hwmon: (max16065) Add chip access warning to documentation The chips supported by the max16065 driver should not be accessed using direct i2ctools commands. Add warning to driver documentation to alert users. Signed-off-by: Guenter Roeck Acked-by: Jean Delvare --- Documentation/hwmon/max16065 | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/hwmon/max16065 b/Documentation/hwmon/max16065 index 44b4f61e04f9..c11f64a1f2ad 100644 --- a/Documentation/hwmon/max16065 +++ b/Documentation/hwmon/max16065 @@ -62,6 +62,13 @@ can be safely used to identify the chip. You will have to instantiate the devices explicitly. Please see Documentation/i2c/instantiating-devices for details. +WARNING: Do not access chip registers using the i2cdump command, and do not use +any of the i2ctools commands on a command register (0xa5 to 0xac). The chips +supported by this driver interpret any access to a command register (including +read commands) as request to execute the command in question. This may result in +power loss, board resets, and/or Flash corruption. Worst case, your board may +turn into a brick. + Sysfs entries ------------- From f020b007d5dd24597f5e985a6309bcb8393797ed Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Wed, 31 Aug 2011 11:53:41 -0400 Subject: [PATCH 149/326] hwmon: (ucd9000/ucd9200) Optimize array walk Rewrite the loop walking the id array during probe. The new code is better adapted to a null-terminated array, and is also clearer and more efficient than the original. Signed-off-by: Jean Delvare Cc: Axel Lin Cc: Guenter Roeck Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/ucd9000.c | 6 ++---- drivers/hwmon/pmbus/ucd9200.c | 6 ++---- 2 files changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/hwmon/pmbus/ucd9000.c b/drivers/hwmon/pmbus/ucd9000.c index ace1c7319734..d0ddb60155c9 100644 --- a/drivers/hwmon/pmbus/ucd9000.c +++ b/drivers/hwmon/pmbus/ucd9000.c @@ -141,13 +141,11 @@ static int ucd9000_probe(struct i2c_client *client, block_buffer[ret] = '\0'; dev_info(&client->dev, "Device ID %s\n", block_buffer); - mid = NULL; - for (i = 0; i < ARRAY_SIZE(ucd9000_id); i++) { - mid = &ucd9000_id[i]; + for (mid = ucd9000_id; mid->name[0]; mid++) { if (!strncasecmp(mid->name, block_buffer, strlen(mid->name))) break; } - if (!mid || !strlen(mid->name)) { + if (!mid->name[0]) { dev_err(&client->dev, "Unsupported device\n"); return -ENODEV; } diff --git a/drivers/hwmon/pmbus/ucd9200.c b/drivers/hwmon/pmbus/ucd9200.c index ffcc1cf3609d..c65e9da707cc 100644 --- a/drivers/hwmon/pmbus/ucd9200.c +++ b/drivers/hwmon/pmbus/ucd9200.c @@ -68,13 +68,11 @@ static int ucd9200_probe(struct i2c_client *client, block_buffer[ret] = '\0'; dev_info(&client->dev, "Device ID %s\n", block_buffer); - mid = NULL; - for (i = 0; i < ARRAY_SIZE(ucd9200_id); i++) { - mid = &ucd9200_id[i]; + for (mid = ucd9200_id; mid->name[0]; mid++) { if (!strncasecmp(mid->name, block_buffer, strlen(mid->name))) break; } - if (!mid || !strlen(mid->name)) { + if (!mid->name[0]) { dev_err(&client->dev, "Unsupported device\n"); return -ENODEV; } From 126caf1376e75ce597f993b66241210c7171b04e Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Thu, 1 Sep 2011 10:59:36 -0700 Subject: [PATCH 150/326] OMAP: omap_device: fix !CONFIG_SUSPEND case in _noirq handlers The suspend/resume _noirq handlers were #ifdef'd out in the !CONFIG_SUSPEND case, but were still assigned to the dev_pm_ops struct. Fix by defining them to NULL in the !CONFIG_SUSPEND case. Reported-by: Arnd Bergmann Acked-by: Arnd Bergmann Signed-off-by: Kevin Hilman --- arch/arm/plat-omap/omap_device.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c index 9a6a53854911..02609eee0562 100644 --- a/arch/arm/plat-omap/omap_device.c +++ b/arch/arm/plat-omap/omap_device.c @@ -615,6 +615,9 @@ static int _od_resume_noirq(struct device *dev) return pm_generic_resume_noirq(dev); } +#else +#define _od_suspend_noirq NULL +#define _od_resume_noirq NULL #endif static struct dev_pm_domain omap_device_pm_domain = { From 7a703aded97e01d7f4a6b8440a431117399666ba Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 30 Aug 2011 14:37:37 +0800 Subject: [PATCH 151/326] i2c-pxa2xx: return proper error code in ce4100_i2c_probe error paths Signed-off-by: Axel Lin Acked-by: Sebastian Andrzej Siewior Signed-off-by: Ben Dooks --- drivers/i2c/busses/i2c-pxa-pci.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-pxa-pci.c b/drivers/i2c/busses/i2c-pxa-pci.c index 6659d269b841..b73da6cd6f91 100644 --- a/drivers/i2c/busses/i2c-pxa-pci.c +++ b/drivers/i2c/busses/i2c-pxa-pci.c @@ -109,12 +109,15 @@ static int __devinit ce4100_i2c_probe(struct pci_dev *dev, return -EINVAL; } sds = kzalloc(sizeof(*sds), GFP_KERNEL); - if (!sds) + if (!sds) { + ret = -ENOMEM; goto err_mem; + } for (i = 0; i < ARRAY_SIZE(sds->pdev); i++) { sds->pdev[i] = add_i2c_device(dev, i); if (IS_ERR(sds->pdev[i])) { + ret = PTR_ERR(sds->pdev[i]); while (--i >= 0) platform_device_unregister(sds->pdev[i]); goto err_dev_add; From 406bd18a7a39ef69f1d60a41d9de74932bcb98d4 Mon Sep 17 00:00:00 2001 From: John Bonesio Date: Tue, 30 Aug 2011 11:46:08 -0600 Subject: [PATCH 152/326] i2c-tegra: Add of_match_table This patch was intended to be part of 7ca2d1a105a239e300b937e9c41a10a4bd08f569 "i2c: Tegra: Add DeviceTree support". However, an early version of that patch, which was missing a chunk, was applied to next-i2c. This change is that missing chunk. Signed-off-by: John Bonesio Signed-off-by: Stephen Warren Signed-off-by: Ben Dooks --- drivers/i2c/busses/i2c-tegra.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c index 2440b7411978..126b4f060231 100644 --- a/drivers/i2c/busses/i2c-tegra.c +++ b/drivers/i2c/busses/i2c-tegra.c @@ -719,6 +719,17 @@ static int tegra_i2c_resume(struct platform_device *pdev) } #endif +#if defined(CONFIG_OF) +/* Match table for of_platform binding */ +static const struct of_device_id tegra_i2c_of_match[] __devinitconst = { + { .compatible = "nvidia,tegra20-i2c", }, + {}, +}; +MODULE_DEVICE_TABLE(of, tegra_i2c_of_match); +#else +#define tegra_i2c_of_match NULL +#endif + static struct platform_driver tegra_i2c_driver = { .probe = tegra_i2c_probe, .remove = tegra_i2c_remove, @@ -729,6 +740,7 @@ static struct platform_driver tegra_i2c_driver = { .driver = { .name = "tegra-i2c", .owner = THIS_MODULE, + .of_match_table = tegra_i2c_of_match, }, }; From 048e29cff95168ea3a9f176e84cc0bae54d0ae64 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Tue, 30 Aug 2011 11:46:09 -0600 Subject: [PATCH 153/326] i2c-tegra: add I2C_FUNC_SMBUS_EMUL Signed-off-by: Mike Rapoport Signed-off-by: Stephen Warren Signed-off-by: Ben Dooks --- drivers/i2c/busses/i2c-tegra.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c index 126b4f060231..17ded1d2f11d 100644 --- a/drivers/i2c/busses/i2c-tegra.c +++ b/drivers/i2c/busses/i2c-tegra.c @@ -531,7 +531,7 @@ static int tegra_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], static u32 tegra_i2c_func(struct i2c_adapter *adap) { - return I2C_FUNC_I2C; + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; } static const struct i2c_algorithm tegra_i2c_algo = { From 96219c3a257cc8ba3b3cae67efdc88be37cf7c9d Mon Sep 17 00:00:00 2001 From: Doug Anderson Date: Tue, 30 Aug 2011 11:46:10 -0600 Subject: [PATCH 154/326] i2c-tegra: fix possible race condition after tx In tegra_i2c_fill_tx_fifo, once we have finished pushing all the bytes to the I2C hardware controller, the interrupt might happen before we have updated i2c_dev->msg_buf_remaining at the end of the function. Then, in tegra_i2c_isr, we will call again tegra_i2c_fill_tx_fifo triggering weird behaviour. This has been shown to happen under real conditions. Signed-off-by: Doug Anderson Tested-by: Vincent Palatin Acked-by: Rhyland Klein Acked-by: Stephen Warren Signed-off-by: Stephen Warren Signed-off-by: Ben Dooks --- drivers/i2c/busses/i2c-tegra.c | 44 ++++++++++++++++++++++++---------- 1 file changed, 31 insertions(+), 13 deletions(-) diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c index 17ded1d2f11d..3c94c4a81a55 100644 --- a/drivers/i2c/busses/i2c-tegra.c +++ b/drivers/i2c/busses/i2c-tegra.c @@ -270,14 +270,30 @@ static int tegra_i2c_fill_tx_fifo(struct tegra_i2c_dev *i2c_dev) /* Rounds down to not include partial word at the end of buf */ words_to_transfer = buf_remaining / BYTES_PER_FIFO_WORD; - if (words_to_transfer > tx_fifo_avail) - words_to_transfer = tx_fifo_avail; - i2c_writesl(i2c_dev, buf, I2C_TX_FIFO, words_to_transfer); + /* It's very common to have < 4 bytes, so optimize that case. */ + if (words_to_transfer) { + if (words_to_transfer > tx_fifo_avail) + words_to_transfer = tx_fifo_avail; - buf += words_to_transfer * BYTES_PER_FIFO_WORD; - buf_remaining -= words_to_transfer * BYTES_PER_FIFO_WORD; - tx_fifo_avail -= words_to_transfer; + /* + * Update state before writing to FIFO. If this casues us + * to finish writing all bytes (AKA buf_remaining goes to 0) we + * have a potential for an interrupt (PACKET_XFER_COMPLETE is + * not maskable). We need to make sure that the isr sees + * buf_remaining as 0 and doesn't call us back re-entrantly. + */ + buf_remaining -= words_to_transfer * BYTES_PER_FIFO_WORD; + tx_fifo_avail -= words_to_transfer; + i2c_dev->msg_buf_remaining = buf_remaining; + i2c_dev->msg_buf = buf + + words_to_transfer * BYTES_PER_FIFO_WORD; + barrier(); + + i2c_writesl(i2c_dev, buf, I2C_TX_FIFO, words_to_transfer); + + buf += words_to_transfer * BYTES_PER_FIFO_WORD; + } /* * If there is a partial word at the end of buf, handle it manually to @@ -287,14 +303,15 @@ static int tegra_i2c_fill_tx_fifo(struct tegra_i2c_dev *i2c_dev) if (tx_fifo_avail > 0 && buf_remaining > 0) { BUG_ON(buf_remaining > 3); memcpy(&val, buf, buf_remaining); + + /* Again update before writing to FIFO to make sure isr sees. */ + i2c_dev->msg_buf_remaining = 0; + i2c_dev->msg_buf = NULL; + barrier(); + i2c_writel(i2c_dev, val, I2C_TX_FIFO); - buf_remaining = 0; - tx_fifo_avail--; } - BUG_ON(tx_fifo_avail > 0 && buf_remaining > 0); - i2c_dev->msg_buf_remaining = buf_remaining; - i2c_dev->msg_buf = buf; return 0; } @@ -411,9 +428,10 @@ static irqreturn_t tegra_i2c_isr(int irq, void *dev_id) tegra_i2c_mask_irq(i2c_dev, I2C_INT_TX_FIFO_DATA_REQ); } - if ((status & I2C_INT_PACKET_XFER_COMPLETE) && - !i2c_dev->msg_buf_remaining) + if (status & I2C_INT_PACKET_XFER_COMPLETE) { + BUG_ON(i2c_dev->msg_buf_remaining); complete(&i2c_dev->msg_complete); + } i2c_writel(i2c_dev, status, I2C_INT_STATUS); if (i2c_dev->is_dvc) From bb9ea77846620ed2b37e74c852d72c7a476b248c Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 6 Sep 2011 08:08:13 +0100 Subject: [PATCH 155/326] ARM: 7081/1: mach-integrator: fix the clocksource I was intrigued by the fact that the clock stood still on the Integrator, but it wasn't strange at all, because the timer was set up all wrong and probably has been for a while. With this patch the clock starts ticking again: make the timer periodic (reload), |= on the divisor bit and load the timer before starting it. Cc: stable@kernel.org Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/mach-integrator/integrator_ap.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-integrator/integrator_ap.c b/arch/arm/mach-integrator/integrator_ap.c index 2fbbdd5eac35..fcf0ae95651f 100644 --- a/arch/arm/mach-integrator/integrator_ap.c +++ b/arch/arm/mach-integrator/integrator_ap.c @@ -337,15 +337,15 @@ static unsigned long timer_reload; static void integrator_clocksource_init(u32 khz) { void __iomem *base = (void __iomem *)TIMER2_VA_BASE; - u32 ctrl = TIMER_CTRL_ENABLE; + u32 ctrl = TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC; if (khz >= 1500) { khz /= 16; - ctrl = TIMER_CTRL_DIV16; + ctrl |= TIMER_CTRL_DIV16; } - writel(ctrl, base + TIMER_CTRL); writel(0xffff, base + TIMER_LOAD); + writel(ctrl, base + TIMER_CTRL); clocksource_mmio_init(base + TIMER_VALUE, "timer2", khz * 1000, 200, 16, clocksource_mmio_readl_down); From bac7e6ecf60933b68af910eb4c83a775a8b20b19 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 6 Sep 2011 07:45:46 +0100 Subject: [PATCH 156/326] ARM: 7080/1: l2x0: make sure I&D are not locked down on init Fighting unfixed U-Boots and other beasts that may the cache in a locked-down state when starting the kernel, we make sure to disable all cache lock-down when initializing the l2x0 so we are in a known state. Cc: Srinidhi Kasagar Cc: Rabin Vincent Cc: Adrian Bunk Cc: Rob Herring Cc: Catalin Marinas Cc: Will Deacon Reviewed-by: Santosh Shilimkar Reported-by: Jan Rinze Tested-by: Robert Marklund Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/include/asm/hardware/cache-l2x0.h | 9 +++++++-- arch/arm/mm/cache-l2x0.c | 21 +++++++++++++++++++++ 2 files changed, 28 insertions(+), 2 deletions(-) diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h index bfa706ffd968..99a6ed7e1bfd 100644 --- a/arch/arm/include/asm/hardware/cache-l2x0.h +++ b/arch/arm/include/asm/hardware/cache-l2x0.h @@ -45,8 +45,13 @@ #define L2X0_CLEAN_INV_LINE_PA 0x7F0 #define L2X0_CLEAN_INV_LINE_IDX 0x7F8 #define L2X0_CLEAN_INV_WAY 0x7FC -#define L2X0_LOCKDOWN_WAY_D 0x900 -#define L2X0_LOCKDOWN_WAY_I 0x904 +/* + * The lockdown registers repeat 8 times for L310, the L210 has only one + * D and one I lockdown register at 0x0900 and 0x0904. + */ +#define L2X0_LOCKDOWN_WAY_D_BASE 0x900 +#define L2X0_LOCKDOWN_WAY_I_BASE 0x904 +#define L2X0_LOCKDOWN_STRIDE 0x08 #define L2X0_TEST_OPERATION 0xF00 #define L2X0_LINE_DATA 0xF10 #define L2X0_LINE_TAG 0xF30 diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c index 44c086710d2b..9ecfdb511951 100644 --- a/arch/arm/mm/cache-l2x0.c +++ b/arch/arm/mm/cache-l2x0.c @@ -277,6 +277,25 @@ static void l2x0_disable(void) spin_unlock_irqrestore(&l2x0_lock, flags); } +static void __init l2x0_unlock(__u32 cache_id) +{ + int lockregs; + int i; + + if (cache_id == L2X0_CACHE_ID_PART_L310) + lockregs = 8; + else + /* L210 and unknown types */ + lockregs = 1; + + for (i = 0; i < lockregs; i++) { + writel_relaxed(0x0, l2x0_base + L2X0_LOCKDOWN_WAY_D_BASE + + i * L2X0_LOCKDOWN_STRIDE); + writel_relaxed(0x0, l2x0_base + L2X0_LOCKDOWN_WAY_I_BASE + + i * L2X0_LOCKDOWN_STRIDE); + } +} + void __init l2x0_init(void __iomem *base, __u32 aux_val, __u32 aux_mask) { __u32 aux; @@ -328,6 +347,8 @@ void __init l2x0_init(void __iomem *base, __u32 aux_val, __u32 aux_mask) * accessing the below registers will fault. */ if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & 1)) { + /* Make sure that I&D is not locked down when starting */ + l2x0_unlock(cache_id); /* l2x0 controller is disabled */ writel_relaxed(aux, l2x0_base + L2X0_AUX_CTRL); From 810198bc9c109489dfadc57131c5183ce6ad2d7d Mon Sep 17 00:00:00 2001 From: "Rajashekhara, Sudhakar" Date: Tue, 12 Jul 2011 15:58:53 +0530 Subject: [PATCH 157/326] ARM: davinci: da850 EVM: read mac address from SPI flash DA850/OMAP-L138 EMAC driver uses random mac address instead of a fixed one because the mac address is not stuffed into EMAC platform data. This patch provides a function which reads the mac address stored in SPI flash (registered as MTD device) and populates the EMAC platform data. The function which reads the mac address is registered as a callback which gets called upon addition of MTD device. NOTE: In case the MAC address stored in SPI flash is erased, follow the instructions at [1] to restore it. [1] http://processors.wiki.ti.com/index.php/GSG:_OMAP-L138_DVEVM_Additional_Procedures#Restoring_MAC_address_on_SPI_Flash Modifications in v2: Guarded registering the mtd_notifier only when MTD is enabled. Earlier this was handled using mtd_has_partitions() call, but this has been removed in Linux v3.0. Modifications in v3: a. Guarded da850_evm_m25p80_notify_add() function and da850evm_spi_notifier structure with CONFIG_MTD macros. b. Renamed da850_evm_register_mtd_user() function to da850_evm_setup_mac_addr() and removed the struct mtd_notifier argument to this function. c. Passed the da850evm_spi_notifier structure to register_mtd_user() function. Modifications in v4: Moved the da850_evm_setup_mac_addr() function within the first CONFIG_MTD ifdef construct. Signed-off-by: Rajashekhara, Sudhakar Signed-off-by: Sekhar Nori Cc: stable@kernel.org --- arch/arm/mach-davinci/board-da850-evm.c | 28 +++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/arch/arm/mach-davinci/board-da850-evm.c b/arch/arm/mach-davinci/board-da850-evm.c index bd5394537c88..008d51407cd7 100644 --- a/arch/arm/mach-davinci/board-da850-evm.c +++ b/arch/arm/mach-davinci/board-da850-evm.c @@ -115,6 +115,32 @@ static struct spi_board_info da850evm_spi_info[] = { }, }; +#ifdef CONFIG_MTD +static void da850_evm_m25p80_notify_add(struct mtd_info *mtd) +{ + char *mac_addr = davinci_soc_info.emac_pdata->mac_addr; + size_t retlen; + + if (!strcmp(mtd->name, "MAC-Address")) { + mtd->read(mtd, 0, ETH_ALEN, &retlen, mac_addr); + if (retlen == ETH_ALEN) + pr_info("Read MAC addr from SPI Flash: %pM\n", + mac_addr); + } +} + +static struct mtd_notifier da850evm_spi_notifier = { + .add = da850_evm_m25p80_notify_add, +}; + +static void da850_evm_setup_mac_addr(void) +{ + register_mtd_user(&da850evm_spi_notifier); +} +#else +static void da850_evm_setup_mac_addr(void) { } +#endif + static struct mtd_partition da850_evm_norflash_partition[] = { { .name = "bootloaders + env", @@ -1244,6 +1270,8 @@ static __init void da850_evm_init(void) if (ret) pr_warning("da850_evm_init: sata registration failed: %d\n", ret); + + da850_evm_setup_mac_addr(); } #ifdef CONFIG_SERIAL_8250_CONSOLE From c08748005d56084a205f5c5db5f211b07a97a9be Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Fri, 8 Jul 2011 19:24:57 +0400 Subject: [PATCH 158/326] ARM: davinci: correct MDSTAT_STATE_MASK MDSTAT.STATE occupies bits 0..5 according to all available documentation, so fix the #define MDSTAT_STATE_MASK at last. Using the wrong value seems to have been harmless though... Signed-off-by: Sergei Shtylyov Signed-off-by: Sekhar Nori --- arch/arm/mach-davinci/include/mach/psc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-davinci/include/mach/psc.h b/arch/arm/mach-davinci/include/mach/psc.h index 47fd0bc3d3e7..fa59c097223d 100644 --- a/arch/arm/mach-davinci/include/mach/psc.h +++ b/arch/arm/mach-davinci/include/mach/psc.h @@ -243,7 +243,7 @@ #define PSC_STATE_DISABLE 2 #define PSC_STATE_ENABLE 3 -#define MDSTAT_STATE_MASK 0x1f +#define MDSTAT_STATE_MASK 0x3f #define MDCTL_FORCE BIT(31) #ifndef __ASSEMBLER__ From dde58cfcc3b6dd2f160ffd355f76ae526155a4df Mon Sep 17 00:00:00 2001 From: David Herrmann Date: Mon, 5 Sep 2011 18:45:28 +0200 Subject: [PATCH 159/326] HID: wacom: Fix error path of power-supply initialization power_supply_unregister() must not be called if power_supply_register() failed. The wdata->psy.dev pointer may point to invalid memory after a failed power_supply_register() and hence wacom_remove() will fail while calling power_supply_unregister(). This changes the wacom_probe function to fail if it cannot register the power_supply devices. If we would want to keep the previous behaviour we had to keep some flag about the power_supply state and check it on wacom_remove, but this seems inappropriate here. Hence, we simply fail, too, if power_supply_register fails. Signed-off-by: David Herrmann Signed-off-by: Jiri Kosina --- drivers/hid/hid-wacom.c | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c index 06888323828c..f66a597cff63 100644 --- a/drivers/hid/hid-wacom.c +++ b/drivers/hid/hid-wacom.c @@ -353,11 +353,7 @@ static int wacom_probe(struct hid_device *hdev, if (ret) { hid_warn(hdev, "can't create sysfs battery attribute, err: %d\n", ret); - /* - * battery attribute is not critical for the tablet, but if it - * failed then there is no need to create ac attribute - */ - goto move_on; + goto err_battery; } wdata->ac.properties = wacom_ac_props; @@ -371,14 +367,8 @@ static int wacom_probe(struct hid_device *hdev, if (ret) { hid_warn(hdev, "can't create ac battery attribute, err: %d\n", ret); - /* - * ac attribute is not critical for the tablet, but if it - * failed then we don't want to battery attribute to exist - */ - power_supply_unregister(&wdata->battery); + goto err_ac; } - -move_on: #endif hidinput = list_entry(hdev->inputs.next, struct hid_input, list); input = hidinput->input; @@ -416,6 +406,13 @@ static int wacom_probe(struct hid_device *hdev, return 0; +#ifdef CONFIG_HID_WACOM_POWER_SUPPLY +err_ac: + power_supply_unregister(&wdata->battery); +err_battery: + device_remove_file(&hdev->dev, &dev_attr_speed); + hid_hw_stop(hdev); +#endif err_free: kfree(wdata); return ret; From 9086617ea3a7f3e574ca64392b827bdd56f607eb Mon Sep 17 00:00:00 2001 From: David Herrmann Date: Mon, 5 Sep 2011 18:45:29 +0200 Subject: [PATCH 160/326] HID: wacom: Unregister sysfs attributes on remove HID devices can be hotplugged so we should unregister all sysfs attributes when removing a driver. Otherwise, manually unloading the wacom-driver will not remove the sysfs attributes. Only when the device is disconnected, they are removed, eventually. Signed-off-by: David Herrmann Signed-off-by: Jiri Kosina --- drivers/hid/hid-wacom.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c index f66a597cff63..a597039d0755 100644 --- a/drivers/hid/hid-wacom.c +++ b/drivers/hid/hid-wacom.c @@ -423,6 +423,7 @@ static void wacom_remove(struct hid_device *hdev) #ifdef CONFIG_HID_WACOM_POWER_SUPPLY struct wacom_data *wdata = hid_get_drvdata(hdev); #endif + device_remove_file(&hdev->dev, &dev_attr_speed); hid_hw_stop(hdev); #ifdef CONFIG_HID_WACOM_POWER_SUPPLY From 897a6a1a14837d6d582bfd1fd7aba00be44b6469 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 2 Aug 2011 17:48:38 +0200 Subject: [PATCH 161/326] ARM: davinci: fix cache flush build error The TNET variant of DaVinci compiles some code that it shares with other DaVinci variants, however it has a V6 CPU rather than an ARM926T, thus the hardcoded call to arm926_flush_kern_cache_all() in sleep.S will obviously fail, and we need to build with the v6_flush_kern_cache_all() call instead. This was triggered by manually altering the DaVinci config to build the TNET version. Cc: Dave Martin Cc: Arnd Bergmann Signed-off-by: Linus Walleij Signed-off-by: Sekhar Nori Cc: stable@kernel.org --- arch/arm/mach-davinci/sleep.S | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-davinci/sleep.S b/arch/arm/mach-davinci/sleep.S index fb5e72b532b0..5f1e045a3ad1 100644 --- a/arch/arm/mach-davinci/sleep.S +++ b/arch/arm/mach-davinci/sleep.S @@ -217,7 +217,11 @@ ddr2clk_stop_done: ENDPROC(davinci_ddr_psc_config) CACHE_FLUSH: - .word arm926_flush_kern_cache_all +#ifdef CONFIG_CPU_V6 + .word v6_flush_kern_cache_all +#else + .word arm926_flush_kern_cache_all +#endif ENTRY(davinci_cpu_suspend_sz) .word . - davinci_cpu_suspend From 3512069eefd3c3424b12f21a68fd473c3fd57220 Mon Sep 17 00:00:00 2001 From: Jason Gerecke Date: Thu, 8 Sep 2011 09:38:14 -0700 Subject: [PATCH 162/326] Input: wacom - add POINTER and DIRECT device properties Adds INPUT_PROP_POINTER or INPUT_PROP_DIRECT as necessary to the hardware supported by the Wacom driver. The DIRECT property is assigned to devices with an embedded screen (i.e. touchscreens and display tablets). The POINTER property is assigned to those without embedded screens. Signed-off-by: Jason Gerecke Reviewed-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- drivers/hid/hid-wacom.c | 2 ++ drivers/input/tablet/wacom_wac.c | 25 ++++++++++++++++++++++++- drivers/input/touchscreen/wacom_w8001.c | 2 ++ 3 files changed, 28 insertions(+), 1 deletion(-) diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c index 06888323828c..5b9267dd3d66 100644 --- a/drivers/hid/hid-wacom.c +++ b/drivers/hid/hid-wacom.c @@ -383,6 +383,8 @@ static int wacom_probe(struct hid_device *hdev, hidinput = list_entry(hdev->inputs.next, struct hid_input, list); input = hidinput->input; + __set_bit(INPUT_PROP_POINTER, input->propbit); + /* Basics */ input->evbit[0] |= BIT(EV_KEY) | BIT(EV_ABS) | BIT(EV_REL); diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index 2d88316d0e54..c31e4e9f2690 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -1098,6 +1098,8 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, __set_bit(BTN_TOOL_MOUSE, input_dev->keybit); __set_bit(BTN_STYLUS, input_dev->keybit); __set_bit(BTN_STYLUS2, input_dev->keybit); + + __set_bit(INPUT_PROP_POINTER, input_dev->propbit); break; case WACOM_21UX2: @@ -1126,6 +1128,9 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, } input_set_abs_params(input_dev, ABS_Z, -900, 899, 0, 0); + + __set_bit(INPUT_PROP_DIRECT, input_dev->propbit); + wacom_setup_cintiq(wacom_wac); break; @@ -1150,6 +1155,8 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, /* fall through */ case INTUOS: + __set_bit(INPUT_PROP_POINTER, input_dev->propbit); + wacom_setup_intuos(wacom_wac); break; @@ -1165,6 +1172,8 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, input_set_abs_params(input_dev, ABS_Z, -900, 899, 0, 0); wacom_setup_intuos(wacom_wac); + + __set_bit(INPUT_PROP_POINTER, input_dev->propbit); break; case TABLETPC2FG: @@ -1183,14 +1192,24 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, case TABLETPC: __clear_bit(ABS_MISC, input_dev->absbit); + __set_bit(INPUT_PROP_DIRECT, input_dev->propbit); + if (features->device_type != BTN_TOOL_PEN) break; /* no need to process stylus stuff */ /* fall through */ case PL: - case PTU: case DTU: + __set_bit(BTN_TOOL_PEN, input_dev->keybit); + __set_bit(BTN_TOOL_RUBBER, input_dev->keybit); + __set_bit(BTN_STYLUS, input_dev->keybit); + __set_bit(BTN_STYLUS2, input_dev->keybit); + + __set_bit(INPUT_PROP_DIRECT, input_dev->propbit); + break; + + case PTU: __set_bit(BTN_STYLUS2, input_dev->keybit); /* fall through */ @@ -1198,11 +1217,15 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, __set_bit(BTN_TOOL_PEN, input_dev->keybit); __set_bit(BTN_TOOL_RUBBER, input_dev->keybit); __set_bit(BTN_STYLUS, input_dev->keybit); + + __set_bit(INPUT_PROP_POINTER, input_dev->propbit); break; case BAMBOO_PT: __clear_bit(ABS_MISC, input_dev->absbit); + __set_bit(INPUT_PROP_POINTER, input_dev->propbit); + if (features->device_type == BTN_TOOL_DOUBLETAP) { __set_bit(BTN_LEFT, input_dev->keybit); __set_bit(BTN_FORWARD, input_dev->keybit); diff --git a/drivers/input/touchscreen/wacom_w8001.c b/drivers/input/touchscreen/wacom_w8001.c index c14412ef4648..9941d39df43d 100644 --- a/drivers/input/touchscreen/wacom_w8001.c +++ b/drivers/input/touchscreen/wacom_w8001.c @@ -383,6 +383,8 @@ static int w8001_setup(struct w8001 *w8001) dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS); strlcat(w8001->name, "Wacom Serial", sizeof(w8001->name)); + __set_bit(INPUT_PROP_DIRECT, dev->propbit); + /* penabled? */ error = w8001_command(w8001, W8001_CMD_QUERY, true); if (!error) { From f10cd522c5fbfec9ae3cc01967868c9c2401ed23 Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Tue, 6 Sep 2011 17:41:47 +0100 Subject: [PATCH 163/326] xen: disable PV spinlocks on HVM PV spinlocks cannot possibly work with the current code because they are enabled after pvops patching has already been done, and because PV spinlocks use a different data structure than native spinlocks so we cannot switch between them dynamically. A spinlock that has been taken once by the native code (__ticket_spin_lock) cannot be taken by __xen_spin_lock even after it has been released. Reported-and-Tested-by: Stefan Bader Signed-off-by: Stefano Stabellini Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/smp.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index d4fc6d454f8d..041d4fe9dfe4 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -532,7 +532,6 @@ static void __init xen_hvm_smp_prepare_cpus(unsigned int max_cpus) WARN_ON(xen_smp_intr_init(0)); xen_init_lock_cpu(0); - xen_init_spinlocks(); } static int __cpuinit xen_hvm_cpu_up(unsigned int cpu) From d2f152878d457a84f3708acee5f682322386a79b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 8 Sep 2011 10:16:50 -0700 Subject: [PATCH 164/326] wireless: fix kernel-doc warning in net/cfg80211.h Fix kernel-doc warning in net/cfg80211.h: Warning(include/net/cfg80211.h:1884): No description found for parameter 'registered' Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds --- include/net/cfg80211.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 408ae4882d22..401d73bd151f 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1744,6 +1744,8 @@ struct wiphy_wowlan_support { * by default for perm_addr. In this case, the mask should be set to * all-zeroes. In this case it is assumed that the device can handle * the same number of arbitrary MAC addresses. + * @registered: protects ->resume and ->suspend sysfs callbacks against + * unregister hardware * @debugfsdir: debugfs directory used for this wiphy, will be renamed * automatically on wiphy renames * @dev: (virtual) struct device for this wiphy From bff747c58cf97bf4fc8b499ee0f419b59d6b226d Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 8 Sep 2011 10:16:47 -0700 Subject: [PATCH 165/326] regulator: fix kernel-doc warning in consumer.h Fix kernel-doc warning about internal/private data by marking it as "private:" so that kernel-doc will ignore it. Warning(include/linux/regulator/consumer.h:128): No description found for parameter 'ret' Signed-off-by: Randy Dunlap Acked-by: Mark Brown Signed-off-by: Linus Torvalds --- include/linux/regulator/consumer.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/regulator/consumer.h b/include/linux/regulator/consumer.h index 26f6ea4444e3..b47771aa5718 100644 --- a/include/linux/regulator/consumer.h +++ b/include/linux/regulator/consumer.h @@ -123,7 +123,7 @@ struct regulator_bulk_data { const char *supply; struct regulator *consumer; - /* Internal use */ + /* private: Internal use */ int ret; }; From ffbc559b0699891c6deb9fd2b4750671eab94999 Mon Sep 17 00:00:00 2001 From: Emil Velikov Date: Sun, 21 Aug 2011 22:48:12 +0100 Subject: [PATCH 166/326] drm/nv50/crtc: Bail out if FB is not bound to crtc Fixes possbile NULL pointer dereference Resolves 'kernel crash in nv50_crtc_do_mode_set_base during shutdown' https://bugs.freedesktop.org/show_bug.cgi?id=40005 Signed-off-by: Emil Velikov Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nv50_crtc.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nv50_crtc.c b/drivers/gpu/drm/nouveau/nv50_crtc.c index 46ad59ea2185..5d989073ba6e 100644 --- a/drivers/gpu/drm/nouveau/nv50_crtc.c +++ b/drivers/gpu/drm/nouveau/nv50_crtc.c @@ -519,12 +519,18 @@ nv50_crtc_do_mode_set_base(struct drm_crtc *crtc, struct drm_device *dev = nv_crtc->base.dev; struct drm_nouveau_private *dev_priv = dev->dev_private; struct nouveau_channel *evo = nv50_display(dev)->master; - struct drm_framebuffer *drm_fb = nv_crtc->base.fb; - struct nouveau_framebuffer *fb = nouveau_framebuffer(drm_fb); + struct drm_framebuffer *drm_fb; + struct nouveau_framebuffer *fb; int ret; NV_DEBUG_KMS(dev, "index %d\n", nv_crtc->index); + /* no fb bound */ + if (!atomic && !crtc->fb) { + NV_DEBUG_KMS(dev, "No FB bound\n"); + return 0; + } + /* If atomic, we want to switch to the fb we were passed, so * now we update pointers to do that. (We don't pin; just * assume we're already pinned and update the base address.) @@ -533,6 +539,8 @@ nv50_crtc_do_mode_set_base(struct drm_crtc *crtc, drm_fb = passed_fb; fb = nouveau_framebuffer(passed_fb); } else { + drm_fb = crtc->fb; + fb = nouveau_framebuffer(crtc->fb); /* If not atomic, we can go ahead and pin, and unpin the * old fb we were passed. */ From cfd8be088e97a762902a4820f501fb13102984e9 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Tue, 23 Aug 2011 10:23:11 +1000 Subject: [PATCH 167/326] drm/nouveau: fix oops on pre-semaphore hardware Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_fence.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 8d02d875376d..c919cfc8f2fd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -530,7 +530,8 @@ nouveau_fence_channel_init(struct nouveau_channel *chan) nouveau_gpuobj_ref(NULL, &obj); if (ret) return ret; - } else { + } else + if (USE_SEMA(dev)) { /* map fence bo into channel's vm */ ret = nouveau_bo_vma_add(dev_priv->fence.bo, chan->vm, &chan->fence.vma); From 17c8b960930da3599e47801a54ac0ea1070545d2 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 22 Aug 2011 23:14:05 +0200 Subject: [PATCH 168/326] drm/nouveau: properly handle allocation failure in nouveau_sgdma_populate Not cleaning after alloc failure would result in crash on destroy, because nouveau_sgdma_clear assumes "ttm_alloced" to be not null when "pages" is not null. Signed-off-by: Marcin Slusarz Cc: stable@kernel.org Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_sgdma.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c index c444cadbf849..88062de26b00 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c @@ -37,8 +37,11 @@ nouveau_sgdma_populate(struct ttm_backend *be, unsigned long num_pages, return -ENOMEM; nvbe->ttm_alloced = kmalloc(sizeof(bool) * num_pages, GFP_KERNEL); - if (!nvbe->ttm_alloced) + if (!nvbe->ttm_alloced) { + kfree(nvbe->pages); + nvbe->pages = NULL; return -ENOMEM; + } nvbe->nr_pages = 0; while (num_pages--) { From 1bf27066017c820b8ab2a1ac8430ea470c2de0c3 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 22 Aug 2011 23:22:13 +0200 Subject: [PATCH 169/326] drm/nouveau: fix nv04_sgdma_bind on non-"4kB pages" archs nv04_sgdma_bind binds the same page multiple times on architectures where PAGE_SIZE != 4096. Let's fix it. Signed-off-by: Marcin Slusarz Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_sgdma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c index 88062de26b00..2706cb3d871a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c +++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c @@ -129,7 +129,7 @@ nv04_sgdma_bind(struct ttm_backend *be, struct ttm_mem_reg *mem) for (j = 0; j < PAGE_SIZE / NV_CTXDMA_PAGE_SIZE; j++, pte++) { nv_wo32(gpuobj, (pte * 4) + 0, offset_l | 3); - dma_offset += NV_CTXDMA_PAGE_SIZE; + offset_l += NV_CTXDMA_PAGE_SIZE; } } From 0e83bb4eee1c504ab98367b4f7d1bc337ab213d2 Mon Sep 17 00:00:00 2001 From: Emil Velikov Date: Thu, 25 Aug 2011 21:36:51 +0100 Subject: [PATCH 170/326] drm/nv04/crtc: Bail out if FB is not bound to crtc This commit resolves a possible 'NULL pointer dereference' It uses the same approach as radeon, intel and nouveau/nv50 Fixes bug 'Nouveau: Kernel oops when unplugging external monitor' https://bugs.freedesktop.org/show_bug.cgi?id=40336 Signed-off-by: Emil Velikov Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nv04_crtc.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nv04_crtc.c b/drivers/gpu/drm/nouveau/nv04_crtc.c index 118261d4927a..5e45398a9e2d 100644 --- a/drivers/gpu/drm/nouveau/nv04_crtc.c +++ b/drivers/gpu/drm/nouveau/nv04_crtc.c @@ -781,11 +781,20 @@ nv04_crtc_do_mode_set_base(struct drm_crtc *crtc, struct drm_device *dev = crtc->dev; struct drm_nouveau_private *dev_priv = dev->dev_private; struct nv04_crtc_reg *regp = &dev_priv->mode_reg.crtc_reg[nv_crtc->index]; - struct drm_framebuffer *drm_fb = nv_crtc->base.fb; - struct nouveau_framebuffer *fb = nouveau_framebuffer(drm_fb); + struct drm_framebuffer *drm_fb; + struct nouveau_framebuffer *fb; int arb_burst, arb_lwm; int ret; + NV_DEBUG_KMS(dev, "index %d\n", nv_crtc->index); + + /* no fb bound */ + if (!atomic && !crtc->fb) { + NV_DEBUG_KMS(dev, "No FB bound\n"); + return 0; + } + + /* If atomic, we want to switch to the fb we were passed, so * now we update pointers to do that. (We don't pin; just * assume we're already pinned and update the base address.) @@ -794,6 +803,8 @@ nv04_crtc_do_mode_set_base(struct drm_crtc *crtc, drm_fb = passed_fb; fb = nouveau_framebuffer(passed_fb); } else { + drm_fb = crtc->fb; + fb = nouveau_framebuffer(crtc->fb); /* If not atomic, we can go ahead and pin, and unpin the * old fb we were passed. */ From 55a01f6f6840b6310b073afabda649727d2ddb24 Mon Sep 17 00:00:00 2001 From: Lin Ming Date: Wed, 7 Sep 2011 22:58:09 +0800 Subject: [PATCH 171/326] drm: Remove duplicate "return" statement Remove the duplicate "return" statement in drm_fb_helper_panic(). Signed-off-by: Lin Ming Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_fb_helper.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 802b61ac3139..f7c6854eb4dd 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -256,7 +256,6 @@ int drm_fb_helper_panic(struct notifier_block *n, unsigned long ununsed, { printk(KERN_ERR "panic occurred, switching back to text console\n"); return drm_fb_helper_force_kernel_mode(); - return 0; } EXPORT_SYMBOL(drm_fb_helper_panic); From 83a497cab1baec75e3e493a96e3456db14729ce0 Mon Sep 17 00:00:00 2001 From: Tommy Lin Date: Fri, 29 Jul 2011 01:14:46 +0800 Subject: [PATCH 172/326] ARM: cns3xxx: Fix compile error caused by hardware.h removed Commit c9d95fbe59e426eed7f16e7cac812e46ac4772d0 "ARM: convert PCI defines to variables" deleted cns3xxx' hardware.h, but didn't remove references for it, so do it now. This patch removes lines that refer to hardware.h. Signed-off-by: Tommy Lin Signed-off-by: Imre Kaloz Signed-off-by: Anton Vorontsov --- arch/arm/mach-cns3xxx/include/mach/entry-macro.S | 1 - arch/arm/mach-cns3xxx/include/mach/system.h | 1 - arch/arm/mach-cns3xxx/include/mach/uncompress.h | 1 - 3 files changed, 3 deletions(-) diff --git a/arch/arm/mach-cns3xxx/include/mach/entry-macro.S b/arch/arm/mach-cns3xxx/include/mach/entry-macro.S index 6bd83ed90afe..d87bfc397d39 100644 --- a/arch/arm/mach-cns3xxx/include/mach/entry-macro.S +++ b/arch/arm/mach-cns3xxx/include/mach/entry-macro.S @@ -8,7 +8,6 @@ * published by the Free Software Foundation. */ -#include #include .macro disable_fiq diff --git a/arch/arm/mach-cns3xxx/include/mach/system.h b/arch/arm/mach-cns3xxx/include/mach/system.h index 58bb03ae3cf4..4f16c9b79f78 100644 --- a/arch/arm/mach-cns3xxx/include/mach/system.h +++ b/arch/arm/mach-cns3xxx/include/mach/system.h @@ -13,7 +13,6 @@ #include #include -#include static inline void arch_idle(void) { diff --git a/arch/arm/mach-cns3xxx/include/mach/uncompress.h b/arch/arm/mach-cns3xxx/include/mach/uncompress.h index de8ead9b91f7..a91b6058ab4f 100644 --- a/arch/arm/mach-cns3xxx/include/mach/uncompress.h +++ b/arch/arm/mach-cns3xxx/include/mach/uncompress.h @@ -8,7 +8,6 @@ */ #include -#include #include #define AMBA_UART_DR(base) (*(volatile unsigned char *)((base) + 0x00)) From 7caaf7efb98a4f713e4c8b16289e03874aace493 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 9 Sep 2011 23:18:00 +0400 Subject: [PATCH 173/326] ARM: cns3xxx: Fix newly introduced warnings in the PCIe code commit d5341942d784134f2997b3ff82cd63cf71d1f932 ("PCI: Make the struct pci_dev * argument of pci_fixup_irqs const") did not change argument of pdev_to_cnspci(), and thus introduced the following warnings: CHECK arch/arm/mach-cns3xxx/pcie.c pcie.c:177:60: warning: incorrect type in argument 1 (different modifiers) pcie.c:177:60: expected struct pci_dev *dev pcie.c:177:60: got struct pci_dev const *dev CC arch/arm/mach-cns3xxx/pcie.o pcie.c: In function 'cns3xxx_pcie_map_irq': pcie.c:177: warning: passing argument 1 of 'pdev_to_cnspci' discards qualifiers from pointer target type pcie.c:52: note: expected 'struct pci_dev *' but argument is of type 'const struct pci_dev *' This patch fixes the issue. Signed-off-by: Anton Vorontsov --- arch/arm/mach-cns3xxx/pcie.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-cns3xxx/pcie.c b/arch/arm/mach-cns3xxx/pcie.c index 06fd25d70aec..0f8fca48a5ed 100644 --- a/arch/arm/mach-cns3xxx/pcie.c +++ b/arch/arm/mach-cns3xxx/pcie.c @@ -49,7 +49,7 @@ static struct cns3xxx_pcie *sysdata_to_cnspci(void *sysdata) return &cns3xxx_pcie[root->domain]; } -static struct cns3xxx_pcie *pdev_to_cnspci(struct pci_dev *dev) +static struct cns3xxx_pcie *pdev_to_cnspci(const struct pci_dev *dev) { return sysdata_to_cnspci(dev->sysdata); } From 1c601beaf21671b5033169d04efeda462bf58f01 Mon Sep 17 00:00:00 2001 From: Pieter-Augustijn Van Malleghem Date: Fri, 9 Sep 2011 13:29:45 -0700 Subject: [PATCH 174/326] Input: bcm5974 - add MacBookAir4,1 trackpad support This patch adds trackpad support for the MacBookAir4,1, released in July 2011. It is very similar to the MacBookAir4,2 patch submitted by Joshua Dillon and Chase Douglas. Signed-off-by: Pieter-Augustijn Van Malleghem Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/bcm5974.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c index da280189ef07..5ec617e28f7e 100644 --- a/drivers/input/mouse/bcm5974.c +++ b/drivers/input/mouse/bcm5974.c @@ -67,6 +67,10 @@ #define USB_DEVICE_ID_APPLE_WELLSPRING5_ANSI 0x0245 #define USB_DEVICE_ID_APPLE_WELLSPRING5_ISO 0x0246 #define USB_DEVICE_ID_APPLE_WELLSPRING5_JIS 0x0247 +/* MacbookAir4,1 (unibody, July 2011) */ +#define USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI 0x0249 +#define USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO 0x024a +#define USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS 0x024b /* MacbookAir4,2 (unibody, July 2011) */ #define USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI 0x024c #define USB_DEVICE_ID_APPLE_WELLSPRING6_ISO 0x024d @@ -112,6 +116,10 @@ static const struct usb_device_id bcm5974_table[] = { BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING5_ANSI), BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING5_ISO), BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING5_JIS), + /* MacbookAir4,1 */ + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI), + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO), + BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS), /* MacbookAir4,2 */ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING6_ANSI), BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING6_ISO), @@ -334,6 +342,18 @@ static const struct bcm5974_config bcm5974_config_table[] = { { DIM_X, DIM_X / SN_COORD, -4750, 5280 }, { DIM_Y, DIM_Y / SN_COORD, -150, 6730 } }, + { + USB_DEVICE_ID_APPLE_WELLSPRING6A_ANSI, + USB_DEVICE_ID_APPLE_WELLSPRING6A_ISO, + USB_DEVICE_ID_APPLE_WELLSPRING6A_JIS, + HAS_INTEGRATED_BUTTON, + 0x84, sizeof(struct bt_data), + 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS, + { DIM_PRESSURE, DIM_PRESSURE / SN_PRESSURE, 0, 300 }, + { DIM_WIDTH, DIM_WIDTH / SN_WIDTH, 0, 2048 }, + { DIM_X, DIM_X / SN_COORD, -4620, 5140 }, + { DIM_Y, DIM_Y / SN_COORD, -150, 6600 } + }, {} }; From 0ec26fd0698285b31248e34bf1abb022c00f23d6 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 5 Sep 2011 18:06:26 +0200 Subject: [PATCH 175/326] vfs: automount should ignore LOOKUP_FOLLOW Prior to 2.6.38 automount would not trigger on either stat(2) or lstat(2) on the automount point. After 2.6.38, with the introduction of the ->d_automount() infrastructure, stat(2) and others would start triggering automount while lstat(2), etc. still would not. This is a regression and a userspace ABI change. Problem originally reported here: http://thread.gmane.org/gmane.linux.kernel.autofs/6098 It appears that there was an attempt at fixing various userspace tools to not trigger the automount. But since the stat system call is rather common it is impossible to "fix" all userspace. This patch reverts the original behavior, which is to not trigger on stat(2) and other symlink following syscalls. [ It's not really clear what the right behavior is. Apparently Solaris does the "automount on stat, leave alone on lstat". And some programs can get unhappy when "stat+open+fstat" ends up giving a different result from the fstat than from the initial stat. But the change in 2.6.38 resulted in problems for some people, so we're going back to old behavior. Maybe we can re-visit this discussion at some future date - Linus ] Reported-by: Leonardo Chiquitto Signed-off-by: Miklos Szeredi Acked-by: Ian Kent Cc: David Howells Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- fs/namei.c | 33 +++++++++++++++------------------ 1 file changed, 15 insertions(+), 18 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index 2826db35dc25..b52bc685465f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -727,25 +727,22 @@ static int follow_automount(struct path *path, unsigned flags, if ((flags & LOOKUP_NO_AUTOMOUNT) && !(flags & LOOKUP_PARENT)) return -EISDIR; /* we actually want to stop here */ - /* - * We don't want to mount if someone's just doing a stat and they've - * set AT_SYMLINK_NOFOLLOW - unless they're stat'ing a directory and - * appended a '/' to the name. + /* We don't want to mount if someone's just doing a stat - + * unless they're stat'ing a directory and appended a '/' to + * the name. + * + * We do, however, want to mount if someone wants to open or + * create a file of any type under the mountpoint, wants to + * traverse through the mountpoint or wants to open the + * mounted directory. Also, autofs may mark negative dentries + * as being automount points. These will need the attentions + * of the daemon to instantiate them before they can be used. */ - if (!(flags & LOOKUP_FOLLOW)) { - /* We do, however, want to mount if someone wants to open or - * create a file of any type under the mountpoint, wants to - * traverse through the mountpoint or wants to open the mounted - * directory. - * Also, autofs may mark negative dentries as being automount - * points. These will need the attentions of the daemon to - * instantiate them before they can be used. - */ - if (!(flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY | - LOOKUP_OPEN | LOOKUP_CREATE)) && - path->dentry->d_inode) - return -EISDIR; - } + if (!(flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY | + LOOKUP_OPEN | LOOKUP_CREATE)) && + path->dentry->d_inode) + return -EISDIR; + current->total_link_count++; if (current->total_link_count >= 40) return -ELOOP; From 5307f6d5fb12fd01f9f321bc4a8fd77e74858647 Mon Sep 17 00:00:00 2001 From: Shyam Iyer Date: Thu, 8 Sep 2011 16:41:17 -0500 Subject: [PATCH 176/326] Fix pointer dereference before call to pcie_bus_configure_settings Commit b03e7495a862 ("PCI: Set PCI-E Max Payload Size on fabric") introduced a potential NULL pointer dereference in calls to pcie_bus_configure_settings due to attempts to access pci_bus self variables when the self pointer is NULL. To correct this, verify that the self pointer in pci_bus is non-NULL before dereferencing it. Reported-by: Stanislaw Gruszka Signed-off-by: Shyam Iyer Signed-off-by: Jon Mason Acked-by: Jesse Barnes Signed-off-by: Linus Torvalds --- arch/x86/pci/acpi.c | 9 +++++++-- drivers/pci/hotplug/pcihp_slot.c | 4 +++- drivers/pci/probe.c | 3 --- 3 files changed, 10 insertions(+), 6 deletions(-) diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index c95330267f08..039d91315bc5 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -365,8 +365,13 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root) */ if (bus) { struct pci_bus *child; - list_for_each_entry(child, &bus->children, node) - pcie_bus_configure_settings(child, child->self->pcie_mpss); + list_for_each_entry(child, &bus->children, node) { + struct pci_dev *self = child->self; + if (!self) + continue; + + pcie_bus_configure_settings(child, self->pcie_mpss); + } } if (!bus) diff --git a/drivers/pci/hotplug/pcihp_slot.c b/drivers/pci/hotplug/pcihp_slot.c index 753b21aaea61..3ffd9c1acc0a 100644 --- a/drivers/pci/hotplug/pcihp_slot.c +++ b/drivers/pci/hotplug/pcihp_slot.c @@ -169,7 +169,9 @@ void pci_configure_slot(struct pci_dev *dev) (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI))) return; - pcie_bus_configure_settings(dev->bus, dev->bus->self->pcie_mpss); + if (dev->bus && dev->bus->self) + pcie_bus_configure_settings(dev->bus, + dev->bus->self->pcie_mpss); memset(&hpp, 0, sizeof(hpp)); ret = pci_get_hp_params(dev, &hpp); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 8473727b29fa..0820fc1544e8 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1456,9 +1456,6 @@ void pcie_bus_configure_settings(struct pci_bus *bus, u8 mpss) { u8 smpss = mpss; - if (!bus->self) - return; - if (!pci_is_pcie(bus->self)) return; From ed2888e906b56769b4ffabb9c577190438aa68b8 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 8 Sep 2011 16:41:18 -0500 Subject: [PATCH 177/326] PCI: Remove MRRS modification from MPS setting code Modifying the Maximum Read Request Size to 0 (value of 128Bytes) has massive negative ramifications on some devices. Without knowing which devices have this issue, do not modify from the default value when walking the PCI-E bus in pcie_bus_safe mode. Also, make pcie_bus_safe the default procedure. Tested-by: Sven Schnelle Tested-by: Simon Kirby Tested-by: Stephen M. Cameron Reported-and-tested-by: Eric Dumazet Reported-and-tested-by: Niels Ole Salscheider References: https://bugzilla.kernel.org/show_bug.cgi?id=42162 Signed-off-by: Jon Mason Acked-by: Jesse Barnes Signed-off-by: Linus Torvalds --- drivers/pci/pci.c | 2 +- drivers/pci/probe.c | 41 ++++++++++++++++++++++------------------- 2 files changed, 23 insertions(+), 20 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 0ce67423a0a3..4e84fd4a4312 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -77,7 +77,7 @@ unsigned long pci_cardbus_mem_size = DEFAULT_CARDBUS_MEM_SIZE; unsigned long pci_hotplug_io_size = DEFAULT_HOTPLUG_IO_SIZE; unsigned long pci_hotplug_mem_size = DEFAULT_HOTPLUG_MEM_SIZE; -enum pcie_bus_config_types pcie_bus_config = PCIE_BUS_PERFORMANCE; +enum pcie_bus_config_types pcie_bus_config = PCIE_BUS_SAFE; /* * The default CLS is used if arch didn't set CLS explicitly and not diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 0820fc1544e8..b1187ff31d89 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1396,34 +1396,37 @@ static void pcie_write_mps(struct pci_dev *dev, int mps) static void pcie_write_mrrs(struct pci_dev *dev, int mps) { - int rc, mrrs; + int rc, mrrs, dev_mpss; - if (pcie_bus_config == PCIE_BUS_PERFORMANCE) { - int dev_mpss = 128 << dev->pcie_mpss; + /* In the "safe" case, do not configure the MRRS. There appear to be + * issues with setting MRRS to 0 on a number of devices. + */ - /* For Max performance, the MRRS must be set to the largest - * supported value. However, it cannot be configured larger - * than the MPS the device or the bus can support. This assumes - * that the largest MRRS available on the device cannot be - * smaller than the device MPSS. - */ - mrrs = mps < dev_mpss ? mps : dev_mpss; - } else - /* In the "safe" case, configure the MRRS for fairness on the - * bus by making all devices have the same size - */ - mrrs = mps; + if (pcie_bus_config != PCIE_BUS_PERFORMANCE) + return; + dev_mpss = 128 << dev->pcie_mpss; + + /* For Max performance, the MRRS must be set to the largest supported + * value. However, it cannot be configured larger than the MPS the + * device or the bus can support. This assumes that the largest MRRS + * available on the device cannot be smaller than the device MPSS. + */ + mrrs = min(mps, dev_mpss); /* MRRS is a R/W register. Invalid values can be written, but a - * subsiquent read will verify if the value is acceptable or not. + * subsequent read will verify if the value is acceptable or not. * If the MRRS value provided is not acceptable (e.g., too large), * shrink the value until it is acceptable to the HW. */ while (mrrs != pcie_get_readrq(dev) && mrrs >= 128) { + dev_warn(&dev->dev, "Attempting to modify the PCI-E MRRS value" + " to %d. If any issues are encountered, please try " + "running with pci=pcie_bus_safe\n", mrrs); rc = pcie_set_readrq(dev, mrrs); if (rc) - dev_err(&dev->dev, "Failed attempting to set the MRRS\n"); + dev_err(&dev->dev, + "Failed attempting to set the MRRS\n"); mrrs /= 2; } @@ -1436,13 +1439,13 @@ static int pcie_bus_configure_set(struct pci_dev *dev, void *data) if (!pci_is_pcie(dev)) return 0; - dev_info(&dev->dev, "Dev MPS %d MPSS %d MRRS %d\n", + dev_dbg(&dev->dev, "Dev MPS %d MPSS %d MRRS %d\n", pcie_get_mps(dev), 128<pcie_mpss, pcie_get_readrq(dev)); pcie_write_mps(dev, mps); pcie_write_mrrs(dev, mps); - dev_info(&dev->dev, "Dev MPS %d MPSS %d MRRS %d\n", + dev_dbg(&dev->dev, "Dev MPS %d MPSS %d MRRS %d\n", pcie_get_mps(dev), 128<pcie_mpss, pcie_get_readrq(dev)); return 0; From 94007751bb02797ba87bac7aacee2731ac2039a3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 10 Sep 2011 17:20:21 +1000 Subject: [PATCH 178/326] Avoid dereferencing a 'request_queue' after last close. On the last close of an 'md' device which as been stopped, the device is destroyed and in particular the request_queue is freed. The free is done in a separate thread so it might happen a short time later. __blkdev_put calls bdev_inode_switch_bdi *after* ->release has been called. Since commit f758eeabeb96f878c860e8f110f94ec8820822a9 bdev_inode_switch_bdi will dereference the 'old' bdi, which lives inside a request_queue, to get a spin lock. This causes the last close on an md device to sometime take a spin_lock which lives in freed memory - which results in an oops. So move the called to bdev_inode_switch_bdi before the call to ->release. Cc: Christoph Hellwig Cc: Hugh Dickins Cc: Andrew Morton Cc: Wu Fengguang Acked-by: Wu Fengguang Cc: stable@kernel.org Signed-off-by: NeilBrown --- fs/block_dev.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index ff77262e887c..95f786ec7f08 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1429,6 +1429,11 @@ static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) WARN_ON_ONCE(bdev->bd_holders); sync_blockdev(bdev); kill_bdev(bdev); + /* ->release can cause the old bdi to disappear, + * so must switch it out first + */ + bdev_inode_switch_bdi(bdev->bd_inode, + &default_backing_dev_info); } if (bdev->bd_contains == bdev) { if (disk->fops->release) @@ -1442,8 +1447,6 @@ static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) disk_put_part(bdev->bd_part); bdev->bd_part = NULL; bdev->bd_disk = NULL; - bdev_inode_switch_bdi(bdev->bd_inode, - &default_backing_dev_info); if (bdev != bdev->bd_contains) victim = bdev->bd_contains; bdev->bd_contains = NULL; From 19d5f834d6aff7efb1c9353523865c5bce869470 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 10 Sep 2011 17:21:17 +1000 Subject: [PATCH 179/326] md/raid10: unify handling of write completion. A write can complete at two different places: 1/ when the last member-device write completes, through raid10_end_write_request 2/ in make_request() when we remove the initial bias from ->remaining. These two should do exactly the same thing and the comment says they do, but they don't. So factor the correct code out into a function and call it in both places. This makes the code much more similar to RAID1. The difference is only significant if there is an error, and they usually take a while, so it is unlikely that there will be an error already when make_request is completing, so this is unlikely to cause real problems. Signed-off-by: NeilBrown --- drivers/md/raid10.c | 38 ++++++++++++++++++-------------------- 1 file changed, 18 insertions(+), 20 deletions(-) diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 8b29cd4f01c8..f6873fc8e5ee 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -337,6 +337,21 @@ static void close_write(r10bio_t *r10_bio) md_write_end(r10_bio->mddev); } +static void one_write_done(r10bio_t *r10_bio) +{ + if (atomic_dec_and_test(&r10_bio->remaining)) { + if (test_bit(R10BIO_WriteError, &r10_bio->state)) + reschedule_retry(r10_bio); + else { + close_write(r10_bio); + if (test_bit(R10BIO_MadeGood, &r10_bio->state)) + reschedule_retry(r10_bio); + else + raid_end_bio_io(r10_bio); + } + } +} + static void raid10_end_write_request(struct bio *bio, int error) { int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); @@ -387,17 +402,7 @@ static void raid10_end_write_request(struct bio *bio, int error) * Let's see if all mirrored write operations have finished * already. */ - if (atomic_dec_and_test(&r10_bio->remaining)) { - if (test_bit(R10BIO_WriteError, &r10_bio->state)) - reschedule_retry(r10_bio); - else { - close_write(r10_bio); - if (test_bit(R10BIO_MadeGood, &r10_bio->state)) - reschedule_retry(r10_bio); - else - raid_end_bio_io(r10_bio); - } - } + one_write_done(r10_bio); if (dec_rdev) rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev); } @@ -1127,15 +1132,8 @@ static int make_request(mddev_t *mddev, struct bio * bio) spin_unlock_irqrestore(&conf->device_lock, flags); } - if (atomic_dec_and_test(&r10_bio->remaining)) { - /* This matches the end of raid10_end_write_request() */ - bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector, - r10_bio->sectors, - !test_bit(R10BIO_Degraded, &r10_bio->state), - 0); - md_write_end(mddev); - raid_end_bio_io(r10_bio); - } + /* Remove the bias on 'remaining' */ + one_write_done(r10_bio); /* In case raid10d snuck in to freeze_array */ wake_up(&conf->wait_barrier); From 079fa166a2874985ae58b2e21e26e1cbc91127d4 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 10 Sep 2011 17:21:23 +1000 Subject: [PATCH 180/326] md/raid1,10: Remove use-after-free bug in make_request. A single request to RAID1 or RAID10 might result in multiple requests if there are known bad blocks that need to be avoided. To detect if we need to submit another write request we test: if (sectors_handled < (bio->bi_size >> 9)) { However this is after we call **_write_done() so the 'bio' no longer belongs to us - the writes could have completed and the bio freed. So move the **_write_done call until after the test against bio->bi_size. This addresses https://bugzilla.kernel.org/show_bug.cgi?id=41862 Reported-by: Bruno Wolff III Tested-by: Bruno Wolff III Signed-off-by: NeilBrown --- drivers/md/raid1.c | 14 +++++++++----- drivers/md/raid10.c | 13 ++++++++----- 2 files changed, 17 insertions(+), 10 deletions(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 32323f0afd89..f4622dd8fc59 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1099,12 +1099,11 @@ static int make_request(mddev_t *mddev, struct bio * bio) bio_list_add(&conf->pending_bio_list, mbio); spin_unlock_irqrestore(&conf->device_lock, flags); } - r1_bio_write_done(r1_bio); - - /* In case raid1d snuck in to freeze_array */ - wake_up(&conf->wait_barrier); - + /* Mustn't call r1_bio_write_done before this next test, + * as it could result in the bio being freed. + */ if (sectors_handled < (bio->bi_size >> 9)) { + r1_bio_write_done(r1_bio); /* We need another r1_bio. It has already been counted * in bio->bi_phys_segments */ @@ -1117,6 +1116,11 @@ static int make_request(mddev_t *mddev, struct bio * bio) goto retry_write; } + r1_bio_write_done(r1_bio); + + /* In case raid1d snuck in to freeze_array */ + wake_up(&conf->wait_barrier); + if (do_sync || !bitmap || !plugged) md_wakeup_thread(mddev->thread); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index f6873fc8e5ee..d7a8468ddeab 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1132,13 +1132,12 @@ static int make_request(mddev_t *mddev, struct bio * bio) spin_unlock_irqrestore(&conf->device_lock, flags); } - /* Remove the bias on 'remaining' */ - one_write_done(r10_bio); - - /* In case raid10d snuck in to freeze_array */ - wake_up(&conf->wait_barrier); + /* Don't remove the bias on 'remaining' (one_write_done) until + * after checking if we need to go around again. + */ if (sectors_handled < (bio->bi_size >> 9)) { + one_write_done(r10_bio); /* We need another r10_bio. It has already been counted * in bio->bi_phys_segments. */ @@ -1152,6 +1151,10 @@ static int make_request(mddev_t *mddev, struct bio * bio) r10_bio->state = 0; goto retry_write; } + one_write_done(r10_bio); + + /* In case raid10d snuck in to freeze_array */ + wake_up(&conf->wait_barrier); if (do_sync || !mddev->bitmap || !plugged) md_wakeup_thread(mddev->thread); From 27a7b260f71439c40546b43588448faac01adb93 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 10 Sep 2011 17:21:28 +1000 Subject: [PATCH 181/326] md: Fix handling for devices from 2TB to 4TB in 0.90 metadata. 0.90 metadata uses an unsigned 32bit number to count the number of kilobytes used from each device. This should allow up to 4TB per device. However we multiply this by 2 (to get sectors) before casting to a larger type, so sizes above 2TB get truncated. Also we allow rdev->sectors to be larger than 4TB, so it is possible for the array to be resized larger than the metadata can handle. So make sure rdev->sectors never exceeds 4TB when 0.90 metadata is in used. Also the sanity check at the end of super_90_load should include level 1 as it used ->size too. (RAID0 and Linear don't use ->size at all). Reported-by: Pim Zandbergen Cc: stable@kernel.org Signed-off-by: NeilBrown --- drivers/md/md.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 3742ce8b0acf..5404b2295820 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1138,8 +1138,11 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version ret = 0; } rdev->sectors = rdev->sb_start; + /* Limit to 4TB as metadata cannot record more than that */ + if (rdev->sectors >= (2ULL << 32)) + rdev->sectors = (2ULL << 32) - 2; - if (rdev->sectors < sb->size * 2 && sb->level > 1) + if (rdev->sectors < ((sector_t)sb->size) * 2 && sb->level >= 1) /* "this cannot possibly happen" ... */ ret = -EINVAL; @@ -1173,7 +1176,7 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev) mddev->clevel[0] = 0; mddev->layout = sb->layout; mddev->raid_disks = sb->raid_disks; - mddev->dev_sectors = sb->size * 2; + mddev->dev_sectors = ((sector_t)sb->size) * 2; mddev->events = ev1; mddev->bitmap_info.offset = 0; mddev->bitmap_info.default_offset = MD_SB_BYTES >> 9; @@ -1415,6 +1418,11 @@ super_90_rdev_size_change(mdk_rdev_t *rdev, sector_t num_sectors) rdev->sb_start = calc_dev_sboffset(rdev); if (!num_sectors || num_sectors > rdev->sb_start) num_sectors = rdev->sb_start; + /* Limit to 4TB as metadata cannot record more than that. + * 4TB == 2^32 KB, or 2*2^32 sectors. + */ + if (num_sectors >= (2ULL << 32)) + num_sectors = (2ULL << 32) - 2; md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size, rdev->sb_page); md_super_wait(rdev->mddev); From c338bfb5ecf6c36b2112479691d69db4c2b5a78a Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Sat, 10 Sep 2011 20:13:01 +0200 Subject: [PATCH 182/326] backlight: Declare backlight_types[] const Since backlight_types[] isn't modified, let's declare it const. That was probably the intention of the author of commit bb7ca747f8d6 ("backlight: add backlight type"), via which the "const char const *" construct was introduced. The duplicate const was detected by sparse. Signed-off-by: Bart Van Assche Cc: Matthew Garrett Cc: Richard Purdie Cc: Florian Tobias Schandinat Cc: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/backlight.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c index 80d292fb92d8..7363c1b169e8 100644 --- a/drivers/video/backlight/backlight.c +++ b/drivers/video/backlight/backlight.c @@ -19,7 +19,7 @@ #include #endif -static const char const *backlight_types[] = { +static const char *const backlight_types[] = { [BACKLIGHT_RAW] = "raw", [BACKLIGHT_PLATFORM] = "platform", [BACKLIGHT_FIRMWARE] = "firmware", From 6c6d8deb5d95a0675a8edd588bbc2249cbce5b34 Mon Sep 17 00:00:00 2001 From: Janusz Krzysztofik Date: Thu, 8 Sep 2011 18:45:40 +0100 Subject: [PATCH 183/326] ARM: 7088/1: entry: fix wrong parameter name used in do_thumb_abort Commit be020f8618ca, "ARM: entry: abort-macro: specify registers to be used for macros", while replacing register numbers with macro parameter names, mismatched the name used for r1. For me, this resulted in user space built for EABI with -march=armv4t -mtune=arm920t -mthumb-interwork -mthumb broken on my OMAP1510 based Amstrad Delta (old ABI and no thumb still worked for me though). Fix this by using correct parameter name fsr instead of mismatched psr, used by callers for another purpose. Signed-off-by: Janusz Krzysztofik Signed-off-by: Russell King --- arch/arm/mm/abort-macro.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mm/abort-macro.S b/arch/arm/mm/abort-macro.S index 52162d59407a..2cbf68ef0e83 100644 --- a/arch/arm/mm/abort-macro.S +++ b/arch/arm/mm/abort-macro.S @@ -17,7 +17,7 @@ cmp \tmp, # 0x5600 @ Is it ldrsb? orreq \tmp, \tmp, #1 << 11 @ Set L-bit if yes tst \tmp, #1 << 11 @ L = 0 -> write - orreq \psr, \psr, #1 << 11 @ yes. + orreq \fsr, \fsr, #1 << 11 @ yes. b do_DataAbort not_thumb: .endm From d7a210f3d356371677cf553ce6241b620e389844 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 10 Sep 2011 17:13:34 -0700 Subject: [PATCH 184/326] scsi: qla4xxx driver depends on NET When CONFIG_NET is disabled, SCSI_QLA_ISCSI selects SCSI_ISCSI_ATTRS, which uses network interfaces, so the build fails with multiple errors: warning: (ISCSI_TCP && SCSI_CXGB3_ISCSI && SCSI_CXGB4_ISCSI && SCSI_QLA_ISCSI && INFINIBAND_ISER) selects SCSI_ISCSI_ATTRS which has unmet direct dependencies (SCSI && NET) ERROR: "skb_trim" [drivers/scsi/scsi_transport_iscsi.ko] undefined! ERROR: "netlink_kernel_create" [drivers/scsi/scsi_transport_iscsi.ko] undefined! ERROR: "netlink_kernel_release" [drivers/scsi/scsi_transport_iscsi.ko] undefined! ... so make SCSI_QLA_ISCSI also depend on NET to prevent the build errors. Signed-off-by: Randy Dunlap Cc: Ravi Anand Cc: Vikas Chaudhary Cc: iscsi-driver@qlogic.com Signed-off-by: Linus Torvalds --- drivers/scsi/qla4xxx/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/qla4xxx/Kconfig b/drivers/scsi/qla4xxx/Kconfig index 2c33ce6eac1e..0f5599e0abf6 100644 --- a/drivers/scsi/qla4xxx/Kconfig +++ b/drivers/scsi/qla4xxx/Kconfig @@ -1,6 +1,6 @@ config SCSI_QLA_ISCSI tristate "QLogic ISP4XXX and ISP82XX host adapter family support" - depends on PCI && SCSI + depends on PCI && SCSI && NET select SCSI_ISCSI_ATTRS ---help--- This driver supports the QLogic 40xx (ISP4XXX) and 8022 (ISP82XX) From 6a53747be5f02ba7b37ff5131330363270782dc3 Mon Sep 17 00:00:00 2001 From: Jamie Iles Date: Mon, 1 Aug 2011 21:09:36 +0100 Subject: [PATCH 185/326] ARM: CSR: add missing sentinels to of_device_id tables The of_device_id tables used for matching should be terminated with empty sentinel values. Signed-off-by: Jamie Iles Signed-off-by: Barry Song --- arch/arm/mach-prima2/clock.c | 1 + arch/arm/mach-prima2/irq.c | 1 + arch/arm/mach-prima2/rstc.c | 1 + arch/arm/mach-prima2/timer.c | 1 + 4 files changed, 4 insertions(+) diff --git a/arch/arm/mach-prima2/clock.c b/arch/arm/mach-prima2/clock.c index f9a2aaf63f71..615a4e75ceab 100644 --- a/arch/arm/mach-prima2/clock.c +++ b/arch/arm/mach-prima2/clock.c @@ -481,6 +481,7 @@ static void __init sirfsoc_clk_init(void) static struct of_device_id clkc_ids[] = { { .compatible = "sirf,prima2-clkc" }, + {}, }; void __init sirfsoc_of_clk_init(void) diff --git a/arch/arm/mach-prima2/irq.c b/arch/arm/mach-prima2/irq.c index c3404cbb6ff7..7af254d046ba 100644 --- a/arch/arm/mach-prima2/irq.c +++ b/arch/arm/mach-prima2/irq.c @@ -51,6 +51,7 @@ static __init void sirfsoc_irq_init(void) static struct of_device_id intc_ids[] = { { .compatible = "sirf,prima2-intc" }, + {}, }; void __init sirfsoc_of_irq_init(void) diff --git a/arch/arm/mach-prima2/rstc.c b/arch/arm/mach-prima2/rstc.c index d074786e83d4..492cfa8d2610 100644 --- a/arch/arm/mach-prima2/rstc.c +++ b/arch/arm/mach-prima2/rstc.c @@ -19,6 +19,7 @@ static DEFINE_MUTEX(rstc_lock); static struct of_device_id rstc_ids[] = { { .compatible = "sirf,prima2-rstc" }, + {}, }; static int __init sirfsoc_of_rstc_init(void) diff --git a/arch/arm/mach-prima2/timer.c b/arch/arm/mach-prima2/timer.c index 44027f34a88a..ed7ec48d11da 100644 --- a/arch/arm/mach-prima2/timer.c +++ b/arch/arm/mach-prima2/timer.c @@ -190,6 +190,7 @@ static void __init sirfsoc_timer_init(void) static struct of_device_id timer_ids[] = { { .compatible = "sirf,prima2-tick" }, + {}, }; static void __init sirfsoc_of_timer_map(void) From 707b38a00bd2e7cac60afc75abe826e68ca83cfb Mon Sep 17 00:00:00 2001 From: Jonas Bonn Date: Mon, 5 Sep 2011 13:47:10 +0200 Subject: [PATCH 186/326] Add missing DMA ops For the initial architecture submission, not all of the DMA ops were implemented. This patch adds the *map_page and *map_sg variants of the DMA mapping ops. This patch is currently of interest mainly to some drivers that haven't been submitted upstream yet. Signed-off-by: Jonas Bonn --- arch/openrisc/include/asm/dma-mapping.h | 59 ++++++++++++++++++++++++- arch/openrisc/kernel/dma.c | 28 +++++++++++- 2 files changed, 84 insertions(+), 3 deletions(-) diff --git a/arch/openrisc/include/asm/dma-mapping.h b/arch/openrisc/include/asm/dma-mapping.h index 052f877b52a5..60b472233900 100644 --- a/arch/openrisc/include/asm/dma-mapping.h +++ b/arch/openrisc/include/asm/dma-mapping.h @@ -31,7 +31,6 @@ #define DMA_ERROR_CODE (~(dma_addr_t)0x0) -int dma_mapping_error(struct device *dev, dma_addr_t dma_addr); #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) @@ -47,6 +46,12 @@ dma_addr_t or1k_map_page(struct device *dev, struct page *page, void or1k_unmap_page(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs); +int or1k_map_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs); +void or1k_unmap_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs); void or1k_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction dir); @@ -98,6 +103,51 @@ static inline void dma_unmap_single(struct device *dev, dma_addr_t addr, debug_dma_unmap_page(dev, addr, size, dir, true); } +static inline int dma_map_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir) +{ + int i, ents; + struct scatterlist *s; + + for_each_sg(sg, s, nents, i) + kmemcheck_mark_initialized(sg_virt(s), s->length); + BUG_ON(!valid_dma_direction(dir)); + ents = or1k_map_sg(dev, sg, nents, dir, NULL); + debug_dma_map_sg(dev, sg, nents, ents, dir); + + return ents; +} + +static inline void dma_unmap_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir) +{ + BUG_ON(!valid_dma_direction(dir)); + debug_dma_unmap_sg(dev, sg, nents, dir); + or1k_unmap_sg(dev, sg, nents, dir, NULL); +} + +static inline dma_addr_t dma_map_page(struct device *dev, struct page *page, + size_t offset, size_t size, + enum dma_data_direction dir) +{ + dma_addr_t addr; + + kmemcheck_mark_initialized(page_address(page) + offset, size); + BUG_ON(!valid_dma_direction(dir)); + addr = or1k_map_page(dev, page, offset, size, dir, NULL); + debug_dma_map_page(dev, page, offset, size, dir, addr, false); + + return addr; +} + +static inline void dma_unmap_page(struct device *dev, dma_addr_t addr, + size_t size, enum dma_data_direction dir) +{ + BUG_ON(!valid_dma_direction(dir)); + or1k_unmap_page(dev, addr, size, dir, NULL); + debug_dma_unmap_page(dev, addr, size, dir, true); +} + static inline void dma_sync_single_for_cpu(struct device *dev, dma_addr_t addr, size_t size, enum dma_data_direction dir) @@ -119,7 +169,12 @@ static inline void dma_sync_single_for_device(struct device *dev, static inline int dma_supported(struct device *dev, u64 dma_mask) { /* Support 32 bit DMA mask exclusively */ - return dma_mask == 0xffffffffULL; + return dma_mask == DMA_BIT_MASK(32); +} + +static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return 0; } static inline int dma_set_mask(struct device *dev, u64 dma_mask) diff --git a/arch/openrisc/kernel/dma.c b/arch/openrisc/kernel/dma.c index 968d3ee477e3..f1c8ee2895d0 100644 --- a/arch/openrisc/kernel/dma.c +++ b/arch/openrisc/kernel/dma.c @@ -154,6 +154,33 @@ void or1k_unmap_page(struct device *dev, dma_addr_t dma_handle, /* Nothing special to do here... */ } +int or1k_map_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + struct scatterlist *s; + int i; + + for_each_sg(sg, s, nents, i) { + s->dma_address = or1k_map_page(dev, sg_page(s), s->offset, + s->length, dir, NULL); + } + + return nents; +} + +void or1k_unmap_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + struct scatterlist *s; + int i; + + for_each_sg(sg, s, nents, i) { + or1k_unmap_page(dev, sg_dma_address(s), sg_dma_len(s), dir, NULL); + } +} + void or1k_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction dir) @@ -187,5 +214,4 @@ static int __init dma_init(void) return 0; } - fs_initcall(dma_init); From 4264a8b6388f5ba16a5c362857cb8bda0b14167f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 23 Jul 2011 15:53:03 -0300 Subject: [PATCH 187/326] [media] pwc: precedence bug in pwc_init_controls() '!' has higher precedence than '&' so we need parenthesis here. Signed-off-by: Dan Carpenter Signed-off-by: Hans de Goede Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/pwc/pwc-v4l.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/video/pwc/pwc-v4l.c b/drivers/media/video/pwc/pwc-v4l.c index e9a0e94b9995..8c70e64444e7 100644 --- a/drivers/media/video/pwc/pwc-v4l.c +++ b/drivers/media/video/pwc/pwc-v4l.c @@ -338,7 +338,7 @@ int pwc_init_controls(struct pwc_device *pdev) if (pdev->restore_factory) pdev->restore_factory->flags = V4L2_CTRL_FLAG_UPDATE; - if (!pdev->features & FEATURE_MOTOR_PANTILT) + if (!(pdev->features & FEATURE_MOTOR_PANTILT)) return hdl->error; /* Motor pan / tilt / reset */ From 55c53e1f24d46fd20e74d3a5089ed9f6e0e9ab14 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jean-Fran=C3=A7ois=20Moine?= Date: Tue, 9 Aug 2011 05:28:17 -0300 Subject: [PATCH 188/326] [media] gspca - ov519: Fix LED inversion of some ov519 webcams MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The list of the webcams which have LED inversion was rebuild scanning ms-win .inf files. Signed-off-by: Jean-François Moine Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/gspca/ov519.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c index 0800433b2092..18305c89083c 100644 --- a/drivers/media/video/gspca/ov519.c +++ b/drivers/media/video/gspca/ov519.c @@ -2858,7 +2858,6 @@ static void ov7xx0_configure(struct sd *sd) case 0x60: PDEBUG(D_PROBE, "Sensor is a OV7660"); sd->sensor = SEN_OV7660; - sd->invert_led = 0; break; default: PDEBUG(D_PROBE, "Unknown sensor: 0x76%x", low); @@ -3337,7 +3336,6 @@ static int sd_config(struct gspca_dev *gspca_dev, case BRIDGE_OV519: cam->cam_mode = ov519_vga_mode; cam->nmodes = ARRAY_SIZE(ov519_vga_mode); - sd->invert_led = !sd->invert_led; break; case BRIDGE_OVFX2: cam->cam_mode = ov519_vga_mode; @@ -5005,24 +5003,24 @@ static const struct sd_desc sd_desc = { /* -- module initialisation -- */ static const struct usb_device_id device_table[] = { {USB_DEVICE(0x041e, 0x4003), .driver_info = BRIDGE_W9968CF }, - {USB_DEVICE(0x041e, 0x4052), .driver_info = BRIDGE_OV519 }, - {USB_DEVICE(0x041e, 0x405f), + {USB_DEVICE(0x041e, 0x4052), .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, + {USB_DEVICE(0x041e, 0x405f), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x041e, 0x4060), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x041e, 0x4061), .driver_info = BRIDGE_OV519 }, - {USB_DEVICE(0x041e, 0x4064), - .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, + {USB_DEVICE(0x041e, 0x4064), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x041e, 0x4067), .driver_info = BRIDGE_OV519 }, - {USB_DEVICE(0x041e, 0x4068), + {USB_DEVICE(0x041e, 0x4068), .driver_info = BRIDGE_OV519 }, + {USB_DEVICE(0x045e, 0x028c), .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, - {USB_DEVICE(0x045e, 0x028c), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x054c, 0x0154), .driver_info = BRIDGE_OV519 }, - {USB_DEVICE(0x054c, 0x0155), - .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, + {USB_DEVICE(0x054c, 0x0155), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x05a9, 0x0511), .driver_info = BRIDGE_OV511 }, {USB_DEVICE(0x05a9, 0x0518), .driver_info = BRIDGE_OV518 }, - {USB_DEVICE(0x05a9, 0x0519), .driver_info = BRIDGE_OV519 }, - {USB_DEVICE(0x05a9, 0x0530), .driver_info = BRIDGE_OV519 }, + {USB_DEVICE(0x05a9, 0x0519), + .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, + {USB_DEVICE(0x05a9, 0x0530), + .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED }, {USB_DEVICE(0x05a9, 0x2800), .driver_info = BRIDGE_OVFX2 }, {USB_DEVICE(0x05a9, 0x4519), .driver_info = BRIDGE_OV519 }, {USB_DEVICE(0x05a9, 0x8519), .driver_info = BRIDGE_OV519 }, From 313c68e644326e88731f03371baa8b5f3d68ef11 Mon Sep 17 00:00:00 2001 From: Luiz Carlos Ramos Date: Tue, 9 Aug 2011 14:36:57 -0300 Subject: [PATCH 189/326] [media] gspca - sonixj: Fix wrong register mask for sensor om6802 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The bug was introduced by git commit 0e4d413af1a9d, giving very dark images. Signed-off-by: Luiz Carlos Ramos Signed-off-by: Jean-François Moine Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/gspca/sonixj.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c index 81b8a600783b..2ad757dc2e1c 100644 --- a/drivers/media/video/gspca/sonixj.c +++ b/drivers/media/video/gspca/sonixj.c @@ -2386,7 +2386,7 @@ static int sd_start(struct gspca_dev *gspca_dev) reg_w1(gspca_dev, 0x01, 0x22); msleep(100); reg01 = SCL_SEL_OD | S_PDN_INV; - reg17 &= MCK_SIZE_MASK; + reg17 &= ~MCK_SIZE_MASK; reg17 |= 0x04; /* clock / 4 */ break; } From 5b5cfc3674756d249cb389bbd2a0be94abae5f7c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jean-Fran=C3=A7ois=20Moine?= Date: Tue, 9 Aug 2011 15:13:50 -0300 Subject: [PATCH 190/326] [media] gspca - sonixj: Fix the darkness of sensor om6802 in 320x240 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The images are clearer with a lower bridge clock. Signed-off-by: Jean-François Moine Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/gspca/sonixj.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c index 2ad757dc2e1c..c477ad11f103 100644 --- a/drivers/media/video/gspca/sonixj.c +++ b/drivers/media/video/gspca/sonixj.c @@ -2532,6 +2532,10 @@ static int sd_start(struct gspca_dev *gspca_dev) if (!mode) { /* if 640x480 */ reg17 &= ~MCK_SIZE_MASK; reg17 |= 0x04; /* clock / 4 */ + } else { + reg01 &= ~SYS_SEL_48M; /* clk 24Mz */ + reg17 &= ~MCK_SIZE_MASK; + reg17 |= 0x02; /* clock / 2 */ } break; case SENSOR_OV7630: From 6a02a3306769c6f0fa8ebd92d5046c841e4e740d Mon Sep 17 00:00:00 2001 From: Kamil Debski Date: Tue, 2 Aug 2011 12:53:49 -0300 Subject: [PATCH 191/326] [media] v4l2: Fix documentation of the codec device controls Fixed missing ids of the codec controls description in the controls.xml file. Signed-off-by: Kamil Debski Signed-off-by: Kyungmin Park Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: Mauro Carvalho Chehab --- Documentation/DocBook/media/v4l/controls.xml | 38 ++++++++++---------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/Documentation/DocBook/media/v4l/controls.xml b/Documentation/DocBook/media/v4l/controls.xml index 85164016ed26..23fdf79f8cf3 100644 --- a/Documentation/DocBook/media/v4l/controls.xml +++ b/Documentation/DocBook/media/v4l/controls.xml @@ -1455,7 +1455,7 @@ Applicable to the H264 encoder. - + V4L2_CID_MPEG_VIDEO_H264_VUI_SAR_IDC  enum v4l2_mpeg_video_h264_vui_sar_idc @@ -1561,7 +1561,7 @@ Applicable to the H264 encoder. - + V4L2_CID_MPEG_VIDEO_H264_LEVEL  enum v4l2_mpeg_video_h264_level @@ -1641,7 +1641,7 @@ Possible values are: - + V4L2_CID_MPEG_VIDEO_MPEG4_LEVEL  enum v4l2_mpeg_video_mpeg4_level @@ -1689,9 +1689,9 @@ Possible values are: - + V4L2_CID_MPEG_VIDEO_H264_PROFILE  - enum v4l2_mpeg_h264_profile + enum v4l2_mpeg_video_h264_profile The profile information for H264. Applicable to the H264 encoder. @@ -1774,9 +1774,9 @@ Possible values are: - + V4L2_CID_MPEG_VIDEO_MPEG4_PROFILE  - enum v4l2_mpeg_mpeg4_profile + enum v4l2_mpeg_video_mpeg4_profile The profile information for MPEG4. Applicable to the MPEG4 encoder. @@ -1820,9 +1820,9 @@ Applicable to the encoder. - + V4L2_CID_MPEG_VIDEO_MULTI_SLICE_MODE  - enum v4l2_mpeg_multi_slice_mode + enum v4l2_mpeg_video_multi_slice_mode Determines how the encoder should handle division of frame into slices. Applicable to the encoder. @@ -1868,9 +1868,9 @@ Applicable to the encoder. - + V4L2_CID_MPEG_VIDEO_H264_LOOP_FILTER_MODE  - enum v4l2_mpeg_h264_loop_filter_mode + enum v4l2_mpeg_video_h264_loop_filter_mode Loop filter mode for H264 encoder. Possible values are: @@ -1913,9 +1913,9 @@ Applicable to the H264 encoder. - + V4L2_CID_MPEG_VIDEO_H264_ENTROPY_MODE  - enum v4l2_mpeg_h264_symbol_mode + enum v4l2_mpeg_video_h264_entropy_mode Entropy coding mode for H264 - CABAC/CAVALC. Applicable to the H264 encoder. @@ -2140,9 +2140,9 @@ previous frames. Applicable to the H264 encoder. - + V4L2_CID_MPEG_VIDEO_HEADER_MODE  - enum v4l2_mpeg_header_mode + enum v4l2_mpeg_video_header_mode Determines whether the header is returned as the first buffer or is it returned together with the first frame. Applicable to encoders. @@ -2320,9 +2320,9 @@ Valid only when H.264 and macroblock level RC is enabled (V4L2_CID_MPE Applicable to the H264 encoder. - + V4L2_CID_MPEG_MFC51_VIDEO_FRAME_SKIP_MODE  - enum v4l2_mpeg_mfc51_frame_skip_mode + enum v4l2_mpeg_mfc51_video_frame_skip_mode Indicates in what conditions the encoder should skip frames. If encoding a frame would cause the encoded stream to be larger then @@ -2361,9 +2361,9 @@ the stream will meet tight bandwidth contraints. Applicable to encoders. - + V4L2_CID_MPEG_MFC51_VIDEO_FORCE_FRAME_TYPE  - enum v4l2_mpeg_mfc51_force_frame_type + enum v4l2_mpeg_mfc51_video_force_frame_type Force a frame type for the next queued buffer. Applicable to encoders. Possible values are: From c8814df3a578895390fe5c05a76328d8d111ed25 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Mon, 1 Aug 2011 18:39:17 -0300 Subject: [PATCH 192/326] [media] [Resend] viacam: Don't explode if pci_find_bus() returns NULL In the unlikely case that pci_find_bus() should return NULL viacam_serial_is_enabled() is going to dereference a NULL pointer and blow up. Better safe than sorry, so be defensive and check the pointer. Signed-off-by: Jesper Juhl Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/via-camera.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/media/video/via-camera.c b/drivers/media/video/via-camera.c index 85d3048c1d67..bb7f17f2a33c 100644 --- a/drivers/media/video/via-camera.c +++ b/drivers/media/video/via-camera.c @@ -1332,6 +1332,8 @@ static __devinit bool viacam_serial_is_enabled(void) struct pci_bus *pbus = pci_find_bus(0, 0); u8 cbyte; + if (!pbus) + return false; pci_bus_read_config_byte(pbus, VIACAM_SERIAL_DEVFN, VIACAM_SERIAL_CREG, &cbyte); if ((cbyte & VIACAM_SERIAL_BIT) == 0) From de4ed0c111ed078b8729a5cc49c23197740f5bad Mon Sep 17 00:00:00 2001 From: Jarod Wilson Date: Mon, 8 Aug 2011 17:20:40 -0300 Subject: [PATCH 193/326] [media] nuvoton-cir: simplify raw IR sample handling The nuvoton-cir driver was storing up consecutive pulse-pulse and space-space samples internally, for no good reason, since ir_raw_event_store_with_filter() already merges back to back like samples types for us. This should also fix a regression introduced late in 3.0 that related to a timeout change, which actually becomes correct when coupled with this change. Tested with RC6 and RC5 on my own nuvoton-cir hardware atop vanilla 3.0.0, after verifying quirky behavior in 3.0 due to the timeout change. Reported-by: Stephan Raue CC: Stephan Raue CC: stable@vger.kernel.org Signed-off-by: Jarod Wilson Signed-off-by: Mauro Carvalho Chehab --- drivers/media/rc/nuvoton-cir.c | 45 ++++++---------------------------- drivers/media/rc/nuvoton-cir.h | 1 - 2 files changed, 8 insertions(+), 38 deletions(-) diff --git a/drivers/media/rc/nuvoton-cir.c b/drivers/media/rc/nuvoton-cir.c index eae05b500476..144f3f55d765 100644 --- a/drivers/media/rc/nuvoton-cir.c +++ b/drivers/media/rc/nuvoton-cir.c @@ -618,7 +618,6 @@ static void nvt_dump_rx_buf(struct nvt_dev *nvt) static void nvt_process_rx_ir_data(struct nvt_dev *nvt) { DEFINE_IR_RAW_EVENT(rawir); - unsigned int count; u32 carrier; u8 sample; int i; @@ -631,65 +630,38 @@ static void nvt_process_rx_ir_data(struct nvt_dev *nvt) if (nvt->carrier_detect_enabled) carrier = nvt_rx_carrier_detect(nvt); - count = nvt->pkts; - nvt_dbg_verbose("Processing buffer of len %d", count); + nvt_dbg_verbose("Processing buffer of len %d", nvt->pkts); init_ir_raw_event(&rawir); - for (i = 0; i < count; i++) { - nvt->pkts--; + for (i = 0; i < nvt->pkts; i++) { sample = nvt->buf[i]; rawir.pulse = ((sample & BUF_PULSE_BIT) != 0); rawir.duration = US_TO_NS((sample & BUF_LEN_MASK) * SAMPLE_PERIOD); - if ((sample & BUF_LEN_MASK) == BUF_LEN_MASK) { - if (nvt->rawir.pulse == rawir.pulse) - nvt->rawir.duration += rawir.duration; - else { - nvt->rawir.duration = rawir.duration; - nvt->rawir.pulse = rawir.pulse; - } - continue; - } + nvt_dbg("Storing %s with duration %d", + rawir.pulse ? "pulse" : "space", rawir.duration); - rawir.duration += nvt->rawir.duration; - - init_ir_raw_event(&nvt->rawir); - nvt->rawir.duration = 0; - nvt->rawir.pulse = rawir.pulse; - - if (sample == BUF_PULSE_BIT) - rawir.pulse = false; - - if (rawir.duration) { - nvt_dbg("Storing %s with duration %d", - rawir.pulse ? "pulse" : "space", - rawir.duration); - - ir_raw_event_store_with_filter(nvt->rdev, &rawir); - } + ir_raw_event_store_with_filter(nvt->rdev, &rawir); /* * BUF_PULSE_BIT indicates end of IR data, BUF_REPEAT_BYTE * indicates end of IR signal, but new data incoming. In both * cases, it means we're ready to call ir_raw_event_handle */ - if ((sample == BUF_PULSE_BIT) && nvt->pkts) { + if ((sample == BUF_PULSE_BIT) && (i + 1 < nvt->pkts)) { nvt_dbg("Calling ir_raw_event_handle (signal end)\n"); ir_raw_event_handle(nvt->rdev); } } + nvt->pkts = 0; + nvt_dbg("Calling ir_raw_event_handle (buffer empty)\n"); ir_raw_event_handle(nvt->rdev); - if (nvt->pkts) { - nvt_dbg("Odd, pkts should be 0 now... (its %u)", nvt->pkts); - nvt->pkts = 0; - } - nvt_dbg_verbose("%s done", __func__); } @@ -1048,7 +1020,6 @@ static int nvt_probe(struct pnp_dev *pdev, const struct pnp_device_id *dev_id) spin_lock_init(&nvt->nvt_lock); spin_lock_init(&nvt->tx.lock); - init_ir_raw_event(&nvt->rawir); ret = -EBUSY; /* now claim resources */ diff --git a/drivers/media/rc/nuvoton-cir.h b/drivers/media/rc/nuvoton-cir.h index 1241fc89a36c..0d5e0872a2ea 100644 --- a/drivers/media/rc/nuvoton-cir.h +++ b/drivers/media/rc/nuvoton-cir.h @@ -67,7 +67,6 @@ static int debug; struct nvt_dev { struct pnp_dev *pdev; struct rc_dev *rdev; - struct ir_raw_event rawir; spinlock_t nvt_lock; From fc61ccd35fd59d5362d37c8bf9c0526c85086c84 Mon Sep 17 00:00:00 2001 From: Florian Mickler Date: Wed, 10 Aug 2011 07:05:20 -0300 Subject: [PATCH 194/326] [media] vp7045: fix buffer setup dvb_usb_device_init calls the frontend_attach method of this driver which uses vp7045_usb_ob. In order to have a buffer ready in vp7045_usb_op, it has to be allocated before that happens. Luckily we can use the whole private data as the buffer as it gets separately allocated on the heap via kzalloc in dvb_usb_device_init and is thus apt for use via usb_control_msg. This fixes a BUG: unable to handle kernel paging request at 0000000000001e78 reported by Tino Keitel and diagnosed by Dan Carpenter. Cc: stable@kernel.org # For v3.0 and upper Tested-by: Tino Keitel Signed-off-by: Florian Mickler Signed-off-by: Mauro Carvalho Chehab --- drivers/media/dvb/dvb-usb/vp7045.c | 26 ++++---------------------- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/drivers/media/dvb/dvb-usb/vp7045.c b/drivers/media/dvb/dvb-usb/vp7045.c index 3db89e3cb0bb..536c16c943bd 100644 --- a/drivers/media/dvb/dvb-usb/vp7045.c +++ b/drivers/media/dvb/dvb-usb/vp7045.c @@ -224,26 +224,8 @@ static struct dvb_usb_device_properties vp7045_properties; static int vp7045_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) { - struct dvb_usb_device *d; - int ret = dvb_usb_device_init(intf, &vp7045_properties, - THIS_MODULE, &d, adapter_nr); - if (ret) - return ret; - - d->priv = kmalloc(20, GFP_KERNEL); - if (!d->priv) { - dvb_usb_device_exit(intf); - return -ENOMEM; - } - - return ret; -} - -static void vp7045_usb_disconnect(struct usb_interface *intf) -{ - struct dvb_usb_device *d = usb_get_intfdata(intf); - kfree(d->priv); - dvb_usb_device_exit(intf); + return dvb_usb_device_init(intf, &vp7045_properties, + THIS_MODULE, NULL, adapter_nr); } static struct usb_device_id vp7045_usb_table [] = { @@ -258,7 +240,7 @@ MODULE_DEVICE_TABLE(usb, vp7045_usb_table); static struct dvb_usb_device_properties vp7045_properties = { .usb_ctrl = CYPRESS_FX2, .firmware = "dvb-usb-vp7045-01.fw", - .size_of_priv = sizeof(u8 *), + .size_of_priv = 20, .num_adapters = 1, .adapter = { @@ -305,7 +287,7 @@ static struct dvb_usb_device_properties vp7045_properties = { static struct usb_driver vp7045_usb_driver = { .name = "dvb_usb_vp7045", .probe = vp7045_usb_probe, - .disconnect = vp7045_usb_disconnect, + .disconnect = dvb_usb_device_exit, .id_table = vp7045_usb_table, }; From 14c7cca780bd210564ae964f57a8bb807d0b3dbf Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 195/326] Btrfs: fix an oops when deleting snapshots We can reproduce this oops via the following steps: $ mkfs.btrfs /dev/sdb7 $ mount /dev/sdb7 /mnt/btrfs $ for ((i=0; i<3; i++)); do btrfs sub snap /mnt/btrfs /mnt/btrfs/s_$i; done $ rm -fr /mnt/btrfs/* $ rm -fr /mnt/btrfs/* then we'll get ------------[ cut here ]------------ kernel BUG at fs/btrfs/inode.c:2264! [...] Call Trace: [] btrfs_rmdir+0xf7/0x1b0 [btrfs] [] vfs_rmdir+0xa5/0xf0 [] do_rmdir+0x123/0x140 [] ? fput+0x197/0x260 [] ? audit_syscall_entry+0x1bf/0x1f0 [] sys_unlinkat+0x2d/0x40 [] system_call_fastpath+0x16/0x1b RIP [] btrfs_orphan_add+0x179/0x1a0 [btrfs] When it comes to btrfs_lookup_dentry, we may set a snapshot's inode->i_ino to BTRFS_EMPTY_SUBVOL_DIR_OBJECTID instead of BTRFS_FIRST_FREE_OBJECTID, while the snapshot's location.objectid remains unchanged. However, btrfs_ino() does not take this into account, and returns a wrong ino, and causes the oops. Signed-off-by: Liu Bo Signed-off-by: Chris Mason --- fs/btrfs/btrfs_inode.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 502b9e988679..d9f99a16edd6 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -176,7 +176,11 @@ static inline u64 btrfs_ino(struct inode *inode) { u64 ino = BTRFS_I(inode)->location.objectid; - if (ino <= BTRFS_FIRST_FREE_OBJECTID) + /* + * !ino: btree_inode + * type == BTRFS_ROOT_ITEM_KEY: subvol dir + */ + if (!ino || BTRFS_I(inode)->location.type == BTRFS_ROOT_ITEM_KEY) ino = inode->i_ino; return ino; } From e0b6d65be57fb37ca67b04ce8964546a74d2125c Mon Sep 17 00:00:00 2001 From: Sergei Trofimovich Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 196/326] btrfs: fix warning in iput for bad-inode iput() shouldn't be called for inodes in I_NEW state. We need to mark inode as constructed first. WARNING: at fs/inode.c:1309 iput+0x20b/0x210() Call Trace: [] warn_slowpath_common+0x7a/0xb0 [] warn_slowpath_null+0x15/0x20 [] iput+0x20b/0x210 [] btrfs_iget+0x1eb/0x4a0 [] btrfs_run_defrag_inodes+0x136/0x210 [] cleaner_kthread+0x17f/0x1a0 [] ? sub_preempt_count+0x9d/0xd0 [] ? transaction_kthread+0x280/0x280 [] kthread+0x96/0xa0 [] kernel_thread_helper+0x4/0x10 [] ? kthread_worker_fn+0x190/0x190 [] ? gs_change+0xb/0xb Signed-off-by: Sergei Trofimovich CC: Konstantin Khlebnikov Tested-by: David Sterba CC: Josef Bacik CC: Chris Mason Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 34195f9fc6bb..edd45f709989 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3952,7 +3952,6 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, struct btrfs_root *root, int *new) { struct inode *inode; - int bad_inode = 0; inode = btrfs_iget_locked(s, location->objectid, root); if (!inode) @@ -3968,15 +3967,12 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, if (new) *new = 1; } else { - bad_inode = 1; + unlock_new_inode(inode); + iput(inode); + inode = ERR_PTR(-ESTALE); } } - if (bad_inode) { - iput(inode); - inode = ERR_PTR(-ESTALE); - } - return inode; } From ddf23b3fc6850bd4654d51ec9457fe7c77cde51e Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 197/326] Btrfs: skip locking if searching the commit root in csum lookup It's not enough to just search the commit root, since we could be cow'ing the very block we need to search through, which would mean that its locked and we'll still deadlock. So use path->skip_locking as well. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/file-item.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index b910694f61ed..a1cb7821becd 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -183,8 +183,10 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root, * read from the commit root and sidestep a nasty deadlock * between reading the free space cache and updating the csum tree. */ - if (btrfs_is_free_space_inode(root, inode)) + if (btrfs_is_free_space_inode(root, inode)) { path->search_commit_root = 1; + path->skip_locking = 1; + } disk_bytenr = (u64)bio->bi_sector << 9; if (dio) From 65450aa645b1ef7ed74e41c34b28d53333744978 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 198/326] Btrfs: reset to appropriate block rsv after orphan operations While truncating free space cache, we forget to change trans->block_rsv back to the original one, but leave it with the orphan_block_rsv, and then with option inode_cache enable, it leads to countless warnings of btrfs_alloc_free_block and btrfs_orphan_commit_root: WARNING: at fs/btrfs/extent-tree.c:5711 btrfs_alloc_free_block+0x180/0x350 [btrfs]() ... WARNING: at fs/btrfs/inode.c:2193 btrfs_orphan_commit_root+0xb0/0xc0 [btrfs]() Signed-off-by: Liu Bo Signed-off-by: Chris Mason --- fs/btrfs/free-space-cache.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 6a265b9f85f2..41ac927401d0 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -190,9 +190,11 @@ int btrfs_truncate_free_space_cache(struct btrfs_root *root, struct btrfs_path *path, struct inode *inode) { + struct btrfs_block_rsv *rsv; loff_t oldsize; int ret = 0; + rsv = trans->block_rsv; trans->block_rsv = root->orphan_block_rsv; ret = btrfs_block_rsv_check(trans, root, root->orphan_block_rsv, @@ -210,6 +212,8 @@ int btrfs_truncate_free_space_cache(struct btrfs_root *root, */ ret = btrfs_truncate_inode_items(trans, root, inode, 0, BTRFS_EXTENT_DATA_KEY); + + trans->block_rsv = rsv; if (ret) { WARN_ON(1); return ret; From 98c9942aca05fff198cd5ca629599cd193444809 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 199/326] Btrfs: fix misuse of trans block rsv At the beginning of create_pending_snapshot, trans->block_rsv is set to pending->block_rsv and is used for snapshot things, however, when it is done, we do not recover it as will. Signed-off-by: Liu Bo Signed-off-by: Chris Mason --- fs/btrfs/transaction.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 7dc36fab4afc..e24b7964a155 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -884,6 +884,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root *tree_root = fs_info->tree_root; struct btrfs_root *root = pending->root; struct btrfs_root *parent_root; + struct btrfs_block_rsv *rsv; struct inode *parent_inode; struct dentry *parent; struct dentry *dentry; @@ -895,6 +896,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, u64 objectid; u64 root_flags; + rsv = trans->block_rsv; + new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS); if (!new_root_item) { pending->error = -ENOMEM; @@ -1002,6 +1005,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_orphan_post_snapshot(trans, pending); fail: kfree(new_root_item); + trans->block_rsv = rsv; btrfs_block_rsv_release(root, &pending->block_rsv, (u64)-1); return 0; } From 5b397377e97d436fc2ed872fc53f85395bb984e0 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 200/326] Btrfs: fix unclosed transaction handle in btrfs_cont_expand The function - btrfs_cont_expand() forgot to close the transaction handle before it jump out the while loop. Fix it. Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index edd45f709989..c257af2ce9cb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3510,15 +3510,19 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) err = btrfs_drop_extents(trans, inode, cur_offset, cur_offset + hole_size, &hint_byte, 1); - if (err) + if (err) { + btrfs_end_transaction(trans, root); break; + } err = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), cur_offset, 0, 0, hole_size, 0, hole_size, 0, 0, 0); - if (err) + if (err) { + btrfs_end_transaction(trans, root); break; + } btrfs_drop_extent_cache(inode, hole_start, last_byte - 1, 0); From 0c1a98c81413e00a6c379d898e06a09350d31926 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Sun, 11 Sep 2011 10:52:24 -0400 Subject: [PATCH 201/326] Btrfs: fix the file extent gap when doing direct IO When we write some data to the place that is beyond the end of the file in direct I/O mode, a data hole will be created. And Btrfs should insert a file extent item that point to this hole into the fs tree. But unfortunately Btrfs forgets doing it. The following is a simple way to reproduce it: # mkfs.btrfs /dev/sdc2 # mount /dev/sdc2 /test4 # touch /test4/a # dd if=/dev/zero of=/test4/a seek=8 count=1 bs=4K oflag=direct conv=nocreat,notrunc # umount /test4 # btrfsck /dev/sdc2 root 5 inode 257 errors 100 Reported-by: Tsutomu Itoh Signed-off-by: Miao Xie Tested-by: Tsutomu Itoh Signed-off-by: Chris Mason --- fs/btrfs/file.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 15e5a1cd8764..98d95bb5f253 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1075,12 +1075,6 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file, start_pos = pos & ~((u64)root->sectorsize - 1); last_pos = ((u64)index + num_pages) << PAGE_CACHE_SHIFT; - if (start_pos > inode->i_size) { - err = btrfs_cont_expand(inode, i_size_read(inode), start_pos); - if (err) - return err; - } - again: for (i = 0; i < num_pages; i++) { pages[i] = find_or_create_page(inode->i_mapping, index + i, @@ -1338,6 +1332,7 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, struct inode *inode = fdentry(file)->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; loff_t *ppos = &iocb->ki_pos; + u64 start_pos; ssize_t num_written = 0; ssize_t err = 0; size_t count, ocount; @@ -1386,6 +1381,15 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, file_update_time(file); BTRFS_I(inode)->sequence++; + start_pos = round_down(pos, root->sectorsize); + if (start_pos > i_size_read(inode)) { + err = btrfs_cont_expand(inode, i_size_read(inode), start_pos); + if (err) { + mutex_unlock(&inode->i_mutex); + goto out; + } + } + if (unlikely(file->f_flags & O_DIRECT)) { num_written = __btrfs_direct_write(iocb, iov, nr_segs, pos, ppos, count, ocount); From a39f75214358d715efa21e2bccf5a709d8649144 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Sun, 11 Sep 2011 10:52:25 -0400 Subject: [PATCH 202/326] Btrfs: fix wrong nbytes information of the inode If we write some data into the data hole of the file(no preallocation for this hole), Btrfs will allocate some disk space, and update nbytes of the inode, but the other element--disk_i_size needn't be updated. At this condition, we must update inode metadata though disk_i_size is not changed(btrfs_ordered_update_i_size() return 1). # mkfs.btrfs /dev/sdb1 # mount /dev/sdb1 /mnt # touch /mnt/a # truncate -s 856002 /mnt/a # dd if=/dev/zero of=/mnt/a bs=4K count=1 conv=nocreat,notrunc # umount /mnt # btrfsck /dev/sdb1 root 5 inode 257 errors 400 found 32768 bytes used err is 1 Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c257af2ce9cb..b94c0da3b43f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1786,7 +1786,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) &ordered_extent->list); ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent); - if (!ret) { + if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) { ret = btrfs_update_inode(trans, root, inode); BUG_ON(ret); } @@ -5788,7 +5788,7 @@ static void btrfs_endio_direct_write(struct bio *bio, int err) add_pending_csums(trans, inode, ordered->file_offset, &ordered->list); ret = btrfs_ordered_update_i_size(inode, 0, ordered); - if (!ret) + if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags)) btrfs_update_inode(trans, root, inode); ret = 0; out_unlock: From 4815053aba7f2304055745df820cd74a39fdaab2 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Sun, 11 Sep 2011 10:52:25 -0400 Subject: [PATCH 203/326] btrfs: xattr: fix attribute removal An attribute is not removed by 'setfattr -x attr file' and remains visible in attr list. This makes xfstests/062 pass again. Signed-off-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/xattr.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index d733b9cfea34..69565e5fc6a0 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -116,6 +116,12 @@ static int do_setxattr(struct btrfs_trans_handle *trans, if (ret) goto out; btrfs_release_path(path); + + /* + * remove the attribute + */ + if (!value) + goto out; } again: @@ -158,6 +164,9 @@ static int do_setxattr(struct btrfs_trans_handle *trans, return ret; } +/* + * @value: "" makes the attribute to empty, NULL removes it + */ int __btrfs_setxattr(struct btrfs_trans_handle *trans, struct inode *inode, const char *name, const void *value, size_t size, int flags) From d72c0842ff0e71342857723bb65f35b71f57b264 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sun, 11 Sep 2011 10:52:25 -0400 Subject: [PATCH 204/326] Btrfs: calc file extent num_bytes correctly in file clone num_bytes should be 4096 not 12288. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index b3d249d6eba7..028a4b8c12cd 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2333,14 +2333,21 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, if (type == BTRFS_FILE_EXTENT_REG || type == BTRFS_FILE_EXTENT_PREALLOC) { + /* + * a | --- range to clone ---| b + * | ------------- extent ------------- | + */ + + /* substract range b */ + if (key.offset + datal > off + len) + datal = off + len - key.offset; + + /* substract range a */ if (off > key.offset) { datao += off - key.offset; datal -= off - key.offset; } - if (key.offset + datal > off + len) - datal = off + len - key.offset; - ret = btrfs_drop_extents(trans, inode, new_key.offset, new_key.offset + datal, From d525e8ab022cb000e6e31a515ba8c3cf0d9c6130 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sun, 11 Sep 2011 10:52:25 -0400 Subject: [PATCH 205/326] Btrfs: add dummy extent if dst offset excceeds file end in You can see there's no file extent with range [0, 4096]. Check this by btrfsck: # btrfsck /dev/sda7 root 5 inode 258 errors 100 ... Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 028a4b8c12cd..63b4de1626d2 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2228,6 +2228,12 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, !IS_ALIGNED(destoff, bs)) goto out_unlock; + if (destoff > inode->i_size) { + ret = btrfs_cont_expand(inode, inode->i_size, destoff); + if (ret) + goto out_unlock; + } + /* do any pending delalloc/csum calc on src, one way or another, and lock file content */ while (1) { From ed585a651681e822089087b426e6ebfb6d3d9873 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 11 Sep 2011 13:59:27 +0200 Subject: [PATCH 206/326] genirq: Make irq_shutdown() symmetric vs. irq_startup again If an irq_chip provides .irq_shutdown(), but neither of .irq_disable() or .irq_mask(), free_irq() crashes when jumping to NULL. Fix this by only trying .irq_disable() and .irq_mask() if there's no .irq_shutdown() provided. This revives the symmetry with irq_startup(), which tries .irq_startup(), .irq_enable(), and irq_unmask(), and makes it consistent with the comment for irq_chip.irq_shutdown() in , which says: * @irq_shutdown: shut down the interrupt (defaults to ->disable if NULL) This is also how __free_irq() behaved before the big overhaul, cfr. e.g. 3b56f0585fd4c02d047dc406668cb40159b2d340 ("genirq: Remove bogus conditional"), where the core interrupt code always overrode .irq_shutdown() to .irq_disable() if .irq_shutdown() was NULL. Signed-off-by: Geert Uytterhoeven Cc: linux-m68k@lists.linux-m68k.org Link: http://lkml.kernel.org/r/1315742394-16036-2-git-send-email-geert@linux-m68k.org Cc: stable@kernel.org Signed-off-by: Thomas Gleixner --- kernel/irq/chip.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index d5a3009da71a..dc5114b4c16c 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -178,7 +178,7 @@ void irq_shutdown(struct irq_desc *desc) desc->depth = 1; if (desc->irq_data.chip->irq_shutdown) desc->irq_data.chip->irq_shutdown(&desc->irq_data); - if (desc->irq_data.chip->irq_disable) + else if (desc->irq_data.chip->irq_disable) desc->irq_data.chip->irq_disable(&desc->irq_data); else desc->irq_data.chip->irq_mask(&desc->irq_data); From 5013951be88e136d9990ef55303276e2779ce8d8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sun, 11 Sep 2011 20:07:30 +0200 Subject: [PATCH 207/326] ASoC: Fix trivial build regression in Kirkwood I2S A fix merged in 3.1-rc2 introduced a small regression, this should get it to build again. Signed-off-by: Arnd Bergmann Acked-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/kirkwood/kirkwood-i2s.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/kirkwood/kirkwood-i2s.c b/sound/soc/kirkwood/kirkwood-i2s.c index 8f16cd37c2af..d0bcf3fcea01 100644 --- a/sound/soc/kirkwood/kirkwood-i2s.c +++ b/sound/soc/kirkwood/kirkwood-i2s.c @@ -424,7 +424,7 @@ static __devinit int kirkwood_i2s_dev_probe(struct platform_device *pdev) if (!priv->mem) { dev_err(&pdev->dev, "request_mem_region failed\n"); err = -EBUSY; - goto error_alloc; + goto err_alloc; } priv->io = ioremap(priv->mem->start, SZ_16K); From 24114504c4d585ec4aae7a2b2acb81bf741f8c8a Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 12 Sep 2011 09:31:49 +0200 Subject: [PATCH 208/326] fuse: fix flock breakage Commit 37fb3a30b4 ("fuse: fix flock") added in 3.1-rc4 caused flock() to fail with ENOSYS with the kernel ABI version 7.16 or earlier. Fix by falling back to testing FUSE_POSIX_LOCKS for ABI versions 7.16 and earlier. Reported-by: Martin Ziegler Signed-off-by: Miklos Szeredi Tested-by: Martin Ziegler Signed-off-by: Linus Torvalds --- fs/fuse/inode.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 12b502929da9..add96f6ffda5 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -812,6 +812,9 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) if (arg->minor >= 17) { if (!(arg->flags & FUSE_FLOCK_LOCKS)) fc->no_flock = 1; + } else { + if (!(arg->flags & FUSE_POSIX_LOCKS)) + fc->no_flock = 1; } if (arg->flags & FUSE_ATOMIC_O_TRUNC) fc->atomic_o_trunc = 1; From 5dfcc87fd79dfb96ed155b524337dbd0da4f5993 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 12 Sep 2011 09:38:03 +0200 Subject: [PATCH 209/326] fuse: fix memory leak kmemleak is reporting that 32 bytes are being leaked by FUSE: unreferenced object 0xe373b270 (size 32): comm "fusermount", pid 1207, jiffies 4294707026 (age 2675.187s) hex dump (first 32 bytes): 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x27/0x50 [] kmem_cache_alloc+0xc5/0x180 [] fuse_alloc_forget+0x1e/0x20 [] fuse_alloc_inode+0xb0/0xd0 [] alloc_inode+0x1c/0x80 [] iget5_locked+0x8f/0x1a0 [] fuse_iget+0x72/0x1a0 [] fuse_get_root_inode+0x8a/0x90 [] fuse_fill_super+0x3ef/0x590 [] mount_nodev+0x3f/0x90 [] fuse_mount+0x15/0x20 [] mount_fs+0x1c/0xc0 [] vfs_kern_mount+0x41/0x90 [] do_kern_mount+0x39/0xd0 [] do_mount+0x2e5/0x660 [] sys_mount+0x66/0xa0 This leak report is consistent and happens once per boot on 3.1.0-rc5-dirty. This happens if a FORGET request is queued after the fuse device was released. Reported-by: Sitsofe Wheeler Signed-off-by: Miklos Szeredi Tested-by: Sitsofe Wheeler Signed-off-by: Linus Torvalds --- fs/fuse/dev.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 168a80f7f12b..5cb8614508c3 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -258,10 +258,14 @@ void fuse_queue_forget(struct fuse_conn *fc, struct fuse_forget_link *forget, forget->forget_one.nlookup = nlookup; spin_lock(&fc->lock); - fc->forget_list_tail->next = forget; - fc->forget_list_tail = forget; - wake_up(&fc->waitq); - kill_fasync(&fc->fasync, SIGIO, POLL_IN); + if (fc->connected) { + fc->forget_list_tail->next = forget; + fc->forget_list_tail = forget; + wake_up(&fc->waitq); + kill_fasync(&fc->fasync, SIGIO, POLL_IN); + } else { + kfree(forget); + } spin_unlock(&fc->lock); } From 14d01ff5341866773e24c5b54a94f7e2520ca271 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Sun, 11 Sep 2011 17:59:04 -0400 Subject: [PATCH 210/326] ioctl: register LTTng ioctl The LTTng 2.0 kernel tracer (stand-alone module package, available at http://lttng.org) uses the 0xF6 ioctl range for tracer control and transport operations. Signed-off-by: Mathieu Desnoyers Signed-off-by: Linus Torvalds --- Documentation/ioctl/ioctl-number.txt | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt index 845a191004b1..54078ed96b37 100644 --- a/Documentation/ioctl/ioctl-number.txt +++ b/Documentation/ioctl/ioctl-number.txt @@ -319,4 +319,6 @@ Code Seq#(hex) Include File Comments 0xF4 00-1F video/mbxfb.h mbxfb +0xF6 all LTTng Linux Trace Toolkit Next Generation + 0xFD all linux/dm-ioctl.h From b6fd41e29dea9c6753b1843a77e50433e6123bcb Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 12 Sep 2011 14:02:02 -0700 Subject: [PATCH 211/326] Linux 3.1-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 03d97aa8c73e..522fa4784e69 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc5 +EXTRAVERSION = -rc6 NAME = "Divemaster Edition" # *DOCUMENTATION* From 8f9068609e8a5b4cbac9e0cf8332b5dcabf05422 Mon Sep 17 00:00:00 2001 From: Chris Bagwell Date: Fri, 9 Sep 2011 13:38:10 -0700 Subject: [PATCH 212/326] Input: wacom - fix touch parsing on newer Bamboos Bamboos with Product ID's > 0xD4 return values unrelated to pressure in touch 1 pressure field. They also report 2nd touch X/Y values shifted down 1 byte (where pressure was). This results in jumpy 1 finger touch and totally invalid 2nd finger data. For touch detection, switch to a Touch Present single bit that all versions of Bamboo support. For touch 2 offset, calculate offset based on a bit that is set different between the two packet layouts. Since touch pressure reports were removed from driver, there was no need to be reading pressure any more. Signed-off-by: Chris Bagwell Reviewed-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- drivers/input/tablet/wacom_wac.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index c31e4e9f2690..0dc97ec15c28 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -800,20 +800,22 @@ static int wacom_bpt_touch(struct wacom_wac *wacom) int i; for (i = 0; i < 2; i++) { - int p = data[9 * i + 2]; - bool touch = p && !wacom->shared->stylus_in_proximity; + int offset = (data[1] & 0x80) ? (8 * i) : (9 * i); + bool touch = data[offset + 3] & 0x80; - input_mt_slot(input, i); - input_mt_report_slot_state(input, MT_TOOL_FINGER, touch); /* * Touch events need to be disabled while stylus is * in proximity because user's hand is resting on touchpad * and sending unwanted events. User expects tablet buttons * to continue working though. */ + touch = touch && !wacom->shared->stylus_in_proximity; + + input_mt_slot(input, i); + input_mt_report_slot_state(input, MT_TOOL_FINGER, touch); if (touch) { - int x = get_unaligned_be16(&data[9 * i + 3]) & 0x7ff; - int y = get_unaligned_be16(&data[9 * i + 5]) & 0x7ff; + int x = get_unaligned_be16(&data[offset + 3]) & 0x7ff; + int y = get_unaligned_be16(&data[offset + 5]) & 0x7ff; if (features->quirks & WACOM_QUIRK_BBTOUCH_LOWRES) { x <<= 5; y <<= 5; From 99e14c9d4140e9ed2c8810322a377e2a51b356eb Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 13 Sep 2011 10:33:16 +0200 Subject: [PATCH 213/326] ALSA: hda - Terminate the recursive connection search properly The recursive search of widget connections in snd_hda_get_conn_index() must be terminated at the pin and the audio-out widgets. Otherwise you'll get "too deep connection" warnings unnecessarily. Reported-by: Francis Moreau Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_codec.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index 3e7850c238c3..f3aefef37216 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -579,9 +579,13 @@ int snd_hda_get_conn_index(struct hda_codec *codec, hda_nid_t mux, return -1; } recursive++; - for (i = 0; i < nums; i++) + for (i = 0; i < nums; i++) { + unsigned int type = get_wcaps_type(get_wcaps(codec, conn[i])); + if (type == AC_WID_PIN || type == AC_WID_AUD_OUT) + continue; if (snd_hda_get_conn_index(codec, conn[i], nid, recursive) >= 0) return i; + } return -1; } EXPORT_SYMBOL_HDA(snd_hda_get_conn_index); From 40257b953fdd519c743138f3fbe3962d54991116 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 10 Sep 2011 06:02:12 -0700 Subject: [PATCH 214/326] hwmon: (pmbus) Fix low limit temperature alarms Temperature alarms are detected by checking the alarm bit and comparing temperature limits against the current temperature. For low limits, this comparison needs to be reversed (temp < limit instead of temp > limit). This was not taken into account, resulting in wrong alarms if a temperature fell below a low limit. Fix by adding a low limit flag in the limit data structure. When creating the sensor entry, the order of registers to compare is now reversed for low limits. Signed-off-by: Guenter Roeck Acked-by: Jean Delvare Cc: stable@kernel.org # 3.0+ --- drivers/hwmon/pmbus/pmbus_core.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/hwmon/pmbus/pmbus_core.c b/drivers/hwmon/pmbus/pmbus_core.c index a561c3a0e916..397fc59b5682 100644 --- a/drivers/hwmon/pmbus/pmbus_core.c +++ b/drivers/hwmon/pmbus/pmbus_core.c @@ -978,6 +978,8 @@ static void pmbus_find_max_attr(struct i2c_client *client, struct pmbus_limit_attr { u16 reg; /* Limit register */ bool update; /* True if register needs updates */ + bool low; /* True if low limit; for limits with compare + functions only */ const char *attr; /* Attribute name */ const char *alarm; /* Alarm attribute name */ u32 sbit; /* Alarm attribute status bit */ @@ -1029,7 +1031,8 @@ static bool pmbus_add_limit_attrs(struct i2c_client *client, if (attr->compare) { pmbus_add_boolean_cmp(data, name, l->alarm, index, - cbase, cindex, + l->low ? cindex : cbase, + l->low ? cbase : cindex, attr->sbase + page, l->sbit); } else { pmbus_add_boolean_reg(data, name, @@ -1366,11 +1369,13 @@ static const struct pmbus_sensor_attr power_attributes[] = { static const struct pmbus_limit_attr temp_limit_attrs[] = { { .reg = PMBUS_UT_WARN_LIMIT, + .low = true, .attr = "min", .alarm = "min_alarm", .sbit = PB_TEMP_UT_WARNING, }, { .reg = PMBUS_UT_FAULT_LIMIT, + .low = true, .attr = "lcrit", .alarm = "lcrit_alarm", .sbit = PB_TEMP_UT_FAULT, @@ -1399,11 +1404,13 @@ static const struct pmbus_limit_attr temp_limit_attrs[] = { static const struct pmbus_limit_attr temp_limit_attrs23[] = { { .reg = PMBUS_UT_WARN_LIMIT, + .low = true, .attr = "min", .alarm = "min_alarm", .sbit = PB_TEMP_UT_WARNING, }, { .reg = PMBUS_UT_FAULT_LIMIT, + .low = true, .attr = "lcrit", .alarm = "lcrit_alarm", .sbit = PB_TEMP_UT_FAULT, From e3b73c4a25e9a5705b4ef28b91676caf01f9bc9f Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Tue, 13 Sep 2011 10:17:32 -0400 Subject: [PATCH 215/326] xen/e820: if there is no dom0_mem=, don't tweak extra_pages. The patch "xen: use maximum reservation to limit amount of usable RAM" (d312ae878b6aed3912e1acaaf5d0b2a9d08a4f11) breaks machines that do not use 'dom0_mem=' argument with: reserve RAM buffer: 000000133f2e2000 - 000000133fffffff (XEN) mm.c:4976:d0 Global bit is set to kernel page fffff8117e (XEN) domain_crash_sync called from entry.S (XEN) Domain 0 (vcpu#0) crashed on cpu#0: ... The reason being that the last E820 entry is created using the 'extra_pages' (which is based on how many pages have been freed). The mentioned git commit sets the initial value of 'extra_pages' using a hypercall which returns the number of pages (if dom0_mem has been used) or -1 otherwise. If the later we return with MAX_DOMAIN_PAGES as basis for calculation: return min(max_pages, MAX_DOMAIN_PAGES); and use it: extra_limit = xen_get_max_pages(); if (extra_limit >= max_pfn) extra_pages = extra_limit - max_pfn; else extra_pages = 0; which means we end up with extra_pages = 128GB in PFNs (33554432) - 8GB in PFNs (2097152, on this specific box, can be larger or smaller), and then we add that value to the E820 making it: Xen: 00000000ff000000 - 0000000100000000 (reserved) Xen: 0000000100000000 - 000000133f2e2000 (usable) which is clearly wrong. It should look as so: Xen: 00000000ff000000 - 0000000100000000 (reserved) Xen: 0000000100000000 - 000000027fbda000 (usable) Naturally this problem does not present itself if dom0_mem=max:X is used. CC: stable@kernel.org Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/setup.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index ff3dfa176814..09688eb4a899 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -305,10 +305,12 @@ char * __init xen_memory_setup(void) sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); extra_limit = xen_get_max_pages(); - if (extra_limit >= max_pfn) - extra_pages = extra_limit - max_pfn; - else - extra_pages = 0; + if (max_pfn + extra_pages > extra_limit) { + if (extra_limit > max_pfn) + extra_pages = extra_limit - max_pfn; + else + extra_pages = 0; + } extra_pages += xen_return_unused_memory(xen_start_info->nr_pages, &e820); From 3401dc6eba788ebc7c14ce51018d775b1c263399 Mon Sep 17 00:00:00 2001 From: George Date: Sat, 3 Sep 2011 10:58:47 -0500 Subject: [PATCH 216/326] rtlwifi: rtl8192su: Fix problem connecting to HT-enabled AP The driver fails to connect to 802.11n-enabled APs. The patch fixes Bug #42262. Signed-off-by: George Signed-off-by: Larry Finger Cc: Stable [2.6.39+] Signed-off-by: John W. Linville --- drivers/net/wireless/rtlwifi/rtl8192cu/trx.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c index 906e7aa55bc3..3e52a5496224 100644 --- a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c +++ b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c @@ -549,15 +549,16 @@ void rtl92cu_tx_fill_desc(struct ieee80211_hw *hw, (tcb_desc->rts_use_shortpreamble ? 1 : 0) : (tcb_desc->rts_use_shortgi ? 1 : 0))); if (mac->bw_40) { - if (tcb_desc->packet_bw) { + if (rate_flag & IEEE80211_TX_RC_DUP_DATA) { SET_TX_DESC_DATA_BW(txdesc, 1); SET_TX_DESC_DATA_SC(txdesc, 3); + } else if(rate_flag & IEEE80211_TX_RC_40_MHZ_WIDTH){ + SET_TX_DESC_DATA_BW(txdesc, 1); + SET_TX_DESC_DATA_SC(txdesc, mac->cur_40_prime_sc); } else { SET_TX_DESC_DATA_BW(txdesc, 0); - if (rate_flag & IEEE80211_TX_RC_DUP_DATA) - SET_TX_DESC_DATA_SC(txdesc, - mac->cur_40_prime_sc); - } + SET_TX_DESC_DATA_SC(txdesc, 0); + } } else { SET_TX_DESC_DATA_BW(txdesc, 0); SET_TX_DESC_DATA_SC(txdesc, 0); From bac2555c6d86387132930af4d14cb47c4dd3f4f7 Mon Sep 17 00:00:00 2001 From: George Date: Sat, 3 Sep 2011 10:58:48 -0500 Subject: [PATCH 217/326] rtlwifi: Fix problem when switching connections The driver fails to clear encryption keys making it impossible to switch connections. Signed-off-by: George Signed-off-by: Larry Finger Cc: Stable [2.6.39+] Signed-off-by: John W. Linville --- drivers/net/wireless/rtlwifi/core.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/wireless/rtlwifi/core.c b/drivers/net/wireless/rtlwifi/core.c index 1bdc1aa305c0..04c4e9eb6ee6 100644 --- a/drivers/net/wireless/rtlwifi/core.c +++ b/drivers/net/wireless/rtlwifi/core.c @@ -610,6 +610,11 @@ static void rtl_op_bss_info_changed(struct ieee80211_hw *hw, mac->link_state = MAC80211_NOLINK; memset(mac->bssid, 0, 6); + + /* reset sec info */ + rtl_cam_reset_sec_info(hw); + + rtl_cam_reset_all_entry(hw); mac->vendor = PEER_UNKNOWN; RT_TRACE(rtlpriv, COMP_MAC80211, DBG_DMESG, @@ -1063,6 +1068,9 @@ static int rtl_op_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd, *or clear all entry here. */ rtl_cam_delete_one_entry(hw, mac_addr, key_idx); + + rtl_cam_reset_sec_info(hw); + break; default: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, From 4bae7d976976fa52d345805ba686934cd548343e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 6 Sep 2011 12:47:39 +0200 Subject: [PATCH 218/326] mac80211: fix missing sta_lock in __sta_info_destroy Since my commit 34e895075e21be3e21e71d6317440d1ee7969ad0 ("mac80211: allow station add/remove to sleep") there is a race in mac80211 when it clears the TIM bit because a sleeping station disconnected, the spinlock isn't held around the relevant code any more. Use the right API to acquire the spinlock correctly. Cc: stable@kernel.org [2.6.34+] Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- net/mac80211/sta_info.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 3db78b696c5c..21070e9bc8d0 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -665,7 +665,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta) BUG_ON(!sdata->bss); atomic_dec(&sdata->bss->num_sta_ps); - __sta_info_clear_tim_bit(sdata->bss, sta); + sta_info_clear_tim_bit(sta); } local->num_sta--; From 6a6b3f3e13decfc4b97263a83ea4e80ac8cc89ae Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Fri, 9 Sep 2011 10:41:08 +0530 Subject: [PATCH 219/326] ath9k: Fix kernel panic on unplugging the device when the device is yanked out ath_pci_remove starts doing the cleanups, unregistering the hardware etc. so we should bail out immediately when we get drv_flush callback from mac80211 when the card is being unplugged. the panic occurs after we had associated to an AP. EIP: 0060:[] EFLAGS: 00010246 CPU: 0 EIP is at ath_reset+0xa0/0x1c0 [ath9k] EAX: 00000000 EBX: 000697c0 ECX: 00000002 EDX: f3c3ccf0 ESI: 00000000 EDI: 00000000 EBP: f43e7b78 ESP: f43e7b50 DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Process kworker/u:2 (pid: 182, ti=f43e6000 task=f3c3c7c0 task.ti=f43e6000) Stack: 0000002a 00000000 00000000 003e7b78 0000000f eaaa8500 ffffffea eaaa97c0 eaaaa000 00000001 f43e7ba8 fb315d23 f99e7721 ecece680 eaaac738 eaaa8500 eaaaa020 000000c8 000000c8 00000000 eaaa8d58 eaaa8500 f43e7bd0 fb080b29 Call Trace: [] ath9k_flush+0x103/0x170 [ath9k] [] __ieee80211_recalc_idle+0x2c9/0x400 [mac80211] [] ieee80211_recalc_idle+0x2e/0x60 [mac80211] [] ieee80211_mgd_deauth+0x173/0x210 [mac80211] [] ieee80211_deauth+0x19/0x20 [mac80211] [] __cfg80211_mlme_deauth+0xf3/0x140 [cfg80211] [] ? __mutex_lock_common+0x1f0/0x380 [] __cfg80211_disconnect+0x18d/0x1f0 [cfg80211] [] cfg80211_netdev_notifier_call+0x159/0x5c0 [cfg80211] [] ? packet_notifier+0x174/0x1f0 [] notifier_call_chain+0x82/0xb0 [] raw_notifier_call_chain+0x1f/0x30 [] call_netdevice_notifiers+0x2c/0x60 [] ? trace_hardirqs_on_caller+0xf4/0x180 [] __dev_close_many+0x4c/0xd0 [] dev_close_many+0x6d/0xc0 [] rollback_registered_many+0x93/0x1c0 [] ? trace_hardirqs_on+0xb/0x10 [] unregister_netdevice_many+0x15/0x50 [] ieee80211_remove_interfaces+0x7b/0xb0 [mac80211] [] ieee80211_unregister_hw+0x4b/0x110 [mac80211] [] ath9k_deinit_device+0x3a/0x60 [ath9k] [] ath_pci_remove+0x46/0x90 [ath9k] [] pci_device_remove+0x44/0x100 [] __device_release_driver+0x64/0xb0 [] device_release_driver+0x27/0x40 [] bus_remove_device+0x7b/0xa0 [] device_del+0xf1/0x180 [] device_unregister+0x10/0x20 [] pci_stop_bus_device+0x6e/0x80 [] pci_remove_bus_device+0x12/0xa0 [] pciehp_unconfigure_device+0x89/0x180 [] ? mark_held_locks+0x64/0x100 [] ? __mutex_unlock_slowpath+0xaf/0x140 [] pciehp_disable_slot+0x64/0x1b0 [] pciehp_power_thread+0xd0/0x100 [] ? process_one_work+0x100/0x4d0 [] process_one_work+0x17c/0x4d0 [] ? process_one_work+0x100/0x4d0 [] ? queue_interrupt_event+0xa0/0xa0 [] worker_thread+0x13b/0x320 [] ? trace_hardirqs_on+0xb/0x10 [] ? manage_workers+0x1e0/0x1e0 [] kthread+0x84/0x90 [] ? __init_kthread_worker+0x60/0x60 [] kernel_thread_helper+0x6/0x10 Cc: Rajkumar Manoharan Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/main.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 6530694a59ae..722967b86cf1 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -2303,6 +2303,12 @@ static void ath9k_flush(struct ieee80211_hw *hw, bool drop) mutex_lock(&sc->mutex); cancel_delayed_work_sync(&sc->tx_complete_work); + if (ah->ah_flags & AH_UNPLUGGED) { + ath_dbg(common, ATH_DBG_ANY, "Device has been unplugged!\n"); + mutex_unlock(&sc->mutex); + return; + } + if (sc->sc_flags & SC_OP_INVALID) { ath_dbg(common, ATH_DBG_ANY, "Device not present\n"); mutex_unlock(&sc->mutex); From 456fc37e4519f3f551830ce01c58ddaa35807204 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 12 Sep 2011 21:08:25 +0200 Subject: [PATCH 220/326] iwlagn: fix stack corruption Alexander reported a strange crash in iwlagn that Meenakshi and Wey couldn't reproduce. I just ran into the same issue and tracked it down to stack corruption. This fixes it. The problem was introduced in commit 4b8b99b6e650d0527f3a123744b7459976581d14 Author: Wey-Yi Guy Date: Fri Jul 8 14:29:48 2011 -0700 iwlagn: radio sensor offset in le16 format Cc: Wey-Yi Guy Cc: Meenakshi Venkataraman Reported-by: Alexander Diewald Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-agn-ucode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-ucode.c b/drivers/net/wireless/iwlwifi/iwl-agn-ucode.c index a895a099d086..56211006a182 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn-ucode.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn-ucode.c @@ -167,7 +167,7 @@ static int iwlagn_set_temperature_offset_calib(struct iwl_priv *priv) memset(&cmd, 0, sizeof(cmd)); iwl_set_calib_hdr(&cmd.hdr, IWL_PHY_CALIBRATE_TEMP_OFFSET_CMD); - memcpy(&cmd.radio_sensor_offset, offset_calib, sizeof(offset_calib)); + memcpy(&cmd.radio_sensor_offset, offset_calib, sizeof(*offset_calib)); if (!(cmd.radio_sensor_offset)) cmd.radio_sensor_offset = DEFAULT_RADIO_SENSOR_OFFSET; From 282cdb325aea4ebbc42ce753b47cc96145eb54bc Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 12 Sep 2011 12:09:10 -0700 Subject: [PATCH 221/326] iwlagn: fix command queue timeout If the command queue is constantly busy, which can happen in P2P, the hangcheck timer will frequently find a command in it and will eventually reset the device because nothing sets the timestamp for this queue when commands are processed. Fix this by setting the timestamp when a command completes. Cc: stable@kernel.org #2.6.39, #3.0.0 #3.1.0 Signed-off-by: Johannes Berg SIgned-off-by: Wey-Yi Guy Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c b/drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c index a6b2b1db0b1d..222d410c586e 100644 --- a/drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c +++ b/drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c @@ -771,6 +771,8 @@ void iwl_tx_cmd_complete(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb) cmd = txq->cmd[cmd_index]; meta = &txq->meta[cmd_index]; + txq->time_stamp = jiffies; + iwlagn_unmap_tfd(priv, meta, &txq->tfds[index], DMA_BIDIRECTIONAL); /* Input error checking is done when commands are added to queue. */ From f13c3620a4d1123dbf032f93f350b856ef292ced Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 12 Sep 2011 11:47:53 -0400 Subject: [PATCH 222/326] NFS: Fix a typo in nfs_flush_multi Fix a typo which causes an Oops in the RPC layer, when using wsize < 4k. Signed-off-by: Trond Myklebust Tested-by: Sricharan R --- fs/nfs/write.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index b39b37f80913..c9bd2a6b7d4b 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -958,7 +958,7 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc, struct list_head if (!data) goto out_bad; data->pagevec[0] = page; - nfs_write_rpcsetup(req, data, wsize, offset, desc->pg_ioflags); + nfs_write_rpcsetup(req, data, len, offset, desc->pg_ioflags); list_add(&data->list, res); requests++; nbytes -= len; From fb2088ccc139ffbf1cf359216883712dab4ae43d Mon Sep 17 00:00:00 2001 From: Sachin Prabhu Date: Mon, 1 Aug 2011 12:10:12 +0100 Subject: [PATCH 223/326] nfs: Do not allow multiple mounts on same mountpoint when using -o noac Do not allow multiple mounts on same mountpoint when using -o noac When you normally attempt to mount a share twice on the same mountpoint, a check in do_add_mount causes it to return an error # mount localhost:/nfsv3 /mnt # mount localhost:/nfsv3 /mnt mount.nfs: /mnt is already mounted or busy However when using the option 'noac', the user is able to mount the same share on the same mountpoint multiple times. This happens because a share mounted with the noac option is automatically assigned the 'sync' flag MS_SYNCHRONOUS in nfs_initialise_sb(). This flag is set after the check for already existing superblocks is done in sget(). The check for the mount flags in nfs_compare_mount_options() does not take into account the 'sync' flag applied later on in the code path. This means that when using 'noac', a new superblock structure is assigned for every new mount of the same share and multiple shares on the same mountpoint are allowed. ie. # mount -onoac localhost:/nfsv3 /mnt can be run multiple times. The patch checks for noac and assigns the sync flag before sget() is called to obtain an already existing superblock structure. Signed-off-by: Sachin Prabhu Reviewed-by: Jeff Layton Signed-off-by: Trond Myklebust --- fs/nfs/super.c | 23 ++++++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index b961ceac66b4..9b7dd7013b15 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -2035,9 +2035,6 @@ static inline void nfs_initialise_sb(struct super_block *sb) sb->s_blocksize = nfs_block_bits(server->wsize, &sb->s_blocksize_bits); - if (server->flags & NFS_MOUNT_NOAC) - sb->s_flags |= MS_SYNCHRONOUS; - sb->s_bdi = &server->backing_dev_info; nfs_super_set_maxbytes(sb, server->maxfilesize); @@ -2249,6 +2246,10 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type, if (server->flags & NFS_MOUNT_UNSHARED) compare_super = NULL; + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + sb_mntdata.mntflags |= MS_SYNCHRONOUS; + /* Get a superblock - note that we may end up sharing one that already exists */ s = sget(fs_type, compare_super, nfs_set_super, &sb_mntdata); if (IS_ERR(s)) { @@ -2361,6 +2362,10 @@ nfs_xdev_mount(struct file_system_type *fs_type, int flags, if (server->flags & NFS_MOUNT_UNSHARED) compare_super = NULL; + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + sb_mntdata.mntflags |= MS_SYNCHRONOUS; + /* Get a superblock - note that we may end up sharing one that already exists */ s = sget(&nfs_fs_type, compare_super, nfs_set_super, &sb_mntdata); if (IS_ERR(s)) { @@ -2628,6 +2633,10 @@ nfs4_remote_mount(struct file_system_type *fs_type, int flags, if (server->flags & NFS4_MOUNT_UNSHARED) compare_super = NULL; + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + sb_mntdata.mntflags |= MS_SYNCHRONOUS; + /* Get a superblock - note that we may end up sharing one that already exists */ s = sget(&nfs4_fs_type, compare_super, nfs_set_super, &sb_mntdata); if (IS_ERR(s)) { @@ -2916,6 +2925,10 @@ nfs4_xdev_mount(struct file_system_type *fs_type, int flags, if (server->flags & NFS4_MOUNT_UNSHARED) compare_super = NULL; + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + sb_mntdata.mntflags |= MS_SYNCHRONOUS; + /* Get a superblock - note that we may end up sharing one that already exists */ s = sget(&nfs4_fs_type, compare_super, nfs_set_super, &sb_mntdata); if (IS_ERR(s)) { @@ -3003,6 +3016,10 @@ nfs4_remote_referral_mount(struct file_system_type *fs_type, int flags, if (server->flags & NFS4_MOUNT_UNSHARED) compare_super = NULL; + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + sb_mntdata.mntflags |= MS_SYNCHRONOUS; + /* Get a superblock - note that we may end up sharing one that already exists */ s = sget(&nfs4_fs_type, compare_super, nfs_set_super, &sb_mntdata); if (IS_ERR(s)) { From 477694e71113fd0694b6bb0bcc2d006b8ac62691 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 19 Jul 2011 16:25:42 +0200 Subject: [PATCH 224/326] x86, iommu: Mark DMAR IRQ as non-threaded Mark this lowlevel IRQ handler as non-threaded. This prevents a boot crash when "threadirqs" is on the kernel commandline. Also the interrupt handler is handling hardware critical events which should not be delayed into a thread. Signed-off-by: Thomas Gleixner Cc: stable@kernel.org Signed-off-by: Ingo Molnar --- drivers/iommu/dmar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c index 3dc9befa5aec..6dcc7e2d54de 100644 --- a/drivers/iommu/dmar.c +++ b/drivers/iommu/dmar.c @@ -1388,7 +1388,7 @@ int dmar_set_interrupt(struct intel_iommu *iommu) return ret; } - ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu); + ret = request_irq(irq, dmar_fault, IRQF_NO_THREAD, iommu->name, iommu); if (ret) printk(KERN_ERR "IOMMU: can't request irq\n"); return ret; From 1a4b1a41b8a3d5256019854e851beed063b34344 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 13 Sep 2011 15:16:33 -0300 Subject: [PATCH 225/326] pci: Don't crash when reading mpss from root complex In pcie_find_smpss(), we have the following statement: if (dev->is_hotplug_bridge && (!list_is_singular(&dev->bus->devices) || dev->bus->self->pcie_type != PCI_EXP_TYPE_ROOT_PORT)) The problem is that at least on my machine, this gets called for the root complex (virtual P2P bridge), and dev->bus->self is NULL since the parent bus for this is not itself anchor to a PCI device. This adds the necessary NULL check. Signed-off-by: Benjamin Herrenschmidt Acked-by: Jon Mason Signed-off-by: Linus Torvalds --- drivers/pci/probe.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index b1187ff31d89..f3f94a5c068f 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1351,7 +1351,8 @@ static int pcie_find_smpss(struct pci_dev *dev, void *data) * will occur as normal. */ if (dev->is_hotplug_bridge && (!list_is_singular(&dev->bus->devices) || - dev->bus->self->pcie_type != PCI_EXP_TYPE_ROOT_PORT)) + (dev->bus->self && + dev->bus->self->pcie_type != PCI_EXP_TYPE_ROOT_PORT))) *smpss = 0; if (*smpss > dev->pcie_mpss) From 003f6c9df54970d8b19578d195b3e2b398cdbde2 Mon Sep 17 00:00:00 2001 From: H Hartley Sweeten Date: Fri, 9 Sep 2011 11:30:27 -0700 Subject: [PATCH 226/326] lib/sha1.c: quiet sparse noise about symbol not declared Include to pickup the declarations for sha_transform and sha_init to quite the sparse noise: warning: symbol 'sha_transform' was not declared. Should it be static? warning: symbol 'sha_init' was not declared. Should it be static? Signed-off-by: H Hartley Sweeten Acked-by: Mandeep Singh Baines Signed-off-by: Linus Torvalds --- lib/sha1.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/sha1.c b/lib/sha1.c index f33271dd00cb..1de509a159c8 100644 --- a/lib/sha1.c +++ b/lib/sha1.c @@ -8,6 +8,7 @@ #include #include #include +#include #include /* From cd5bd3df1a6e7a68454734fb109c409101c20f42 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Wed, 14 Sep 2011 04:43:07 -0400 Subject: [PATCH 227/326] hwmon: (coretemp) Initialize tmin ttarget is initialized when the driver is loaded, but tmin is not. As a result, tempX_max_hyst attributes read 0. Fix this. Also use THERM_*_THRESHOLD* constants in these initializations instead of hard-coding the constants. Signed-off-by: Jean Delvare Cc: "R, Durgadoss" Cc: Guenter Roeck Cc: Fenghua Yu Signed-off-by: Guenter Roeck --- drivers/hwmon/coretemp.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/hwmon/coretemp.c b/drivers/hwmon/coretemp.c index 59d83e83da7f..411257676133 100644 --- a/drivers/hwmon/coretemp.c +++ b/drivers/hwmon/coretemp.c @@ -601,7 +601,12 @@ static int create_core_data(struct platform_data *pdata, err = rdmsr_safe_on_cpu(cpu, tdata->intrpt_reg, &eax, &edx); if (!err) { tdata->attr_size += MAX_THRESH_ATTRS; - tdata->ttarget = tdata->tjmax - ((eax >> 16) & 0x7f) * 1000; + tdata->tmin = tdata->tjmax - + ((eax & THERM_MASK_THRESHOLD0) >> + THERM_SHIFT_THRESHOLD0) * 1000; + tdata->ttarget = tdata->tjmax - + ((eax & THERM_MASK_THRESHOLD1) >> + THERM_SHIFT_THRESHOLD1) * 1000; } pdata->core_data[attr_no] = tdata; From 2e1210bc3d065a6e26ff5fef228a9a7e08921d2c Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Wed, 14 Sep 2011 13:22:54 +0200 Subject: [PATCH 228/326] ALSA: HDA: Cirrus - fix "Surround Speaker" volume control name This patch fixes "Surround Speaker Playback Volume" being cut off. (Commit b4dabfc452a10 was probably meant to fix this, but it fixed only the "Switch" name, not the "Volume" name.) Signed-off-by: David Henningsson Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_cirrus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c index d6c93d92b550..c45f3e69bcf0 100644 --- a/sound/pci/hda/patch_cirrus.c +++ b/sound/pci/hda/patch_cirrus.c @@ -535,7 +535,7 @@ static int add_volume(struct hda_codec *codec, const char *name, int index, unsigned int pval, int dir, struct snd_kcontrol **kctlp) { - char tmp[32]; + char tmp[44]; struct snd_kcontrol_new knew = HDA_CODEC_VOLUME_IDX(tmp, index, 0, 0, HDA_OUTPUT); knew.private_value = pval; From ff02b13f6867af72682d7a9bb9bd705f9af2bab0 Mon Sep 17 00:00:00 2001 From: Ben Skeggs Date: Wed, 14 Sep 2011 06:08:06 +1000 Subject: [PATCH 229/326] drm/ttm: request zeroed system memory pages for new TT buffer objects Fixes an information leak to userspace, we were handing out un-zeroed pages for any newly created TTM_PL_TT buffer. Reported-by: Marcin Slusarz Signed-off-by: Ben Skeggs Tested-by: Marcin Slusarz Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/ttm/ttm_bo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index a4d38d85909a..ef06194c5aa6 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -394,7 +394,8 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo, if (!(new_man->flags & TTM_MEMTYPE_FLAG_FIXED)) { if (bo->ttm == NULL) { - ret = ttm_bo_add_ttm(bo, false); + bool zero = !(old_man->flags & TTM_MEMTYPE_FLAG_FIXED); + ret = ttm_bo_add_ttm(bo, zero); if (ret) goto out_err; } From 87463ff83bcda210d8f0ae440bd64d1548f852e7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Tue, 13 Sep 2011 11:27:35 +0200 Subject: [PATCH 230/326] drm/radeon: Don't read from CP ring write pointer registers. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Apparently this doesn't always work reliably, e.g. at resume time. Just initialize to 0, so the ring is considered empty. Tested with hibernation on Sumo and Cayman cards. Should fix https://bugs.launchpad.net/ubuntu/+source/linux/+bug/820746/ . Signed-off-by: Michel Dänzer Reviewed-by: Alex Deucher cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen.c | 4 ++-- drivers/gpu/drm/radeon/ni.c | 12 ++++++------ drivers/gpu/drm/radeon/r100.c | 6 ++---- drivers/gpu/drm/radeon/r600.c | 4 ++-- 4 files changed, 12 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index dc0a5b56c81a..f10d1c1c2554 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -1404,7 +1404,8 @@ int evergreen_cp_resume(struct radeon_device *rdev) /* Initialize the ring buffer's read and write pointers */ WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA); WREG32(CP_RB_RPTR_WR, 0); - WREG32(CP_RB_WPTR, 0); + rdev->cp.wptr = 0; + WREG32(CP_RB_WPTR, rdev->cp.wptr); /* set the wb address wether it's enabled or not */ WREG32(CP_RB_RPTR_ADDR, @@ -1426,7 +1427,6 @@ int evergreen_cp_resume(struct radeon_device *rdev) WREG32(CP_DEBUG, (1 << 27) | (1 << 28)); rdev->cp.rptr = RREG32(CP_RB_RPTR); - rdev->cp.wptr = RREG32(CP_RB_WPTR); evergreen_cp_start(rdev); rdev->cp.ready = true; diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index cbf57d75d925..99fbd793c08c 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -1187,7 +1187,8 @@ int cayman_cp_resume(struct radeon_device *rdev) /* Initialize the ring buffer's read and write pointers */ WREG32(CP_RB0_CNTL, tmp | RB_RPTR_WR_ENA); - WREG32(CP_RB0_WPTR, 0); + rdev->cp.wptr = 0; + WREG32(CP_RB0_WPTR, rdev->cp.wptr); /* set the wb address wether it's enabled or not */ WREG32(CP_RB0_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC); @@ -1207,7 +1208,6 @@ int cayman_cp_resume(struct radeon_device *rdev) WREG32(CP_RB0_BASE, rdev->cp.gpu_addr >> 8); rdev->cp.rptr = RREG32(CP_RB0_RPTR); - rdev->cp.wptr = RREG32(CP_RB0_WPTR); /* ring1 - compute only */ /* Set ring buffer size */ @@ -1220,7 +1220,8 @@ int cayman_cp_resume(struct radeon_device *rdev) /* Initialize the ring buffer's read and write pointers */ WREG32(CP_RB1_CNTL, tmp | RB_RPTR_WR_ENA); - WREG32(CP_RB1_WPTR, 0); + rdev->cp1.wptr = 0; + WREG32(CP_RB1_WPTR, rdev->cp1.wptr); /* set the wb address wether it's enabled or not */ WREG32(CP_RB1_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP1_RPTR_OFFSET) & 0xFFFFFFFC); @@ -1232,7 +1233,6 @@ int cayman_cp_resume(struct radeon_device *rdev) WREG32(CP_RB1_BASE, rdev->cp1.gpu_addr >> 8); rdev->cp1.rptr = RREG32(CP_RB1_RPTR); - rdev->cp1.wptr = RREG32(CP_RB1_WPTR); /* ring2 - compute only */ /* Set ring buffer size */ @@ -1245,7 +1245,8 @@ int cayman_cp_resume(struct radeon_device *rdev) /* Initialize the ring buffer's read and write pointers */ WREG32(CP_RB2_CNTL, tmp | RB_RPTR_WR_ENA); - WREG32(CP_RB2_WPTR, 0); + rdev->cp2.wptr = 0; + WREG32(CP_RB2_WPTR, rdev->cp2.wptr); /* set the wb address wether it's enabled or not */ WREG32(CP_RB2_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP2_RPTR_OFFSET) & 0xFFFFFFFC); @@ -1257,7 +1258,6 @@ int cayman_cp_resume(struct radeon_device *rdev) WREG32(CP_RB2_BASE, rdev->cp2.gpu_addr >> 8); rdev->cp2.rptr = RREG32(CP_RB2_RPTR); - rdev->cp2.wptr = RREG32(CP_RB2_WPTR); /* start the rings */ cayman_cp_start(rdev); diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index f2204cb1ccdf..11e44a3479e3 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -990,7 +990,8 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size) /* Force read & write ptr to 0 */ WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_RPTR_WR_ENA | RADEON_RB_NO_UPDATE); WREG32(RADEON_CP_RB_RPTR_WR, 0); - WREG32(RADEON_CP_RB_WPTR, 0); + rdev->cp.wptr = 0; + WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr); /* set the wb address whether it's enabled or not */ WREG32(R_00070C_CP_RB_RPTR_ADDR, @@ -1007,9 +1008,6 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size) WREG32(RADEON_CP_RB_CNTL, tmp); udelay(10); rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR); - rdev->cp.wptr = RREG32(RADEON_CP_RB_WPTR); - /* protect against crazy HW on resume */ - rdev->cp.wptr &= rdev->cp.ptr_mask; /* Set cp mode to bus mastering & enable cp*/ WREG32(RADEON_CP_CSQ_MODE, REG_SET(RADEON_INDIRECT2_START, indirect2_start) | diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index aa5571b73aa0..c68427612e3b 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2209,7 +2209,8 @@ int r600_cp_resume(struct radeon_device *rdev) /* Initialize the ring buffer's read and write pointers */ WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA); WREG32(CP_RB_RPTR_WR, 0); - WREG32(CP_RB_WPTR, 0); + rdev->cp.wptr = 0; + WREG32(CP_RB_WPTR, rdev->cp.wptr); /* set the wb address whether it's enabled or not */ WREG32(CP_RB_RPTR_ADDR, @@ -2231,7 +2232,6 @@ int r600_cp_resume(struct radeon_device *rdev) WREG32(CP_DEBUG, (1 << 27) | (1 << 28)); rdev->cp.rptr = RREG32(CP_RB_RPTR); - rdev->cp.wptr = RREG32(CP_RB_WPTR); r600_cp_start(rdev); rdev->cp.ready = true; From db318d7a8a910657f10ffdf223c971af20a9b09c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Tue, 13 Sep 2011 11:29:12 +0200 Subject: [PATCH 231/326] drm/radeon: Unreference GEM object outside of spinlock in page flip error path. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Should fix https://bugzilla.redhat.com/show_bug.cgi?id=726277 . Signed-off-by: Michel Dänzer cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_display.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 6cc17fb96a57..6adb3e58affd 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -473,8 +473,8 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc, spin_lock_irqsave(&dev->event_lock, flags); radeon_crtc->unpin_work = NULL; unlock_free: - drm_gem_object_unreference_unlocked(old_radeon_fb->obj); spin_unlock_irqrestore(&dev->event_lock, flags); + drm_gem_object_unreference_unlocked(old_radeon_fb->obj); radeon_fence_unref(&work->fence); kfree(work); From 2d2422aebc037095f77551119f795449d29befed Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 13 Sep 2011 22:26:00 +0000 Subject: [PATCH 232/326] xfs: fix a use after free in xfs_end_io_direct_write There is a window in which the ioend that we call inode_dio_wake on in xfs_end_io_direct_write is already free. Fix this by storing the inode pointer in a local variable. This is a fix for the regression introduced in 3.1-rc by "fs: move inode_dio_done to the end_io handler". Signed-off-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/xfs_aops.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 63e971e2b837..8c37dde4c521 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1300,6 +1300,7 @@ xfs_end_io_direct_write( bool is_async) { struct xfs_ioend *ioend = iocb->private; + struct inode *inode = ioend->io_inode; /* * blockdev_direct_IO can return an error even after the I/O @@ -1331,7 +1332,7 @@ xfs_end_io_direct_write( } /* XXX: probably should move into the real I/O completion handler */ - inode_dio_done(ioend->io_inode); + inode_dio_done(inode); } STATIC ssize_t From 1d2ef5901483004d74947bbf78d5146c24038fe7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 18:55:41 +0100 Subject: [PATCH 233/326] restore pinning the victim dentry in vfs_rmdir()/vfs_rename_dir() We used to get the victim pinned by dentry_unhash() prior to commit 64252c75a219 ("vfs: remove dget() from dentry_unhash()") and ->rmdir() and ->rename() instances relied on that; most of them don't care, but ones that used d_delete() themselves do. As the result, we are getting rmdir() oopses on NFS now. Just grab the reference before locking the victim and drop it explicitly after unlocking, same as vfs_rename_other() does. Signed-off-by: Al Viro Tested-by: Simon Kirby Cc: stable@kernel.org (3.0.x) Signed-off-by: Linus Torvalds --- fs/namei.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/namei.c b/fs/namei.c index b52bc685465f..f4788365ea22 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2616,6 +2616,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) if (!dir->i_op->rmdir) return -EPERM; + dget(dentry); mutex_lock(&dentry->d_inode->i_mutex); error = -EBUSY; @@ -2636,6 +2637,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) out: mutex_unlock(&dentry->d_inode->i_mutex); + dput(dentry); if (!error) d_delete(dentry); return error; @@ -3025,6 +3027,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, if (error) return error; + dget(new_dentry); if (target) mutex_lock(&target->i_mutex); @@ -3045,6 +3048,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, out: if (target) mutex_unlock(&target->i_mutex); + dput(new_dentry); if (!error) if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE)) d_move(old_dentry,new_dentry); From 8aacc9f550feb09c8b26470498345c192996a68e Mon Sep 17 00:00:00 2001 From: Caspar Zhang Date: Wed, 14 Sep 2011 16:20:58 -0700 Subject: [PATCH 234/326] mm/mempolicy.c: fix pgoff in mbind vma merge commit 9d8cebd4bcd7 ("mm: fix mbind vma merge problem") didn't really fix the mbind vma merge problem due to wrong pgoff value passing to vma_merge(), which made vma_merge() always return NULL. Before the patch applied, we are getting a result like: addr = 0x7fa58f00c000 [snip] 7fa58f00c000-7fa58f00d000 rw-p 00000000 00:00 0 7fa58f00d000-7fa58f00e000 rw-p 00000000 00:00 0 7fa58f00e000-7fa58f00f000 rw-p 00000000 00:00 0 here 7fa58f00c000->7fa58f00f000 we get 3 VMAs which are expected to be merged described as described in commit 9d8cebd. Re-testing the patched kernel with the reproducer provided in commit 9d8cebd, we get the correct result: addr = 0x7ffa5aaa2000 [snip] 7ffa5aaa2000-7ffa5aaa6000 rw-p 00000000 00:00 0 7fffd556f000-7fffd5584000 rw-p 00000000 00:00 0 [stack] Signed-off-by: Caspar Zhang Cc: KOSAKI Motohiro Cc: Christoph Lameter Cc: Hugh Dickins Cc: Mel Gorman Cc: Lee Schermerhorn Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 8b57173c1dd5..b1f70d6eec35 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -636,7 +636,6 @@ static int mbind_range(struct mm_struct *mm, unsigned long start, struct vm_area_struct *prev; struct vm_area_struct *vma; int err = 0; - pgoff_t pgoff; unsigned long vmstart; unsigned long vmend; @@ -649,9 +648,9 @@ static int mbind_range(struct mm_struct *mm, unsigned long start, vmstart = max(start, vma->vm_start); vmend = min(end, vma->vm_end); - pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); prev = vma_merge(mm, prev, vmstart, vmend, vma->vm_flags, - vma->anon_vma, vma->vm_file, pgoff, new_pol); + vma->anon_vma, vma->vm_file, vma->vm_pgoff, + new_pol); if (prev) { vma = prev; next = vma->vm_next; From 2bbff6c761e31b4642d297513cd3e0e89bc68ff7 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 14 Sep 2011 16:21:02 -0700 Subject: [PATCH 235/326] mm/mempolicy.c: make copy_from_user() provably correct When compiling mm/mempolicy.c with struct user copy checks the following warning is shown: In file included from arch/x86/include/asm/uaccess.h:572, from include/linux/uaccess.h:5, from include/linux/highmem.h:7, from include/linux/pagemap.h:10, from include/linux/mempolicy.h:70, from mm/mempolicy.c:68: In function `copy_from_user', inlined from `compat_sys_get_mempolicy' at mm/mempolicy.c:1415: arch/x86/include/asm/uaccess_64.h:64: warning: call to `copy_from_user_overflow' declared with attribute warning: copy_from_user() buffer size is not provably correct LD mm/built-in.o Fix this by passing correct buffer size value. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index b1f70d6eec35..9c51f9f58cac 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1411,7 +1411,9 @@ asmlinkage long compat_sys_get_mempolicy(int __user *policy, err = sys_get_mempolicy(policy, nm, nr_bits+1, addr, flags); if (!err && nmask) { - err = copy_from_user(bm, nm, alloc_size); + unsigned long copy_size; + copy_size = min_t(unsigned long, sizeof(bm), alloc_size); + err = copy_from_user(bm, nm, copy_size); /* ensure entire bitmap is zeroed */ err |= clear_user(nmask, ALIGN(maxnode-1, 8) / 8); err |= compat_put_bitmap(nmask, bm, nr_bits); From 0d6617c7732c083659566117ca620eda6f1a87af Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Wed, 14 Sep 2011 16:21:05 -0700 Subject: [PATCH 236/326] numa: fix NUMA compile error when sysfs and procfs are disabled The vmstat_text array is only defined for CONFIG_SYSFS or CONFIG_PROC_FS, yet it is referenced for per-node vmstat with CONFIG_NUMA: drivers/built-in.o: In function `node_read_vmstat': node.c:(.text+0x1106df): undefined reference to `vmstat_text' Introduced in commit fa25c503dfa2 ("mm: per-node vmstat: show proper vmstats"). Define the array for CONFIG_NUMA as well. [akpm@linux-foundation.org: remove unneeded ifdefs] Signed-off-by: David Rientjes Reported-by: Cong Wang Acked-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmstat.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/vmstat.c b/mm/vmstat.c index 20c18b7694b2..d52b13d28e8f 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -659,7 +659,7 @@ static void walk_zones_in_node(struct seq_file *m, pg_data_t *pgdat, } #endif -#if defined(CONFIG_PROC_FS) || defined(CONFIG_SYSFS) +#if defined(CONFIG_PROC_FS) || defined(CONFIG_SYSFS) || defined(CONFIG_NUMA) #ifdef CONFIG_ZONE_DMA #define TEXT_FOR_DMA(xx) xx "_dma", #else @@ -788,7 +788,7 @@ const char * const vmstat_text[] = { #endif /* CONFIG_VM_EVENTS_COUNTERS */ }; -#endif /* CONFIG_PROC_FS || CONFIG_SYSFS */ +#endif /* CONFIG_PROC_FS || CONFIG_SYSFS || CONFIG_NUMA */ #ifdef CONFIG_PROC_FS From e5f0bdc7840bdb791247cb98dfc1dab6ea6c7da4 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Wed, 14 Sep 2011 16:21:08 -0700 Subject: [PATCH 237/326] um: disable CMPXCHG_DOUBLE as it breaks UML build Commit b789ef518b2 ("slub: Add cmpxchg_double_slab()") tests for cmpxchg_double support in the SLUB code and it breaks UML builds with SLUB. Since UML does not support checking for CPU features, disable CMPXCHG_DOUBLE just like CMPXCHG_LOCAL is disabled for UML. Signed-off-by: Thadeu Lima de Souza Cascardo Reviewed-by: Christoph Lameter Cc: Jeff Dike Cc: Richard Weinberger Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig.x86 | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/um/Kconfig.x86 b/arch/um/Kconfig.x86 index d31ecf346b4e..21bebe63df66 100644 --- a/arch/um/Kconfig.x86 +++ b/arch/um/Kconfig.x86 @@ -10,6 +10,10 @@ config CMPXCHG_LOCAL bool default n +config CMPXCHG_DOUBLE + bool + default n + source "arch/x86/Kconfig.cpu" endmenu From b40997b872cdb70140f127af6069f00a86b6cf81 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jonathan=20Neusch=C3=A4fer?= Date: Wed, 14 Sep 2011 16:21:20 -0700 Subject: [PATCH 238/326] um: drivers/xterm.c: fix a file descriptor leak MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I could use out_close1, but that seems to be the code path to close the fd returned by os_create_unix_socket, and using it to close the fd returned by mkstemp might lead to some confusion, so I don't do it. Signed-off-by: Jonathan Neuschäfer Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/xterm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/drivers/xterm.c b/arch/um/drivers/xterm.c index 8ac7146c237f..2e1de5728604 100644 --- a/arch/um/drivers/xterm.c +++ b/arch/um/drivers/xterm.c @@ -123,6 +123,7 @@ static int xterm_open(int input, int output, int primary, void *d, err = -errno; printk(UM_KERN_ERR "xterm_open : unlink failed, errno = %d\n", errno); + close(fd); return err; } close(fd); From fbfe9c847edf57ac8232aeafb290f272289893a3 Mon Sep 17 00:00:00 2001 From: Ingo van Lil Date: Wed, 14 Sep 2011 16:21:23 -0700 Subject: [PATCH 239/326] um: Save FPU registers between task switches Some time ago Jeff prepared 42daba316557 ("uml: stop saving process FP state") for UML to stop saving the process FP state between task switches. The assumption was that since with SKAS0 every guest process runs inside a host process context the host OS will take care of keeping the proper FP state. Unfortunately this is not true for multi-threaded applications, where all guest threads share a single host process context yet all may use the FPU on their own. Although I haven't verified it I suspect things to be even worse in SKAS3 mode where all guest processes run inside a single host process. The patch reintroduces the saving and restoring of the FP context between task switches. [richard@nod.at: Ingo posted this patch in 2009, sadly it was never applied and got lost. Now in 2011 the problem was reported by Gunnar.] Signed-off-by: Ingo van Lil Signed-off-by: Richard Weinberger Reported-by: Tested-by: Cc: Stanislav Meduna Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/shared/registers.h | 2 +- arch/um/kernel/process.c | 2 +- arch/um/os-Linux/registers.c | 9 ++++++++- arch/um/os-Linux/skas/mem.c | 2 +- arch/um/os-Linux/skas/process.c | 19 ++++++++++++++++++- arch/um/sys-i386/shared/sysdep/ptrace.h | 1 + arch/um/sys-x86_64/shared/sysdep/ptrace.h | 1 + 7 files changed, 31 insertions(+), 5 deletions(-) diff --git a/arch/um/include/shared/registers.h b/arch/um/include/shared/registers.h index b0b4589e0ebc..f1e0aa56c52a 100644 --- a/arch/um/include/shared/registers.h +++ b/arch/um/include/shared/registers.h @@ -16,7 +16,7 @@ extern int restore_fpx_registers(int pid, unsigned long *fp_regs); extern int save_registers(int pid, struct uml_pt_regs *regs); extern int restore_registers(int pid, struct uml_pt_regs *regs); extern int init_registers(int pid); -extern void get_safe_registers(unsigned long *regs); +extern void get_safe_registers(unsigned long *regs, unsigned long *fp_regs); extern unsigned long get_thread_reg(int reg, jmp_buf *buf); extern int get_fp_registers(int pid, unsigned long *regs); extern int put_fp_registers(int pid, unsigned long *regs); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index fab4371184f6..21c1ae7c3d75 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -202,7 +202,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, arch_copy_thread(¤t->thread.arch, &p->thread.arch); } else { - get_safe_registers(p->thread.regs.regs.gp); + get_safe_registers(p->thread.regs.regs.gp, p->thread.regs.regs.fp); p->thread.request.u.thread = current->thread.request.u.thread; handler = new_thread_handler; } diff --git a/arch/um/os-Linux/registers.c b/arch/um/os-Linux/registers.c index 830fe6a1518a..b866b9e3bef9 100644 --- a/arch/um/os-Linux/registers.c +++ b/arch/um/os-Linux/registers.c @@ -8,6 +8,8 @@ #include #include #include "sysdep/ptrace.h" +#include "sysdep/ptrace_user.h" +#include "registers.h" int save_registers(int pid, struct uml_pt_regs *regs) { @@ -32,6 +34,7 @@ int restore_registers(int pid, struct uml_pt_regs *regs) /* This is set once at boot time and not changed thereafter */ static unsigned long exec_regs[MAX_REG_NR]; +static unsigned long exec_fp_regs[FP_SIZE]; int init_registers(int pid) { @@ -42,10 +45,14 @@ int init_registers(int pid) return -errno; arch_init_registers(pid); + get_fp_registers(pid, exec_fp_regs); return 0; } -void get_safe_registers(unsigned long *regs) +void get_safe_registers(unsigned long *regs, unsigned long *fp_regs) { memcpy(regs, exec_regs, sizeof(exec_regs)); + + if (fp_regs) + memcpy(fp_regs, exec_fp_regs, sizeof(exec_fp_regs)); } diff --git a/arch/um/os-Linux/skas/mem.c b/arch/um/os-Linux/skas/mem.c index d261f170d120..e771398be5f3 100644 --- a/arch/um/os-Linux/skas/mem.c +++ b/arch/um/os-Linux/skas/mem.c @@ -39,7 +39,7 @@ static unsigned long syscall_regs[MAX_REG_NR]; static int __init init_syscall_regs(void) { - get_safe_registers(syscall_regs); + get_safe_registers(syscall_regs, NULL); syscall_regs[REGS_IP_INDEX] = STUB_CODE + ((unsigned long) &batch_syscall_stub - (unsigned long) &__syscall_stub_start); diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index d6e0a2234b86..dee0e8cf8ad0 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -373,6 +373,9 @@ void userspace(struct uml_pt_regs *regs) if (ptrace(PTRACE_SETREGS, pid, 0, regs->gp)) fatal_sigsegv(); + if (put_fp_registers(pid, regs->fp)) + fatal_sigsegv(); + /* Now we set local_using_sysemu to be used for one loop */ local_using_sysemu = get_using_sysemu(); @@ -399,6 +402,12 @@ void userspace(struct uml_pt_regs *regs) fatal_sigsegv(); } + if (get_fp_registers(pid, regs->fp)) { + printk(UM_KERN_ERR "userspace - get_fp_registers failed, " + "errno = %d\n", errno); + fatal_sigsegv(); + } + UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */ if (WIFSTOPPED(status)) { @@ -457,10 +466,11 @@ void userspace(struct uml_pt_regs *regs) } static unsigned long thread_regs[MAX_REG_NR]; +static unsigned long thread_fp_regs[FP_SIZE]; static int __init init_thread_regs(void) { - get_safe_registers(thread_regs); + get_safe_registers(thread_regs, thread_fp_regs); /* Set parent's instruction pointer to start of clone-stub */ thread_regs[REGS_IP_INDEX] = STUB_CODE + (unsigned long) stub_clone_handler - @@ -503,6 +513,13 @@ int copy_context_skas0(unsigned long new_stack, int pid) return err; } + err = put_fp_registers(pid, thread_fp_regs); + if (err < 0) { + printk(UM_KERN_ERR "copy_context_skas0 : put_fp_registers " + "failed, pid = %d, err = %d\n", pid, err); + return err; + } + /* set a well known return code for detection of child write failure */ child_data->err = 12345678; diff --git a/arch/um/sys-i386/shared/sysdep/ptrace.h b/arch/um/sys-i386/shared/sysdep/ptrace.h index d50e62e07070..c398a5076111 100644 --- a/arch/um/sys-i386/shared/sysdep/ptrace.h +++ b/arch/um/sys-i386/shared/sysdep/ptrace.h @@ -53,6 +53,7 @@ extern int sysemu_supported; struct uml_pt_regs { unsigned long gp[MAX_REG_NR]; + unsigned long fp[HOST_FPX_SIZE]; struct faultinfo faultinfo; long syscall; int is_user; diff --git a/arch/um/sys-x86_64/shared/sysdep/ptrace.h b/arch/um/sys-x86_64/shared/sysdep/ptrace.h index fdba5457947a..8ee8f8e12af1 100644 --- a/arch/um/sys-x86_64/shared/sysdep/ptrace.h +++ b/arch/um/sys-x86_64/shared/sysdep/ptrace.h @@ -85,6 +85,7 @@ struct uml_pt_regs { unsigned long gp[MAX_REG_NR]; + unsigned long fp[HOST_FP_SIZE]; struct faultinfo faultinfo; long syscall; int is_user; From f71f94845e0126884eca8ce57a92e30b189c8e71 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:25 -0700 Subject: [PATCH 240/326] um: fix oopsable race in line_close() tty->count is decremented only after ->close() had been called and several tasks can hit it in parallel. As the result, using tty->count to check if you are the last one is broken. We end up leaving line->tty not reset to NULL and the next IRQ on that sucker will blow up trying to dereference pointers from kfree'd struct tty. Fix is obvious: we need to use a counter of our own. Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/line.c | 25 ++++++++++++------------- arch/um/include/shared/line.h | 1 + 2 files changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index d51c404239a8..c5bff1ddeabc 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -399,8 +399,8 @@ int line_setup_irq(int fd, int input, int output, struct line *line, void *data) * is done under a spinlock. Checking whether the device is in use is * line->tty->count > 1, also under the spinlock. * - * tty->count serves to decide whether the device should be enabled or - * disabled on the host. If it's equal to 1, then we are doing the + * line->count serves to decide whether the device should be enabled or + * disabled on the host. If it's equal to 0, then we are doing the * first open or last close. Otherwise, open and close just return. */ @@ -414,16 +414,16 @@ int line_open(struct line *lines, struct tty_struct *tty) goto out_unlock; err = 0; - if (tty->count > 1) + if (line->count++) goto out_unlock; - spin_unlock(&line->count_lock); - + BUG_ON(tty->driver_data); tty->driver_data = line; line->tty = tty; + spin_unlock(&line->count_lock); err = enable_chan(line); - if (err) + if (err) /* line_close() will be called by our caller */ return err; INIT_DELAYED_WORK(&line->task, line_timer_cb); @@ -436,7 +436,7 @@ int line_open(struct line *lines, struct tty_struct *tty) chan_window_size(&line->chan_list, &tty->winsize.ws_row, &tty->winsize.ws_col); - return err; + return 0; out_unlock: spin_unlock(&line->count_lock); @@ -460,17 +460,16 @@ void line_close(struct tty_struct *tty, struct file * filp) flush_buffer(line); spin_lock(&line->count_lock); - if (!line->valid) - goto out_unlock; + BUG_ON(!line->valid); - if (tty->count > 1) + if (--line->count) goto out_unlock; - spin_unlock(&line->count_lock); - line->tty = NULL; tty->driver_data = NULL; + spin_unlock(&line->count_lock); + if (line->sigio) { unregister_winch(tty); line->sigio = 0; @@ -498,7 +497,7 @@ static int setup_one_line(struct line *lines, int n, char *init, int init_prio, spin_lock(&line->count_lock); - if (line->tty != NULL) { + if (line->count) { *error_out = "Device is already open"; goto out; } diff --git a/arch/um/include/shared/line.h b/arch/um/include/shared/line.h index 72f4f25af247..63df3ca02ac2 100644 --- a/arch/um/include/shared/line.h +++ b/arch/um/include/shared/line.h @@ -33,6 +33,7 @@ struct line_driver { struct line { struct tty_struct *tty; spinlock_t count_lock; + unsigned long count; int valid; char *init_str; From 45cd5e2d4e632f55af1d6131f33b554c98f8b929 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:28 -0700 Subject: [PATCH 241/326] um: winch_interrupt() can happen inside of free_winch() ... so set winch->fd to -1 before doing free_irq(), to avoid having winch_interrupt() come from/during the latter and attempt to do reactivate_fd() on something that's already gone. Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/line.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index c5bff1ddeabc..91bf18941ca4 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -725,6 +725,8 @@ struct winch { static void free_winch(struct winch *winch, int free_irq_ok) { + int fd = winch->fd; + winch->fd = -1; if (free_irq_ok) free_irq(WINCH_IRQ, winch); @@ -732,8 +734,8 @@ static void free_winch(struct winch *winch, int free_irq_ok) if (winch->pid != -1) os_kill_process(winch->pid, 1); - if (winch->fd != -1) - os_close_file(winch->fd); + if (fd != -1) + os_close_file(fd); if (winch->stack != 0) free_stack(winch->stack, 0); kfree(winch); From 7cf3cf21aac7d75d27e8e7cd039bd33d19fb300d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:31 -0700 Subject: [PATCH 242/326] um: fix free_winch() mess while not doing free_irq() from irq handler is commendable, kfree() on the data passed to said handler before free_irq() is Not Good(tm). Freeing the stack it's being run on is also not nice... Solution: delay actually freeing stuff. Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/line.c | 38 ++++++++++++++++++++++++-------------- 1 file changed, 24 insertions(+), 14 deletions(-) diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 91bf18941ca4..364c8a15c4c3 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -721,43 +721,53 @@ struct winch { int pid; struct tty_struct *tty; unsigned long stack; + struct work_struct work; }; -static void free_winch(struct winch *winch, int free_irq_ok) +static void __free_winch(struct work_struct *work) { - int fd = winch->fd; - winch->fd = -1; - if (free_irq_ok) - free_irq(WINCH_IRQ, winch); - - list_del(&winch->list); + struct winch *winch = container_of(work, struct winch, work); + free_irq(WINCH_IRQ, winch); if (winch->pid != -1) os_kill_process(winch->pid, 1); - if (fd != -1) - os_close_file(fd); if (winch->stack != 0) free_stack(winch->stack, 0); kfree(winch); } +static void free_winch(struct winch *winch) +{ + int fd = winch->fd; + winch->fd = -1; + if (fd != -1) + os_close_file(fd); + list_del(&winch->list); + __free_winch(&winch->work); +} + static irqreturn_t winch_interrupt(int irq, void *data) { struct winch *winch = data; struct tty_struct *tty; struct line *line; + int fd = winch->fd; int err; char c; - if (winch->fd != -1) { - err = generic_read(winch->fd, &c, NULL); + if (fd != -1) { + err = generic_read(fd, &c, NULL); if (err < 0) { if (err != -EAGAIN) { + winch->fd = -1; + list_del(&winch->list); + os_close_file(fd); printk(KERN_ERR "winch_interrupt : " "read failed, errno = %d\n", -err); printk(KERN_ERR "fd %d is losing SIGWINCH " "support\n", winch->tty_fd); - free_winch(winch, 0); + INIT_WORK(&winch->work, __free_winch); + schedule_work(&winch->work); return IRQ_HANDLED; } goto out; @@ -829,7 +839,7 @@ static void unregister_winch(struct tty_struct *tty) list_for_each_safe(ele, next, &winch_handlers) { winch = list_entry(ele, struct winch, list); if (winch->tty == tty) { - free_winch(winch, 1); + free_winch(winch); break; } } @@ -845,7 +855,7 @@ static void winch_cleanup(void) list_for_each_safe(ele, next, &winch_handlers) { winch = list_entry(ele, struct winch, list); - free_winch(winch, 1); + free_winch(winch); } spin_unlock(&winch_handler_lock); From 01599cdc2f891415387aed9921909b3e9f27c801 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:34 -0700 Subject: [PATCH 243/326] um: PTRACE_[GS]ETFPXREGS had been wired on the wrong subarch It's 32bit-only, not 64bit-only... And while we are at it, it's set_fpxregs(), not set_fpregs()... Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/sys-i386/ptrace.c | 14 +++++++++++++- arch/um/sys-x86_64/ptrace.c | 14 +------------- 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/arch/um/sys-i386/ptrace.c b/arch/um/sys-i386/ptrace.c index d23b2d3ea384..49fd25a5f206 100644 --- a/arch/um/sys-i386/ptrace.c +++ b/arch/um/sys-i386/ptrace.c @@ -206,5 +206,17 @@ int set_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) long subarch_ptrace(struct task_struct *child, long request, unsigned long addr, unsigned long data) { - return -EIO; + int ret = -EIO; + void __user *datap = (void __user *) data; + + switch (request) { + case PTRACE_GETFPXREGS: /* Get the child FPU state. */ + ret = get_fpxregs(datap, child); + break; + case PTRACE_SETFPXREGS: /* Set the child FPU state. */ + ret = set_fpxregs(datap, child); + break; + } + + return ret; } diff --git a/arch/um/sys-x86_64/ptrace.c b/arch/um/sys-x86_64/ptrace.c index f43613643cdb..7ed49ac78f88 100644 --- a/arch/um/sys-x86_64/ptrace.c +++ b/arch/um/sys-x86_64/ptrace.c @@ -178,17 +178,5 @@ int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) long subarch_ptrace(struct task_struct *child, long request, unsigned long addr, unsigned long data) { - int ret = -EIO; - void __user *datap = (void __user *) data; - - switch (request) { - case PTRACE_GETFPXREGS: /* Get the child FPU state. */ - ret = get_fpregs(datap, child); - break; - case PTRACE_SETFPXREGS: /* Set the child FPU state. */ - ret = set_fpregs(datap, child); - break; - } - - return ret; + return -EIO; } From f2833aef6a0517e933992c8007f330d0df5d9317 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:37 -0700 Subject: [PATCH 244/326] um: clean arch_ptrace() up a bit 1) take subarch-specific stuff to subarch_ptrace() 2) PTRACE_{PEEK,POKE}{TEXT,DATA} is handled by ptrace_request() just fine... Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/asm/ptrace-generic.h | 4 ---- arch/um/kernel/ptrace.c | 28 ---------------------------- arch/um/sys-i386/asm/ptrace.h | 5 ----- arch/um/sys-i386/ptrace.c | 18 ++++++++++++------ arch/um/sys-x86_64/ptrace.c | 22 +++++++++++++++++++--- 5 files changed, 31 insertions(+), 46 deletions(-) diff --git a/arch/um/include/asm/ptrace-generic.h b/arch/um/include/asm/ptrace-generic.h index ae084ad1a3a0..1a7d2757fe05 100644 --- a/arch/um/include/asm/ptrace-generic.h +++ b/arch/um/include/asm/ptrace-generic.h @@ -42,10 +42,6 @@ extern long subarch_ptrace(struct task_struct *child, long request, unsigned long addr, unsigned long data); extern unsigned long getreg(struct task_struct *child, int regno); extern int putreg(struct task_struct *child, int regno, unsigned long value); -extern int get_fpregs(struct user_i387_struct __user *buf, - struct task_struct *child); -extern int set_fpregs(struct user_i387_struct __user *buf, - struct task_struct *child); extern int arch_copy_tls(struct task_struct *new); extern void clear_flushed_tls(struct task_struct *task); diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 701b672c1122..c9da32b0c707 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -50,23 +50,11 @@ long arch_ptrace(struct task_struct *child, long request, void __user *vp = p; switch (request) { - /* read word at location addr. */ - case PTRACE_PEEKTEXT: - case PTRACE_PEEKDATA: - ret = generic_ptrace_peekdata(child, addr, data); - break; - /* read the word at location addr in the USER area. */ case PTRACE_PEEKUSR: ret = peek_user(child, addr, data); break; - /* write the word at location addr. */ - case PTRACE_POKETEXT: - case PTRACE_POKEDATA: - ret = generic_ptrace_pokedata(child, addr, data); - break; - /* write the word at location addr in the USER area */ case PTRACE_POKEUSR: ret = poke_user(child, addr, data); @@ -106,16 +94,6 @@ long arch_ptrace(struct task_struct *child, long request, ret = 0; break; } -#endif -#ifdef PTRACE_GETFPREGS - case PTRACE_GETFPREGS: /* Get the child FPU state. */ - ret = get_fpregs(vp, child); - break; -#endif -#ifdef PTRACE_SETFPREGS - case PTRACE_SETFPREGS: /* Set the child FPU state. */ - ret = set_fpregs(vp, child); - break; #endif case PTRACE_GET_THREAD_AREA: ret = ptrace_get_thread_area(child, addr, vp); @@ -153,12 +131,6 @@ long arch_ptrace(struct task_struct *child, long request, ret = -EIO; break; } -#endif -#ifdef PTRACE_ARCH_PRCTL - case PTRACE_ARCH_PRCTL: - /* XXX Calls ptrace on the host - needs some SMP thinking */ - ret = arch_prctl(child, data, (void __user *) addr); - break; #endif default: ret = ptrace_request(child, request, addr, data); diff --git a/arch/um/sys-i386/asm/ptrace.h b/arch/um/sys-i386/asm/ptrace.h index 0273e4d09af7..5d2a59112537 100644 --- a/arch/um/sys-i386/asm/ptrace.h +++ b/arch/um/sys-i386/asm/ptrace.h @@ -42,11 +42,6 @@ */ struct user_desc; -extern int get_fpxregs(struct user_fxsr_struct __user *buf, - struct task_struct *child); -extern int set_fpxregs(struct user_fxsr_struct __user *buf, - struct task_struct *tsk); - extern int ptrace_get_thread_area(struct task_struct *child, int idx, struct user_desc __user *user_desc); diff --git a/arch/um/sys-i386/ptrace.c b/arch/um/sys-i386/ptrace.c index 49fd25a5f206..3375c2717851 100644 --- a/arch/um/sys-i386/ptrace.c +++ b/arch/um/sys-i386/ptrace.c @@ -145,7 +145,7 @@ int peek_user(struct task_struct *child, long addr, long data) return put_user(tmp, (unsigned long __user *) data); } -int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) +static int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) { int err, n, cpu = ((struct thread_info *) child->stack)->cpu; struct user_i387_struct fpregs; @@ -161,7 +161,7 @@ int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) return n; } -int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) +static int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) { int n, cpu = ((struct thread_info *) child->stack)->cpu; struct user_i387_struct fpregs; @@ -174,7 +174,7 @@ int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) (unsigned long *) &fpregs); } -int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) +static int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) { int err, n, cpu = ((struct thread_info *) child->stack)->cpu; struct user_fxsr_struct fpregs; @@ -190,7 +190,7 @@ int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) return n; } -int set_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) +static int set_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child) { int n, cpu = ((struct thread_info *) child->stack)->cpu; struct user_fxsr_struct fpregs; @@ -208,15 +208,21 @@ long subarch_ptrace(struct task_struct *child, long request, { int ret = -EIO; void __user *datap = (void __user *) data; - switch (request) { + case PTRACE_GETFPREGS: /* Get the child FPU state. */ + ret = get_fpregs(datap, child); + break; + case PTRACE_SETFPREGS: /* Set the child FPU state. */ + ret = set_fpregs(datap, child); + break; case PTRACE_GETFPXREGS: /* Get the child FPU state. */ ret = get_fpxregs(datap, child); break; case PTRACE_SETFPXREGS: /* Set the child FPU state. */ ret = set_fpxregs(datap, child); break; + default: + ret = -EIO; } - return ret; } diff --git a/arch/um/sys-x86_64/ptrace.c b/arch/um/sys-x86_64/ptrace.c index 7ed49ac78f88..4005506834fd 100644 --- a/arch/um/sys-x86_64/ptrace.c +++ b/arch/um/sys-x86_64/ptrace.c @@ -145,7 +145,7 @@ int is_syscall(unsigned long addr) return instr == 0x050f; } -int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) +static int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) { int err, n, cpu = ((struct thread_info *) child->stack)->cpu; long fpregs[HOST_FP_SIZE]; @@ -162,7 +162,7 @@ int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) return n; } -int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) +static int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) { int n, cpu = ((struct thread_info *) child->stack)->cpu; long fpregs[HOST_FP_SIZE]; @@ -178,5 +178,21 @@ int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child) long subarch_ptrace(struct task_struct *child, long request, unsigned long addr, unsigned long data) { - return -EIO; + int ret = -EIO; + void __user *datap = (void __user *) data; + + switch (request) { + case PTRACE_GETFPREGS: /* Get the child FPU state. */ + ret = get_fpregs(datap, child); + break; + case PTRACE_SETFPREGS: /* Set the child FPU state. */ + ret = set_fpregs(datap, child); + break; + case PTRACE_ARCH_PRCTL: + /* XXX Calls ptrace on the host - needs some SMP thinking */ + ret = arch_prctl(child, data, (void __user *) addr); + break; + } + + return ret; } From 2c51a4bc0233487db81706a0189715a59b18e9d6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 14 Sep 2011 16:21:38 -0700 Subject: [PATCH 245/326] um: fix strrchr() problems richard@nod.at: Fixes: /usr/lib/gcc/x86_64-linux-gnu/4.4.5/../../../../lib/libc.a(strrchr.o): In function `rindex': (.text+0x0): multiple definition of `strrchr' If both STATIC_LINK and UML_NET_VDE are set to "y" libc's strrchr may clash with the kernel implementation. This workaround comes originally from Jeff Dike: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=494995#35 Signed-off-by: Al Viro Signed-off-by: Richard Weinberger Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/um/Makefile b/arch/um/Makefile index fab8121d2b32..c0f712cc7c5f 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -41,7 +41,7 @@ KBUILD_CPPFLAGS += -I$(srctree)/$(ARCH_DIR)/sys-$(SUBARCH) KBUILD_CFLAGS += $(CFLAGS) $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ $(ARCH_INCLUDE) $(MODE_INCLUDE) -Dvmap=kernel_vmap \ -Din6addr_loopback=kernel_in6addr_loopback \ - -Din6addr_any=kernel_in6addr_any + -Din6addr_any=kernel_in6addr_any -Dstrrchr=kernel_strrchr KBUILD_AFLAGS += $(ARCH_INCLUDE) From 8694a1840c71fc7835595ee69139c83f7a8e5543 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Wed, 14 Sep 2011 16:21:42 -0700 Subject: [PATCH 246/326] alpha, gpio: GENERIC_GPIO default must be n Since GPIOLIB is optional on alpha, GENERIC_GPIO must not be selected by default. If GPIOLIB is enabled, it will select GENERIC_GPIO. See for an example of what 'def_bool y' breaks. Reported-by: Michael Cree Signed-off-by: Ben Hutchings Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Matt Turner Cc: Michael Cree Cc: Mark Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 60cde53d266c..8bb936226dee 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -51,7 +51,7 @@ config GENERIC_CMOS_UPDATE def_bool y config GENERIC_GPIO - def_bool y + bool config ZONE_DMA bool From d4c32f355cec2647efb65e4b24e630bd2386f787 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 14 Sep 2011 16:21:47 -0700 Subject: [PATCH 247/326] drivers/rtc/rtc-imxdi.c needs linux/sched.h Include linux/sched.h to fix below build error. CC drivers/rtc/rtc-imxdi.o drivers/rtc/rtc-imxdi.c: In function 'di_write_wait': drivers/rtc/rtc-imxdi.c:168: error: 'TASK_INTERRUPTIBLE' undeclared (first use in this function) drivers/rtc/rtc-imxdi.c:168: error: (Each undeclared identifier is reported only once drivers/rtc/rtc-imxdi.c:168: error: for each function it appears in.) drivers/rtc/rtc-imxdi.c:168: error: implicit declaration of function 'signal_pending' drivers/rtc/rtc-imxdi.c:168: error: implicit declaration of function 'schedule_timeout' drivers/rtc/rtc-imxdi.c: In function 'dryice_norm_irq': drivers/rtc/rtc-imxdi.c:329: error: 'TASK_INTERRUPTIBLE' undeclared (first use in this function) Signed-off-by: Axel Lin Cc: Baruch Siach Cc: Wan ZongShun Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-imxdi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/rtc/rtc-imxdi.c b/drivers/rtc/rtc-imxdi.c index 2dd3c0163272..d93a9608b1f0 100644 --- a/drivers/rtc/rtc-imxdi.c +++ b/drivers/rtc/rtc-imxdi.c @@ -35,6 +35,7 @@ #include #include #include +#include #include /* DryIce Register Definitions */ From a4d3e9e76337059406fcf3ead288c0df22a790e9 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 14 Sep 2011 16:21:52 -0700 Subject: [PATCH 248/326] mm: vmscan: fix force-scanning small targets without swap Without swap, anonymous pages are not scanned. As such, they should not count when considering force-scanning a small target if there is no swap. Otherwise, targets are not force-scanned even when their effective scan number is zero and the other conditions--kswapd/memcg--apply. This fixes 246e87a93934 ("memcg: fix get_scan_count() for small targets"). [akpm@linux-foundation.org: fix comment] Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Reviewed-by: Michal Hocko Cc: Ying Han Cc: Balbir Singh Cc: KOSAKI Motohiro Cc: Daisuke Nishimura Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index b7719ec10dc5..e49bcb6d4948 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1808,23 +1808,15 @@ static void get_scan_count(struct zone *zone, struct scan_control *sc, u64 fraction[2], denominator; enum lru_list l; int noswap = 0; - int force_scan = 0; + bool force_scan = false; unsigned long nr_force_scan[2]; - - anon = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_ANON) + - zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON); - file = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_FILE) + - zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE); - - if (((anon + file) >> priority) < SWAP_CLUSTER_MAX) { - /* kswapd does zone balancing and need to scan this zone */ - if (scanning_global_lru(sc) && current_is_kswapd()) - force_scan = 1; - /* memcg may have small limit and need to avoid priority drop */ - if (!scanning_global_lru(sc)) - force_scan = 1; - } + /* kswapd does zone balancing and needs to scan this zone */ + if (scanning_global_lru(sc) && current_is_kswapd()) + force_scan = true; + /* memcg may have small limit and need to avoid priority drop */ + if (!scanning_global_lru(sc)) + force_scan = true; /* If we have no swap space, do not bother scanning anon pages. */ if (!sc->may_swap || (nr_swap_pages <= 0)) { @@ -1837,6 +1829,11 @@ static void get_scan_count(struct zone *zone, struct scan_control *sc, goto out; } + anon = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_ANON) + + zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON); + file = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_FILE) + + zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE); + if (scanning_global_lru(sc)) { free = zone_page_state(zone, NR_FREE_PAGES); /* If we have very few page cache pages, From 185efc0f9a1f2d6ad6d4782c5d9e529f3290567f Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 14 Sep 2011 16:21:58 -0700 Subject: [PATCH 249/326] memcg: Revert "memcg: add memory.vmscan_stat" Revert the post-3.0 commit 82f9d486e59f5 ("memcg: add memory.vmscan_stat"). The implementation of per-memcg reclaim statistics violates how memcg hierarchies usually behave: hierarchically. The reclaim statistics are accounted to child memcgs and the parent hitting the limit, but not to hierarchy levels in between. Usually, hierarchical statistics are perfectly recursive, with each level representing the sum of itself and all its children. Since this exports statistics to userspace, this may lead to confusion and problems with changing things after the release, so revert it now, we can try again later. Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Cc: Daisuke Nishimura Cc: Michal Hocko Cc: Ying Han Cc: Balbir Singh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/cgroups/memory.txt | 85 +-------------- include/linux/memcontrol.h | 19 ---- include/linux/swap.h | 6 ++ mm/memcontrol.c | 172 ++----------------------------- mm/vmscan.c | 39 +------ 5 files changed, 18 insertions(+), 303 deletions(-) diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt index 6f3c598971fc..06eb6d957c83 100644 --- a/Documentation/cgroups/memory.txt +++ b/Documentation/cgroups/memory.txt @@ -380,7 +380,7 @@ will be charged as a new owner of it. 5.2 stat file -5.2.1 memory.stat file includes following statistics +memory.stat file includes following statistics # per-memory cgroup local status cache - # of bytes of page cache memory. @@ -438,89 +438,6 @@ Note: file_mapped is accounted only when the memory cgroup is owner of page cache.) -5.2.2 memory.vmscan_stat - -memory.vmscan_stat includes statistics information for memory scanning and -freeing, reclaiming. The statistics shows memory scanning information since -memory cgroup creation and can be reset to 0 by writing 0 as - - #echo 0 > ../memory.vmscan_stat - -This file contains following statistics. - -[param]_[file_or_anon]_pages_by_[reason]_[under_heararchy] -[param]_elapsed_ns_by_[reason]_[under_hierarchy] - -For example, - - scanned_file_pages_by_limit indicates the number of scanned - file pages at vmscan. - -Now, 3 parameters are supported - - scanned - the number of pages scanned by vmscan - rotated - the number of pages activated at vmscan - freed - the number of pages freed by vmscan - -If "rotated" is high against scanned/freed, the memcg seems busy. - -Now, 2 reason are supported - - limit - the memory cgroup's limit - system - global memory pressure + softlimit - (global memory pressure not under softlimit is not handled now) - -When under_hierarchy is added in the tail, the number indicates the -total memcg scan of its children and itself. - -elapsed_ns is a elapsed time in nanosecond. This may include sleep time -and not indicates CPU usage. So, please take this as just showing -latency. - -Here is an example. - -# cat /cgroup/memory/A/memory.vmscan_stat -scanned_pages_by_limit 9471864 -scanned_anon_pages_by_limit 6640629 -scanned_file_pages_by_limit 2831235 -rotated_pages_by_limit 4243974 -rotated_anon_pages_by_limit 3971968 -rotated_file_pages_by_limit 272006 -freed_pages_by_limit 2318492 -freed_anon_pages_by_limit 962052 -freed_file_pages_by_limit 1356440 -elapsed_ns_by_limit 351386416101 -scanned_pages_by_system 0 -scanned_anon_pages_by_system 0 -scanned_file_pages_by_system 0 -rotated_pages_by_system 0 -rotated_anon_pages_by_system 0 -rotated_file_pages_by_system 0 -freed_pages_by_system 0 -freed_anon_pages_by_system 0 -freed_file_pages_by_system 0 -elapsed_ns_by_system 0 -scanned_pages_by_limit_under_hierarchy 9471864 -scanned_anon_pages_by_limit_under_hierarchy 6640629 -scanned_file_pages_by_limit_under_hierarchy 2831235 -rotated_pages_by_limit_under_hierarchy 4243974 -rotated_anon_pages_by_limit_under_hierarchy 3971968 -rotated_file_pages_by_limit_under_hierarchy 272006 -freed_pages_by_limit_under_hierarchy 2318492 -freed_anon_pages_by_limit_under_hierarchy 962052 -freed_file_pages_by_limit_under_hierarchy 1356440 -elapsed_ns_by_limit_under_hierarchy 351386416101 -scanned_pages_by_system_under_hierarchy 0 -scanned_anon_pages_by_system_under_hierarchy 0 -scanned_file_pages_by_system_under_hierarchy 0 -rotated_pages_by_system_under_hierarchy 0 -rotated_anon_pages_by_system_under_hierarchy 0 -rotated_file_pages_by_system_under_hierarchy 0 -freed_pages_by_system_under_hierarchy 0 -freed_anon_pages_by_system_under_hierarchy 0 -freed_file_pages_by_system_under_hierarchy 0 -elapsed_ns_by_system_under_hierarchy 0 - 5.3 swappiness Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only. diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 3b535db00a94..343bd7661f2a 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -39,16 +39,6 @@ extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, struct mem_cgroup *mem_cont, int active, int file); -struct memcg_scanrecord { - struct mem_cgroup *mem; /* scanend memory cgroup */ - struct mem_cgroup *root; /* scan target hierarchy root */ - int context; /* scanning context (see memcontrol.c) */ - unsigned long nr_scanned[2]; /* the number of scanned pages */ - unsigned long nr_rotated[2]; /* the number of rotated pages */ - unsigned long nr_freed[2]; /* the number of freed pages */ - unsigned long elapsed; /* nsec of time elapsed while scanning */ -}; - #ifdef CONFIG_CGROUP_MEM_RES_CTLR /* * All "charge" functions with gfp_mask should use GFP_KERNEL or @@ -127,15 +117,6 @@ mem_cgroup_get_reclaim_stat_from_page(struct page *page); extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p); -extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem, - gfp_t gfp_mask, bool noswap, - struct memcg_scanrecord *rec); -extern unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, - gfp_t gfp_mask, bool noswap, - struct zone *zone, - struct memcg_scanrecord *rec, - unsigned long *nr_scanned); - #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP extern int do_swap_account; #endif diff --git a/include/linux/swap.h b/include/linux/swap.h index 14d62490922e..c71f84bb62ec 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -252,6 +252,12 @@ static inline void lru_cache_add_file(struct page *page) extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *mask); extern int __isolate_lru_page(struct page *page, int mode, int file); +extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem, + gfp_t gfp_mask, bool noswap); +extern unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, + gfp_t gfp_mask, bool noswap, + struct zone *zone, + unsigned long *nr_scanned); extern unsigned long shrink_all_memory(unsigned long nr_pages); extern int vm_swappiness; extern int remove_mapping(struct address_space *mapping, struct page *page); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index ebd1e86bef1c..3508777837c7 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -204,50 +204,6 @@ struct mem_cgroup_eventfd_list { static void mem_cgroup_threshold(struct mem_cgroup *mem); static void mem_cgroup_oom_notify(struct mem_cgroup *mem); -enum { - SCAN_BY_LIMIT, - SCAN_BY_SYSTEM, - NR_SCAN_CONTEXT, - SCAN_BY_SHRINK, /* not recorded now */ -}; - -enum { - SCAN, - SCAN_ANON, - SCAN_FILE, - ROTATE, - ROTATE_ANON, - ROTATE_FILE, - FREED, - FREED_ANON, - FREED_FILE, - ELAPSED, - NR_SCANSTATS, -}; - -struct scanstat { - spinlock_t lock; - unsigned long stats[NR_SCAN_CONTEXT][NR_SCANSTATS]; - unsigned long rootstats[NR_SCAN_CONTEXT][NR_SCANSTATS]; -}; - -const char *scanstat_string[NR_SCANSTATS] = { - "scanned_pages", - "scanned_anon_pages", - "scanned_file_pages", - "rotated_pages", - "rotated_anon_pages", - "rotated_file_pages", - "freed_pages", - "freed_anon_pages", - "freed_file_pages", - "elapsed_ns", -}; -#define SCANSTAT_WORD_LIMIT "_by_limit" -#define SCANSTAT_WORD_SYSTEM "_by_system" -#define SCANSTAT_WORD_HIERARCHY "_under_hierarchy" - - /* * The memory controller data structure. The memory controller controls both * page cache and RSS per cgroup. We would eventually like to provide @@ -313,8 +269,7 @@ struct mem_cgroup { /* For oom notifier event fd */ struct list_head oom_notify; - /* For recording LRU-scan statistics */ - struct scanstat scanstat; + /* * Should we move charges of a task when a task is moved into this * mem_cgroup ? And what type of charges should we move ? @@ -1678,44 +1633,6 @@ bool mem_cgroup_reclaimable(struct mem_cgroup *mem, bool noswap) } #endif -static void __mem_cgroup_record_scanstat(unsigned long *stats, - struct memcg_scanrecord *rec) -{ - - stats[SCAN] += rec->nr_scanned[0] + rec->nr_scanned[1]; - stats[SCAN_ANON] += rec->nr_scanned[0]; - stats[SCAN_FILE] += rec->nr_scanned[1]; - - stats[ROTATE] += rec->nr_rotated[0] + rec->nr_rotated[1]; - stats[ROTATE_ANON] += rec->nr_rotated[0]; - stats[ROTATE_FILE] += rec->nr_rotated[1]; - - stats[FREED] += rec->nr_freed[0] + rec->nr_freed[1]; - stats[FREED_ANON] += rec->nr_freed[0]; - stats[FREED_FILE] += rec->nr_freed[1]; - - stats[ELAPSED] += rec->elapsed; -} - -static void mem_cgroup_record_scanstat(struct memcg_scanrecord *rec) -{ - struct mem_cgroup *mem; - int context = rec->context; - - if (context >= NR_SCAN_CONTEXT) - return; - - mem = rec->mem; - spin_lock(&mem->scanstat.lock); - __mem_cgroup_record_scanstat(mem->scanstat.stats[context], rec); - spin_unlock(&mem->scanstat.lock); - - mem = rec->root; - spin_lock(&mem->scanstat.lock); - __mem_cgroup_record_scanstat(mem->scanstat.rootstats[context], rec); - spin_unlock(&mem->scanstat.lock); -} - /* * Scan the hierarchy if needed to reclaim memory. We remember the last child * we reclaimed from, so that we don't end up penalizing one child extensively @@ -1740,9 +1657,8 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem, bool noswap = reclaim_options & MEM_CGROUP_RECLAIM_NOSWAP; bool shrink = reclaim_options & MEM_CGROUP_RECLAIM_SHRINK; bool check_soft = reclaim_options & MEM_CGROUP_RECLAIM_SOFT; - struct memcg_scanrecord rec; unsigned long excess; - unsigned long scanned; + unsigned long nr_scanned; excess = res_counter_soft_limit_excess(&root_mem->res) >> PAGE_SHIFT; @@ -1750,15 +1666,6 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem, if (!check_soft && !shrink && root_mem->memsw_is_minimum) noswap = true; - if (shrink) - rec.context = SCAN_BY_SHRINK; - else if (check_soft) - rec.context = SCAN_BY_SYSTEM; - else - rec.context = SCAN_BY_LIMIT; - - rec.root = root_mem; - while (1) { victim = mem_cgroup_select_victim(root_mem); if (victim == root_mem) { @@ -1799,23 +1706,14 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem, css_put(&victim->css); continue; } - rec.mem = victim; - rec.nr_scanned[0] = 0; - rec.nr_scanned[1] = 0; - rec.nr_rotated[0] = 0; - rec.nr_rotated[1] = 0; - rec.nr_freed[0] = 0; - rec.nr_freed[1] = 0; - rec.elapsed = 0; /* we use swappiness of local cgroup */ if (check_soft) { ret = mem_cgroup_shrink_node_zone(victim, gfp_mask, - noswap, zone, &rec, &scanned); - *total_scanned += scanned; + noswap, zone, &nr_scanned); + *total_scanned += nr_scanned; } else ret = try_to_free_mem_cgroup_pages(victim, gfp_mask, - noswap, &rec); - mem_cgroup_record_scanstat(&rec); + noswap); css_put(&victim->css); /* * At shrinking usage, we can't check we should stop here or @@ -3854,18 +3752,14 @@ static int mem_cgroup_force_empty(struct mem_cgroup *mem, bool free_all) /* try to free all pages in this cgroup */ shrink = 1; while (nr_retries && mem->res.usage > 0) { - struct memcg_scanrecord rec; int progress; if (signal_pending(current)) { ret = -EINTR; goto out; } - rec.context = SCAN_BY_SHRINK; - rec.mem = mem; - rec.root = mem; progress = try_to_free_mem_cgroup_pages(mem, GFP_KERNEL, - false, &rec); + false); if (!progress) { nr_retries--; /* maybe some writeback is necessary */ @@ -4709,54 +4603,6 @@ static int mem_control_numa_stat_open(struct inode *unused, struct file *file) } #endif /* CONFIG_NUMA */ -static int mem_cgroup_vmscan_stat_read(struct cgroup *cgrp, - struct cftype *cft, - struct cgroup_map_cb *cb) -{ - struct mem_cgroup *mem = mem_cgroup_from_cont(cgrp); - char string[64]; - int i; - - for (i = 0; i < NR_SCANSTATS; i++) { - strcpy(string, scanstat_string[i]); - strcat(string, SCANSTAT_WORD_LIMIT); - cb->fill(cb, string, mem->scanstat.stats[SCAN_BY_LIMIT][i]); - } - - for (i = 0; i < NR_SCANSTATS; i++) { - strcpy(string, scanstat_string[i]); - strcat(string, SCANSTAT_WORD_SYSTEM); - cb->fill(cb, string, mem->scanstat.stats[SCAN_BY_SYSTEM][i]); - } - - for (i = 0; i < NR_SCANSTATS; i++) { - strcpy(string, scanstat_string[i]); - strcat(string, SCANSTAT_WORD_LIMIT); - strcat(string, SCANSTAT_WORD_HIERARCHY); - cb->fill(cb, string, mem->scanstat.rootstats[SCAN_BY_LIMIT][i]); - } - for (i = 0; i < NR_SCANSTATS; i++) { - strcpy(string, scanstat_string[i]); - strcat(string, SCANSTAT_WORD_SYSTEM); - strcat(string, SCANSTAT_WORD_HIERARCHY); - cb->fill(cb, string, mem->scanstat.rootstats[SCAN_BY_SYSTEM][i]); - } - return 0; -} - -static int mem_cgroup_reset_vmscan_stat(struct cgroup *cgrp, - unsigned int event) -{ - struct mem_cgroup *mem = mem_cgroup_from_cont(cgrp); - - spin_lock(&mem->scanstat.lock); - memset(&mem->scanstat.stats, 0, sizeof(mem->scanstat.stats)); - memset(&mem->scanstat.rootstats, 0, sizeof(mem->scanstat.rootstats)); - spin_unlock(&mem->scanstat.lock); - return 0; -} - - static struct cftype mem_cgroup_files[] = { { .name = "usage_in_bytes", @@ -4827,11 +4673,6 @@ static struct cftype mem_cgroup_files[] = { .mode = S_IRUGO, }, #endif - { - .name = "vmscan_stat", - .read_map = mem_cgroup_vmscan_stat_read, - .trigger = mem_cgroup_reset_vmscan_stat, - }, }; #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP @@ -5095,7 +4936,6 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont) atomic_set(&mem->refcnt, 1); mem->move_charge_at_immigrate = 0; mutex_init(&mem->thresholds_lock); - spin_lock_init(&mem->scanstat.lock); return &mem->css; free_out: __mem_cgroup_free(mem); diff --git a/mm/vmscan.c b/mm/vmscan.c index e49bcb6d4948..b55699cd9067 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -105,7 +105,6 @@ struct scan_control { /* Which cgroup do we reclaim from */ struct mem_cgroup *mem_cgroup; - struct memcg_scanrecord *memcg_record; /* * Nodemask of nodes allowed by the caller. If NULL, all nodes @@ -1349,8 +1348,6 @@ putback_lru_pages(struct zone *zone, struct scan_control *sc, int file = is_file_lru(lru); int numpages = hpage_nr_pages(page); reclaim_stat->recent_rotated[file] += numpages; - if (!scanning_global_lru(sc)) - sc->memcg_record->nr_rotated[file] += numpages; } if (!pagevec_add(&pvec, page)) { spin_unlock_irq(&zone->lru_lock); @@ -1394,10 +1391,6 @@ static noinline_for_stack void update_isolated_counts(struct zone *zone, reclaim_stat->recent_scanned[0] += *nr_anon; reclaim_stat->recent_scanned[1] += *nr_file; - if (!scanning_global_lru(sc)) { - sc->memcg_record->nr_scanned[0] += *nr_anon; - sc->memcg_record->nr_scanned[1] += *nr_file; - } } /* @@ -1511,9 +1504,6 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone, nr_reclaimed += shrink_page_list(&page_list, zone, sc); } - if (!scanning_global_lru(sc)) - sc->memcg_record->nr_freed[file] += nr_reclaimed; - local_irq_disable(); if (current_is_kswapd()) __count_vm_events(KSWAPD_STEAL, nr_reclaimed); @@ -1613,8 +1603,6 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, } reclaim_stat->recent_scanned[file] += nr_taken; - if (!scanning_global_lru(sc)) - sc->memcg_record->nr_scanned[file] += nr_taken; __count_zone_vm_events(PGREFILL, zone, pgscanned); if (file) @@ -1666,8 +1654,6 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, * get_scan_ratio. */ reclaim_stat->recent_rotated[file] += nr_rotated; - if (!scanning_global_lru(sc)) - sc->memcg_record->nr_rotated[file] += nr_rotated; move_active_pages_to_lru(zone, &l_active, LRU_ACTIVE + file * LRU_FILE); @@ -2265,10 +2251,9 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order, #ifdef CONFIG_CGROUP_MEM_RES_CTLR unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, - gfp_t gfp_mask, bool noswap, - struct zone *zone, - struct memcg_scanrecord *rec, - unsigned long *scanned) + gfp_t gfp_mask, bool noswap, + struct zone *zone, + unsigned long *nr_scanned) { struct scan_control sc = { .nr_scanned = 0, @@ -2278,9 +2263,7 @@ unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, .may_swap = !noswap, .order = 0, .mem_cgroup = mem, - .memcg_record = rec, }; - ktime_t start, end; sc.gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) | (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK); @@ -2289,7 +2272,6 @@ unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, sc.may_writepage, sc.gfp_mask); - start = ktime_get(); /* * NOTE: Although we can get the priority field, using it * here is not a good idea, since it limits the pages we can scan. @@ -2298,25 +2280,19 @@ unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem, * the priority and make it zero. */ shrink_zone(0, zone, &sc); - end = ktime_get(); - - if (rec) - rec->elapsed += ktime_to_ns(ktime_sub(end, start)); - *scanned = sc.nr_scanned; trace_mm_vmscan_memcg_softlimit_reclaim_end(sc.nr_reclaimed); + *nr_scanned = sc.nr_scanned; return sc.nr_reclaimed; } unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, gfp_t gfp_mask, - bool noswap, - struct memcg_scanrecord *rec) + bool noswap) { struct zonelist *zonelist; unsigned long nr_reclaimed; - ktime_t start, end; int nid; struct scan_control sc = { .may_writepage = !laptop_mode, @@ -2325,7 +2301,6 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, .nr_to_reclaim = SWAP_CLUSTER_MAX, .order = 0, .mem_cgroup = mem_cont, - .memcg_record = rec, .nodemask = NULL, /* we don't care the placement */ .gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) | (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK), @@ -2334,7 +2309,6 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, .gfp_mask = sc.gfp_mask, }; - start = ktime_get(); /* * Unlike direct reclaim via alloc_pages(), memcg's reclaim doesn't * take care of from where we get pages. So the node where we start the @@ -2349,9 +2323,6 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, sc.gfp_mask); nr_reclaimed = do_try_to_free_pages(zonelist, &sc, &shrink); - end = ktime_get(); - if (rec) - rec->elapsed += ktime_to_ns(ktime_sub(end, start)); trace_mm_vmscan_memcg_reclaim_end(nr_reclaimed); From 461ae488ecb125b140d7ea29ceeedbcce9327003 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Wed, 14 Sep 2011 16:22:02 -0700 Subject: [PATCH 250/326] mm: sync vmalloc address space page tables in alloc_vm_area() Xen backend drivers (e.g., blkback and netback) would sometimes fail to map grant pages into the vmalloc address space allocated with alloc_vm_area(). The GNTTABOP_map_grant_ref would fail because Xen could not find the page (in the L2 table) containing the PTEs it needed to update. (XEN) mm.c:3846:d0 Could not find L1 PTE for address fbb42000 netback and blkback were making the hypercall from a kernel thread where task->active_mm != &init_mm and alloc_vm_area() was only updating the page tables for init_mm. The usual method of deferring the update to the page tables of other processes (i.e., after taking a fault) doesn't work as a fault cannot occur during the hypercall. This would work on some systems depending on what else was using vmalloc. Fix this by reverting ef691947d8a3 ("vmalloc: remove vmalloc_sync_all() from alloc_vm_area()") and add a comment to explain why it's needed. Signed-off-by: David Vrabel Cc: Jeremy Fitzhardinge Cc: Konrad Rzeszutek Wilk Cc: Ian Campbell Cc: Keir Fraser Cc: [3.0.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmalloc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 7ef0903058ee..5016f19e1661 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -2140,6 +2140,14 @@ struct vm_struct *alloc_vm_area(size_t size) return NULL; } + /* + * If the allocated address space is passed to a hypercall + * before being used then we cannot rely on a page fault to + * trigger an update of the page tables. So sync all the page + * tables here. + */ + vmalloc_sync_all(); + return area; } EXPORT_SYMBOL_GPL(alloc_vm_area); From 83ede96e98f5a7eb3ed07c78cb1dd166581eb864 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 14 Sep 2011 16:22:06 -0700 Subject: [PATCH 251/326] cris: fix a build error in drivers/tty/serial/crisv10.c Fix these errors: drivers/tty/serial/crisv10.c:4453: error: 'if_ser0' undeclared (first use in this function): 2 errors in 2 logs drivers/tty/serial/crisv10.c:4453: error: (Each undeclared identifier is reported only once: 2 errors in 2 logs drivers/tty/serial/crisv10.c:4453: error: for each function it appears in.): 2 errors in 2 logs "if_ser0" is a typo, it should be "if_serial_0". Signed-off-by: WANG Cong Cc: Mikael Starvik Cc: Jesper Nilsson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/tty/serial/crisv10.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/tty/serial/crisv10.c b/drivers/tty/serial/crisv10.c index 225123b37f19..58be715913cd 100644 --- a/drivers/tty/serial/crisv10.c +++ b/drivers/tty/serial/crisv10.c @@ -4450,7 +4450,7 @@ static int __init rs_init(void) #if defined(CONFIG_ETRAX_RS485) #if defined(CONFIG_ETRAX_RS485_ON_PA) - if (cris_io_interface_allocate_pins(if_ser0, 'a', rs485_pa_bit, + if (cris_io_interface_allocate_pins(if_serial_0, 'a', rs485_pa_bit, rs485_pa_bit)) { printk(KERN_CRIT "ETRAX100LX serial: Could not allocate " "RS485 pin\n"); @@ -4459,7 +4459,7 @@ static int __init rs_init(void) } #endif #if defined(CONFIG_ETRAX_RS485_ON_PORT_G) - if (cris_io_interface_allocate_pins(if_ser0, 'g', rs485_pa_bit, + if (cris_io_interface_allocate_pins(if_serial_0, 'g', rs485_pa_bit, rs485_port_g_bit)) { printk(KERN_CRIT "ETRAX100LX serial: Could not allocate " "RS485 pin\n"); From 1ebe9dad947d3158676f5ae55fc8b4f05b85c527 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Wed, 14 Sep 2011 16:22:12 -0700 Subject: [PATCH 252/326] drivers/misc/pti.c: give 'comm' function scope in pti_control_frame_built_and_sent() In drivers/misc/pti.c::pti_control_frame_built_and_sent() we assign 'comm' to 'thread_name_p' if (!thread_name). The problem is that 'comm' then goes out of scope and later we use 'thread_name_p' which now refers to an out-of-scope variable. To fix that, simply move 'comm' up to have function scope. Signed-off-by: Jesper Juhl Cc: Greg Kroah-Hartman Cc: J Freyensee Cc: Jeremy Rocher Cc: Sergei Trofimovich Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/pti.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/misc/pti.c b/drivers/misc/pti.c index 06df1877ad0f..0b56e3f43573 100644 --- a/drivers/misc/pti.c +++ b/drivers/misc/pti.c @@ -165,6 +165,11 @@ static void pti_write_to_aperture(struct pti_masterchannel *mc, static void pti_control_frame_built_and_sent(struct pti_masterchannel *mc, const char *thread_name) { + /* + * Since we access the comm member in current's task_struct, we only + * need to be as large as what 'comm' in that structure is. + */ + char comm[TASK_COMM_LEN]; struct pti_masterchannel mccontrol = {.master = CONTROL_ID, .channel = 0}; const char *thread_name_p; @@ -172,13 +177,6 @@ static void pti_control_frame_built_and_sent(struct pti_masterchannel *mc, u8 control_frame[CONTROL_FRAME_LEN]; if (!thread_name) { - /* - * Since we access the comm member in current's task_struct, - * we only need to be as large as what 'comm' in that - * structure is. - */ - char comm[TASK_COMM_LEN]; - if (!in_interrupt()) get_task_comm(comm, current); else From 7a5caabd090b8f7d782c40fc1c048d798f2b6fd7 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 14 Sep 2011 16:22:16 -0700 Subject: [PATCH 253/326] drivers/leds/ledtrig-timer.c: fix broken sysfs delay handling Fix regression introduced by commit 5ada28bf7675 ("led-class: always implement blinking") which broke sysfs delay handling by not storing the updated value. Consequently it was only possible to set one of the delays through the sysfs interface as the other delay was automatically restored to it's default value. Reading the parameters always gave the defaults. Signed-off-by: Johan Hovold Acked-by: Florian Fainelli Acked-by: Richard Purdie Cc: [2.6.37+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/leds/ledtrig-timer.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/leds/ledtrig-timer.c b/drivers/leds/ledtrig-timer.c index d87c9d02f786..328c64c0841c 100644 --- a/drivers/leds/ledtrig-timer.c +++ b/drivers/leds/ledtrig-timer.c @@ -41,6 +41,7 @@ static ssize_t led_delay_on_store(struct device *dev, if (count == size) { led_blink_set(led_cdev, &state, &led_cdev->blink_delay_off); + led_cdev->blink_delay_on = state; ret = count; } @@ -69,6 +70,7 @@ static ssize_t led_delay_off_store(struct device *dev, if (count == size) { led_blink_set(led_cdev, &led_cdev->blink_delay_on, &state); + led_cdev->blink_delay_off = state; ret = count; } From 88cee8fd77af28d414b983798dd30c8950c71e31 Mon Sep 17 00:00:00 2001 From: Donggeun Kim Date: Wed, 14 Sep 2011 16:22:19 -0700 Subject: [PATCH 254/326] drivers/rtc/rtc-s3c.c: fix no occurrence of alarm interrupt The driver does not generate an alarm interrupt even though a time for an alarm is set. This results from disabling rtc_clk after setting the alarm time. To generate an alarm interrupt the driver should maintain its enabled state for rtc_clk the until alarm interrupt occurs. This patch permits generation of an alarm interrupt. [akpm@linux-foundation.org: make s3c_rtc_alarm_clk_lock local to s3c_rtc_alarm_clk_enable()] Signed-off-by: Donggeun Kim Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-s3c.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c index 4e7c04e773e0..7639ab906f02 100644 --- a/drivers/rtc/rtc-s3c.c +++ b/drivers/rtc/rtc-s3c.c @@ -51,6 +51,27 @@ static enum s3c_cpu_type s3c_rtc_cpu_type; static DEFINE_SPINLOCK(s3c_rtc_pie_lock); +static void s3c_rtc_alarm_clk_enable(bool enable) +{ + static DEFINE_SPINLOCK(s3c_rtc_alarm_clk_lock); + static bool alarm_clk_enabled; + unsigned long irq_flags; + + spin_lock_irqsave(&s3c_rtc_alarm_clk_lock, irq_flags); + if (enable) { + if (!alarm_clk_enabled) { + clk_enable(rtc_clk); + alarm_clk_enabled = true; + } + } else { + if (alarm_clk_enabled) { + clk_disable(rtc_clk); + alarm_clk_enabled = false; + } + } + spin_unlock_irqrestore(&s3c_rtc_alarm_clk_lock, irq_flags); +} + /* IRQ Handlers */ static irqreturn_t s3c_rtc_alarmirq(int irq, void *id) @@ -64,6 +85,9 @@ static irqreturn_t s3c_rtc_alarmirq(int irq, void *id) writeb(S3C2410_INTP_ALM, s3c_rtc_base + S3C2410_INTP); clk_disable(rtc_clk); + + s3c_rtc_alarm_clk_enable(false); + return IRQ_HANDLED; } @@ -97,6 +121,8 @@ static int s3c_rtc_setaie(struct device *dev, unsigned int enabled) writeb(tmp, s3c_rtc_base + S3C2410_RTCALM); clk_disable(rtc_clk); + s3c_rtc_alarm_clk_enable(enabled); + return 0; } From e71f5cc402ecb42b407ae52add7b173bf1c53daa Mon Sep 17 00:00:00 2001 From: Naga Chumbalkar Date: Wed, 14 Sep 2011 16:22:23 -0700 Subject: [PATCH 255/326] drivers/cpufreq/pcc-cpufreq.c: avoid NULL pointer dereference per_cpu(processors, n) can be NULL, resulting in: Loading CPUFreq modules[ 437.661360] BUG: unable to handle kernel NULL pointer dereference at (null) IP: [] pcc_cpufreq_cpu_init+0x74/0x220 [pcc_cpufreq] It's better to avoid the oops by failing the driver, and allowing the system to boot. Signed-off-by: Naga Chumbalkar Cc: Dave Jones Cc: Len Brown Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/cpufreq/pcc-cpufreq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/cpufreq/pcc-cpufreq.c b/drivers/cpufreq/pcc-cpufreq.c index 7b0603eb0129..cdc02ac8f41a 100644 --- a/drivers/cpufreq/pcc-cpufreq.c +++ b/drivers/cpufreq/pcc-cpufreq.c @@ -261,6 +261,9 @@ static int pcc_get_offset(int cpu) pr = per_cpu(processors, cpu); pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + if (!pr) + return -ENODEV; + status = acpi_evaluate_object(pr->handle, "PCCP", NULL, &buffer); if (ACPI_FAILURE(status)) return -ENODEV; From df4e33ad249b7480d6ba5dd11fc1dac5cfaffce1 Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Wed, 14 Sep 2011 16:22:26 -0700 Subject: [PATCH 256/326] MAINTAINERS: update e-mail address of Benny Halevy Acked-by: Boaz Harrosh Signed-off-by: Benny Halevy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7336c15abd43..7658e1ffd58e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4785,7 +4785,7 @@ F: drivers/net/wireless/orinoco/ OSD LIBRARY and FILESYSTEM M: Boaz Harrosh -M: Benny Halevy +M: Benny Halevy L: osd-dev@open-osd.org W: http://open-osd.org T: git git://git.open-osd.org/open-osd.git From fa2563e41c3d6d6e8af437643981ed28ae0cb56d Mon Sep 17 00:00:00 2001 From: Thomas Tuttle Date: Wed, 14 Sep 2011 16:22:28 -0700 Subject: [PATCH 257/326] workqueue: lock cwq access in drain_workqueue Take cwq->gcwq->lock to avoid racing between drain_workqueue checking to make sure the workqueues are empty and cwq_dec_nr_in_flight decrementing and then incrementing nr_active when it activates a delayed work. We discovered this when a corner case in one of our drivers resulted in us trying to destroy a workqueue in which the remaining work would always requeue itself again in the same workqueue. We would hit this race condition and trip the BUG_ON on workqueue.c:3080. Signed-off-by: Thomas Tuttle Acked-by: Tejun Heo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/workqueue.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 25fb1b0e53fa..1783aabc6128 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2412,8 +2412,13 @@ void drain_workqueue(struct workqueue_struct *wq) for_each_cwq_cpu(cpu, wq) { struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + bool drained; - if (!cwq->nr_active && list_empty(&cwq->delayed_works)) + spin_lock_irq(&cwq->gcwq->lock); + drained = !cwq->nr_active && list_empty(&cwq->delayed_works); + spin_unlock_irq(&cwq->gcwq->lock); + + if (drained) continue; if (++flush_cnt == 10 || From 4f5b04800a224aadb6cffcbbc3d3fa26e2367c7f Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 14 Sep 2011 16:22:29 -0700 Subject: [PATCH 258/326] drivers/gpio/gpio-generic.c: fix build errors Building a kernel with hotplug disabled results in a link failure: `bgpio_remove' referenced in section `___ksymtab_gpl+bgpio_remove' of drivers/built-in.o: defined in discarded section `.devexit.text' of drivers/built-in.o This is because of bgpio_remove() is exported. It is illegal to export symbols which are discarded either at link time or as part of an init/exit section. Fix this by dropping the __devexit attributation from bgpio_remove(). Also drop the __devinit attributation from bgpio_init(). Signed-off-by: Russell King Cc: Grant Likely Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpio/gpio-generic.c | 15 +++++---------- include/linux/basic_mmio_gpio.h | 15 +++++---------- 2 files changed, 10 insertions(+), 20 deletions(-) diff --git a/drivers/gpio/gpio-generic.c b/drivers/gpio/gpio-generic.c index 231714def4d2..4e24436b0f82 100644 --- a/drivers/gpio/gpio-generic.c +++ b/drivers/gpio/gpio-generic.c @@ -351,7 +351,7 @@ static int bgpio_setup_direction(struct bgpio_chip *bgc, return 0; } -int __devexit bgpio_remove(struct bgpio_chip *bgc) +int bgpio_remove(struct bgpio_chip *bgc) { int err = gpiochip_remove(&bgc->gc); @@ -361,15 +361,10 @@ int __devexit bgpio_remove(struct bgpio_chip *bgc) } EXPORT_SYMBOL_GPL(bgpio_remove); -int __devinit bgpio_init(struct bgpio_chip *bgc, - struct device *dev, - unsigned long sz, - void __iomem *dat, - void __iomem *set, - void __iomem *clr, - void __iomem *dirout, - void __iomem *dirin, - bool big_endian) +int bgpio_init(struct bgpio_chip *bgc, struct device *dev, + unsigned long sz, void __iomem *dat, void __iomem *set, + void __iomem *clr, void __iomem *dirout, void __iomem *dirin, + bool big_endian) { int ret; diff --git a/include/linux/basic_mmio_gpio.h b/include/linux/basic_mmio_gpio.h index 98999cf107ce..feb912196745 100644 --- a/include/linux/basic_mmio_gpio.h +++ b/include/linux/basic_mmio_gpio.h @@ -63,15 +63,10 @@ static inline struct bgpio_chip *to_bgpio_chip(struct gpio_chip *gc) return container_of(gc, struct bgpio_chip, gc); } -int __devexit bgpio_remove(struct bgpio_chip *bgc); -int __devinit bgpio_init(struct bgpio_chip *bgc, - struct device *dev, - unsigned long sz, - void __iomem *dat, - void __iomem *set, - void __iomem *clr, - void __iomem *dirout, - void __iomem *dirin, - bool big_endian); +int bgpio_remove(struct bgpio_chip *bgc); +int bgpio_init(struct bgpio_chip *bgc, struct device *dev, + unsigned long sz, void __iomem *dat, void __iomem *set, + void __iomem *clr, void __iomem *dirout, void __iomem *dirin, + bool big_endian); #endif /* __BASIC_MMIO_GPIO_H */ From cc39c6a9bbdebfcf1a7dee64d83bf302bc38d941 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 15 Sep 2011 08:45:19 +0800 Subject: [PATCH 259/326] mm: account skipped entries to avoid looping in find_get_pages The found entries by find_get_pages() could be all swap entries. In this case we skip the entries, but make sure the skipped entries are accounted, so we don't keep looping. Using nr_found > nr_skip to simplify code as suggested by Eric. Reported-and-tested-by: Eric Dumazet Signed-off-by: Shaohua Li Signed-off-by: Linus Torvalds --- mm/filemap.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 645a080ba4df..7771871fa353 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -827,13 +827,14 @@ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, { unsigned int i; unsigned int ret; - unsigned int nr_found; + unsigned int nr_found, nr_skip; rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, (void ***)pages, NULL, start, nr_pages); ret = 0; + nr_skip = 0; for (i = 0; i < nr_found; i++) { struct page *page; repeat: @@ -856,6 +857,7 @@ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, * here as an exceptional entry: so skip over it - * we only reach this from invalidate_mapping_pages(). */ + nr_skip++; continue; } @@ -876,7 +878,7 @@ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, * If all entries were removed before we could secure them, * try again, because callers stop trying once 0 is returned. */ - if (unlikely(!ret && nr_found)) + if (unlikely(!ret && nr_found > nr_skip)) goto restart; rcu_read_unlock(); return ret; From 763437a9e7737535b2fc72175ad4974048769be6 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Thu, 15 Sep 2011 08:49:25 +0200 Subject: [PATCH 260/326] ALSA: pcm - fix race condition in wait_for_avail() wait_for_avail() in pcm_lib.c has a race in it (observed in practice by an Intel validation group). The function is supposed to return once space in the buffer has become available, or if some timeout happens. The entity that creates space (irq handler of sound driver and some such) will do a wake up on a waitqueue that this function registers for. However there are two races in the existing code 1) If space became available between the caller noticing there was no space and this function actually sleeping, the wakeup is missed and the timeout condition will happen instead 2) If a wakeup happened but not sufficient space became available, the code will loop again and wait for more space. However, if the second wake comes in prior to hitting the schedule_timeout_interruptible(), it will be missed, and potentially you'll wait out until the timeout happens. The fix consists of using more careful setting of the current state (so that if a wakeup happens in the main loop window, the schedule_timeout() falls through) and by checking for available space prior to going into the schedule_timeout() loop, but after being on the waitqueue and having the state set to interruptible. [tiwai: the following changes have been added to Arjan's original patch: - merged akpm's fix for waitqueue adding order into a single patch - reduction of duplicated code of avail check ] Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Cc: Signed-off-by: Takashi Iwai --- sound/core/pcm_lib.c | 33 ++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c index 86d0caf91b35..62e90b862a0d 100644 --- a/sound/core/pcm_lib.c +++ b/sound/core/pcm_lib.c @@ -1761,6 +1761,10 @@ static int wait_for_avail(struct snd_pcm_substream *substream, snd_pcm_uframes_t avail = 0; long wait_time, tout; + init_waitqueue_entry(&wait, current); + set_current_state(TASK_INTERRUPTIBLE); + add_wait_queue(&runtime->tsleep, &wait); + if (runtime->no_period_wakeup) wait_time = MAX_SCHEDULE_TIMEOUT; else { @@ -1771,16 +1775,32 @@ static int wait_for_avail(struct snd_pcm_substream *substream, } wait_time = msecs_to_jiffies(wait_time * 1000); } - init_waitqueue_entry(&wait, current); - add_wait_queue(&runtime->tsleep, &wait); + for (;;) { if (signal_pending(current)) { err = -ERESTARTSYS; break; } + + /* + * We need to check if space became available already + * (and thus the wakeup happened already) first to close + * the race of space already having become available. + * This check must happen after been added to the waitqueue + * and having current state be INTERRUPTIBLE. + */ + if (is_playback) + avail = snd_pcm_playback_avail(runtime); + else + avail = snd_pcm_capture_avail(runtime); + if (avail >= runtime->twake) + break; snd_pcm_stream_unlock_irq(substream); - tout = schedule_timeout_interruptible(wait_time); + + tout = schedule_timeout(wait_time); + snd_pcm_stream_lock_irq(substream); + set_current_state(TASK_INTERRUPTIBLE); switch (runtime->status->state) { case SNDRV_PCM_STATE_SUSPENDED: err = -ESTRPIPE; @@ -1806,14 +1826,9 @@ static int wait_for_avail(struct snd_pcm_substream *substream, err = -EIO; break; } - if (is_playback) - avail = snd_pcm_playback_avail(runtime); - else - avail = snd_pcm_capture_avail(runtime); - if (avail >= runtime->twake) - break; } _endloop: + set_current_state(TASK_RUNNING); remove_wait_queue(&runtime->tsleep, &wait); *availp = avail; return err; From 773659483685d652970583384a0294948e57f8b3 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 14 Sep 2011 05:10:00 -0400 Subject: [PATCH 261/326] xen/irq: Alter the locking to use a mutex instead of a spinlock. When we allocate/change the IRQ informations, we do not need to use spinlocks. We can use a mutex (which is what the generic IRQ code does for allocations/changes). Fixes a slew of: BUG: sleeping function called from invalid context at /linux/kernel/mutex.c:271 in_atomic(): 1, irqs_disabled(): 0, pid: 3216, name: xenstored 2 locks held by xenstored/3216: #0: (&u->bind_mutex){......}, at: [] evtchn_ioctl+0x30/0x3a0 [xen_evtchn] #1: (irq_mapping_update_lock){......}, at: [] bind_evtchn_to_irq+0x24/0x90 Pid: 3216, comm: xenstored Not tainted 3.1.0-rc6-00021-g437a3d1 #2 Call Trace: [] __might_sleep+0x100/0x130 [] mutex_lock_nested+0x2f/0x50 [] __irq_alloc_descs+0x49/0x200 [] ? evtchn_ioctl+0x30/0x3a0 [xen_evtchn] [] xen_allocate_irq_dynamic+0x34/0x70 [] bind_evtchn_to_irq+0x5d/0x90 [] ? evtchn_bind_to_user+0x60/0x60 [xen_evtchn] [] bind_evtchn_to_irqhandler+0x32/0x80 [] evtchn_bind_to_user+0x49/0x60 [xen_evtchn] [] evtchn_ioctl+0x144/0x3a0 [xen_evtchn] [] ? vfsmount_lock_local_unlock+0x50/0x80 [] do_vfs_ioctl+0x9a/0x5e0 [] ? mntput+0x1f/0x30 [] ? fput+0x199/0x240 [] sys_ioctl+0xa1/0xb0 [] system_call_fastpath+0x16/0x1b Reported-by: Jim Burns Acked-by: Ian Campbell Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/events.c | 40 ++++++++++++++++++++-------------------- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/drivers/xen/events.c b/drivers/xen/events.c index da70f5c32eb9..7523719bf8a4 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -54,7 +54,7 @@ * This lock protects updates to the following mapping and reference-count * arrays. The lock does not need to be acquired to read the mapping tables. */ -static DEFINE_SPINLOCK(irq_mapping_update_lock); +static DEFINE_MUTEX(irq_mapping_update_lock); static LIST_HEAD(xen_irq_list_head); @@ -631,7 +631,7 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi, int irq = -1; struct physdev_irq irq_op; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); irq = find_irq_by_gsi(gsi); if (irq != -1) { @@ -684,7 +684,7 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi, handle_edge_irq, name); out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; } @@ -710,7 +710,7 @@ int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc, { int irq, ret; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); irq = xen_allocate_irq_dynamic(); if (irq == -1) @@ -724,10 +724,10 @@ int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc, if (ret < 0) goto error_irq; out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; error_irq: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); xen_free_irq(irq); return -1; } @@ -740,7 +740,7 @@ int xen_destroy_irq(int irq) struct irq_info *info = info_for_irq(irq); int rc = -ENOENT; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); desc = irq_to_desc(irq); if (!desc) @@ -766,7 +766,7 @@ int xen_destroy_irq(int irq) xen_free_irq(irq); out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return rc; } @@ -776,7 +776,7 @@ int xen_irq_from_pirq(unsigned pirq) struct irq_info *info; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); list_for_each_entry(info, &xen_irq_list_head, list) { if (info == NULL || info->type != IRQT_PIRQ) @@ -787,7 +787,7 @@ int xen_irq_from_pirq(unsigned pirq) } irq = -1; out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; } @@ -802,7 +802,7 @@ int bind_evtchn_to_irq(unsigned int evtchn) { int irq; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); irq = evtchn_to_irq[evtchn]; @@ -818,7 +818,7 @@ int bind_evtchn_to_irq(unsigned int evtchn) } out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; } @@ -829,7 +829,7 @@ static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu) struct evtchn_bind_ipi bind_ipi; int evtchn, irq; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); irq = per_cpu(ipi_to_irq, cpu)[ipi]; @@ -853,7 +853,7 @@ static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu) } out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; } @@ -878,7 +878,7 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu) struct evtchn_bind_virq bind_virq; int evtchn, irq; - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); irq = per_cpu(virq_to_irq, cpu)[virq]; @@ -903,7 +903,7 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu) } out: - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); return irq; } @@ -913,7 +913,7 @@ static void unbind_from_irq(unsigned int irq) struct evtchn_close close; int evtchn = evtchn_from_irq(irq); - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); if (VALID_EVTCHN(evtchn)) { close.port = evtchn; @@ -943,7 +943,7 @@ static void unbind_from_irq(unsigned int irq) xen_free_irq(irq); - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); } int bind_evtchn_to_irqhandler(unsigned int evtchn, @@ -1279,7 +1279,7 @@ void rebind_evtchn_irq(int evtchn, int irq) will also be masked. */ disable_irq(irq); - spin_lock(&irq_mapping_update_lock); + mutex_lock(&irq_mapping_update_lock); /* After resume the irq<->evtchn mappings are all cleared out */ BUG_ON(evtchn_to_irq[evtchn] != -1); @@ -1289,7 +1289,7 @@ void rebind_evtchn_irq(int evtchn, int irq) xen_irq_info_evtchn_init(irq, evtchn); - spin_unlock(&irq_mapping_update_lock); + mutex_unlock(&irq_mapping_update_lock); /* new event channels are always bound to cpu 0 */ irq_set_affinity(irq, cpumask_of(0)); From 61cca2fab7ecba18f9b9680cd736ef5fa82ad3b1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 15 Sep 2011 08:52:40 +0100 Subject: [PATCH 262/326] xen/i386: follow-up to "replace order-based range checking of M2P table by linear one" The numbers obtained from the hypervisor really can't ever lead to an overflow here, only the original calculation going through the order of the range could have. This avoids the (as Jeremy points outs) somewhat ugly NULL-based calculation here. Signed-off-by: Jan Beulich Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/mmu.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 24abc1f50dc5..a3872f7632e8 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -1721,10 +1721,8 @@ void __init xen_setup_machphys_mapping(void) machine_to_phys_nr = MACH2PHYS_NR_ENTRIES; } #ifdef CONFIG_X86_32 - if ((machine_to_phys_mapping + machine_to_phys_nr) - < machine_to_phys_mapping) - machine_to_phys_nr = (unsigned long *)NULL - - machine_to_phys_mapping; + WARN_ON((machine_to_phys_mapping + (machine_to_phys_nr - 1)) + < machine_to_phys_mapping); #endif } From f588c960fcaa6fa8bf82930bb819c9aca4eb9347 Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Thu, 15 Sep 2011 10:48:27 -0400 Subject: [PATCH 263/326] hfsplus: Fix kfree of wrong pointers in hfsplus_fill_super() error path Commit 6596528e391a ("hfsplus: ensure bio requests are not smaller than the hardware sectors") changed the pointers used for volume header allocations but failed to free the correct pointers in the error path path of hfsplus_fill_super() and hfsplus_read_wrapper. The second hunk came from a separate patch by Pavel Ivanov. Reported-by: Pavel Ivanov Signed-off-by: Seth Forshee Signed-off-by: Christoph Hellwig Cc: Signed-off-by: Linus Torvalds --- fs/hfsplus/super.c | 4 ++-- fs/hfsplus/wrapper.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index c106ca22e812..cadbb8c81887 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -525,8 +525,8 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) out_close_ext_tree: hfs_btree_close(sbi->ext_tree); out_free_vhdr: - kfree(sbi->s_vhdr); - kfree(sbi->s_backup_vhdr); + kfree(sbi->s_vhdr_buf); + kfree(sbi->s_backup_vhdr_buf); out_unload_nls: unload_nls(sbi->nls); unload_nls(nls); diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index 10e515a0d452..7daf4b852d1c 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -272,9 +272,9 @@ int hfsplus_read_wrapper(struct super_block *sb) return 0; out_free_backup_vhdr: - kfree(sbi->s_backup_vhdr); + kfree(sbi->s_backup_vhdr_buf); out_free_vhdr: - kfree(sbi->s_vhdr); + kfree(sbi->s_vhdr_buf); out: return error; } From f1fcd9f0e96d12498afb5543107f560f196cfcf3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 15 Sep 2011 10:48:40 -0400 Subject: [PATCH 264/326] hfsplus: fix filesystem size checks generic_check_addressable can't deal with hfsplus's larger than page size allocation blocks, so simply opencode the checks that we actually need in hfsplus_fill_super. Signed-off-by: Christoph Hellwig Reported-by: Pavel Ivanov Tested-by: Pavel Ivanov Signed-off-by: Linus Torvalds --- fs/hfsplus/super.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index cadbb8c81887..d24a9b666a23 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -344,6 +344,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) struct inode *root, *inode; struct qstr str; struct nls_table *nls = NULL; + u64 last_fs_block, last_fs_page; int err; err = -EINVAL; @@ -399,9 +400,13 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) if (!sbi->rsrc_clump_blocks) sbi->rsrc_clump_blocks = 1; - err = generic_check_addressable(sbi->alloc_blksz_shift, - sbi->total_blocks); - if (err) { + err = -EFBIG; + last_fs_block = sbi->total_blocks - 1; + last_fs_page = (last_fs_block << sbi->alloc_blksz_shift) >> + PAGE_CACHE_SHIFT; + + if ((last_fs_block > (sector_t)(~0ULL) >> (sbi->alloc_blksz_shift - 9)) || + (last_fs_page > (pgoff_t)(~0ULL))) { printk(KERN_ERR "hfs: filesystem size too large.\n"); goto out_free_vhdr; } From dfacf1387ceb6d7d6df614b18016fd1f347a1996 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:39 +0000 Subject: [PATCH 265/326] bnx2x: fix BRB thresholds for dropless_fc mode Fix the thresholds according to 5778x HW and increase rx_ring size to suit new thresholds in dropless_fc mode. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x.h | 84 ++++++++++++++++++++++++++++++---- drivers/net/bnx2x/bnx2x_cmn.c | 10 ++-- drivers/net/bnx2x/bnx2x_main.c | 33 +++++++++---- 3 files changed, 102 insertions(+), 25 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h index c423504a755f..85297326506c 100644 --- a/drivers/net/bnx2x/bnx2x.h +++ b/drivers/net/bnx2x/bnx2x.h @@ -315,6 +315,14 @@ union db_prod { u32 raw; }; +/* dropless fc FW/HW related params */ +#define BRB_SIZE(bp) (CHIP_IS_E3(bp) ? 1024 : 512) +#define MAX_AGG_QS(bp) (CHIP_IS_E1(bp) ? \ + ETH_MAX_AGGREGATION_QUEUES_E1 :\ + ETH_MAX_AGGREGATION_QUEUES_E1H_E2) +#define FW_DROP_LEVEL(bp) (3 + MAX_SPQ_PENDING + MAX_AGG_QS(bp)) +#define FW_PREFETCH_CNT 16 +#define DROPLESS_FC_HEADROOM 100 /* MC hsi */ #define BCM_PAGE_SHIFT 12 @@ -331,15 +339,35 @@ union db_prod { /* SGE ring related macros */ #define NUM_RX_SGE_PAGES 2 #define RX_SGE_CNT (BCM_PAGE_SIZE / sizeof(struct eth_rx_sge)) -#define MAX_RX_SGE_CNT (RX_SGE_CNT - 2) +#define NEXT_PAGE_SGE_DESC_CNT 2 +#define MAX_RX_SGE_CNT (RX_SGE_CNT - NEXT_PAGE_SGE_DESC_CNT) /* RX_SGE_CNT is promised to be a power of 2 */ #define RX_SGE_MASK (RX_SGE_CNT - 1) #define NUM_RX_SGE (RX_SGE_CNT * NUM_RX_SGE_PAGES) #define MAX_RX_SGE (NUM_RX_SGE - 1) #define NEXT_SGE_IDX(x) ((((x) & RX_SGE_MASK) == \ - (MAX_RX_SGE_CNT - 1)) ? (x) + 3 : (x) + 1) + (MAX_RX_SGE_CNT - 1)) ? \ + (x) + 1 + NEXT_PAGE_SGE_DESC_CNT : \ + (x) + 1) #define RX_SGE(x) ((x) & MAX_RX_SGE) +/* + * Number of required SGEs is the sum of two: + * 1. Number of possible opened aggregations (next packet for + * these aggregations will probably consume SGE immidiatelly) + * 2. Rest of BRB blocks divided by 2 (block will consume new SGE only + * after placement on BD for new TPA aggregation) + * + * Takes into account NEXT_PAGE_SGE_DESC_CNT "next" elements on each page + */ +#define NUM_SGE_REQ (MAX_AGG_QS(bp) + \ + (BRB_SIZE(bp) - MAX_AGG_QS(bp)) / 2) +#define NUM_SGE_PG_REQ ((NUM_SGE_REQ + MAX_RX_SGE_CNT - 1) / \ + MAX_RX_SGE_CNT) +#define SGE_TH_LO(bp) (NUM_SGE_REQ + \ + NUM_SGE_PG_REQ * NEXT_PAGE_SGE_DESC_CNT) +#define SGE_TH_HI(bp) (SGE_TH_LO(bp) + DROPLESS_FC_HEADROOM) + /* Manipulate a bit vector defined as an array of u64 */ /* Number of bits in one sge_mask array element */ @@ -551,24 +579,43 @@ struct bnx2x_fastpath { #define NUM_TX_RINGS 16 #define TX_DESC_CNT (BCM_PAGE_SIZE / sizeof(union eth_tx_bd_types)) -#define MAX_TX_DESC_CNT (TX_DESC_CNT - 1) +#define NEXT_PAGE_TX_DESC_CNT 1 +#define MAX_TX_DESC_CNT (TX_DESC_CNT - NEXT_PAGE_TX_DESC_CNT) #define NUM_TX_BD (TX_DESC_CNT * NUM_TX_RINGS) #define MAX_TX_BD (NUM_TX_BD - 1) #define MAX_TX_AVAIL (MAX_TX_DESC_CNT * NUM_TX_RINGS - 2) #define NEXT_TX_IDX(x) ((((x) & MAX_TX_DESC_CNT) == \ - (MAX_TX_DESC_CNT - 1)) ? (x) + 2 : (x) + 1) + (MAX_TX_DESC_CNT - 1)) ? \ + (x) + 1 + NEXT_PAGE_TX_DESC_CNT : \ + (x) + 1) #define TX_BD(x) ((x) & MAX_TX_BD) #define TX_BD_POFF(x) ((x) & MAX_TX_DESC_CNT) /* The RX BD ring is special, each bd is 8 bytes but the last one is 16 */ #define NUM_RX_RINGS 8 #define RX_DESC_CNT (BCM_PAGE_SIZE / sizeof(struct eth_rx_bd)) -#define MAX_RX_DESC_CNT (RX_DESC_CNT - 2) +#define NEXT_PAGE_RX_DESC_CNT 2 +#define MAX_RX_DESC_CNT (RX_DESC_CNT - NEXT_PAGE_RX_DESC_CNT) #define RX_DESC_MASK (RX_DESC_CNT - 1) #define NUM_RX_BD (RX_DESC_CNT * NUM_RX_RINGS) #define MAX_RX_BD (NUM_RX_BD - 1) #define MAX_RX_AVAIL (MAX_RX_DESC_CNT * NUM_RX_RINGS - 2) -#define MIN_RX_AVAIL 128 + +/* dropless fc calculations for BDs + * + * Number of BDs should as number of buffers in BRB: + * Low threshold takes into account NEXT_PAGE_RX_DESC_CNT + * "next" elements on each page + */ +#define NUM_BD_REQ BRB_SIZE(bp) +#define NUM_BD_PG_REQ ((NUM_BD_REQ + MAX_RX_DESC_CNT - 1) / \ + MAX_RX_DESC_CNT) +#define BD_TH_LO(bp) (NUM_BD_REQ + \ + NUM_BD_PG_REQ * NEXT_PAGE_RX_DESC_CNT + \ + FW_DROP_LEVEL(bp)) +#define BD_TH_HI(bp) (BD_TH_LO(bp) + DROPLESS_FC_HEADROOM) + +#define MIN_RX_AVAIL ((bp)->dropless_fc ? BD_TH_HI(bp) + 128 : 128) #define MIN_RX_SIZE_TPA_HW (CHIP_IS_E1(bp) ? \ ETH_MIN_RX_CQES_WITH_TPA_E1 : \ @@ -579,7 +626,9 @@ struct bnx2x_fastpath { MIN_RX_AVAIL)) #define NEXT_RX_IDX(x) ((((x) & RX_DESC_MASK) == \ - (MAX_RX_DESC_CNT - 1)) ? (x) + 3 : (x) + 1) + (MAX_RX_DESC_CNT - 1)) ? \ + (x) + 1 + NEXT_PAGE_RX_DESC_CNT : \ + (x) + 1) #define RX_BD(x) ((x) & MAX_RX_BD) /* @@ -589,14 +638,31 @@ struct bnx2x_fastpath { #define CQE_BD_REL (sizeof(union eth_rx_cqe) / sizeof(struct eth_rx_bd)) #define NUM_RCQ_RINGS (NUM_RX_RINGS * CQE_BD_REL) #define RCQ_DESC_CNT (BCM_PAGE_SIZE / sizeof(union eth_rx_cqe)) -#define MAX_RCQ_DESC_CNT (RCQ_DESC_CNT - 1) +#define NEXT_PAGE_RCQ_DESC_CNT 1 +#define MAX_RCQ_DESC_CNT (RCQ_DESC_CNT - NEXT_PAGE_RCQ_DESC_CNT) #define NUM_RCQ_BD (RCQ_DESC_CNT * NUM_RCQ_RINGS) #define MAX_RCQ_BD (NUM_RCQ_BD - 1) #define MAX_RCQ_AVAIL (MAX_RCQ_DESC_CNT * NUM_RCQ_RINGS - 2) #define NEXT_RCQ_IDX(x) ((((x) & MAX_RCQ_DESC_CNT) == \ - (MAX_RCQ_DESC_CNT - 1)) ? (x) + 2 : (x) + 1) + (MAX_RCQ_DESC_CNT - 1)) ? \ + (x) + 1 + NEXT_PAGE_RCQ_DESC_CNT : \ + (x) + 1) #define RCQ_BD(x) ((x) & MAX_RCQ_BD) +/* dropless fc calculations for RCQs + * + * Number of RCQs should be as number of buffers in BRB: + * Low threshold takes into account NEXT_PAGE_RCQ_DESC_CNT + * "next" elements on each page + */ +#define NUM_RCQ_REQ BRB_SIZE(bp) +#define NUM_RCQ_PG_REQ ((NUM_BD_REQ + MAX_RCQ_DESC_CNT - 1) / \ + MAX_RCQ_DESC_CNT) +#define RCQ_TH_LO(bp) (NUM_RCQ_REQ + \ + NUM_RCQ_PG_REQ * NEXT_PAGE_RCQ_DESC_CNT + \ + FW_DROP_LEVEL(bp)) +#define RCQ_TH_HI(bp) (RCQ_TH_LO(bp) + DROPLESS_FC_HEADROOM) + /* This is needed for determining of last_max */ #define SUB_S16(a, b) (s16)((s16)(a) - (s16)(b)) diff --git a/drivers/net/bnx2x/bnx2x_cmn.c b/drivers/net/bnx2x/bnx2x_cmn.c index 37e5790681ad..2a33d2433c31 100644 --- a/drivers/net/bnx2x/bnx2x_cmn.c +++ b/drivers/net/bnx2x/bnx2x_cmn.c @@ -987,8 +987,6 @@ void __bnx2x_link_report(struct bnx2x *bp) void bnx2x_init_rx_rings(struct bnx2x *bp) { int func = BP_FUNC(bp); - int max_agg_queues = CHIP_IS_E1(bp) ? ETH_MAX_AGGREGATION_QUEUES_E1 : - ETH_MAX_AGGREGATION_QUEUES_E1H_E2; u16 ring_prod; int i, j; @@ -1001,7 +999,7 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) if (!fp->disable_tpa) { /* Fill the per-aggregtion pool */ - for (i = 0; i < max_agg_queues; i++) { + for (i = 0; i < MAX_AGG_QS(bp); i++) { struct bnx2x_agg_info *tpa_info = &fp->tpa_info[i]; struct sw_rx_bd *first_buf = @@ -1041,7 +1039,7 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) bnx2x_free_rx_sge_range(bp, fp, ring_prod); bnx2x_free_tpa_pool(bp, fp, - max_agg_queues); + MAX_AGG_QS(bp)); fp->disable_tpa = 1; ring_prod = 0; break; @@ -1137,9 +1135,7 @@ static void bnx2x_free_rx_skbs(struct bnx2x *bp) bnx2x_free_rx_bds(fp); if (!fp->disable_tpa) - bnx2x_free_tpa_pool(bp, fp, CHIP_IS_E1(bp) ? - ETH_MAX_AGGREGATION_QUEUES_E1 : - ETH_MAX_AGGREGATION_QUEUES_E1H_E2); + bnx2x_free_tpa_pool(bp, fp, MAX_AGG_QS(bp)); } } diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index f74582a22c68..3f93e8666104 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -2756,8 +2756,14 @@ static void bnx2x_pf_rx_q_prep(struct bnx2x *bp, u16 tpa_agg_size = 0; if (!fp->disable_tpa) { - pause->sge_th_hi = 250; - pause->sge_th_lo = 150; + pause->sge_th_lo = SGE_TH_LO(bp); + pause->sge_th_hi = SGE_TH_HI(bp); + + /* validate SGE ring has enough to cross high threshold */ + WARN_ON(bp->dropless_fc && + pause->sge_th_hi + FW_PREFETCH_CNT > + MAX_RX_SGE_CNT * NUM_RX_SGE_PAGES); + tpa_agg_size = min_t(u32, (min_t(u32, 8, MAX_SKB_FRAGS) * SGE_PAGE_SIZE * PAGES_PER_SGE), 0xffff); @@ -2771,10 +2777,21 @@ static void bnx2x_pf_rx_q_prep(struct bnx2x *bp, /* pause - not for e1 */ if (!CHIP_IS_E1(bp)) { - pause->bd_th_hi = 350; - pause->bd_th_lo = 250; - pause->rcq_th_hi = 350; - pause->rcq_th_lo = 250; + pause->bd_th_lo = BD_TH_LO(bp); + pause->bd_th_hi = BD_TH_HI(bp); + + pause->rcq_th_lo = RCQ_TH_LO(bp); + pause->rcq_th_hi = RCQ_TH_HI(bp); + /* + * validate that rings have enough entries to cross + * high thresholds + */ + WARN_ON(bp->dropless_fc && + pause->bd_th_hi + FW_PREFETCH_CNT > + bp->rx_ring_size); + WARN_ON(bp->dropless_fc && + pause->rcq_th_hi + FW_PREFETCH_CNT > + NUM_RCQ_RINGS * MAX_RCQ_DESC_CNT); pause->pri_map = 1; } @@ -2802,9 +2819,7 @@ static void bnx2x_pf_rx_q_prep(struct bnx2x *bp, * For PF Clients it should be the maximum avaliable number. * VF driver(s) may want to define it to a smaller value. */ - rxq_init->max_tpa_queues = - (CHIP_IS_E1(bp) ? ETH_MAX_AGGREGATION_QUEUES_E1 : - ETH_MAX_AGGREGATION_QUEUES_E1H_E2); + rxq_init->max_tpa_queues = MAX_AGG_QS(bp); rxq_init->cache_line_log = BNX2X_RX_ALIGN_SHIFT; rxq_init->fw_sb_id = fp->fw_sb_id; From 5f837363457a2280530373267f86092625d15a4d Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:40 +0000 Subject: [PATCH 266/326] bnx2x: decrease print level to debug It may happen every link toggle. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_stats.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/bnx2x/bnx2x_stats.c b/drivers/net/bnx2x/bnx2x_stats.c index 771f6803b238..f5d9b4213cad 100644 --- a/drivers/net/bnx2x/bnx2x_stats.c +++ b/drivers/net/bnx2x/bnx2x_stats.c @@ -710,7 +710,8 @@ static int bnx2x_hw_stats_update(struct bnx2x *bp) break; case MAC_TYPE_NONE: /* unreached */ - BNX2X_ERR("stats updated by DMAE but no MAC active\n"); + DP(BNX2X_MSG_STATS, + "stats updated by DMAE but no MAC active\n"); return -1; default: /* unreached */ From c2188952fc7d2ca54bb8aca1bc502618a7488baf Mon Sep 17 00:00:00 2001 From: Vladislav Zolotarov Date: Tue, 30 Aug 2011 00:08:41 +0000 Subject: [PATCH 267/326] bnx2x: fix rx ring size report Store the size in bp, read from bp when queried. Signed-off-by: Dmitry Kravkov Signed-off-by: Vladislav Zolotarov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_cmn.c | 17 +++++++++++------ drivers/net/bnx2x/bnx2x_ethtool.c | 5 +---- 2 files changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_cmn.c b/drivers/net/bnx2x/bnx2x_cmn.c index 2a33d2433c31..c4cbf9736414 100644 --- a/drivers/net/bnx2x/bnx2x_cmn.c +++ b/drivers/net/bnx2x/bnx2x_cmn.c @@ -3091,15 +3091,20 @@ static int bnx2x_alloc_fp_mem_at(struct bnx2x *bp, int index) struct bnx2x_fastpath *fp = &bp->fp[index]; int ring_size = 0; u8 cos; + int rx_ring_size = 0; /* if rx_ring_size specified - use it */ - int rx_ring_size = bp->rx_ring_size ? bp->rx_ring_size : - MAX_RX_AVAIL/BNX2X_NUM_RX_QUEUES(bp); + if (!bp->rx_ring_size) { - /* allocate at least number of buffers required by FW */ - rx_ring_size = max_t(int, bp->disable_tpa ? MIN_RX_SIZE_NONTPA : - MIN_RX_SIZE_TPA, - rx_ring_size); + rx_ring_size = MAX_RX_AVAIL/BNX2X_NUM_RX_QUEUES(bp); + + /* allocate at least number of buffers required by FW */ + rx_ring_size = max_t(int, bp->disable_tpa ? MIN_RX_SIZE_NONTPA : + MIN_RX_SIZE_TPA, rx_ring_size); + + bp->rx_ring_size = rx_ring_size; + } else + rx_ring_size = bp->rx_ring_size; /* Common */ sb = &bnx2x_fp(bp, index, status_blk); diff --git a/drivers/net/bnx2x/bnx2x_ethtool.c b/drivers/net/bnx2x/bnx2x_ethtool.c index 221863059dae..0ceb6c7b1238 100644 --- a/drivers/net/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/bnx2x/bnx2x_ethtool.c @@ -1310,10 +1310,7 @@ static void bnx2x_get_ringparam(struct net_device *dev, if (bp->rx_ring_size) ering->rx_pending = bp->rx_ring_size; else - if (bp->state == BNX2X_STATE_OPEN && bp->num_queues) - ering->rx_pending = MAX_RX_AVAIL/bp->num_queues; - else - ering->rx_pending = MAX_RX_AVAIL; + ering->rx_pending = MAX_RX_AVAIL; ering->rx_mini_pending = 0; ering->rx_jumbo_pending = 0; From 3395a033a7c2f1a089fae7e89bf108764b59529c Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:42 +0000 Subject: [PATCH 268/326] bnx2x: fix MF for 4-port devices Number of VNs for 4-port devices is 2 instead of 4 Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x.h | 15 ++++++------ drivers/net/bnx2x/bnx2x_main.c | 43 ++++++++++++++++++++------------- drivers/net/bnx2x/bnx2x_stats.c | 4 +-- 3 files changed, 36 insertions(+), 26 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h index 85297326506c..2621a1c56358 100644 --- a/drivers/net/bnx2x/bnx2x.h +++ b/drivers/net/bnx2x/bnx2x.h @@ -1166,11 +1166,12 @@ struct bnx2x { #define BP_PORT(bp) (bp->pfid & 1) #define BP_FUNC(bp) (bp->pfid) #define BP_ABS_FUNC(bp) (bp->pf_num) -#define BP_E1HVN(bp) (bp->pfid >> 1) -#define BP_VN(bp) (BP_E1HVN(bp)) /*remove when approved*/ -#define BP_L_ID(bp) (BP_E1HVN(bp) << 2) -#define BP_FW_MB_IDX(bp) (BP_PORT(bp) +\ - BP_VN(bp) * ((CHIP_IS_E1x(bp) || (CHIP_MODE_IS_4_PORT(bp))) ? 2 : 1)) +#define BP_VN(bp) ((bp)->pfid >> 1) +#define BP_MAX_VN_NUM(bp) (CHIP_MODE_IS_4_PORT(bp) ? 2 : 4) +#define BP_L_ID(bp) (BP_VN(bp) << 2) +#define BP_FW_MB_IDX_VN(bp, vn) (BP_PORT(bp) +\ + (vn) * ((CHIP_IS_E1x(bp) || (CHIP_MODE_IS_4_PORT(bp))) ? 2 : 1)) +#define BP_FW_MB_IDX(bp) BP_FW_MB_IDX_VN(bp, BP_VN(bp)) struct net_device *dev; struct pci_dev *pdev; @@ -1833,7 +1834,7 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms, #define MAX_DMAE_C_PER_PORT 8 #define INIT_DMAE_C(bp) (BP_PORT(bp) * MAX_DMAE_C_PER_PORT + \ - BP_E1HVN(bp)) + BP_VN(bp)) #define PMF_DMAE_C(bp) (BP_PORT(bp) * MAX_DMAE_C_PER_PORT + \ E1HVN_MAX) @@ -1859,7 +1860,7 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms, /* must be used on a CID before placing it on a HW ring */ #define HW_CID(bp, x) ((BP_PORT(bp) << 23) | \ - (BP_E1HVN(bp) << BNX2X_SWCID_SHIFT) | \ + (BP_VN(bp) << BNX2X_SWCID_SHIFT) | \ (x)) #define SP_DESC_CNT (BCM_PAGE_SIZE / sizeof(struct eth_spe)) diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index 3f93e8666104..9633e9b6853c 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -407,8 +407,8 @@ u32 bnx2x_dmae_opcode(struct bnx2x *bp, u8 src_type, u8 dst_type, opcode |= (DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET); opcode |= (BP_PORT(bp) ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0); - opcode |= ((BP_E1HVN(bp) << DMAE_CMD_E1HVN_SHIFT) | - (BP_E1HVN(bp) << DMAE_COMMAND_DST_VN_SHIFT)); + opcode |= ((BP_VN(bp) << DMAE_CMD_E1HVN_SHIFT) | + (BP_VN(bp) << DMAE_COMMAND_DST_VN_SHIFT)); opcode |= (DMAE_COM_SET_ERR << DMAE_COMMAND_ERR_POLICY_SHIFT); #ifdef __BIG_ENDIAN @@ -1419,7 +1419,7 @@ static void bnx2x_hc_int_enable(struct bnx2x *bp) if (!CHIP_IS_E1(bp)) { /* init leading/trailing edge */ if (IS_MF(bp)) { - val = (0xee0f | (1 << (BP_E1HVN(bp) + 4))); + val = (0xee0f | (1 << (BP_VN(bp) + 4))); if (bp->port.pmf) /* enable nig and gpio3 attention */ val |= 0x1100; @@ -1471,7 +1471,7 @@ static void bnx2x_igu_int_enable(struct bnx2x *bp) /* init leading/trailing edge */ if (IS_MF(bp)) { - val = (0xee0f | (1 << (BP_E1HVN(bp) + 4))); + val = (0xee0f | (1 << (BP_VN(bp) + 4))); if (bp->port.pmf) /* enable nig and gpio3 attention */ val |= 0x1100; @@ -2287,7 +2287,7 @@ static void bnx2x_calc_vn_weight_sum(struct bnx2x *bp) int vn; bp->vn_weight_sum = 0; - for (vn = VN_0; vn < E1HVN_MAX; vn++) { + for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) { u32 vn_cfg = bp->mf_config[vn]; u32 vn_min_rate = ((vn_cfg & FUNC_MF_CFG_MIN_BW_MASK) >> FUNC_MF_CFG_MIN_BW_SHIFT) * 100; @@ -2320,12 +2320,18 @@ static void bnx2x_calc_vn_weight_sum(struct bnx2x *bp) CMNG_FLAGS_PER_PORT_FAIRNESS_VN; } +/* returns func by VN for current port */ +static inline int func_by_vn(struct bnx2x *bp, int vn) +{ + return 2 * vn + BP_PORT(bp); +} + static void bnx2x_init_vn_minmax(struct bnx2x *bp, int vn) { struct rate_shaping_vars_per_vn m_rs_vn; struct fairness_vars_per_vn m_fair_vn; u32 vn_cfg = bp->mf_config[vn]; - int func = 2*vn + BP_PORT(bp); + int func = func_by_vn(bp, vn); u16 vn_min_rate, vn_max_rate; int i; @@ -2422,7 +2428,7 @@ void bnx2x_read_mf_cfg(struct bnx2x *bp) * * and there are 2 functions per port */ - for (vn = VN_0; vn < E1HVN_MAX; vn++) { + for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) { int /*abs*/func = n * (2 * vn + BP_PORT(bp)) + BP_PATH(bp); if (func >= E1H_FUNC_MAX) @@ -2454,7 +2460,7 @@ static void bnx2x_cmng_fns_init(struct bnx2x *bp, u8 read_cfg, u8 cmng_type) /* calculate and set min-max rate for each vn */ if (bp->port.pmf) - for (vn = VN_0; vn < E1HVN_MAX; vn++) + for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) bnx2x_init_vn_minmax(bp, vn); /* always enable rate shaping and fairness */ @@ -2473,16 +2479,15 @@ static void bnx2x_cmng_fns_init(struct bnx2x *bp, u8 read_cfg, u8 cmng_type) static inline void bnx2x_link_sync_notify(struct bnx2x *bp) { - int port = BP_PORT(bp); int func; int vn; /* Set the attention towards other drivers on the same port */ - for (vn = VN_0; vn < E1HVN_MAX; vn++) { - if (vn == BP_E1HVN(bp)) + for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) { + if (vn == BP_VN(bp)) continue; - func = ((vn << 1) | port); + func = func_by_vn(bp, vn); REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_0 + (LINK_SYNC_ATTENTION_BIT_FUNC_0 + func)*4, 1); } @@ -2577,7 +2582,7 @@ static void bnx2x_pmf_update(struct bnx2x *bp) bnx2x_dcbx_pmf_update(bp); /* enable nig attention */ - val = (0xff0f | (1 << (BP_E1HVN(bp) + 4))); + val = (0xff0f | (1 << (BP_VN(bp) + 4))); if (bp->common.int_block == INT_BLOCK_HC) { REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val); REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val); @@ -6686,12 +6691,16 @@ static int bnx2x_init_hw_func(struct bnx2x *bp) if (CHIP_MODE_IS_4_PORT(bp)) dsb_idx = BP_FUNC(bp); else - dsb_idx = BP_E1HVN(bp); + dsb_idx = BP_VN(bp); prod_offset = (CHIP_INT_MODE_IS_BC(bp) ? IGU_BC_BASE_DSB_PROD + dsb_idx : IGU_NORM_BASE_DSB_PROD + dsb_idx); + /* + * igu prods come in chunks of E1HVN_MAX (4) - + * does not matters what is the current chip mode + */ for (i = 0; i < (num_segs * E1HVN_MAX); i += E1HVN_MAX) { addr = IGU_REG_PROD_CONS_MEMORY + @@ -7585,7 +7594,7 @@ u32 bnx2x_send_unload_req(struct bnx2x *bp, int unload_mode) u32 val; /* The mac address is written to entries 1-4 to preserve entry 0 which is used by the PMF */ - u8 entry = (BP_E1HVN(bp) + 1)*8; + u8 entry = (BP_VN(bp) + 1)*8; val = (mac_addr[0] << 8) | mac_addr[1]; EMAC_WR(bp, EMAC_REG_EMAC_MAC_MATCH + entry, val); @@ -8792,13 +8801,13 @@ static void __devinit bnx2x_get_common_hwinfo(struct bnx2x *bp) static void __devinit bnx2x_get_igu_cam_info(struct bnx2x *bp) { int pfid = BP_FUNC(bp); - int vn = BP_E1HVN(bp); int igu_sb_id; u32 val; u8 fid, igu_sb_cnt = 0; bp->igu_base_sb = 0xff; if (CHIP_INT_MODE_IS_BC(bp)) { + int vn = BP_VN(bp); igu_sb_cnt = bp->igu_sb_cnt; bp->igu_base_sb = (CHIP_MODE_IS_4_PORT(bp) ? pfid : vn) * FP_SB_MAX_E1x; @@ -9488,7 +9497,7 @@ static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp) bp->mf_ov = 0; bp->mf_mode = 0; - vn = BP_E1HVN(bp); + vn = BP_VN(bp); if (!CHIP_IS_E1(bp) && !BP_NOMCP(bp)) { BNX2X_DEV_INFO("shmem2base 0x%x, size %d, mfcfg offset %d\n", diff --git a/drivers/net/bnx2x/bnx2x_stats.c b/drivers/net/bnx2x/bnx2x_stats.c index f5d9b4213cad..9908f2bbcf73 100644 --- a/drivers/net/bnx2x/bnx2x_stats.c +++ b/drivers/net/bnx2x/bnx2x_stats.c @@ -1392,7 +1392,7 @@ static void bnx2x_port_stats_base_init(struct bnx2x *bp) static void bnx2x_func_stats_base_init(struct bnx2x *bp) { - int vn, vn_max = IS_MF(bp) ? E1HVN_MAX : E1VN_MAX; + int vn, vn_max = IS_MF(bp) ? BP_MAX_VN_NUM(bp) : E1VN_MAX; u32 func_stx; /* sanity */ @@ -1405,7 +1405,7 @@ static void bnx2x_func_stats_base_init(struct bnx2x *bp) func_stx = bp->func_stx; for (vn = VN_0; vn < vn_max; vn++) { - int mb_idx = CHIP_IS_E1x(bp) ? 2*vn + BP_PORT(bp) : vn; + int mb_idx = BP_FW_MB_IDX_VN(bp, vn); bp->func_stx = SHMEM_RD(bp, func_mb[mb_idx].fw_mb_param); bnx2x_func_stats_init(bp); From 7a06a122322c89544774e789a11aa671423e9362 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:43 +0000 Subject: [PATCH 269/326] bnx2x: don't reset device while reading its configuration. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_main.c | 24 +++++++++++++++--------- drivers/net/bnx2x/bnx2x_reg.h | 2 +- 2 files changed, 16 insertions(+), 10 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index 9633e9b6853c..00dc8f0fc3af 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -5822,7 +5822,7 @@ static int bnx2x_init_hw_common(struct bnx2x *bp) * take the UNDI lock to protect undi_unload flow from accessing * registers while we're resetting the chip */ - bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_UNDI); + bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_RESET); bnx2x_reset_common(bp); REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0xffffffff); @@ -5834,7 +5834,7 @@ static int bnx2x_init_hw_common(struct bnx2x *bp) } REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET, val); - bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_UNDI); + bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_RESET); bnx2x_init_block(bp, BLOCK_MISC, PHASE_COMMON); @@ -8570,10 +8570,12 @@ static void __devinit bnx2x_undi_unload(struct bnx2x *bp) /* Check if there is any driver already loaded */ val = REG_RD(bp, MISC_REG_UNPREPARED); if (val == 0x1) { - /* Check if it is the UNDI driver + + bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_RESET); + /* + * Check if it is the UNDI driver * UNDI driver initializes CID offset for normal bell to 0x7 */ - bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_UNDI); val = REG_RD(bp, DORQ_REG_NORM_CID_OFST); if (val == 0x7) { u32 reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS; @@ -8611,9 +8613,6 @@ static void __devinit bnx2x_undi_unload(struct bnx2x *bp) bnx2x_fw_command(bp, reset_code, 0); } - /* now it's safe to release the lock */ - bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_UNDI); - bnx2x_undi_int_disable(bp); port = BP_PORT(bp); @@ -8663,8 +8662,10 @@ static void __devinit bnx2x_undi_unload(struct bnx2x *bp) bp->fw_seq = (SHMEM_RD(bp, func_mb[bp->pf_num].drv_mb_header) & DRV_MSG_SEQ_NUMBER_MASK); - } else - bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_UNDI); + } + + /* now it's safe to release the lock */ + bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_RESET); } } @@ -9440,6 +9441,10 @@ static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp) bp->igu_base_sb = 0; } else { bp->common.int_block = INT_BLOCK_IGU; + + /* do not allow device reset during IGU info preocessing */ + bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_RESET); + val = REG_RD(bp, IGU_REG_BLOCK_CONFIGURATION); if (val & IGU_BLOCK_CONFIGURATION_REG_BACKWARD_COMP_EN) { @@ -9471,6 +9476,7 @@ static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp) bnx2x_get_igu_cam_info(bp); + bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_RESET); } /* diff --git a/drivers/net/bnx2x/bnx2x_reg.h b/drivers/net/bnx2x/bnx2x_reg.h index 40266c14e6dc..dac217d478f2 100644 --- a/drivers/net/bnx2x/bnx2x_reg.h +++ b/drivers/net/bnx2x/bnx2x_reg.h @@ -5766,7 +5766,7 @@ #define HW_LOCK_RESOURCE_RECOVERY_LEADER_0 8 #define HW_LOCK_RESOURCE_RECOVERY_LEADER_1 9 #define HW_LOCK_RESOURCE_SPIO 2 -#define HW_LOCK_RESOURCE_UNDI 5 +#define HW_LOCK_RESOURCE_RESET 5 #define AEU_INPUTS_ATTN_BITS_ATC_HW_INTERRUPT (0x1<<4) #define AEU_INPUTS_ATTN_BITS_ATC_PARITY_ERROR (0x1<<5) #define AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR (0x1<<18) From 0735f2fc8c49f1fbbbb245d038582922984ed3d5 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:44 +0000 Subject: [PATCH 270/326] bnx2x: init fw_seq after undi_unload is done Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_main.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index 00dc8f0fc3af..94382a78c951 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -9623,13 +9623,6 @@ static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp) /* port info */ bnx2x_get_port_hwinfo(bp); - if (!BP_NOMCP(bp)) { - bp->fw_seq = - (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) & - DRV_MSG_SEQ_NUMBER_MASK); - BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq); - } - /* Get MAC addresses */ bnx2x_get_mac_hwinfo(bp); @@ -9795,6 +9788,14 @@ static int __devinit bnx2x_init_bp(struct bnx2x *bp) if (!BP_NOMCP(bp)) bnx2x_undi_unload(bp); + /* init fw_seq after undi_unload! */ + if (!BP_NOMCP(bp)) { + bp->fw_seq = + (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) & + DRV_MSG_SEQ_NUMBER_MASK); + BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq); + } + if (CHIP_REV_IS_FPGA(bp)) dev_err(&bp->pdev->dev, "FPGA detected\n"); From a5c53dbcde9a156e8303acc6ecb2296bf609fe38 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Tue, 30 Aug 2011 00:08:45 +0000 Subject: [PATCH 271/326] bnx2x: don't access removed registers on 57712 and above Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_main.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index 94382a78c951..0b68d02fe455 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -10290,17 +10290,21 @@ static int __devinit bnx2x_init_dev(struct pci_dev *pdev, /* clean indirect addresses */ pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS, PCICFG_VENDOR_ID_OFFSET); - /* Clean the following indirect addresses for all functions since it + /* + * Clean the following indirect addresses for all functions since it * is not used by the driver. */ REG_WR(bp, PXP2_REG_PGL_ADDR_88_F0, 0); REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F0, 0); REG_WR(bp, PXP2_REG_PGL_ADDR_90_F0, 0); REG_WR(bp, PXP2_REG_PGL_ADDR_94_F0, 0); - REG_WR(bp, PXP2_REG_PGL_ADDR_88_F1, 0); - REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F1, 0); - REG_WR(bp, PXP2_REG_PGL_ADDR_90_F1, 0); - REG_WR(bp, PXP2_REG_PGL_ADDR_94_F1, 0); + + if (CHIP_IS_E1x(bp)) { + REG_WR(bp, PXP2_REG_PGL_ADDR_88_F1, 0); + REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F1, 0); + REG_WR(bp, PXP2_REG_PGL_ADDR_90_F1, 0); + REG_WR(bp, PXP2_REG_PGL_ADDR_94_F1, 0); + } /* * Enable internal target-read (in case we are probed after PF FLR). From 150966ad56291776a1f3fed86000a027e0794922 Mon Sep 17 00:00:00 2001 From: Ariel Elior Date: Tue, 30 Aug 2011 00:08:46 +0000 Subject: [PATCH 272/326] bnx2x: Fix for a host coalescing bug which impared latency. Seperated Rx and Tx coalescing to different state machines. Signed-off-by: Ariel Elior Signed-off-by: Eilon Greenstein Signed-off-by: Dmitry Kravkov Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x.h | 23 ++++++++--------------- drivers/net/bnx2x/bnx2x_main.c | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 41 insertions(+), 15 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h index 2621a1c56358..e46df5331c55 100644 --- a/drivers/net/bnx2x/bnx2x.h +++ b/drivers/net/bnx2x/bnx2x.h @@ -751,25 +751,18 @@ struct bnx2x_fastpath { #define FP_CSB_FUNC_OFF \ offsetof(struct cstorm_status_block_c, func) -#define HC_INDEX_TOE_RX_CQ_CONS 0 /* Formerly Ustorm TOE CQ index */ - /* (HC_INDEX_U_TOE_RX_CQ_CONS) */ -#define HC_INDEX_ETH_RX_CQ_CONS 1 /* Formerly Ustorm ETH CQ index */ - /* (HC_INDEX_U_ETH_RX_CQ_CONS) */ -#define HC_INDEX_ETH_RX_BD_CONS 2 /* Formerly Ustorm ETH BD index */ - /* (HC_INDEX_U_ETH_RX_BD_CONS) */ +#define HC_INDEX_ETH_RX_CQ_CONS 1 -#define HC_INDEX_TOE_TX_CQ_CONS 4 /* Formerly Cstorm TOE CQ index */ - /* (HC_INDEX_C_TOE_TX_CQ_CONS) */ -#define HC_INDEX_ETH_TX_CQ_CONS_COS0 5 /* Formerly Cstorm ETH CQ index */ - /* (HC_INDEX_C_ETH_TX_CQ_CONS) */ -#define HC_INDEX_ETH_TX_CQ_CONS_COS1 6 /* Formerly Cstorm ETH CQ index */ - /* (HC_INDEX_C_ETH_TX_CQ_CONS) */ -#define HC_INDEX_ETH_TX_CQ_CONS_COS2 7 /* Formerly Cstorm ETH CQ index */ - /* (HC_INDEX_C_ETH_TX_CQ_CONS) */ +#define HC_INDEX_OOO_TX_CQ_CONS 4 + +#define HC_INDEX_ETH_TX_CQ_CONS_COS0 5 + +#define HC_INDEX_ETH_TX_CQ_CONS_COS1 6 + +#define HC_INDEX_ETH_TX_CQ_CONS_COS2 7 #define HC_INDEX_ETH_FIRST_TX_CQ_CONS HC_INDEX_ETH_TX_CQ_CONS_COS0 - #define BNX2X_RX_SB_INDEX \ (&fp->sb_index_values[HC_INDEX_ETH_RX_CQ_CONS]) diff --git a/drivers/net/bnx2x/bnx2x_main.c b/drivers/net/bnx2x/bnx2x_main.c index 0b68d02fe455..c027e9341a1a 100644 --- a/drivers/net/bnx2x/bnx2x_main.c +++ b/drivers/net/bnx2x/bnx2x_main.c @@ -4828,6 +4828,37 @@ void bnx2x_setup_ndsb_state_machine(struct hc_status_block_sm *hc_sm, hc_sm->time_to_expire = 0xFFFFFFFF; } + +/* allocates state machine ids. */ +static inline +void bnx2x_map_sb_state_machines(struct hc_index_data *index_data) +{ + /* zero out state machine indices */ + /* rx indices */ + index_data[HC_INDEX_ETH_RX_CQ_CONS].flags &= ~HC_INDEX_DATA_SM_ID; + + /* tx indices */ + index_data[HC_INDEX_OOO_TX_CQ_CONS].flags &= ~HC_INDEX_DATA_SM_ID; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS0].flags &= ~HC_INDEX_DATA_SM_ID; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS1].flags &= ~HC_INDEX_DATA_SM_ID; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS2].flags &= ~HC_INDEX_DATA_SM_ID; + + /* map indices */ + /* rx indices */ + index_data[HC_INDEX_ETH_RX_CQ_CONS].flags |= + SM_RX_ID << HC_INDEX_DATA_SM_ID_SHIFT; + + /* tx indices */ + index_data[HC_INDEX_OOO_TX_CQ_CONS].flags |= + SM_TX_ID << HC_INDEX_DATA_SM_ID_SHIFT; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS0].flags |= + SM_TX_ID << HC_INDEX_DATA_SM_ID_SHIFT; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS1].flags |= + SM_TX_ID << HC_INDEX_DATA_SM_ID_SHIFT; + index_data[HC_INDEX_ETH_TX_CQ_CONS_COS2].flags |= + SM_TX_ID << HC_INDEX_DATA_SM_ID_SHIFT; +} + static void bnx2x_init_sb(struct bnx2x *bp, dma_addr_t mapping, int vfid, u8 vf_valid, int fw_sb_id, int igu_sb_id) { @@ -4859,6 +4890,7 @@ static void bnx2x_init_sb(struct bnx2x *bp, dma_addr_t mapping, int vfid, hc_sm_p = sb_data_e2.common.state_machine; sb_data_p = (u32 *)&sb_data_e2; data_size = sizeof(struct hc_status_block_data_e2)/sizeof(u32); + bnx2x_map_sb_state_machines(sb_data_e2.index_data); } else { memset(&sb_data_e1x, 0, sizeof(struct hc_status_block_data_e1x)); @@ -4873,6 +4905,7 @@ static void bnx2x_init_sb(struct bnx2x *bp, dma_addr_t mapping, int vfid, hc_sm_p = sb_data_e1x.common.state_machine; sb_data_p = (u32 *)&sb_data_e1x; data_size = sizeof(struct hc_status_block_data_e1x)/sizeof(u32); + bnx2x_map_sb_state_machines(sb_data_e1x.index_data); } bnx2x_setup_ndsb_state_machine(&hc_sm_p[SM_RX_ID], From 27e95a8c670e0c587990ec5b9a87a7ea17873d28 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Igor=20Maravi=C4=87?= Date: Tue, 30 Aug 2011 03:12:55 +0000 Subject: [PATCH 273/326] pkt_sched: cls_rsvp.h was outdated MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit File cls_rsvp.h in /net/sched was outdated. I'm sending you patch for this file. [ tb[] array should be indexed by X not X-1 -DaveM ] Signed-off-by: Igor Maravić Signed-off-by: David S. Miller --- net/sched/cls_rsvp.h | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h index be4505ee67a9..b01427924f81 100644 --- a/net/sched/cls_rsvp.h +++ b/net/sched/cls_rsvp.h @@ -425,7 +425,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, struct rsvp_filter *f, **fp; struct rsvp_session *s, **sp; struct tc_rsvp_pinfo *pinfo = NULL; - struct nlattr *opt = tca[TCA_OPTIONS-1]; + struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_RSVP_MAX + 1]; struct tcf_exts e; unsigned int h1, h2; @@ -439,7 +439,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, if (err < 0) return err; - err = tcf_exts_validate(tp, tb, tca[TCA_RATE-1], &e, &rsvp_ext_map); + err = tcf_exts_validate(tp, tb, tca[TCA_RATE], &e, &rsvp_ext_map); if (err < 0) return err; @@ -449,8 +449,8 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, if (f->handle != handle && handle) goto errout2; - if (tb[TCA_RSVP_CLASSID-1]) { - f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID-1]); + if (tb[TCA_RSVP_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); tcf_bind_filter(tp, &f->res, base); } @@ -462,7 +462,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, err = -EINVAL; if (handle) goto errout2; - if (tb[TCA_RSVP_DST-1] == NULL) + if (tb[TCA_RSVP_DST] == NULL) goto errout2; err = -ENOBUFS; @@ -471,19 +471,19 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base, goto errout2; h2 = 16; - if (tb[TCA_RSVP_SRC-1]) { - memcpy(f->src, nla_data(tb[TCA_RSVP_SRC-1]), sizeof(f->src)); + if (tb[TCA_RSVP_SRC]) { + memcpy(f->src, nla_data(tb[TCA_RSVP_SRC]), sizeof(f->src)); h2 = hash_src(f->src); } - if (tb[TCA_RSVP_PINFO-1]) { - pinfo = nla_data(tb[TCA_RSVP_PINFO-1]); + if (tb[TCA_RSVP_PINFO]) { + pinfo = nla_data(tb[TCA_RSVP_PINFO]); f->spi = pinfo->spi; f->tunnelhdr = pinfo->tunnelhdr; } - if (tb[TCA_RSVP_CLASSID-1]) - f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID-1]); + if (tb[TCA_RSVP_CLASSID]) + f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); - dst = nla_data(tb[TCA_RSVP_DST-1]); + dst = nla_data(tb[TCA_RSVP_DST]); h1 = hash_dst(dst, pinfo ? pinfo->protocol : 0, pinfo ? pinfo->tunnelid : 0); err = -ENOMEM; @@ -642,8 +642,7 @@ static int rsvp_dump(struct tcf_proto *tp, unsigned long fh, return -1; } -static struct tcf_proto_ops RSVP_OPS = { - .next = NULL, +static struct tcf_proto_ops RSVP_OPS __read_mostly = { .kind = RSVP_ID, .classify = rsvp_classify, .init = rsvp_init, From 946cedccbd7387488d2cee5da92cdfeb28d2e670 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 30 Aug 2011 03:21:44 +0000 Subject: [PATCH 274/326] tcp: Change possible SYN flooding messages "Possible SYN flooding on port xxxx " messages can fill logs on servers. Change logic to log the message only once per listener, and add two new SNMP counters to track : TCPReqQFullDoCookies : number of times a SYNCOOKIE was replied to client TCPReqQFullDrop : number of times a SYN request was dropped because syncookies were not enabled. Based on a prior patch from Tom Herbert, and suggestions from David. Signed-off-by: Eric Dumazet CC: Tom Herbert Signed-off-by: David S. Miller --- include/linux/snmp.h | 2 ++ include/net/request_sock.h | 3 ++- include/net/tcp.h | 3 +++ net/ipv4/proc.c | 2 ++ net/ipv4/tcp_ipv4.c | 49 ++++++++++++++++++++++---------------- net/ipv6/tcp_ipv6.c | 31 +++--------------------- 6 files changed, 40 insertions(+), 50 deletions(-) diff --git a/include/linux/snmp.h b/include/linux/snmp.h index 12b2b18e50c1..e16557a357e5 100644 --- a/include/linux/snmp.h +++ b/include/linux/snmp.h @@ -231,6 +231,8 @@ enum LINUX_MIB_TCPDEFERACCEPTDROP, LINUX_MIB_IPRPFILTER, /* IP Reverse Path Filter (rp_filter) */ LINUX_MIB_TCPTIMEWAITOVERFLOW, /* TCPTimeWaitOverflow */ + LINUX_MIB_TCPREQQFULLDOCOOKIES, /* TCPReqQFullDoCookies */ + LINUX_MIB_TCPREQQFULLDROP, /* TCPReqQFullDrop */ __LINUX_MIB_MAX }; diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 99e6e19b57c2..4c0766e201e3 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -96,7 +96,8 @@ extern int sysctl_max_syn_backlog; */ struct listen_sock { u8 max_qlen_log; - /* 3 bytes hole, try to use */ + u8 synflood_warned; + /* 2 bytes hole, try to use */ int qlen; int qlen_young; int clock_hand; diff --git a/include/net/tcp.h b/include/net/tcp.h index 149a415d1e0a..e9b48b094683 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -460,6 +460,9 @@ extern int tcp_write_wakeup(struct sock *); extern void tcp_send_fin(struct sock *sk); extern void tcp_send_active_reset(struct sock *sk, gfp_t priority); extern int tcp_send_synack(struct sock *); +extern int tcp_syn_flood_action(struct sock *sk, + const struct sk_buff *skb, + const char *proto); extern void tcp_push_one(struct sock *, unsigned int mss_now); extern void tcp_send_ack(struct sock *sk); extern void tcp_send_delayed_ack(struct sock *sk); diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index b14ec7d03b6e..4bfad5da94f4 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -254,6 +254,8 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("TCPDeferAcceptDrop", LINUX_MIB_TCPDEFERACCEPTDROP), SNMP_MIB_ITEM("IPReversePathFilter", LINUX_MIB_IPRPFILTER), SNMP_MIB_ITEM("TCPTimeWaitOverflow", LINUX_MIB_TCPTIMEWAITOVERFLOW), + SNMP_MIB_ITEM("TCPReqQFullDoCookies", LINUX_MIB_TCPREQQFULLDOCOOKIES), + SNMP_MIB_ITEM("TCPReqQFullDrop", LINUX_MIB_TCPREQQFULLDROP), SNMP_MIB_SENTINEL }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1c12b8ec849d..c34f01513945 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -808,20 +808,38 @@ static void tcp_v4_reqsk_destructor(struct request_sock *req) kfree(inet_rsk(req)->opt); } -static void syn_flood_warning(const struct sk_buff *skb) +/* + * Return 1 if a syncookie should be sent + */ +int tcp_syn_flood_action(struct sock *sk, + const struct sk_buff *skb, + const char *proto) { - const char *msg; + const char *msg = "Dropping request"; + int want_cookie = 0; + struct listen_sock *lopt; + + #ifdef CONFIG_SYN_COOKIES - if (sysctl_tcp_syncookies) + if (sysctl_tcp_syncookies) { msg = "Sending cookies"; - else + want_cookie = 1; + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES); + } else #endif - msg = "Dropping request"; + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP); - pr_info("TCP: Possible SYN flooding on port %d. %s.\n", - ntohs(tcp_hdr(skb)->dest), msg); + lopt = inet_csk(sk)->icsk_accept_queue.listen_opt; + if (!lopt->synflood_warned) { + lopt->synflood_warned = 1; + pr_info("%s: Possible SYN flooding on port %d. %s. " + " Check SNMP counters.\n", + proto, ntohs(tcp_hdr(skb)->dest), msg); + } + return want_cookie; } +EXPORT_SYMBOL(tcp_syn_flood_action); /* * Save and compile IPv4 options into the request_sock if needed. @@ -1235,11 +1253,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) __be32 saddr = ip_hdr(skb)->saddr; __be32 daddr = ip_hdr(skb)->daddr; __u32 isn = TCP_SKB_CB(skb)->when; -#ifdef CONFIG_SYN_COOKIES int want_cookie = 0; -#else -#define want_cookie 0 /* Argh, why doesn't gcc optimize this :( */ -#endif /* Never answer to SYNs send to broadcast or multicast */ if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) @@ -1250,14 +1264,9 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) * evidently real one. */ if (inet_csk_reqsk_queue_is_full(sk) && !isn) { - if (net_ratelimit()) - syn_flood_warning(skb); -#ifdef CONFIG_SYN_COOKIES - if (sysctl_tcp_syncookies) { - want_cookie = 1; - } else -#endif - goto drop; + want_cookie = tcp_syn_flood_action(sk, skb, "TCP"); + if (!want_cookie) + goto drop; } /* Accept backlog is full. If we have already queued enough @@ -1303,9 +1312,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) while (l-- > 0) *c++ ^= *hash_location++; -#ifdef CONFIG_SYN_COOKIES want_cookie = 0; /* not our kind of cookie */ -#endif tmp_ext.cookie_out_never = 0; /* false */ tmp_ext.cookie_plus = tmp_opt.cookie_plus; } else if (!tp->rx_opt.cookie_in_always) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d1fb63f4aeb7..3c9fa618b69d 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -531,20 +531,6 @@ static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req, return tcp_v6_send_synack(sk, req, rvp); } -static inline void syn_flood_warning(struct sk_buff *skb) -{ -#ifdef CONFIG_SYN_COOKIES - if (sysctl_tcp_syncookies) - printk(KERN_INFO - "TCPv6: Possible SYN flooding on port %d. " - "Sending cookies.\n", ntohs(tcp_hdr(skb)->dest)); - else -#endif - printk(KERN_INFO - "TCPv6: Possible SYN flooding on port %d. " - "Dropping request.\n", ntohs(tcp_hdr(skb)->dest)); -} - static void tcp_v6_reqsk_destructor(struct request_sock *req) { kfree_skb(inet6_rsk(req)->pktopts); @@ -1179,11 +1165,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); __u32 isn = TCP_SKB_CB(skb)->when; struct dst_entry *dst = NULL; -#ifdef CONFIG_SYN_COOKIES int want_cookie = 0; -#else -#define want_cookie 0 -#endif if (skb->protocol == htons(ETH_P_IP)) return tcp_v4_conn_request(sk, skb); @@ -1192,14 +1174,9 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) goto drop; if (inet_csk_reqsk_queue_is_full(sk) && !isn) { - if (net_ratelimit()) - syn_flood_warning(skb); -#ifdef CONFIG_SYN_COOKIES - if (sysctl_tcp_syncookies) - want_cookie = 1; - else -#endif - goto drop; + want_cookie = tcp_syn_flood_action(sk, skb, "TCPv6"); + if (!want_cookie) + goto drop; } if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1) @@ -1249,9 +1226,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) while (l-- > 0) *c++ ^= *hash_location++; -#ifdef CONFIG_SYN_COOKIES want_cookie = 0; /* not our kind of cookie */ -#endif tmp_ext.cookie_out_never = 0; /* false */ tmp_ext.cookie_plus = tmp_opt.cookie_plus; } else if (!tp->rx_opt.cookie_in_always) { From 6b59e3191daade2b975eeec1c71c591eb5c86b7b Mon Sep 17 00:00:00 2001 From: Marcos Paulo de Souza Date: Tue, 30 Aug 2011 05:33:57 +0000 Subject: [PATCH 275/326] Documentation: networking: dmfe.txt: Remove the maintainer of orphan networking driver The dmfe module is a orphan driver, and with this was removed the maintainer of the documentation. Signed-off-by: Marcos Paulo de Souza Signed-off-by: David S. Miller --- Documentation/networking/dmfe.txt | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/dmfe.txt b/Documentation/networking/dmfe.txt index 8006c227fda2..25320bf19c86 100644 --- a/Documentation/networking/dmfe.txt +++ b/Documentation/networking/dmfe.txt @@ -1,3 +1,5 @@ +Note: This driver doesn't have a maintainer. + Davicom DM9102(A)/DM9132/DM9801 fast ethernet driver for Linux. This program is free software; you can redistribute it and/or @@ -55,7 +57,6 @@ Test and make sure PCI latency is now correct for all cases. Authors: Sten Wang : Original Author -Tobias Ringstrom : Current Maintainer Contributors: From d9e64f83ebb8f563810b10536b23516d3bd30e80 Mon Sep 17 00:00:00 2001 From: "rajan.aggarwal85@gmail.com" Date: Tue, 30 Aug 2011 23:57:38 +0000 Subject: [PATCH 276/326] net/can/af_can.c: Change del_timer to del_timer_sync This is important for SMP platform to check if timer function is executing on other CPU with deleting the timer. Signed-off-by: Rajan Aggarwal Acked-by: Oliver Hartkopp Signed-off-by: David S. Miller --- net/can/af_can.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/can/af_can.c b/net/can/af_can.c index 8ce926d3b2cb..9b0c32a2690c 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -857,7 +857,7 @@ static __exit void can_exit(void) struct net_device *dev; if (stats_timer) - del_timer(&can_stattimer); + del_timer_sync(&can_stattimer); can_remove_proc(); From 02009afc223aae43b8e18918fc816e4520791537 Mon Sep 17 00:00:00 2001 From: Kavan Smith Date: Wed, 31 Aug 2011 05:12:05 +0000 Subject: [PATCH 277/326] ipheth: iPhone 4 Verizon CDMA USB Product ID add Add USB product ID for iPhone 4 CDMA Verizon Tested on at least 2 devices Signed-off-by: Kavan Smith Signed-off-by: David S. Miller --- drivers/net/usb/ipheth.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/usb/ipheth.c b/drivers/net/usb/ipheth.c index 15772b1b6a91..13c1f044b40d 100644 --- a/drivers/net/usb/ipheth.c +++ b/drivers/net/usb/ipheth.c @@ -59,6 +59,7 @@ #define USB_PRODUCT_IPHONE_3G 0x1292 #define USB_PRODUCT_IPHONE_3GS 0x1294 #define USB_PRODUCT_IPHONE_4 0x1297 +#define USB_PRODUCT_IPHONE_4_VZW 0x129c #define IPHETH_USBINTF_CLASS 255 #define IPHETH_USBINTF_SUBCLASS 253 @@ -98,6 +99,10 @@ static struct usb_device_id ipheth_table[] = { USB_VENDOR_APPLE, USB_PRODUCT_IPHONE_4, IPHETH_USBINTF_CLASS, IPHETH_USBINTF_SUBCLASS, IPHETH_USBINTF_PROTO) }, + { USB_DEVICE_AND_INTERFACE_INFO( + USB_VENDOR_APPLE, USB_PRODUCT_IPHONE_4_VZW, + IPHETH_USBINTF_CLASS, IPHETH_USBINTF_SUBCLASS, + IPHETH_USBINTF_PROTO) }, { } }; MODULE_DEVICE_TABLE(usb, ipheth_table); From 0542b69e2c57fc9668ce6a03155bea6e1f557901 Mon Sep 17 00:00:00 2001 From: dpward Date: Wed, 31 Aug 2011 06:05:27 +0000 Subject: [PATCH 278/326] net: Make flow cache namespace-aware flow_cache_lookup will return a cached object (or null pointer) that the resolver (i.e. xfrm_policy_lookup) previously found for another namespace using the same key/family/dir. Instead, make the namespace part of what identifies entries in the cache. As before, flow_entry_valid will return 0 for entries where the namespace has been deleted, and they will be removed from the cache the next time flow_cache_gc_task is run. Reported-by: Andrew Dickinson Signed-off-by: David Ward Signed-off-by: David S. Miller --- net/core/flow.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/core/flow.c b/net/core/flow.c index bf32c33cad3b..47b6d26c2afb 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -30,6 +30,7 @@ struct flow_cache_entry { struct hlist_node hlist; struct list_head gc_list; } u; + struct net *net; u16 family; u8 dir; u32 genid; @@ -232,7 +233,8 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, hash = flow_hash_code(fc, fcp, key); hlist_for_each_entry(tfle, entry, &fcp->hash_table[hash], u.hlist) { - if (tfle->family == family && + if (tfle->net == net && + tfle->family == family && tfle->dir == dir && flow_key_compare(key, &tfle->key) == 0) { fle = tfle; @@ -246,6 +248,7 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, fle = kmem_cache_alloc(flow_cachep, GFP_ATOMIC); if (fle) { + fle->net = net; fle->family = family; fle->dir = dir; memcpy(&fle->key, key, sizeof(*key)); From 48c830120f2a20b44220aa26feda9ed15f49eaab Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 31 Aug 2011 08:03:29 +0000 Subject: [PATCH 279/326] net: copy userspace buffers on device forwarding dev_forward_skb loops an skb back into host networking stack which might hang on the memory indefinitely. In particular, this can happen in macvtap in bridged mode. Copy the userspace fragments to avoid blocking the sender in that case. As this patch makes skb_copy_ubufs extern now, I also added some documentation and made it clear the SKBTX_DEV_ZEROCOPY flag automatically instead of doing it in all callers. This can be made into a separate patch if people feel it's worth it. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- include/linux/skbuff.h | 1 + net/core/dev.c | 8 ++++++++ net/core/skbuff.c | 22 +++++++++++++++++----- 3 files changed, 26 insertions(+), 5 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7b996ed86d5b..8bd383caa363 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -524,6 +524,7 @@ static inline struct sk_buff *alloc_skb_fclone(unsigned int size, extern bool skb_recycle_check(struct sk_buff *skb, int skb_size); extern struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src); +extern int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask); extern struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t priority); extern struct sk_buff *skb_copy(const struct sk_buff *skb, diff --git a/net/core/dev.c b/net/core/dev.c index 17d67b579beb..b10ff0a71855 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1515,6 +1515,14 @@ static inline bool is_skb_forwardable(struct net_device *dev, */ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) { + if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { + if (skb_copy_ubufs(skb, GFP_ATOMIC)) { + atomic_long_inc(&dev->rx_dropped); + kfree_skb(skb); + return NET_RX_DROP; + } + } + skb_orphan(skb); nf_reset(skb); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 27002dffe7ed..387703f56fce 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -611,8 +611,21 @@ struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src) } EXPORT_SYMBOL_GPL(skb_morph); -/* skb frags copy userspace buffers to kernel */ -static int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) +/* skb_copy_ubufs - copy userspace skb frags buffers to kernel + * @skb: the skb to modify + * @gfp_mask: allocation priority + * + * This must be called on SKBTX_DEV_ZEROCOPY skb. + * It will copy all frags into kernel and drop the reference + * to userspace pages. + * + * If this function is called from an interrupt gfp_mask() must be + * %GFP_ATOMIC. + * + * Returns 0 on success or a negative error code on failure + * to allocate kernel memory to copy to. + */ +int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) { int i; int num_frags = skb_shinfo(skb)->nr_frags; @@ -652,6 +665,8 @@ static int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) skb_shinfo(skb)->frags[i - 1].page = head; head = (struct page *)head->private; } + + skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; return 0; } @@ -677,7 +692,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { if (skb_copy_ubufs(skb, gfp_mask)) return NULL; - skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; } n = skb + 1; @@ -803,7 +817,6 @@ struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) n = NULL; goto out; } - skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; } for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { skb_shinfo(n)->frags[i] = skb_shinfo(skb)->frags[i]; @@ -896,7 +909,6 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { if (skb_copy_ubufs(skb, gfp_mask)) goto nofrags; - skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY; } for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) get_page(skb_shinfo(skb)->frags[i].page); From 31dda0ae933bb9fea9cfe000b698c41af0417cac Mon Sep 17 00:00:00 2001 From: nhorman Date: Wed, 14 Sep 2011 03:05:02 +0000 Subject: [PATCH 280/326] net: don't clear IFF_XMIT_DST_RELEASE in ether_setup d88733150 introduced the IFF_SKB_TX_SHARING flag, which I unilaterally set in ether_setup. In doing this I didn't realize that other flags (such as IFF_XMIT_DST_RELEASE) might be set prior to calling the ether_setup routine. This patch changes ether_setup to or in SKB_TX_SHARING so as not to inadvertently clear other existing flags. Thanks to Pekka Riikonen for pointing out my error Signed-off-by: Neil Horman Reported-by: Pekka Riikonen CC: "David S. Miller" Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ethernet/eth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 27997d35ebd3..a2468363978e 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -340,7 +340,7 @@ void ether_setup(struct net_device *dev) dev->addr_len = ETH_ALEN; dev->tx_queue_len = 1000; /* Ethernet wants good queues */ dev->flags = IFF_BROADCAST|IFF_MULTICAST; - dev->priv_flags = IFF_TX_SKB_SHARING; + dev->priv_flags |= IFF_TX_SKB_SHARING; memset(dev->broadcast, 0xFF, ETH_ALEN); From c482e6c064613b3fd40758ef6c33318462b83789 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:47:49 +0000 Subject: [PATCH 281/326] bnx2x: Fix ETS bandwidth ETS bandwidth of 0% is not allowed by driver, so provide alternative HW configuration for this case. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index d45b1555a602..9d381db16516 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -778,9 +778,9 @@ static int bnx2x_ets_e3b0_set_cos_bw(struct bnx2x *bp, { u32 nig_reg_adress_crd_weight = 0; u32 pbf_reg_adress_crd_weight = 0; - /* Calculate and set BW for this COS*/ - const u32 cos_bw_nig = (bw * min_w_val_nig) / total_bw; - const u32 cos_bw_pbf = (bw * min_w_val_pbf) / total_bw; + /* Calculate and set BW for this COS - use 1 instead of 0 for BW */ + const u32 cos_bw_nig = ((bw ? bw : 1) * min_w_val_nig) / total_bw; + const u32 cos_bw_pbf = ((bw ? bw : 1) * min_w_val_pbf) / total_bw; switch (cos_entry) { case 0: @@ -852,18 +852,12 @@ static int bnx2x_ets_e3b0_get_total_bw( /* Calculate total BW requested */ for (cos_idx = 0; cos_idx < ets_params->num_of_cos; cos_idx++) { if (bnx2x_cos_state_bw == ets_params->cos[cos_idx].state) { - - if (0 == ets_params->cos[cos_idx].params.bw_params.bw) { - DP(NETIF_MSG_LINK, "bnx2x_ets_E3B0_config BW" - "was set to 0\n"); - return -EINVAL; + *total_bw += + ets_params->cos[cos_idx].params.bw_params.bw; } - *total_bw += - ets_params->cos[cos_idx].params.bw_params.bw; - } } - /*Check taotl BW is valid */ + /* Check total BW is valid */ if ((100 != *total_bw) || (0 == *total_bw)) { if (0 == *total_bw) { DP(NETIF_MSG_LINK, "bnx2x_ets_E3B0_config toatl BW" From 6b1f3900fc0909fbf3bd672242378015f76b3df8 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:47:54 +0000 Subject: [PATCH 282/326] bnx2x: Enable FEC for 57810-KR Enable FEC(Forward Error Correction) for 57810-KR to reduce link errors. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 6 ++++++ drivers/net/bnx2x/bnx2x_reg.h | 3 +++ 2 files changed, 9 insertions(+) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 9d381db16516..f7a7ac3e889c 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -3624,6 +3624,12 @@ static void bnx2x_warpcore_enable_AN_KR(struct bnx2x_phy *phy, bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, MDIO_WC_REG_AN_IEEE1BLK_AN_ADVERTISEMENT1, val16); + /* Advertised and set FEC (Forward Error Correction) */ + bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, + MDIO_WC_REG_AN_IEEE1BLK_AN_ADVERTISEMENT2, + (MDIO_WC_REG_AN_IEEE1BLK_AN_ADV2_FEC_ABILITY | + MDIO_WC_REG_AN_IEEE1BLK_AN_ADV2_FEC_REQ)); + /* Enable CL37 BAM */ if (REG_RD(bp, params->shmem_base + offsetof(struct shmem_region, dev_info. diff --git a/drivers/net/bnx2x/bnx2x_reg.h b/drivers/net/bnx2x/bnx2x_reg.h index dac217d478f2..057738623ba4 100644 --- a/drivers/net/bnx2x/bnx2x_reg.h +++ b/drivers/net/bnx2x/bnx2x_reg.h @@ -6853,6 +6853,9 @@ Theotherbitsarereservedandshouldbezero*/ #define MDIO_WC_REG_IEEE0BLK_AUTONEGNP 0x7 #define MDIO_WC_REG_AN_IEEE1BLK_AN_ADVERTISEMENT0 0x10 #define MDIO_WC_REG_AN_IEEE1BLK_AN_ADVERTISEMENT1 0x11 +#define MDIO_WC_REG_AN_IEEE1BLK_AN_ADVERTISEMENT2 0x12 +#define MDIO_WC_REG_AN_IEEE1BLK_AN_ADV2_FEC_ABILITY 0x4000 +#define MDIO_WC_REG_AN_IEEE1BLK_AN_ADV2_FEC_REQ 0x8000 #define MDIO_WC_REG_PMD_IEEE9BLK_TENGBASE_KR_PMD_CONTROL_REGISTER_150 0x96 #define MDIO_WC_REG_XGXSBLK0_XGXSCONTROL 0x8000 #define MDIO_WC_REG_XGXSBLK0_MISCCONTROL1 0x800e From 0582242049c67d59c3a95cd1cba8995fa955c858 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:47:58 +0000 Subject: [PATCH 283/326] bnx2x: Remove fiber remote fault detection Remove remote fault detection as a tactic retreat due to link issues involved with it. Once issue is resolved, this feature will be restored again. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index f7a7ac3e889c..db5913da5527 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -10638,8 +10638,7 @@ static struct bnx2x_phy phy_warpcore = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT, .addr = 0xff, .def_md_devad = 0, - .flags = (FLAGS_HW_LOCK_REQUIRED | - FLAGS_TX_ERROR_CHECK), + .flags = FLAGS_HW_LOCK_REQUIRED, .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10765,8 +10764,7 @@ static struct bnx2x_phy phy_8706 = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706, .addr = 0xff, .def_md_devad = 0, - .flags = (FLAGS_INIT_XGXS_FIRST | - FLAGS_TX_ERROR_CHECK), + .flags = FLAGS_INIT_XGXS_FIRST, .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10797,8 +10795,7 @@ static struct bnx2x_phy phy_8726 = { .addr = 0xff, .def_md_devad = 0, .flags = (FLAGS_HW_LOCK_REQUIRED | - FLAGS_INIT_XGXS_FIRST | - FLAGS_TX_ERROR_CHECK), + FLAGS_INIT_XGXS_FIRST), .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10829,8 +10826,7 @@ static struct bnx2x_phy phy_8727 = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8727, .addr = 0xff, .def_md_devad = 0, - .flags = (FLAGS_FAN_FAILURE_DET_REQ | - FLAGS_TX_ERROR_CHECK), + .flags = FLAGS_FAN_FAILURE_DET_REQ, .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, From 4d7e25d6cc4312b1f949123fea7026fd56441513 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:48:03 +0000 Subject: [PATCH 284/326] bnx2x: Fix XMAC loopback test Change XMAC loopback type from CORE LOCAL to LINE LOCAL for the BCM578xx due to intermittent problem with the loopback with this configuration. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 2 +- drivers/net/bnx2x/bnx2x_reg.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index db5913da5527..342807585c2b 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -1720,7 +1720,7 @@ static int bnx2x_xmac_enable(struct link_params *params, /* Check loopback mode */ if (lb) - val |= XMAC_CTRL_REG_CORE_LOCAL_LPBK; + val |= XMAC_CTRL_REG_LINE_LOCAL_LPBK; REG_WR(bp, xmac_base + XMAC_REG_CTRL, val); bnx2x_set_xumac_nig(params, ((vars->flow_ctrl & BNX2X_FLOW_CTRL_TX) != 0), 1); diff --git a/drivers/net/bnx2x/bnx2x_reg.h b/drivers/net/bnx2x/bnx2x_reg.h index 057738623ba4..750e8445dac4 100644 --- a/drivers/net/bnx2x/bnx2x_reg.h +++ b/drivers/net/bnx2x/bnx2x_reg.h @@ -5320,7 +5320,7 @@ #define XCM_REG_XX_OVFL_EVNT_ID 0x20058 #define XMAC_CLEAR_RX_LSS_STATUS_REG_CLEAR_LOCAL_FAULT_STATUS (0x1<<0) #define XMAC_CLEAR_RX_LSS_STATUS_REG_CLEAR_REMOTE_FAULT_STATUS (0x1<<1) -#define XMAC_CTRL_REG_CORE_LOCAL_LPBK (0x1<<3) +#define XMAC_CTRL_REG_LINE_LOCAL_LPBK (0x1<<2) #define XMAC_CTRL_REG_RX_EN (0x1<<1) #define XMAC_CTRL_REG_SOFT_RESET (0x1<<6) #define XMAC_CTRL_REG_TX_EN (0x1<<0) From ab505dec96340946079d1288f49041bea9f259ff Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:48:06 +0000 Subject: [PATCH 285/326] bnx2x: Fix 578xx link LED Fix 1G link LED for the BCM578xx-SFI/KR. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 342807585c2b..ba15bdc5a1a9 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -5924,7 +5924,7 @@ int bnx2x_set_led(struct link_params *params, (tmp | EMAC_LED_OVERRIDE)); /* * return here without enabling traffic - * LED blink andsetting rate in ON mode. + * LED blink and setting rate in ON mode. * In oper mode, enabling LED blink * and setting rate is needed. */ @@ -5936,7 +5936,11 @@ int bnx2x_set_led(struct link_params *params, * This is a work-around for HW issue found when link * is up in CL73 */ - REG_WR(bp, NIG_REG_LED_10G_P0 + port*4, 1); + if ((!CHIP_IS_E3(bp)) || + (CHIP_IS_E3(bp) && + mode == LED_MODE_ON)) + REG_WR(bp, NIG_REG_LED_10G_P0 + port*4, 1); + if (CHIP_IS_E1x(bp) || CHIP_IS_E2(bp) || (mode == LED_MODE_ON)) From 8d661637407963d1990e53c36d53ace123219da3 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Wed, 7 Sep 2011 00:48:11 +0000 Subject: [PATCH 286/326] bnx2x: Fix ethtool advertisement Enable changing advertisement settings via ethtool. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Reviewed-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_ethtool.c | 43 ++++++++++++++++++++++++++++--- 1 file changed, 40 insertions(+), 3 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_ethtool.c b/drivers/net/bnx2x/bnx2x_ethtool.c index 0ceb6c7b1238..cf3e47914dd7 100644 --- a/drivers/net/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/bnx2x/bnx2x_ethtool.c @@ -363,13 +363,50 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) } /* advertise the requested speed and duplex if supported */ - cmd->advertising &= bp->port.supported[cfg_idx]; + if (cmd->advertising & ~(bp->port.supported[cfg_idx])) { + DP(NETIF_MSG_LINK, "Advertisement parameters " + "are not supported\n"); + return -EINVAL; + } bp->link_params.req_line_speed[cfg_idx] = SPEED_AUTO_NEG; - bp->link_params.req_duplex[cfg_idx] = DUPLEX_FULL; - bp->port.advertising[cfg_idx] |= (ADVERTISED_Autoneg | + bp->link_params.req_duplex[cfg_idx] = cmd->duplex; + bp->port.advertising[cfg_idx] = (ADVERTISED_Autoneg | cmd->advertising); + if (cmd->advertising) { + bp->link_params.speed_cap_mask[cfg_idx] = 0; + if (cmd->advertising & ADVERTISED_10baseT_Half) { + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF; + } + if (cmd->advertising & ADVERTISED_10baseT_Full) + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL; + + if (cmd->advertising & ADVERTISED_100baseT_Full) + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL; + + if (cmd->advertising & ADVERTISED_100baseT_Half) { + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF; + } + if (cmd->advertising & ADVERTISED_1000baseT_Half) { + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_1G; + } + if (cmd->advertising & (ADVERTISED_1000baseT_Full | + ADVERTISED_1000baseKX_Full)) + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_1G; + + if (cmd->advertising & (ADVERTISED_10000baseT_Full | + ADVERTISED_10000baseKX4_Full | + ADVERTISED_10000baseKR_Full)) + bp->link_params.speed_cap_mask[cfg_idx] |= + PORT_HW_CFG_SPEED_CAPABILITY_D0_10G; + } } else { /* forced speed */ /* advertise the requested speed and duplex if supported */ switch (speed) { From a7f934d4f16144cb9521b62e9b8c9ac0118097da Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 15 Sep 2011 13:28:33 -0700 Subject: [PATCH 287/326] asm alternatives: remove incorrect alignment notes On x86-64, they were just wasteful: with the explicitly added (now unnecessary) padding, the size of the alternatives structure was 16 bytes, and an alignment of 8 bytes didn't hurt much. However, it was still silly, since the natural size and alignment for the structure is actually just 12 bytes, 4-byte aligned since commit 59e97e4d6fbc ("x86: Make alternative instruction pointers relative"). So removing the padding, and removing the extra alignment is just a good idea. On x86-32, the alignment of 4 bytes was correct, but was incorrectly hardcoded as 8 bytes in . That header file had used to be an x86-64 only header file, but various unification efforts have made it be used for x86-32 too (ie the unification of rwlock and rwsem). That in turn caused x86-32 boot failures, because the extra alignment would result in random zero-filled words in the altinstructions section, causing oopses early at boot when doing alternative instruction replacement. So just remove all the alignment noise entirely. It's wrong, and it's unnecessary. The section itself is already properly aligned by the linker scripts, and all additions to the section had better be of the proper 12-byte format, keeping it aligned. So if the align directive were to ever make a difference, that would be an indication of a serious bug to begin with. Reported-by: Werner Landgraf Acked-by: Andrew Lutomirski Signed-off-by: Linus Torvalds --- arch/x86/include/asm/alternative-asm.h | 1 - arch/x86/include/asm/alternative.h | 4 ---- arch/x86/include/asm/cpufeature.h | 2 -- 3 files changed, 7 deletions(-) diff --git a/arch/x86/include/asm/alternative-asm.h b/arch/x86/include/asm/alternative-asm.h index 4554cc6fb96a..091508b533b4 100644 --- a/arch/x86/include/asm/alternative-asm.h +++ b/arch/x86/include/asm/alternative-asm.h @@ -16,7 +16,6 @@ #endif .macro altinstruction_entry orig alt feature orig_len alt_len - .align 8 .long \orig - . .long \alt - . .word \feature diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 23fb6d79f209..37ad100a2210 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -48,9 +48,6 @@ struct alt_instr { u16 cpuid; /* cpuid bit set for replacement */ u8 instrlen; /* length of original instruction */ u8 replacementlen; /* length of new instruction, <= instrlen */ -#ifdef CONFIG_X86_64 - u32 pad2; -#endif }; extern void alternative_instructions(void); @@ -83,7 +80,6 @@ static inline int alternatives_text_reserved(void *start, void *end) \ "661:\n\t" oldinstr "\n662:\n" \ ".section .altinstructions,\"a\"\n" \ - _ASM_ALIGN "\n" \ " .long 661b - .\n" /* label */ \ " .long 663f - .\n" /* new instruction */ \ " .word " __stringify(feature) "\n" /* feature bit */ \ diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 4258aac99a6e..88b23a43f340 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -332,7 +332,6 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) asm goto("1: jmp %l[t_no]\n" "2:\n" ".section .altinstructions,\"a\"\n" - _ASM_ALIGN "\n" " .long 1b - .\n" " .long 0\n" /* no replacement */ " .word %P0\n" /* feature bit */ @@ -350,7 +349,6 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) asm volatile("1: movb $0,%0\n" "2:\n" ".section .altinstructions,\"a\"\n" - _ASM_ALIGN "\n" " .long 1b - .\n" " .long 3f - .\n" " .word %P1\n" /* feature bit */ From 86c432ca5d6da90a26ac8d3e680f2268b502d9c5 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 1 Sep 2011 12:09:29 +0000 Subject: [PATCH 288/326] Revert "sfc: Use write-combining to reduce TX latency" and follow-ups This reverts commits 65f0b417dee94f779ce9b77102b7d73c93723b39, d88d6b05fee3cc78e5b0273eb58c31201dcc6b76, fcfa060468a4edcf776f0c1211d826d5de1668c1, 747df2258b1b9a2e25929ef496262c339c380009 and 867955f5682f7157fdafe8670804b9f8ea077bc7. Depending on the processor model, write-combining may result in reordering that the NIC will not tolerate. This typically results in a DMA error event and reset by the driver, logged as: sfc 0000:0e:00.0: eth2: TX DMA Q reports TX_EV_PKT_ERR. sfc 0000:0e:00.0: eth2: resetting (ALL) Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/sfc/efx.c | 18 ++------------ drivers/net/sfc/io.h | 15 +++--------- drivers/net/sfc/mcdi.c | 46 +++++++++++++---------------------- drivers/net/sfc/nic.c | 7 ------ drivers/net/sfc/nic.h | 2 -- drivers/net/sfc/siena.c | 25 +++---------------- drivers/net/sfc/workarounds.h | 2 -- 7 files changed, 27 insertions(+), 88 deletions(-) diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c index faca764aa21b..b59abc706d93 100644 --- a/drivers/net/sfc/efx.c +++ b/drivers/net/sfc/efx.c @@ -1050,7 +1050,6 @@ static int efx_init_io(struct efx_nic *efx) { struct pci_dev *pci_dev = efx->pci_dev; dma_addr_t dma_mask = efx->type->max_dma_mask; - bool use_wc; int rc; netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n"); @@ -1101,21 +1100,8 @@ static int efx_init_io(struct efx_nic *efx) rc = -EIO; goto fail3; } - - /* bug22643: If SR-IOV is enabled then tx push over a write combined - * mapping is unsafe. We need to disable write combining in this case. - * MSI is unsupported when SR-IOV is enabled, and the firmware will - * have removed the MSI capability. So write combining is safe if - * there is an MSI capability. - */ - use_wc = (!EFX_WORKAROUND_22643(efx) || - pci_find_capability(pci_dev, PCI_CAP_ID_MSI)); - if (use_wc) - efx->membase = ioremap_wc(efx->membase_phys, - efx->type->mem_map_size); - else - efx->membase = ioremap_nocache(efx->membase_phys, - efx->type->mem_map_size); + efx->membase = ioremap_nocache(efx->membase_phys, + efx->type->mem_map_size); if (!efx->membase) { netif_err(efx, probe, efx->net_dev, "could not map memory BAR at %llx+%x\n", diff --git a/drivers/net/sfc/io.h b/drivers/net/sfc/io.h index cc978803d484..dc45110b2456 100644 --- a/drivers/net/sfc/io.h +++ b/drivers/net/sfc/io.h @@ -48,9 +48,9 @@ * replacing the low 96 bits with zero does not affect functionality. * - If the host writes to the last dword address of such a register * (i.e. the high 32 bits) the underlying register will always be - * written. If the collector and the current write together do not - * provide values for all 128 bits of the register, the low 96 bits - * will be written as zero. + * written. If the collector does not hold values for the low 96 + * bits of the register, they will be written as zero. Writing to + * the last qword does not have this effect and must not be done. * - If the host writes to the address of any other part of such a * register while the collector already holds values for some other * register, the write is discarded and the collector maintains its @@ -103,7 +103,6 @@ static inline void efx_writeo(struct efx_nic *efx, efx_oword_t *value, _efx_writed(efx, value->u32[2], reg + 8); _efx_writed(efx, value->u32[3], reg + 12); #endif - wmb(); mmiowb(); spin_unlock_irqrestore(&efx->biu_lock, flags); } @@ -126,7 +125,6 @@ static inline void efx_sram_writeq(struct efx_nic *efx, void __iomem *membase, __raw_writel((__force u32)value->u32[0], membase + addr); __raw_writel((__force u32)value->u32[1], membase + addr + 4); #endif - wmb(); mmiowb(); spin_unlock_irqrestore(&efx->biu_lock, flags); } @@ -141,7 +139,6 @@ static inline void efx_writed(struct efx_nic *efx, efx_dword_t *value, /* No lock required */ _efx_writed(efx, value->u32[0], reg); - wmb(); } /* Read a 128-bit CSR, locking as appropriate. */ @@ -152,7 +149,6 @@ static inline void efx_reado(struct efx_nic *efx, efx_oword_t *value, spin_lock_irqsave(&efx->biu_lock, flags); value->u32[0] = _efx_readd(efx, reg + 0); - rmb(); value->u32[1] = _efx_readd(efx, reg + 4); value->u32[2] = _efx_readd(efx, reg + 8); value->u32[3] = _efx_readd(efx, reg + 12); @@ -175,7 +171,6 @@ static inline void efx_sram_readq(struct efx_nic *efx, void __iomem *membase, value->u64[0] = (__force __le64)__raw_readq(membase + addr); #else value->u32[0] = (__force __le32)__raw_readl(membase + addr); - rmb(); value->u32[1] = (__force __le32)__raw_readl(membase + addr + 4); #endif spin_unlock_irqrestore(&efx->biu_lock, flags); @@ -242,14 +237,12 @@ static inline void _efx_writeo_page(struct efx_nic *efx, efx_oword_t *value, #ifdef EFX_USE_QWORD_IO _efx_writeq(efx, value->u64[0], reg + 0); - _efx_writeq(efx, value->u64[1], reg + 8); #else _efx_writed(efx, value->u32[0], reg + 0); _efx_writed(efx, value->u32[1], reg + 4); +#endif _efx_writed(efx, value->u32[2], reg + 8); _efx_writed(efx, value->u32[3], reg + 12); -#endif - wmb(); } #define efx_writeo_page(efx, value, reg, page) \ _efx_writeo_page(efx, value, \ diff --git a/drivers/net/sfc/mcdi.c b/drivers/net/sfc/mcdi.c index 3dd45ed61f0a..81a425397468 100644 --- a/drivers/net/sfc/mcdi.c +++ b/drivers/net/sfc/mcdi.c @@ -50,20 +50,6 @@ static inline struct efx_mcdi_iface *efx_mcdi(struct efx_nic *efx) return &nic_data->mcdi; } -static inline void -efx_mcdi_readd(struct efx_nic *efx, efx_dword_t *value, unsigned reg) -{ - struct siena_nic_data *nic_data = efx->nic_data; - value->u32[0] = (__force __le32)__raw_readl(nic_data->mcdi_smem + reg); -} - -static inline void -efx_mcdi_writed(struct efx_nic *efx, const efx_dword_t *value, unsigned reg) -{ - struct siena_nic_data *nic_data = efx->nic_data; - __raw_writel((__force u32)value->u32[0], nic_data->mcdi_smem + reg); -} - void efx_mcdi_init(struct efx_nic *efx) { struct efx_mcdi_iface *mcdi; @@ -84,8 +70,8 @@ static void efx_mcdi_copyin(struct efx_nic *efx, unsigned cmd, const u8 *inbuf, size_t inlen) { struct efx_mcdi_iface *mcdi = efx_mcdi(efx); - unsigned pdu = MCDI_PDU(efx); - unsigned doorbell = MCDI_DOORBELL(efx); + unsigned pdu = FR_CZ_MC_TREG_SMEM + MCDI_PDU(efx); + unsigned doorbell = FR_CZ_MC_TREG_SMEM + MCDI_DOORBELL(efx); unsigned int i; efx_dword_t hdr; u32 xflags, seqno; @@ -106,28 +92,29 @@ static void efx_mcdi_copyin(struct efx_nic *efx, unsigned cmd, MCDI_HEADER_SEQ, seqno, MCDI_HEADER_XFLAGS, xflags); - efx_mcdi_writed(efx, &hdr, pdu); + efx_writed(efx, &hdr, pdu); for (i = 0; i < inlen; i += 4) - efx_mcdi_writed(efx, (const efx_dword_t *)(inbuf + i), - pdu + 4 + i); + _efx_writed(efx, *((__le32 *)(inbuf + i)), pdu + 4 + i); + + /* Ensure the payload is written out before the header */ + wmb(); /* ring the doorbell with a distinctive value */ - EFX_POPULATE_DWORD_1(hdr, EFX_DWORD_0, 0x45789abc); - efx_mcdi_writed(efx, &hdr, doorbell); + _efx_writed(efx, (__force __le32) 0x45789abc, doorbell); } static void efx_mcdi_copyout(struct efx_nic *efx, u8 *outbuf, size_t outlen) { struct efx_mcdi_iface *mcdi = efx_mcdi(efx); - unsigned int pdu = MCDI_PDU(efx); + unsigned int pdu = FR_CZ_MC_TREG_SMEM + MCDI_PDU(efx); int i; BUG_ON(atomic_read(&mcdi->state) == MCDI_STATE_QUIESCENT); BUG_ON(outlen & 3 || outlen >= 0x100); for (i = 0; i < outlen; i += 4) - efx_mcdi_readd(efx, (efx_dword_t *)(outbuf + i), pdu + 4 + i); + *((__le32 *)(outbuf + i)) = _efx_readd(efx, pdu + 4 + i); } static int efx_mcdi_poll(struct efx_nic *efx) @@ -135,7 +122,7 @@ static int efx_mcdi_poll(struct efx_nic *efx) struct efx_mcdi_iface *mcdi = efx_mcdi(efx); unsigned int time, finish; unsigned int respseq, respcmd, error; - unsigned int pdu = MCDI_PDU(efx); + unsigned int pdu = FR_CZ_MC_TREG_SMEM + MCDI_PDU(efx); unsigned int rc, spins; efx_dword_t reg; @@ -161,7 +148,8 @@ static int efx_mcdi_poll(struct efx_nic *efx) time = get_seconds(); - efx_mcdi_readd(efx, ®, pdu); + rmb(); + efx_readd(efx, ®, pdu); /* All 1's indicates that shared memory is in reset (and is * not a valid header). Wait for it to come out reset before @@ -188,7 +176,7 @@ static int efx_mcdi_poll(struct efx_nic *efx) respseq, mcdi->seqno); rc = EIO; } else if (error) { - efx_mcdi_readd(efx, ®, pdu + 4); + efx_readd(efx, ®, pdu + 4); switch (EFX_DWORD_FIELD(reg, EFX_DWORD_0)) { #define TRANSLATE_ERROR(name) \ case MC_CMD_ERR_ ## name: \ @@ -222,21 +210,21 @@ static int efx_mcdi_poll(struct efx_nic *efx) /* Test and clear MC-rebooted flag for this port/function */ int efx_mcdi_poll_reboot(struct efx_nic *efx) { - unsigned int addr = MCDI_REBOOT_FLAG(efx); + unsigned int addr = FR_CZ_MC_TREG_SMEM + MCDI_REBOOT_FLAG(efx); efx_dword_t reg; uint32_t value; if (efx_nic_rev(efx) < EFX_REV_SIENA_A0) return false; - efx_mcdi_readd(efx, ®, addr); + efx_readd(efx, ®, addr); value = EFX_DWORD_FIELD(reg, EFX_DWORD_0); if (value == 0) return 0; EFX_ZERO_DWORD(reg); - efx_mcdi_writed(efx, ®, addr); + efx_writed(efx, ®, addr); if (value == MC_STATUS_DWORD_ASSERT) return -EINTR; diff --git a/drivers/net/sfc/nic.c b/drivers/net/sfc/nic.c index bafa23a6874c..3edfbaf5f022 100644 --- a/drivers/net/sfc/nic.c +++ b/drivers/net/sfc/nic.c @@ -1936,13 +1936,6 @@ void efx_nic_get_regs(struct efx_nic *efx, void *buf) size = min_t(size_t, table->step, 16); - if (table->offset >= efx->type->mem_map_size) { - /* No longer mapped; return dummy data */ - memcpy(buf, "\xde\xc0\xad\xde", 4); - buf += table->rows * size; - continue; - } - for (i = 0; i < table->rows; i++) { switch (table->step) { case 4: /* 32-bit register or SRAM */ diff --git a/drivers/net/sfc/nic.h b/drivers/net/sfc/nic.h index 4bd1f2839dfe..7443f99c977f 100644 --- a/drivers/net/sfc/nic.h +++ b/drivers/net/sfc/nic.h @@ -143,12 +143,10 @@ static inline struct falcon_board *falcon_board(struct efx_nic *efx) /** * struct siena_nic_data - Siena NIC state * @mcdi: Management-Controller-to-Driver Interface - * @mcdi_smem: MCDI shared memory mapping. The mapping is always uncacheable. * @wol_filter_id: Wake-on-LAN packet filter id */ struct siena_nic_data { struct efx_mcdi_iface mcdi; - void __iomem *mcdi_smem; int wol_filter_id; }; diff --git a/drivers/net/sfc/siena.c b/drivers/net/sfc/siena.c index 5735e84c69de..2c3bd93fab54 100644 --- a/drivers/net/sfc/siena.c +++ b/drivers/net/sfc/siena.c @@ -250,26 +250,12 @@ static int siena_probe_nic(struct efx_nic *efx) efx_reado(efx, ®, FR_AZ_CS_DEBUG); efx->net_dev->dev_id = EFX_OWORD_FIELD(reg, FRF_CZ_CS_PORT_NUM) - 1; - /* Initialise MCDI */ - nic_data->mcdi_smem = ioremap_nocache(efx->membase_phys + - FR_CZ_MC_TREG_SMEM, - FR_CZ_MC_TREG_SMEM_STEP * - FR_CZ_MC_TREG_SMEM_ROWS); - if (!nic_data->mcdi_smem) { - netif_err(efx, probe, efx->net_dev, - "could not map MCDI at %llx+%x\n", - (unsigned long long)efx->membase_phys + - FR_CZ_MC_TREG_SMEM, - FR_CZ_MC_TREG_SMEM_STEP * FR_CZ_MC_TREG_SMEM_ROWS); - rc = -ENOMEM; - goto fail1; - } efx_mcdi_init(efx); /* Recover from a failed assertion before probing */ rc = efx_mcdi_handle_assertion(efx); if (rc) - goto fail2; + goto fail1; /* Let the BMC know that the driver is now in charge of link and * filter settings. We must do this before we reset the NIC */ @@ -324,7 +310,6 @@ static int siena_probe_nic(struct efx_nic *efx) fail3: efx_mcdi_drv_attach(efx, false, NULL); fail2: - iounmap(nic_data->mcdi_smem); fail1: kfree(efx->nic_data); return rc; @@ -404,8 +389,6 @@ static int siena_init_nic(struct efx_nic *efx) static void siena_remove_nic(struct efx_nic *efx) { - struct siena_nic_data *nic_data = efx->nic_data; - efx_nic_free_buffer(efx, &efx->irq_status); siena_reset_hw(efx, RESET_TYPE_ALL); @@ -415,8 +398,7 @@ static void siena_remove_nic(struct efx_nic *efx) efx_mcdi_drv_attach(efx, false, NULL); /* Tear down the private nic state */ - iounmap(nic_data->mcdi_smem); - kfree(nic_data); + kfree(efx->nic_data); efx->nic_data = NULL; } @@ -656,7 +638,8 @@ const struct efx_nic_type siena_a0_nic_type = { .default_mac_ops = &efx_mcdi_mac_operations, .revision = EFX_REV_SIENA_A0, - .mem_map_size = FR_CZ_MC_TREG_SMEM, /* MC_TREG_SMEM mapped separately */ + .mem_map_size = (FR_CZ_MC_TREG_SMEM + + FR_CZ_MC_TREG_SMEM_STEP * FR_CZ_MC_TREG_SMEM_ROWS), .txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL, .rxd_ptr_tbl_base = FR_BZ_RX_DESC_PTR_TBL, .buf_tbl_base = FR_BZ_BUF_FULL_TBL, diff --git a/drivers/net/sfc/workarounds.h b/drivers/net/sfc/workarounds.h index 99ff11400cef..e4dd3a7f304b 100644 --- a/drivers/net/sfc/workarounds.h +++ b/drivers/net/sfc/workarounds.h @@ -38,8 +38,6 @@ #define EFX_WORKAROUND_15783 EFX_WORKAROUND_ALWAYS /* Legacy interrupt storm when interrupt fifo fills */ #define EFX_WORKAROUND_17213 EFX_WORKAROUND_SIENA -/* Write combining and sriov=enabled are incompatible */ -#define EFX_WORKAROUND_22643 EFX_WORKAROUND_SIENA /* Spurious parity errors in TSORT buffers */ #define EFX_WORKAROUND_5129 EFX_WORKAROUND_FALCON_A From 483f97f8b2b7f0ab09e14c06fe327d5e346fac28 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 1 Sep 2011 12:09:59 +0000 Subject: [PATCH 289/326] sfc: Use 64-bit writes for TX push where possible This was originally done as part of commit 65f0b417dee94f779ce9b77102b7d73c93723b39 ("sfc: Use write-combining to reduce TX latency"), but that had to be reverted. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- drivers/net/sfc/io.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/sfc/io.h b/drivers/net/sfc/io.h index dc45110b2456..751d1ec112cc 100644 --- a/drivers/net/sfc/io.h +++ b/drivers/net/sfc/io.h @@ -48,9 +48,9 @@ * replacing the low 96 bits with zero does not affect functionality. * - If the host writes to the last dword address of such a register * (i.e. the high 32 bits) the underlying register will always be - * written. If the collector does not hold values for the low 96 - * bits of the register, they will be written as zero. Writing to - * the last qword does not have this effect and must not be done. + * written. If the collector and the current write together do not + * provide values for all 128 bits of the register, the low 96 bits + * will be written as zero. * - If the host writes to the address of any other part of such a * register while the collector already holds values for some other * register, the write is discarded and the collector maintains its @@ -237,12 +237,13 @@ static inline void _efx_writeo_page(struct efx_nic *efx, efx_oword_t *value, #ifdef EFX_USE_QWORD_IO _efx_writeq(efx, value->u64[0], reg + 0); + _efx_writeq(efx, value->u64[1], reg + 8); #else _efx_writed(efx, value->u32[0], reg + 0); _efx_writed(efx, value->u32[1], reg + 4); -#endif _efx_writed(efx, value->u32[2], reg + 8); _efx_writed(efx, value->u32[3], reg + 12); +#endif } #define efx_writeo_page(efx, value, reg, page) \ _efx_writeo_page(efx, value, \ From 5229d87edcd80a3bceb0708ebd767faff2e589a9 Mon Sep 17 00:00:00 2001 From: Toshiharu Okada Date: Thu, 1 Sep 2011 14:20:07 +0000 Subject: [PATCH 290/326] pch_gbe: fixed the issue which receives an unnecessary packet. This patch fixed the issue which receives an unnecessary packet before link When using PHY of GMII, an unnecessary packet is received, And it becomes impossible to receive a packet after link up. Signed-off-by: Toshiharu Okada Signed-off-by: David S. Miller --- drivers/net/pch_gbe/pch_gbe_main.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/net/pch_gbe/pch_gbe_main.c b/drivers/net/pch_gbe/pch_gbe_main.c index eac3c5ca9731..48ff87c455ae 100644 --- a/drivers/net/pch_gbe/pch_gbe_main.c +++ b/drivers/net/pch_gbe/pch_gbe_main.c @@ -717,13 +717,6 @@ static void pch_gbe_configure_rx(struct pch_gbe_adapter *adapter) iowrite32(rdba, &hw->reg->RX_DSC_BASE); iowrite32(rdlen, &hw->reg->RX_DSC_SIZE); iowrite32((rdba + rdlen), &hw->reg->RX_DSC_SW_P); - - /* Enables Receive DMA */ - rxdma = ioread32(&hw->reg->DMA_CTRL); - rxdma |= PCH_GBE_RX_DMA_EN; - iowrite32(rxdma, &hw->reg->DMA_CTRL); - /* Enables Receive */ - iowrite32(PCH_GBE_MRE_MAC_RX_EN, &hw->reg->MAC_RX_EN); } /** @@ -1097,6 +1090,19 @@ void pch_gbe_update_stats(struct pch_gbe_adapter *adapter) spin_unlock_irqrestore(&adapter->stats_lock, flags); } +static void pch_gbe_start_receive(struct pch_gbe_hw *hw) +{ + u32 rxdma; + + /* Enables Receive DMA */ + rxdma = ioread32(&hw->reg->DMA_CTRL); + rxdma |= PCH_GBE_RX_DMA_EN; + iowrite32(rxdma, &hw->reg->DMA_CTRL); + /* Enables Receive */ + iowrite32(PCH_GBE_MRE_MAC_RX_EN, &hw->reg->MAC_RX_EN); + return; +} + /** * pch_gbe_intr - Interrupt Handler * @irq: Interrupt number @@ -1717,6 +1723,7 @@ int pch_gbe_up(struct pch_gbe_adapter *adapter) pch_gbe_alloc_tx_buffers(adapter, tx_ring); pch_gbe_alloc_rx_buffers(adapter, rx_ring, rx_ring->count); adapter->tx_queue_len = netdev->tx_queue_len; + pch_gbe_start_receive(&adapter->hw); mod_timer(&adapter->watchdog_timer, jiffies); From 124d770a6459be21b84445f6ebf7dbfb60d43585 Mon Sep 17 00:00:00 2001 From: Toshiharu Okada Date: Thu, 1 Sep 2011 14:20:08 +0000 Subject: [PATCH 291/326] pch_gbe: added the process of FIFO over run error This patch added the processing which should be done to hardware, when a FIFO over run error occurred. Signed-off-by: Toshiharu Okada Signed-off-by: David S. Miller --- drivers/net/pch_gbe/pch_gbe.h | 12 +- drivers/net/pch_gbe/pch_gbe_main.c | 269 ++++++++++++++++++----------- 2 files changed, 178 insertions(+), 103 deletions(-) diff --git a/drivers/net/pch_gbe/pch_gbe.h b/drivers/net/pch_gbe/pch_gbe.h index 59fac77d0dbb..a09a07197eb5 100644 --- a/drivers/net/pch_gbe/pch_gbe.h +++ b/drivers/net/pch_gbe/pch_gbe.h @@ -127,8 +127,8 @@ struct pch_gbe_regs { /* Reset */ #define PCH_GBE_ALL_RST 0x80000000 /* All reset */ -#define PCH_GBE_TX_RST 0x40000000 /* TX MAC, TX FIFO, TX DMA reset */ -#define PCH_GBE_RX_RST 0x04000000 /* RX MAC, RX FIFO, RX DMA reset */ +#define PCH_GBE_TX_RST 0x00008000 /* TX MAC, TX FIFO, TX DMA reset */ +#define PCH_GBE_RX_RST 0x00004000 /* RX MAC, RX FIFO, RX DMA reset */ /* TCP/IP Accelerator Control */ #define PCH_GBE_EX_LIST_EN 0x00000008 /* External List Enable */ @@ -276,6 +276,9 @@ struct pch_gbe_regs { #define PCH_GBE_RX_DMA_EN 0x00000002 /* Enables Receive DMA */ #define PCH_GBE_TX_DMA_EN 0x00000001 /* Enables Transmission DMA */ +/* RX DMA STATUS */ +#define PCH_GBE_IDLE_CHECK 0xFFFFFFFE + /* Wake On LAN Status */ #define PCH_GBE_WLS_BR 0x00000008 /* Broadcas Address */ #define PCH_GBE_WLS_MLT 0x00000004 /* Multicast Address */ @@ -471,6 +474,7 @@ struct pch_gbe_tx_desc { struct pch_gbe_buffer { struct sk_buff *skb; dma_addr_t dma; + unsigned char *rx_buffer; unsigned long time_stamp; u16 length; bool mapped; @@ -511,6 +515,9 @@ struct pch_gbe_tx_ring { struct pch_gbe_rx_ring { struct pch_gbe_rx_desc *desc; dma_addr_t dma; + unsigned char *rx_buff_pool; + dma_addr_t rx_buff_pool_logic; + unsigned int rx_buff_pool_size; unsigned int size; unsigned int count; unsigned int next_to_use; @@ -622,6 +629,7 @@ struct pch_gbe_adapter { unsigned long rx_buffer_len; unsigned long tx_queue_len; bool have_msi; + bool rx_stop_flag; }; extern const char pch_driver_version[]; diff --git a/drivers/net/pch_gbe/pch_gbe_main.c b/drivers/net/pch_gbe/pch_gbe_main.c index 48ff87c455ae..39ce0ee44ad7 100644 --- a/drivers/net/pch_gbe/pch_gbe_main.c +++ b/drivers/net/pch_gbe/pch_gbe_main.c @@ -20,7 +20,6 @@ #include "pch_gbe.h" #include "pch_gbe_api.h" -#include #define DRV_VERSION "1.00" const char pch_driver_version[] = DRV_VERSION; @@ -34,6 +33,7 @@ const char pch_driver_version[] = DRV_VERSION; #define PCH_GBE_WATCHDOG_PERIOD (1 * HZ) /* watchdog time */ #define PCH_GBE_COPYBREAK_DEFAULT 256 #define PCH_GBE_PCI_BAR 1 +#define PCH_GBE_RESERVE_MEMORY 0x200000 /* 2MB */ /* Macros for ML7223 */ #define PCI_VENDOR_ID_ROHM 0x10db @@ -52,6 +52,7 @@ const char pch_driver_version[] = DRV_VERSION; ) /* Ethertype field values */ +#define PCH_GBE_MAX_RX_BUFFER_SIZE 0x2880 #define PCH_GBE_MAX_JUMBO_FRAME_SIZE 10318 #define PCH_GBE_FRAME_SIZE_2048 2048 #define PCH_GBE_FRAME_SIZE_4096 4096 @@ -83,10 +84,12 @@ const char pch_driver_version[] = DRV_VERSION; #define PCH_GBE_INT_ENABLE_MASK ( \ PCH_GBE_INT_RX_DMA_CMPLT | \ PCH_GBE_INT_RX_DSC_EMP | \ + PCH_GBE_INT_RX_FIFO_ERR | \ PCH_GBE_INT_WOL_DET | \ PCH_GBE_INT_TX_CMPLT \ ) +#define PCH_GBE_INT_DISABLE_ALL 0 static unsigned int copybreak __read_mostly = PCH_GBE_COPYBREAK_DEFAULT; @@ -138,6 +141,27 @@ static void pch_gbe_wait_clr_bit(void *reg, u32 bit) if (!tmp) pr_err("Error: busy bit is not cleared\n"); } + +/** + * pch_gbe_wait_clr_bit_irq - Wait to clear a bit for interrupt context + * @reg: Pointer of register + * @busy: Busy bit + */ +static int pch_gbe_wait_clr_bit_irq(void *reg, u32 bit) +{ + u32 tmp; + int ret = -1; + /* wait busy */ + tmp = 20; + while ((ioread32(reg) & bit) && --tmp) + udelay(5); + if (!tmp) + pr_err("Error: busy bit is not cleared\n"); + else + ret = 0; + return ret; +} + /** * pch_gbe_mac_mar_set - Set MAC address register * @hw: Pointer to the HW structure @@ -189,6 +213,17 @@ static void pch_gbe_mac_reset_hw(struct pch_gbe_hw *hw) return; } +static void pch_gbe_mac_reset_rx(struct pch_gbe_hw *hw) +{ + /* Read the MAC address. and store to the private data */ + pch_gbe_mac_read_mac_addr(hw); + iowrite32(PCH_GBE_RX_RST, &hw->reg->RESET); + pch_gbe_wait_clr_bit_irq(&hw->reg->RESET, PCH_GBE_RX_RST); + /* Setup the MAC address */ + pch_gbe_mac_mar_set(hw, hw->mac.addr, 0); + return; +} + /** * pch_gbe_mac_init_rx_addrs - Initialize receive address's * @hw: Pointer to the HW structure @@ -671,13 +706,8 @@ static void pch_gbe_setup_rctl(struct pch_gbe_adapter *adapter) tcpip = ioread32(&hw->reg->TCPIP_ACC); - if (netdev->features & NETIF_F_RXCSUM) { - tcpip &= ~PCH_GBE_RX_TCPIPACC_OFF; - tcpip |= PCH_GBE_RX_TCPIPACC_EN; - } else { - tcpip |= PCH_GBE_RX_TCPIPACC_OFF; - tcpip &= ~PCH_GBE_RX_TCPIPACC_EN; - } + tcpip |= PCH_GBE_RX_TCPIPACC_OFF; + tcpip &= ~PCH_GBE_RX_TCPIPACC_EN; iowrite32(tcpip, &hw->reg->TCPIP_ACC); return; } @@ -1090,6 +1120,35 @@ void pch_gbe_update_stats(struct pch_gbe_adapter *adapter) spin_unlock_irqrestore(&adapter->stats_lock, flags); } +static void pch_gbe_stop_receive(struct pch_gbe_adapter *adapter) +{ + struct pch_gbe_hw *hw = &adapter->hw; + u32 rxdma; + u16 value; + int ret; + + /* Disable Receive DMA */ + rxdma = ioread32(&hw->reg->DMA_CTRL); + rxdma &= ~PCH_GBE_RX_DMA_EN; + iowrite32(rxdma, &hw->reg->DMA_CTRL); + /* Wait Rx DMA BUS is IDLE */ + ret = pch_gbe_wait_clr_bit_irq(&hw->reg->RX_DMA_ST, PCH_GBE_IDLE_CHECK); + if (ret) { + /* Disable Bus master */ + pci_read_config_word(adapter->pdev, PCI_COMMAND, &value); + value &= ~PCI_COMMAND_MASTER; + pci_write_config_word(adapter->pdev, PCI_COMMAND, value); + /* Stop Receive */ + pch_gbe_mac_reset_rx(hw); + /* Enable Bus master */ + value |= PCI_COMMAND_MASTER; + pci_write_config_word(adapter->pdev, PCI_COMMAND, value); + } else { + /* Stop Receive */ + pch_gbe_mac_reset_rx(hw); + } +} + static void pch_gbe_start_receive(struct pch_gbe_hw *hw) { u32 rxdma; @@ -1129,7 +1188,15 @@ static irqreturn_t pch_gbe_intr(int irq, void *data) if (int_st & PCH_GBE_INT_RX_FRAME_ERR) adapter->stats.intr_rx_frame_err_count++; if (int_st & PCH_GBE_INT_RX_FIFO_ERR) - adapter->stats.intr_rx_fifo_err_count++; + if (!adapter->rx_stop_flag) { + adapter->stats.intr_rx_fifo_err_count++; + pr_debug("Rx fifo over run\n"); + adapter->rx_stop_flag = true; + int_en = ioread32(&hw->reg->INT_EN); + iowrite32((int_en & ~PCH_GBE_INT_RX_FIFO_ERR), + &hw->reg->INT_EN); + pch_gbe_stop_receive(adapter); + } if (int_st & PCH_GBE_INT_RX_DMA_ERR) adapter->stats.intr_rx_dma_err_count++; if (int_st & PCH_GBE_INT_TX_FIFO_ERR) @@ -1141,7 +1208,7 @@ static irqreturn_t pch_gbe_intr(int irq, void *data) /* When Rx descriptor is empty */ if ((int_st & PCH_GBE_INT_RX_DSC_EMP)) { adapter->stats.intr_rx_dsc_empty_count++; - pr_err("Rx descriptor is empty\n"); + pr_debug("Rx descriptor is empty\n"); int_en = ioread32(&hw->reg->INT_EN); iowrite32((int_en & ~PCH_GBE_INT_RX_DSC_EMP), &hw->reg->INT_EN); if (hw->mac.tx_fc_enable) { @@ -1191,29 +1258,23 @@ pch_gbe_alloc_rx_buffers(struct pch_gbe_adapter *adapter, unsigned int i; unsigned int bufsz; - bufsz = adapter->rx_buffer_len + PCH_GBE_DMA_ALIGN; + bufsz = adapter->rx_buffer_len + NET_IP_ALIGN; i = rx_ring->next_to_use; while ((cleaned_count--)) { buffer_info = &rx_ring->buffer_info[i]; - skb = buffer_info->skb; - if (skb) { - skb_trim(skb, 0); - } else { - skb = netdev_alloc_skb(netdev, bufsz); - if (unlikely(!skb)) { - /* Better luck next round */ - adapter->stats.rx_alloc_buff_failed++; - break; - } - /* 64byte align */ - skb_reserve(skb, PCH_GBE_DMA_ALIGN); - - buffer_info->skb = skb; - buffer_info->length = adapter->rx_buffer_len; + skb = netdev_alloc_skb(netdev, bufsz); + if (unlikely(!skb)) { + /* Better luck next round */ + adapter->stats.rx_alloc_buff_failed++; + break; } + /* align */ + skb_reserve(skb, NET_IP_ALIGN); + buffer_info->skb = skb; + buffer_info->dma = dma_map_single(&pdev->dev, - skb->data, + buffer_info->rx_buffer, buffer_info->length, DMA_FROM_DEVICE); if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)) { @@ -1246,6 +1307,36 @@ pch_gbe_alloc_rx_buffers(struct pch_gbe_adapter *adapter, return; } +static int +pch_gbe_alloc_rx_buffers_pool(struct pch_gbe_adapter *adapter, + struct pch_gbe_rx_ring *rx_ring, int cleaned_count) +{ + struct pci_dev *pdev = adapter->pdev; + struct pch_gbe_buffer *buffer_info; + unsigned int i; + unsigned int bufsz; + unsigned int size; + + bufsz = adapter->rx_buffer_len; + + size = rx_ring->count * bufsz + PCH_GBE_RESERVE_MEMORY; + rx_ring->rx_buff_pool = dma_alloc_coherent(&pdev->dev, size, + &rx_ring->rx_buff_pool_logic, + GFP_KERNEL); + if (!rx_ring->rx_buff_pool) { + pr_err("Unable to allocate memory for the receive poll buffer\n"); + return -ENOMEM; + } + memset(rx_ring->rx_buff_pool, 0, size); + rx_ring->rx_buff_pool_size = size; + for (i = 0; i < rx_ring->count; i++) { + buffer_info = &rx_ring->buffer_info[i]; + buffer_info->rx_buffer = rx_ring->rx_buff_pool + bufsz * i; + buffer_info->length = bufsz; + } + return 0; +} + /** * pch_gbe_alloc_tx_buffers - Allocate transmit buffers * @adapter: Board private structure @@ -1386,7 +1477,7 @@ pch_gbe_clean_rx(struct pch_gbe_adapter *adapter, unsigned int i; unsigned int cleaned_count = 0; bool cleaned = false; - struct sk_buff *skb, *new_skb; + struct sk_buff *skb; u8 dma_status; u16 gbec_status; u32 tcp_ip_status; @@ -1407,13 +1498,12 @@ pch_gbe_clean_rx(struct pch_gbe_adapter *adapter, rx_desc->gbec_status = DSC_INIT16; buffer_info = &rx_ring->buffer_info[i]; skb = buffer_info->skb; + buffer_info->skb = NULL; /* unmap dma */ dma_unmap_single(&pdev->dev, buffer_info->dma, buffer_info->length, DMA_FROM_DEVICE); buffer_info->mapped = false; - /* Prefetch the packet */ - prefetch(skb->data); pr_debug("RxDecNo = 0x%04x Status[DMA:0x%02x GBE:0x%04x " "TCP:0x%08x] BufInf = 0x%p\n", @@ -1433,70 +1523,16 @@ pch_gbe_clean_rx(struct pch_gbe_adapter *adapter, pr_err("Receive CRC Error\n"); } else { /* get receive length */ - /* length convert[-3] */ - length = (rx_desc->rx_words_eob) - 3; + /* length convert[-3], length includes FCS length */ + length = (rx_desc->rx_words_eob) - 3 - ETH_FCS_LEN; + if (rx_desc->rx_words_eob & 0x02) + length = length - 4; + /* + * buffer_info->rx_buffer: [Header:14][payload] + * skb->data: [Reserve:2][Header:14][payload] + */ + memcpy(skb->data, buffer_info->rx_buffer, length); - /* Decide the data conversion method */ - if (!(netdev->features & NETIF_F_RXCSUM)) { - /* [Header:14][payload] */ - if (NET_IP_ALIGN) { - /* Because alignment differs, - * the new_skb is newly allocated, - * and data is copied to new_skb.*/ - new_skb = netdev_alloc_skb(netdev, - length + NET_IP_ALIGN); - if (!new_skb) { - /* dorrop error */ - pr_err("New skb allocation " - "Error\n"); - goto dorrop; - } - skb_reserve(new_skb, NET_IP_ALIGN); - memcpy(new_skb->data, skb->data, - length); - skb = new_skb; - } else { - /* DMA buffer is used as SKB as it is.*/ - buffer_info->skb = NULL; - } - } else { - /* [Header:14][padding:2][payload] */ - /* The length includes padding length */ - length = length - PCH_GBE_DMA_PADDING; - if ((length < copybreak) || - (NET_IP_ALIGN != PCH_GBE_DMA_PADDING)) { - /* Because alignment differs, - * the new_skb is newly allocated, - * and data is copied to new_skb. - * Padding data is deleted - * at the time of a copy.*/ - new_skb = netdev_alloc_skb(netdev, - length + NET_IP_ALIGN); - if (!new_skb) { - /* dorrop error */ - pr_err("New skb allocation " - "Error\n"); - goto dorrop; - } - skb_reserve(new_skb, NET_IP_ALIGN); - memcpy(new_skb->data, skb->data, - ETH_HLEN); - memcpy(&new_skb->data[ETH_HLEN], - &skb->data[ETH_HLEN + - PCH_GBE_DMA_PADDING], - length - ETH_HLEN); - skb = new_skb; - } else { - /* Padding data is deleted - * by moving header data.*/ - memmove(&skb->data[PCH_GBE_DMA_PADDING], - &skb->data[0], ETH_HLEN); - skb_reserve(skb, NET_IP_ALIGN); - buffer_info->skb = NULL; - } - } - /* The length includes FCS length */ - length = length - ETH_FCS_LEN; /* update status of driver */ adapter->stats.rx_bytes += length; adapter->stats.rx_packets++; @@ -1515,7 +1551,6 @@ pch_gbe_clean_rx(struct pch_gbe_adapter *adapter, pr_debug("Receive skb->ip_summed: %d length: %d\n", skb->ip_summed, length); } -dorrop: /* return some buffers to hardware, one at a time is too slow */ if (unlikely(cleaned_count >= PCH_GBE_RX_BUFFER_WRITE)) { pch_gbe_alloc_rx_buffers(adapter, rx_ring, @@ -1720,6 +1755,11 @@ int pch_gbe_up(struct pch_gbe_adapter *adapter) pr_err("Error: can't bring device up\n"); return err; } + err = pch_gbe_alloc_rx_buffers_pool(adapter, rx_ring, rx_ring->count); + if (err) { + pr_err("Error: can't bring device up\n"); + return err; + } pch_gbe_alloc_tx_buffers(adapter, tx_ring); pch_gbe_alloc_rx_buffers(adapter, rx_ring, rx_ring->count); adapter->tx_queue_len = netdev->tx_queue_len; @@ -1741,6 +1781,7 @@ int pch_gbe_up(struct pch_gbe_adapter *adapter) void pch_gbe_down(struct pch_gbe_adapter *adapter) { struct net_device *netdev = adapter->netdev; + struct pch_gbe_rx_ring *rx_ring = adapter->rx_ring; /* signal that we're down so the interrupt handler does not * reschedule our watchdog timer */ @@ -1759,6 +1800,12 @@ void pch_gbe_down(struct pch_gbe_adapter *adapter) pch_gbe_reset(adapter); pch_gbe_clean_tx_ring(adapter, adapter->tx_ring); pch_gbe_clean_rx_ring(adapter, adapter->rx_ring); + + pci_free_consistent(adapter->pdev, rx_ring->rx_buff_pool_size, + rx_ring->rx_buff_pool, rx_ring->rx_buff_pool_logic); + rx_ring->rx_buff_pool_logic = 0; + rx_ring->rx_buff_pool_size = 0; + rx_ring->rx_buff_pool = NULL; } /** @@ -2011,6 +2058,8 @@ static int pch_gbe_change_mtu(struct net_device *netdev, int new_mtu) { struct pch_gbe_adapter *adapter = netdev_priv(netdev); int max_frame; + unsigned long old_rx_buffer_len = adapter->rx_buffer_len; + int err; max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN; if ((max_frame < ETH_ZLEN + ETH_FCS_LEN) || @@ -2025,14 +2074,24 @@ static int pch_gbe_change_mtu(struct net_device *netdev, int new_mtu) else if (max_frame <= PCH_GBE_FRAME_SIZE_8192) adapter->rx_buffer_len = PCH_GBE_FRAME_SIZE_8192; else - adapter->rx_buffer_len = PCH_GBE_MAX_JUMBO_FRAME_SIZE; - netdev->mtu = new_mtu; - adapter->hw.mac.max_frame_size = max_frame; + adapter->rx_buffer_len = PCH_GBE_MAX_RX_BUFFER_SIZE; - if (netif_running(netdev)) - pch_gbe_reinit_locked(adapter); - else + if (netif_running(netdev)) { + pch_gbe_down(adapter); + err = pch_gbe_up(adapter); + if (err) { + adapter->rx_buffer_len = old_rx_buffer_len; + pch_gbe_up(adapter); + return -ENOMEM; + } else { + netdev->mtu = new_mtu; + adapter->hw.mac.max_frame_size = max_frame; + } + } else { pch_gbe_reset(adapter); + netdev->mtu = new_mtu; + adapter->hw.mac.max_frame_size = max_frame; + } pr_debug("max_frame : %d rx_buffer_len : %d mtu : %d max_frame_size : %d\n", max_frame, (u32) adapter->rx_buffer_len, netdev->mtu, @@ -2110,6 +2169,7 @@ static int pch_gbe_napi_poll(struct napi_struct *napi, int budget) int work_done = 0; bool poll_end_flag = false; bool cleaned = false; + u32 int_en; pr_debug("budget : %d\n", budget); @@ -2117,8 +2177,15 @@ static int pch_gbe_napi_poll(struct napi_struct *napi, int budget) if (!netif_carrier_ok(netdev)) { poll_end_flag = true; } else { - cleaned = pch_gbe_clean_tx(adapter, adapter->tx_ring); pch_gbe_clean_rx(adapter, adapter->rx_ring, &work_done, budget); + if (adapter->rx_stop_flag) { + adapter->rx_stop_flag = false; + pch_gbe_start_receive(&adapter->hw); + int_en = ioread32(&adapter->hw.reg->INT_EN); + iowrite32((int_en | PCH_GBE_INT_RX_FIFO_ERR), + &adapter->hw.reg->INT_EN); + } + cleaned = pch_gbe_clean_tx(adapter, adapter->tx_ring); if (cleaned) work_done = budget; From 7756332f5b64c9c1535712b9679792e8bd4f0019 Mon Sep 17 00:00:00 2001 From: Toshiharu Okada Date: Thu, 1 Sep 2011 14:20:09 +0000 Subject: [PATCH 292/326] pch_gbe: support ML7831 IOH Support new device OKI SEMICONDUCTOR ML7831 IOH(Input/Output Hub) ML7831 is for general purpose use. ML7831 is companion chip for Intel Atom E6xx series. ML7831 is completely compatible for Intel EG20T PCH. Signed-off-by: Toshiharu Okada Signed-off-by: David S. Miller --- drivers/net/Kconfig | 11 ++++++----- drivers/net/pch_gbe/pch_gbe_main.c | 10 ++++++++++ 2 files changed, 16 insertions(+), 5 deletions(-) diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 8d0314dbd946..a44874e24f2a 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2535,7 +2535,7 @@ config S6GMAC source "drivers/net/stmmac/Kconfig" config PCH_GBE - tristate "Intel EG20T PCH / OKI SEMICONDUCTOR ML7223 IOH GbE" + tristate "Intel EG20T PCH/OKI SEMICONDUCTOR IOH(ML7223/ML7831) GbE" depends on PCI select MII ---help--- @@ -2548,10 +2548,11 @@ config PCH_GBE This driver enables Gigabit Ethernet function. This driver also can be used for OKI SEMICONDUCTOR IOH(Input/ - Output Hub), ML7223. - ML7223 IOH is for MP(Media Phone) use. - ML7223 is companion chip for Intel Atom E6xx series. - ML7223 is completely compatible for Intel EG20T PCH. + Output Hub), ML7223/ML7831. + ML7223 IOH is for MP(Media Phone) use. ML7831 IOH is for general + purpose use. + ML7223/ML7831 is companion chip for Intel Atom E6xx series. + ML7223/ML7831 is completely compatible for Intel EG20T PCH. config FTGMAC100 tristate "Faraday FTGMAC100 Gigabit Ethernet support" diff --git a/drivers/net/pch_gbe/pch_gbe_main.c b/drivers/net/pch_gbe/pch_gbe_main.c index 39ce0ee44ad7..567ff10889be 100644 --- a/drivers/net/pch_gbe/pch_gbe_main.c +++ b/drivers/net/pch_gbe/pch_gbe_main.c @@ -39,6 +39,9 @@ const char pch_driver_version[] = DRV_VERSION; #define PCI_VENDOR_ID_ROHM 0x10db #define PCI_DEVICE_ID_ROHM_ML7223_GBE 0x8013 +/* Macros for ML7831 */ +#define PCI_DEVICE_ID_ROHM_ML7831_GBE 0x8802 + #define PCH_GBE_TX_WEIGHT 64 #define PCH_GBE_RX_WEIGHT 64 #define PCH_GBE_RX_BUFFER_WRITE 16 @@ -2526,6 +2529,13 @@ static DEFINE_PCI_DEVICE_TABLE(pch_gbe_pcidev_id) = { .class = (PCI_CLASS_NETWORK_ETHERNET << 8), .class_mask = (0xFFFF00) }, + {.vendor = PCI_VENDOR_ID_ROHM, + .device = PCI_DEVICE_ID_ROHM_ML7831_GBE, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .class = (PCI_CLASS_NETWORK_ETHERNET << 8), + .class_mask = (0xFFFF00) + }, /* required last entry */ {0} }; From bcac364a24c894c4cf8cf219b7863c192cd34079 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sat, 27 Aug 2011 21:33:16 -0700 Subject: [PATCH 293/326] target: Fix race between multiple invocations of target_qf_do_work() When work is scheduled with schedule_work(), the work can end up running on multiple CPUs at the same time -- this happens if the work is already running on one CPU and schedule_work() is called on another CPU. This leads to list corruption with target_qf_do_work(), which is roughly doing: spin_lock(...); list_for_each_entry_safe(...) { list_del(...); spin_unlock(...); // do stuff spin_lock(...); } With multiple CPUs running this code, one CPU can end up deleting the list entry that the other CPU is about to work on. Fix this by splicing the list entries onto a local list and then operating on that in the work function. This way, each invocation of target_qf_do_work() operates on its own local list and so multiple invocations don't corrupt each other's list. This also avoids dropping and reacquiring the lock for each list entry. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 8d0c58ea6316..a4b0a8d27f25 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -977,15 +977,17 @@ static void target_qf_do_work(struct work_struct *work) { struct se_device *dev = container_of(work, struct se_device, qf_work_queue); + LIST_HEAD(qf_cmd_list); struct se_cmd *cmd, *cmd_tmp; spin_lock_irq(&dev->qf_cmd_lock); - list_for_each_entry_safe(cmd, cmd_tmp, &dev->qf_cmd_list, se_qf_node) { + list_splice_init(&dev->qf_cmd_list, &qf_cmd_list); + spin_unlock_irq(&dev->qf_cmd_lock); + list_for_each_entry_safe(cmd, cmd_tmp, &qf_cmd_list, se_qf_node) { list_del(&cmd->se_qf_node); atomic_dec(&dev->dev_qf_count); smp_mb__after_atomic_dec(); - spin_unlock_irq(&dev->qf_cmd_lock); pr_debug("Processing %s cmd: %p QUEUE_FULL in work queue" " context: %s\n", cmd->se_tfo->get_fabric_name(), cmd, @@ -997,10 +999,7 @@ static void target_qf_do_work(struct work_struct *work) * has been added to head of queue */ transport_add_cmd_to_queue(cmd, cmd->t_state); - - spin_lock_irq(&dev->qf_cmd_lock); } - spin_unlock_irq(&dev->qf_cmd_lock); } unsigned char *transport_dump_cmd_direction(struct se_cmd *cmd) From 079587b4eb4d3b78a4d65d142f662aa9d7eedab4 Mon Sep 17 00:00:00 2001 From: Kiran Patil Date: Fri, 26 Aug 2011 09:25:25 -0700 Subject: [PATCH 294/326] tcm_fc: Invalidation of DDP context for FCoE target in error conditions Problem: HW DDP context wasn;t invalidated in case of ABORTS, etc... This leads to the problem where memory pages which are used for DDP as user descriptor could get reused for some other purpose (such as to satisfy new memory allocation request either by kernel or user mode threads) and since HW DDP context was not invalidated, HW continue to write to those pages, hence causing memory corruption. Fix: Either on incoming ABORTS or due to exchange time out, allowed the target to cleanup HW DDP context if it was setup for respective ft_cmd. Added new function to perform this cleanup, furthur it can be enhanced for other cleanup activity. Additinal Notes: To avoid calling ddp_done from multiple places, composed the functionality in helper function "ft_invl_hw_context" and it is being called from multiple places. Cleaned up code in function "ft_recv_write_data" w.r.t DDP. Signed-off-by: Kiran Patil Signed-off-by: Nicholas Bellinger --- drivers/target/tcm_fc/tfc_io.c | 62 ++++++++++++++++------------------ 1 file changed, 30 insertions(+), 32 deletions(-) diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c index c37f4cd96452..d35ea5a3d56c 100644 --- a/drivers/target/tcm_fc/tfc_io.c +++ b/drivers/target/tcm_fc/tfc_io.c @@ -219,43 +219,41 @@ void ft_recv_write_data(struct ft_cmd *cmd, struct fc_frame *fp) if (cmd->was_ddp_setup) { BUG_ON(!ep); BUG_ON(!lport); - } - - /* - * Doesn't expect payload if DDP is setup. Payload - * is expected to be copied directly to user buffers - * due to DDP (Large Rx offload), - */ - buf = fc_frame_payload_get(fp, 1); - if (buf) - pr_err("%s: xid 0x%x, f_ctl 0x%x, cmd->sg %p, " + /* + * Since DDP (Large Rx offload) was setup for this request, + * payload is expected to be copied directly to user buffers. + */ + buf = fc_frame_payload_get(fp, 1); + if (buf) + pr_err("%s: xid 0x%x, f_ctl 0x%x, cmd->sg %p, " "cmd->sg_cnt 0x%x. DDP was setup" " hence not expected to receive frame with " - "payload, Frame will be dropped if " - "'Sequence Initiative' bit in f_ctl is " + "payload, Frame will be dropped if" + "'Sequence Initiative' bit in f_ctl is" "not set\n", __func__, ep->xid, f_ctl, cmd->sg, cmd->sg_cnt); - /* - * Invalidate HW DDP context if it was setup for respective - * command. Invalidation of HW DDP context is requited in both - * situation (success and error). - */ - ft_invl_hw_context(cmd); + /* + * Invalidate HW DDP context if it was setup for respective + * command. Invalidation of HW DDP context is requited in both + * situation (success and error). + */ + ft_invl_hw_context(cmd); - /* - * If "Sequence Initiative (TSI)" bit set in f_ctl, means last - * write data frame is received successfully where payload is - * posted directly to user buffer and only the last frame's - * header is posted in receive queue. - * - * If "Sequence Initiative (TSI)" bit is not set, means error - * condition w.r.t. DDP, hence drop the packet and let explict - * ABORTS from other end of exchange timer trigger the recovery. - */ - if (f_ctl & FC_FC_SEQ_INIT) - goto last_frame; - else - goto drop; + /* + * If "Sequence Initiative (TSI)" bit set in f_ctl, means last + * write data frame is received successfully where payload is + * posted directly to user buffer and only the last frame's + * header is posted in receive queue. + * + * If "Sequence Initiative (TSI)" bit is not set, means error + * condition w.r.t. DDP, hence drop the packet and let explict + * ABORTS from other end of exchange timer trigger the recovery. + */ + if (f_ctl & FC_FC_SEQ_INIT) + goto last_frame; + else + goto drop; + } rel_off = ntohl(fh->fh_parm_offset); frame_len = fr_len(fp); From 58fc73d10f3e92bfcd1e9a8391eb3e49b68df8e5 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 26 Aug 2011 09:25:38 -0700 Subject: [PATCH 295/326] tcm_fc: Work queue based approach instead of managing own thread and event based mechanism Problem: Changed from wake_up_interruptible -> wake_up_process and wait_event_interruptible-> schedule_timeout_interruptible broke the FCoE target. Earlier approach of wake_up_interruptible was also looking at 'queue_cnt' which is not necessary, because it increment of 'queue_cnt' with wake_up_inetrriptible / waker_up_process introduces race condition. Fix: Instead of fixing the code which used wake_up_process and remove 'queue_cnt', using work_queue based approach is cleaner and acheives same result. As well, work queue based approach has less programming overhead and OS manages threads which processes work queues. This patch is developed by Christoph Hellwig and reviwed+validated by Kiran Patil. Signed-off-by: Christoph Hellwig Signed-off-by: Kiran Patil Signed-off-by: Nicholas Bellinger --- drivers/target/tcm_fc/tcm_fc.h | 12 +---- drivers/target/tcm_fc/tfc_cmd.c | 90 ++++---------------------------- drivers/target/tcm_fc/tfc_conf.c | 7 ++- 3 files changed, 16 insertions(+), 93 deletions(-) diff --git a/drivers/target/tcm_fc/tcm_fc.h b/drivers/target/tcm_fc/tcm_fc.h index bd4fe21a23b8..3749d8b4b423 100644 --- a/drivers/target/tcm_fc/tcm_fc.h +++ b/drivers/target/tcm_fc/tcm_fc.h @@ -98,8 +98,7 @@ struct ft_tpg { struct list_head list; /* linkage in ft_lport_acl tpg_list */ struct list_head lun_list; /* head of LUNs */ struct se_portal_group se_tpg; - struct task_struct *thread; /* processing thread */ - struct se_queue_obj qobj; /* queue for processing thread */ + struct workqueue_struct *workqueue; }; struct ft_lport_acl { @@ -110,16 +109,10 @@ struct ft_lport_acl { struct se_wwn fc_lport_wwn; }; -enum ft_cmd_state { - FC_CMD_ST_NEW = 0, - FC_CMD_ST_REJ -}; - /* * Commands */ struct ft_cmd { - enum ft_cmd_state state; u32 lun; /* LUN from request */ struct ft_sess *sess; /* session held for cmd */ struct fc_seq *seq; /* sequence in exchange mgr */ @@ -127,7 +120,7 @@ struct ft_cmd { struct fc_frame *req_frame; unsigned char *cdb; /* pointer to CDB inside frame */ u32 write_data_len; /* data received on writes */ - struct se_queue_req se_req; + struct work_struct work; /* Local sense buffer */ unsigned char ft_sense_buffer[TRANSPORT_SENSE_BUFFER]; u32 was_ddp_setup:1; /* Set only if ddp is setup */ @@ -177,7 +170,6 @@ int ft_is_state_remove(struct se_cmd *); /* * other internal functions. */ -int ft_thread(void *); void ft_recv_req(struct ft_sess *, struct fc_frame *); struct ft_tpg *ft_lport_find_tpg(struct fc_lport *); struct ft_node_acl *ft_acl_get(struct ft_tpg *, struct fc_rport_priv *); diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c index 5654dc22f7ae..80fbcde00cb6 100644 --- a/drivers/target/tcm_fc/tfc_cmd.c +++ b/drivers/target/tcm_fc/tfc_cmd.c @@ -62,8 +62,8 @@ void ft_dump_cmd(struct ft_cmd *cmd, const char *caller) int count; se_cmd = &cmd->se_cmd; - pr_debug("%s: cmd %p state %d sess %p seq %p se_cmd %p\n", - caller, cmd, cmd->state, cmd->sess, cmd->seq, se_cmd); + pr_debug("%s: cmd %p sess %p seq %p se_cmd %p\n", + caller, cmd, cmd->sess, cmd->seq, se_cmd); pr_debug("%s: cmd %p cdb %p\n", caller, cmd, cmd->cdb); pr_debug("%s: cmd %p lun %d\n", caller, cmd, cmd->lun); @@ -90,38 +90,6 @@ void ft_dump_cmd(struct ft_cmd *cmd, const char *caller) 16, 4, cmd->cdb, MAX_COMMAND_SIZE, 0); } -static void ft_queue_cmd(struct ft_sess *sess, struct ft_cmd *cmd) -{ - struct ft_tpg *tpg = sess->tport->tpg; - struct se_queue_obj *qobj = &tpg->qobj; - unsigned long flags; - - qobj = &sess->tport->tpg->qobj; - spin_lock_irqsave(&qobj->cmd_queue_lock, flags); - list_add_tail(&cmd->se_req.qr_list, &qobj->qobj_list); - atomic_inc(&qobj->queue_cnt); - spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags); - - wake_up_process(tpg->thread); -} - -static struct ft_cmd *ft_dequeue_cmd(struct se_queue_obj *qobj) -{ - unsigned long flags; - struct se_queue_req *qr; - - spin_lock_irqsave(&qobj->cmd_queue_lock, flags); - if (list_empty(&qobj->qobj_list)) { - spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags); - return NULL; - } - qr = list_first_entry(&qobj->qobj_list, struct se_queue_req, qr_list); - list_del(&qr->qr_list); - atomic_dec(&qobj->queue_cnt); - spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags); - return container_of(qr, struct ft_cmd, se_req); -} - static void ft_free_cmd(struct ft_cmd *cmd) { struct fc_frame *fp; @@ -282,9 +250,7 @@ u32 ft_get_task_tag(struct se_cmd *se_cmd) int ft_get_cmd_state(struct se_cmd *se_cmd) { - struct ft_cmd *cmd = container_of(se_cmd, struct ft_cmd, se_cmd); - - return cmd->state; + return 0; } int ft_is_state_remove(struct se_cmd *se_cmd) @@ -505,6 +471,8 @@ int ft_queue_tm_resp(struct se_cmd *se_cmd) return 0; } +static void ft_send_work(struct work_struct *work); + /* * Handle incoming FCP command. */ @@ -523,7 +491,9 @@ static void ft_recv_cmd(struct ft_sess *sess, struct fc_frame *fp) goto busy; } cmd->req_frame = fp; /* hold frame during cmd */ - ft_queue_cmd(sess, cmd); + + INIT_WORK(&cmd->work, ft_send_work); + queue_work(sess->tport->tpg->workqueue, &cmd->work); return; busy: @@ -563,12 +533,13 @@ void ft_recv_req(struct ft_sess *sess, struct fc_frame *fp) /* * Send new command to target. */ -static void ft_send_cmd(struct ft_cmd *cmd) +static void ft_send_work(struct work_struct *work) { + struct ft_cmd *cmd = container_of(work, struct ft_cmd, work); struct fc_frame_header *fh = fc_frame_header_get(cmd->req_frame); struct se_cmd *se_cmd; struct fcp_cmnd *fcp; - int data_dir; + int data_dir = 0; u32 data_len; int task_attr; int ret; @@ -675,42 +646,3 @@ static void ft_send_cmd(struct ft_cmd *cmd) err: ft_send_resp_code_and_free(cmd, FCP_CMND_FIELDS_INVALID); } - -/* - * Handle request in the command thread. - */ -static void ft_exec_req(struct ft_cmd *cmd) -{ - pr_debug("cmd state %x\n", cmd->state); - switch (cmd->state) { - case FC_CMD_ST_NEW: - ft_send_cmd(cmd); - break; - default: - break; - } -} - -/* - * Processing thread. - * Currently one thread per tpg. - */ -int ft_thread(void *arg) -{ - struct ft_tpg *tpg = arg; - struct se_queue_obj *qobj = &tpg->qobj; - struct ft_cmd *cmd; - - while (!kthread_should_stop()) { - schedule_timeout_interruptible(MAX_SCHEDULE_TIMEOUT); - if (kthread_should_stop()) - goto out; - - cmd = ft_dequeue_cmd(qobj); - if (cmd) - ft_exec_req(cmd); - } - -out: - return 0; -} diff --git a/drivers/target/tcm_fc/tfc_conf.c b/drivers/target/tcm_fc/tfc_conf.c index b15879d43e22..8fa39b74f22c 100644 --- a/drivers/target/tcm_fc/tfc_conf.c +++ b/drivers/target/tcm_fc/tfc_conf.c @@ -327,7 +327,6 @@ static struct se_portal_group *ft_add_tpg( tpg->index = index; tpg->lport_acl = lacl; INIT_LIST_HEAD(&tpg->lun_list); - transport_init_queue_obj(&tpg->qobj); ret = core_tpg_register(&ft_configfs->tf_ops, wwn, &tpg->se_tpg, tpg, TRANSPORT_TPG_TYPE_NORMAL); @@ -336,8 +335,8 @@ static struct se_portal_group *ft_add_tpg( return NULL; } - tpg->thread = kthread_run(ft_thread, tpg, "ft_tpg%lu", index); - if (IS_ERR(tpg->thread)) { + tpg->workqueue = alloc_workqueue("tcm_fc", 0, 1); + if (!tpg->workqueue) { kfree(tpg); return NULL; } @@ -356,7 +355,7 @@ static void ft_del_tpg(struct se_portal_group *se_tpg) pr_debug("del tpg %s\n", config_item_name(&tpg->se_tpg.tpg_group.cg_item)); - kthread_stop(tpg->thread); + destroy_workqueue(tpg->workqueue); /* Wait for sessions to be freed thru RCU, for BUG_ON below */ synchronize_rcu(); From 33a48ab105a75d37021e422a0a3283241099b142 Mon Sep 17 00:00:00 2001 From: Brian King Date: Wed, 7 Sep 2011 14:41:03 +0000 Subject: [PATCH 296/326] ibmveth: Fix DMA unmap error Commit 6e8ab30ec677 (ibmveth: Add scatter-gather support) introduced a DMA mapping API inconsistency resulting in dma_unmap_page getting called on memory mapped via dma_map_single. This was seen when CONFIG_DMA_API_DEBUG was enabled. Fix up this API usage inconsistency. Signed-off-by: Brian King Acked-by: Anton Blanchard Cc: # v2.6.37+ Signed-off-by: David S. Miller --- drivers/net/ibmveth.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 3e6679269400..dcf65d8f10d2 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -1026,7 +1026,12 @@ static netdev_tx_t ibmveth_start_xmit(struct sk_buff *skb, netdev->stats.tx_bytes += skb->len; } - for (i = 0; i < skb_shinfo(skb)->nr_frags + 1; i++) + dma_unmap_single(&adapter->vdev->dev, + descs[0].fields.address, + descs[0].fields.flags_len & IBMVETH_BUF_LEN_MASK, + DMA_TO_DEVICE); + + for (i = 1; i < skb_shinfo(skb)->nr_frags + 1; i++) dma_unmap_page(&adapter->vdev->dev, descs[i].fields.address, descs[i].fields.flags_len & IBMVETH_BUF_LEN_MASK, DMA_TO_DEVICE); From b93da27f5234198433345e40b39ff59797bc6f6e Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 7 Sep 2011 14:41:04 +0000 Subject: [PATCH 297/326] ibmveth: Fix issue with DMA mapping failure descs[].fields.address is 32bit which truncates any dma mapping errors so dma_mapping_error() fails to catch it. Use a dma_addr_t to do the comparison. With this patch I was able to transfer many gigabytes of data with IOMMU fault injection set at 10% probability. Signed-off-by: Anton Blanchard Cc: # v2.6.37+ Signed-off-by: David S. Miller --- drivers/net/ibmveth.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index dcf65d8f10d2..5b8b411d5a80 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -930,6 +930,7 @@ static netdev_tx_t ibmveth_start_xmit(struct sk_buff *skb, union ibmveth_buf_desc descs[6]; int last, i; int force_bounce = 0; + dma_addr_t dma_addr; /* * veth handles a maximum of 6 segments including the header, so @@ -994,17 +995,16 @@ static netdev_tx_t ibmveth_start_xmit(struct sk_buff *skb, } /* Map the header */ - descs[0].fields.address = dma_map_single(&adapter->vdev->dev, skb->data, - skb_headlen(skb), - DMA_TO_DEVICE); - if (dma_mapping_error(&adapter->vdev->dev, descs[0].fields.address)) + dma_addr = dma_map_single(&adapter->vdev->dev, skb->data, + skb_headlen(skb), DMA_TO_DEVICE); + if (dma_mapping_error(&adapter->vdev->dev, dma_addr)) goto map_failed; descs[0].fields.flags_len = desc_flags | skb_headlen(skb); + descs[0].fields.address = dma_addr; /* Map the frags */ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { - unsigned long dma_addr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; dma_addr = dma_map_page(&adapter->vdev->dev, frag->page, From 91aae1e5c407d4fc79f6983e6c6ba04756c004cb Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 7 Sep 2011 14:41:05 +0000 Subject: [PATCH 298/326] ibmveth: Checksum offload is always disabled Commit b9367bf3ee6d (net: ibmveth: convert to hw_features) reversed a check in ibmveth_set_csum_offload that results in checksum offload never being enabled. Signed-off-by: Anton Blanchard Cc: # 3.0+ Signed-off-by: David S. Miller --- drivers/net/ibmveth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 5b8b411d5a80..07830f918aed 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -812,7 +812,7 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data) } else adapter->fw_ipv6_csum_support = data; - if (ret != H_SUCCESS || ret6 != H_SUCCESS) + if (ret == H_SUCCESS || ret6 == H_SUCCESS) adapter->rx_csum = data; else rc1 = -EIO; From fb82fd204b6e6c67661bbd37df032edafb2da56e Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Wed, 7 Sep 2011 14:41:06 +0000 Subject: [PATCH 299/326] ibmveth: Fix checksum offload failure handling Fix a number of issues in ibmveth_set_csum_offload: - set_attr6 and clr_attr6 may be used uninitialised - We store the result of the IPV4 checksum change in ret but overwrite it in a couple of places before checking it again later. Add ret4 to make it obvious what we are doing. - We weren't clearing the NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM flags if the enable of that hypervisor feature failed. Signed-off-by: Anton Blanchard Signed-off-by: David S. Miller --- drivers/net/ibmveth.c | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 07830f918aed..8dd5fccef725 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -757,7 +757,7 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data) struct ibmveth_adapter *adapter = netdev_priv(dev); unsigned long set_attr, clr_attr, ret_attr; unsigned long set_attr6, clr_attr6; - long ret, ret6; + long ret, ret4, ret6; int rc1 = 0, rc2 = 0; int restart = 0; @@ -770,6 +770,8 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data) set_attr = 0; clr_attr = 0; + set_attr6 = 0; + clr_attr6 = 0; if (data) { set_attr = IBMVETH_ILLAN_IPV4_TCP_CSUM; @@ -784,16 +786,20 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data) if (ret == H_SUCCESS && !(ret_attr & IBMVETH_ILLAN_ACTIVE_TRUNK) && !(ret_attr & IBMVETH_ILLAN_TRUNK_PRI_MASK) && (ret_attr & IBMVETH_ILLAN_PADDED_PKT_CSUM)) { - ret = h_illan_attributes(adapter->vdev->unit_address, clr_attr, + ret4 = h_illan_attributes(adapter->vdev->unit_address, clr_attr, set_attr, &ret_attr); - if (ret != H_SUCCESS) { + if (ret4 != H_SUCCESS) { netdev_err(dev, "unable to change IPv4 checksum " "offload settings. %d rc=%ld\n", - data, ret); + data, ret4); + + h_illan_attributes(adapter->vdev->unit_address, + set_attr, clr_attr, &ret_attr); + + if (data == 1) + dev->features &= ~NETIF_F_IP_CSUM; - ret = h_illan_attributes(adapter->vdev->unit_address, - set_attr, clr_attr, &ret_attr); } else { adapter->fw_ipv4_csum_support = data; } @@ -804,15 +810,18 @@ static int ibmveth_set_csum_offload(struct net_device *dev, u32 data) if (ret6 != H_SUCCESS) { netdev_err(dev, "unable to change IPv6 checksum " "offload settings. %d rc=%ld\n", - data, ret); + data, ret6); + + h_illan_attributes(adapter->vdev->unit_address, + set_attr6, clr_attr6, &ret_attr); + + if (data == 1) + dev->features &= ~NETIF_F_IPV6_CSUM; - ret = h_illan_attributes(adapter->vdev->unit_address, - set_attr6, clr_attr6, - &ret_attr); } else adapter->fw_ipv6_csum_support = data; - if (ret == H_SUCCESS || ret6 == H_SUCCESS) + if (ret4 == H_SUCCESS || ret6 == H_SUCCESS) adapter->rx_csum = data; else rc1 = -EIO; From 784eb99ebad91db4c8c231c4b17f203147ab827b Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 16 Sep 2011 01:31:28 -0700 Subject: [PATCH 300/326] target: Skip non hex characters for VPD=0x83 NAA IEEE Registered Extended This patch adds target_parse_naa_6h_vendor_specific() to address a bug where the conversion of PRODUCT SERIAL NUMBER to use hex2bin() in target_emulate_evpd_83() was not doing proper isxdigit() checking. This conversion of the vpd_unit_serial configifs attribute is done while generating a VPD=0x83 NAA IEEE Registered Extended DESIGNATOR format's 100 bits of unique VENDOR SPECIFIC IDENTIFIER + VENDOR SPECIFIC IDENTIFIER EXTENSION area. This patch allows vpd_unit_serial (VPD=0x80) and the T10 Vendor ID DESIGNATOR format (VPD=0x83) to continue to use free-form variable length ASCII values, and now skips any non hex characters for fixed length NAA IEEE Registered Extended DESIGNATOR format (VPD=0x83) requring the binary conversion. This was originally reported by Martin after the v3.1-rc1 change to use hex2bin() in commit 11650b859681e03fdbf26277fcfc5f1f62186703 where the use of non hex characters in vpd_unit_serial generated different values than the original v3.0 internal hex -> binary code. This v3.1 change caused a problem with filesystems who write a NAA DESIGNATOR onto it's ondisk metadata, and this patch will (again) change existing values to ensure that non hex characters are not included in the fixed length NAA DESIGNATOR. Note this patch still expects vpd_unit_serial to be set via existing userspace methods of uuid generation, and does not do strict formatting via configfs input. The original bug report and thread can be found here: NAA breakage http://www.spinics.net/lists/target-devel/msg00477.html The v3.1-rc1 formatting of VPD=0x83 w/o this patch: VPD INQUIRY: Device Identification page Designation descriptor number 1, descriptor length: 20 designator_type: NAA, code_set: Binary associated with the addressed logical unit NAA 6, IEEE Company_id: 0x1405 Vendor Specific Identifier: 0xffde35ebf Vendor Specific Identifier Extension: 0x3092f498ffa820f9 [0x6001405ffde35ebf3092f498ffa820f9] Designation descriptor number 2, descriptor length: 56 designator_type: T10 vendor identification, code_set: ASCII associated with the addressed logical unit vendor id: LIO-ORG vendor specific: IBLOCK:ffde35ec-3092-4980-a820-917636ca54f1 The v3.1-final formatting of VPD=0x83 w/ this patch: VPD INQUIRY: Device Identification page Designation descriptor number 1, descriptor length: 20 designator_type: NAA, code_set: Binary associated with the addressed logical unit NAA 6, IEEE Company_id: 0x1405 Vendor Specific Identifier: 0xffde35ec3 Vendor Specific Identifier Extension: 0x924980a82091763 [0x6001405ffde35ec30924980a82091763] Designation descriptor number 2, descriptor length: 56 designator_type: T10 vendor identification, code_set: ASCII associated with the addressed logical unit vendor id: LIO-ORG vendor specific: IBLOCK:ffde35ec-3092-4980-a820-917636ca54f1 (v2: Fix parsing code to dereference + check for string terminator instead of null pointer to ensure a zeroed payload for vpd_unit_serial less than 100 bits of NAA DESIGNATOR VENDOR SPECIFIC area. Also, remove the unnecessary bitwise assignment) Reported-by: Martin Svec Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_cdb.c | 35 ++++++++++++++++++++++++++++++-- 1 file changed, 33 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_cdb.c b/drivers/target/target_core_cdb.c index 89ae923c5da6..f04d4ef99dca 100644 --- a/drivers/target/target_core_cdb.c +++ b/drivers/target/target_core_cdb.c @@ -24,6 +24,7 @@ */ #include +#include #include #include @@ -154,6 +155,37 @@ target_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf) return 0; } +static void +target_parse_naa_6h_vendor_specific(struct se_device *dev, unsigned char *buf_off) +{ + unsigned char *p = &dev->se_sub_dev->t10_wwn.unit_serial[0]; + unsigned char *buf = buf_off; + int cnt = 0, next = 1; + /* + * Generate up to 36 bits of VENDOR SPECIFIC IDENTIFIER starting on + * byte 3 bit 3-0 for NAA IEEE Registered Extended DESIGNATOR field + * format, followed by 64 bits of VENDOR SPECIFIC IDENTIFIER EXTENSION + * to complete the payload. These are based from VPD=0x80 PRODUCT SERIAL + * NUMBER set via vpd_unit_serial in target_core_configfs.c to ensure + * per device uniqeness. + */ + while (*p != '\0') { + if (cnt >= 13) + break; + if (!isxdigit(*p)) { + p++; + continue; + } + if (next != 0) { + buf[cnt++] |= hex_to_bin(*p++); + next = 0; + } else { + buf[cnt] = hex_to_bin(*p++) << 4; + next = 1; + } + } +} + /* * Device identification VPD, for a complete list of * DESIGNATOR TYPEs see spc4r17 Table 459. @@ -219,8 +251,7 @@ target_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf) * VENDOR_SPECIFIC_IDENTIFIER and * VENDOR_SPECIFIC_IDENTIFIER_EXTENTION */ - buf[off++] |= hex_to_bin(dev->se_sub_dev->t10_wwn.unit_serial[0]); - hex2bin(&buf[off], &dev->se_sub_dev->t10_wwn.unit_serial[1], 12); + target_parse_naa_6h_vendor_specific(dev, &buf[off]); len = 20; off = (len + 4); From 2ff017f5b4299e24a7f22d9a336dd162bf52bb54 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 16 Sep 2011 01:44:54 -0700 Subject: [PATCH 301/326] iscsi-target: Disable markers + remove dangerous local scope array usage This patch makes iscsi-target explictly disable OFMarker=Yes and IFMarker=yes parameter key usage during iscsi login by setting IFMarkInt_Reject and OFMarkInt_Reject values in iscsi_enforce_integrity_rules() to effectively disable iscsi marker usage. With this patch, an initiator proposer asking to enable either marker parameter keys will be issued a 'No' response, and the target sets OFMarkInt + IFMarkInt parameter key response to 'Irrelevant'. With markers disabled during iscsi login, this patch removes the problematic on-stack local-scope array for marker intervals in iscsit_do_rx_data() + iscsit_do_tx_data(), and other related marker code in iscsi_target_util.c. This fixes a potentional stack smashing scenario with small range markers enabled and a large MRDSL as reported by DanC here: [bug report] target: stack can be smashed http://www.spinics.net/lists/target-devel/msg00453.html Reported-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- .../target/iscsi/iscsi_target_parameters.c | 2 +- drivers/target/iscsi/iscsi_target_util.c | 248 +----------------- 2 files changed, 7 insertions(+), 243 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target_parameters.c b/drivers/target/iscsi/iscsi_target_parameters.c index 497b2e718a76..5b773160200f 100644 --- a/drivers/target/iscsi/iscsi_target_parameters.c +++ b/drivers/target/iscsi/iscsi_target_parameters.c @@ -1430,7 +1430,7 @@ static int iscsi_enforce_integrity_rules( u8 DataSequenceInOrder = 0; u8 ErrorRecoveryLevel = 0, SessionType = 0; u8 IFMarker = 0, OFMarker = 0; - u8 IFMarkInt_Reject = 0, OFMarkInt_Reject = 0; + u8 IFMarkInt_Reject = 1, OFMarkInt_Reject = 1; u32 FirstBurstLength = 0, MaxBurstLength = 0; struct iscsi_param *param = NULL; diff --git a/drivers/target/iscsi/iscsi_target_util.c b/drivers/target/iscsi/iscsi_target_util.c index a0d23bc0fc98..1d1b4fe33e43 100644 --- a/drivers/target/iscsi/iscsi_target_util.c +++ b/drivers/target/iscsi/iscsi_target_util.c @@ -874,40 +874,6 @@ void iscsit_inc_session_usage_count(struct iscsi_session *sess) spin_unlock_bh(&sess->session_usage_lock); } -/* - * Used before iscsi_do[rx,tx]_data() to determine iov and [rx,tx]_marker - * array counts needed for sync and steering. - */ -static int iscsit_determine_sync_and_steering_counts( - struct iscsi_conn *conn, - struct iscsi_data_count *count) -{ - u32 length = count->data_length; - u32 marker, markint; - - count->sync_and_steering = 1; - - marker = (count->type == ISCSI_RX_DATA) ? - conn->of_marker : conn->if_marker; - markint = (count->type == ISCSI_RX_DATA) ? - (conn->conn_ops->OFMarkInt * 4) : - (conn->conn_ops->IFMarkInt * 4); - count->ss_iov_count = count->iov_count; - - while (length > 0) { - if (length >= marker) { - count->ss_iov_count += 3; - count->ss_marker_count += 2; - - length -= marker; - marker = markint; - } else - length = 0; - } - - return 0; -} - /* * Setup conn->if_marker and conn->of_marker values based upon * the initial marker-less interval. (see iSCSI v19 A.2) @@ -1431,8 +1397,7 @@ static int iscsit_do_rx_data( struct iscsi_data_count *count) { int data = count->data_length, rx_loop = 0, total_rx = 0, iov_len; - u32 rx_marker_val[count->ss_marker_count], rx_marker_iov = 0; - struct kvec iov[count->ss_iov_count], *iov_p; + struct kvec *iov_p; struct msghdr msg; if (!conn || !conn->sock || !conn->conn_ops) @@ -1440,93 +1405,8 @@ static int iscsit_do_rx_data( memset(&msg, 0, sizeof(struct msghdr)); - if (count->sync_and_steering) { - int size = 0; - u32 i, orig_iov_count = 0; - u32 orig_iov_len = 0, orig_iov_loc = 0; - u32 iov_count = 0, per_iov_bytes = 0; - u32 *rx_marker, old_rx_marker = 0; - struct kvec *iov_record; - - memset(&rx_marker_val, 0, - count->ss_marker_count * sizeof(u32)); - memset(&iov, 0, count->ss_iov_count * sizeof(struct kvec)); - - iov_record = count->iov; - orig_iov_count = count->iov_count; - rx_marker = &conn->of_marker; - - i = 0; - size = data; - orig_iov_len = iov_record[orig_iov_loc].iov_len; - while (size > 0) { - pr_debug("rx_data: #1 orig_iov_len %u," - " orig_iov_loc %u\n", orig_iov_len, orig_iov_loc); - pr_debug("rx_data: #2 rx_marker %u, size" - " %u\n", *rx_marker, size); - - if (orig_iov_len >= *rx_marker) { - iov[iov_count].iov_len = *rx_marker; - iov[iov_count++].iov_base = - (iov_record[orig_iov_loc].iov_base + - per_iov_bytes); - - iov[iov_count].iov_len = (MARKER_SIZE / 2); - iov[iov_count++].iov_base = - &rx_marker_val[rx_marker_iov++]; - iov[iov_count].iov_len = (MARKER_SIZE / 2); - iov[iov_count++].iov_base = - &rx_marker_val[rx_marker_iov++]; - old_rx_marker = *rx_marker; - - /* - * OFMarkInt is in 32-bit words. - */ - *rx_marker = (conn->conn_ops->OFMarkInt * 4); - size -= old_rx_marker; - orig_iov_len -= old_rx_marker; - per_iov_bytes += old_rx_marker; - - pr_debug("rx_data: #3 new_rx_marker" - " %u, size %u\n", *rx_marker, size); - } else { - iov[iov_count].iov_len = orig_iov_len; - iov[iov_count++].iov_base = - (iov_record[orig_iov_loc].iov_base + - per_iov_bytes); - - per_iov_bytes = 0; - *rx_marker -= orig_iov_len; - size -= orig_iov_len; - - if (size) - orig_iov_len = - iov_record[++orig_iov_loc].iov_len; - - pr_debug("rx_data: #4 new_rx_marker" - " %u, size %u\n", *rx_marker, size); - } - } - data += (rx_marker_iov * (MARKER_SIZE / 2)); - - iov_p = &iov[0]; - iov_len = iov_count; - - if (iov_count > count->ss_iov_count) { - pr_err("iov_count: %d, count->ss_iov_count:" - " %d\n", iov_count, count->ss_iov_count); - return -1; - } - if (rx_marker_iov > count->ss_marker_count) { - pr_err("rx_marker_iov: %d, count->ss_marker" - "_count: %d\n", rx_marker_iov, - count->ss_marker_count); - return -1; - } - } else { - iov_p = count->iov; - iov_len = count->iov_count; - } + iov_p = count->iov; + iov_len = count->iov_count; while (total_rx < data) { rx_loop = kernel_recvmsg(conn->sock, &msg, iov_p, iov_len, @@ -1541,16 +1421,6 @@ static int iscsit_do_rx_data( rx_loop, total_rx, data); } - if (count->sync_and_steering) { - int j; - for (j = 0; j < rx_marker_iov; j++) { - pr_debug("rx_data: #5 j: %d, offset: %d\n", - j, rx_marker_val[j]); - conn->of_marker_offset = rx_marker_val[j]; - } - total_rx -= (rx_marker_iov * (MARKER_SIZE / 2)); - } - return total_rx; } @@ -1559,8 +1429,7 @@ static int iscsit_do_tx_data( struct iscsi_data_count *count) { int data = count->data_length, total_tx = 0, tx_loop = 0, iov_len; - u32 tx_marker_val[count->ss_marker_count], tx_marker_iov = 0; - struct kvec iov[count->ss_iov_count], *iov_p; + struct kvec *iov_p; struct msghdr msg; if (!conn || !conn->sock || !conn->conn_ops) @@ -1573,98 +1442,8 @@ static int iscsit_do_tx_data( memset(&msg, 0, sizeof(struct msghdr)); - if (count->sync_and_steering) { - int size = 0; - u32 i, orig_iov_count = 0; - u32 orig_iov_len = 0, orig_iov_loc = 0; - u32 iov_count = 0, per_iov_bytes = 0; - u32 *tx_marker, old_tx_marker = 0; - struct kvec *iov_record; - - memset(&tx_marker_val, 0, - count->ss_marker_count * sizeof(u32)); - memset(&iov, 0, count->ss_iov_count * sizeof(struct kvec)); - - iov_record = count->iov; - orig_iov_count = count->iov_count; - tx_marker = &conn->if_marker; - - i = 0; - size = data; - orig_iov_len = iov_record[orig_iov_loc].iov_len; - while (size > 0) { - pr_debug("tx_data: #1 orig_iov_len %u," - " orig_iov_loc %u\n", orig_iov_len, orig_iov_loc); - pr_debug("tx_data: #2 tx_marker %u, size" - " %u\n", *tx_marker, size); - - if (orig_iov_len >= *tx_marker) { - iov[iov_count].iov_len = *tx_marker; - iov[iov_count++].iov_base = - (iov_record[orig_iov_loc].iov_base + - per_iov_bytes); - - tx_marker_val[tx_marker_iov] = - (size - *tx_marker); - iov[iov_count].iov_len = (MARKER_SIZE / 2); - iov[iov_count++].iov_base = - &tx_marker_val[tx_marker_iov++]; - iov[iov_count].iov_len = (MARKER_SIZE / 2); - iov[iov_count++].iov_base = - &tx_marker_val[tx_marker_iov++]; - old_tx_marker = *tx_marker; - - /* - * IFMarkInt is in 32-bit words. - */ - *tx_marker = (conn->conn_ops->IFMarkInt * 4); - size -= old_tx_marker; - orig_iov_len -= old_tx_marker; - per_iov_bytes += old_tx_marker; - - pr_debug("tx_data: #3 new_tx_marker" - " %u, size %u\n", *tx_marker, size); - pr_debug("tx_data: #4 offset %u\n", - tx_marker_val[tx_marker_iov-1]); - } else { - iov[iov_count].iov_len = orig_iov_len; - iov[iov_count++].iov_base - = (iov_record[orig_iov_loc].iov_base + - per_iov_bytes); - - per_iov_bytes = 0; - *tx_marker -= orig_iov_len; - size -= orig_iov_len; - - if (size) - orig_iov_len = - iov_record[++orig_iov_loc].iov_len; - - pr_debug("tx_data: #5 new_tx_marker" - " %u, size %u\n", *tx_marker, size); - } - } - - data += (tx_marker_iov * (MARKER_SIZE / 2)); - - iov_p = &iov[0]; - iov_len = iov_count; - - if (iov_count > count->ss_iov_count) { - pr_err("iov_count: %d, count->ss_iov_count:" - " %d\n", iov_count, count->ss_iov_count); - return -1; - } - if (tx_marker_iov > count->ss_marker_count) { - pr_err("tx_marker_iov: %d, count->ss_marker" - "_count: %d\n", tx_marker_iov, - count->ss_marker_count); - return -1; - } - } else { - iov_p = count->iov; - iov_len = count->iov_count; - } + iov_p = count->iov; + iov_len = count->iov_count; while (total_tx < data) { tx_loop = kernel_sendmsg(conn->sock, &msg, iov_p, iov_len, @@ -1679,9 +1458,6 @@ static int iscsit_do_tx_data( tx_loop, total_tx, data); } - if (count->sync_and_steering) - total_tx -= (tx_marker_iov * (MARKER_SIZE / 2)); - return total_tx; } @@ -1702,12 +1478,6 @@ int rx_data( c.data_length = data; c.type = ISCSI_RX_DATA; - if (conn->conn_ops->OFMarker && - (conn->conn_state >= TARG_CONN_STATE_LOGGED_IN)) { - if (iscsit_determine_sync_and_steering_counts(conn, &c) < 0) - return -1; - } - return iscsit_do_rx_data(conn, &c); } @@ -1728,12 +1498,6 @@ int tx_data( c.data_length = data; c.type = ISCSI_TX_DATA; - if (conn->conn_ops->IFMarker && - (conn->conn_state >= TARG_CONN_STATE_LOGGED_IN)) { - if (iscsit_determine_sync_and_steering_counts(conn, &c) < 0) - return -1; - } - return iscsit_do_tx_data(conn, &c); } From f39aa30d7741f40ad964341e9243dbbd7f8ff057 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Wed, 31 Aug 2011 10:45:46 +0800 Subject: [PATCH 302/326] firewire: ohci: add no MSI quirk for O2Micro controller This fixes https://bugs.launchpad.net/ubuntu/+source/linux/+bug/801719 . An O2Micro PCI Express FireWire controller, "FireWire (IEEE 1394) [0c00]: O2 Micro, Inc. Device [1217:11f7] (rev 05)" which is a combination device together with an SDHCI controller and some sort of storage controller, misses SBP-2 status writes from an attached FireWire HDD. This problem goes away if MSI is disabled for this FireWire controller. The device reportedly does not require QUIRK_CYCLE_TIMER. Signed-off-by: Ming Lei Signed-off-by: Stefan Richter (amended changelog) Cc: --- drivers/firewire/ohci.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c index 57cd3a406edf..fd7170a9ad2c 100644 --- a/drivers/firewire/ohci.c +++ b/drivers/firewire/ohci.c @@ -290,6 +290,9 @@ static const struct { {PCI_VENDOR_ID_NEC, PCI_ANY_ID, PCI_ANY_ID, QUIRK_CYCLE_TIMER}, + {PCI_VENDOR_ID_O2, PCI_ANY_ID, PCI_ANY_ID, + QUIRK_NO_MSI}, + {PCI_VENDOR_ID_RICOH, PCI_ANY_ID, PCI_ANY_ID, QUIRK_CYCLE_TIMER}, From d5ccd496601b8776a516d167a6485754575dc38f Mon Sep 17 00:00:00 2001 From: Max Matveev Date: Mon, 29 Aug 2011 21:02:24 +0000 Subject: [PATCH 303/326] sctp: deal with multiple COOKIE_ECHO chunks Attempt to reduce the number of IP packets emitted in response to single SCTP packet (2e3216cd) introduced a complication - if a packet contains two COOKIE_ECHO chunks and nothing else then SCTP state machine corks the socket while processing first COOKIE_ECHO and then loses the association and forgets to uncork the socket. To deal with the issue add new SCTP command which can be used to set association explictly. Use this new command when processing second COOKIE_ECHO chunk to restore the context for SCTP state machine. Signed-off-by: Max Matveev Signed-off-by: David S. Miller --- include/net/sctp/command.h | 1 + net/sctp/sm_sideeffect.c | 5 +++++ net/sctp/sm_statefuns.c | 6 ++++++ 3 files changed, 12 insertions(+) diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 6506458ccd33..712b3bebeda7 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -109,6 +109,7 @@ typedef enum { SCTP_CMD_SEND_MSG, /* Send the whole use message */ SCTP_CMD_SEND_NEXT_ASCONF, /* Send the next ASCONF after ACK */ SCTP_CMD_PURGE_ASCONF_QUEUE, /* Purge all asconf queues.*/ + SCTP_CMD_SET_ASOC, /* Restore association context */ SCTP_CMD_LAST } sctp_verb_t; diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 167c880cf8da..76388b083f28 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -1689,6 +1689,11 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, case SCTP_CMD_PURGE_ASCONF_QUEUE: sctp_asconf_queue_teardown(asoc); break; + + case SCTP_CMD_SET_ASOC: + asoc = cmd->obj.asoc; + break; + default: pr_warn("Impossible command: %u, %p\n", cmd->verb, cmd->obj.ptr); diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 49b847b00f99..a0f31e6c1c63 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -2047,6 +2047,12 @@ sctp_disposition_t sctp_sf_do_5_2_4_dupcook(const struct sctp_endpoint *ep, sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(new_asoc)); sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + /* Restore association pointer to provide SCTP command interpeter + * with a valid context in case it needs to manipulate + * the queues */ + sctp_add_cmd_sf(commands, SCTP_CMD_SET_ASOC, + SCTP_ASOC((struct sctp_association *)asoc)); + return retval; nomem: From 4fb66b8210c7d7147b164e19b1b44da916a75691 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 2 Sep 2011 02:19:23 +0000 Subject: [PATCH 304/326] caif: fix a potential NULL dereference MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit bd30ce4bc0b7 (caif: Use RCU instead of spin-lock in caif_dev.c) added a potential NULL dereference in case alloc_percpu() fails. caif_device_alloc() can also use GFP_KERNEL instead of GFP_ATOMIC. Signed-off-by: Eric Dumazet CC: Sjur Brændeland Acked-by: Sjur Brændeland Signed-off-by: David S. Miller --- net/caif/caif_dev.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c index 7c2fa0a08148..7f9ac0742d19 100644 --- a/net/caif/caif_dev.c +++ b/net/caif/caif_dev.c @@ -93,10 +93,14 @@ static struct caif_device_entry *caif_device_alloc(struct net_device *dev) caifdevs = caif_device_list(dev_net(dev)); BUG_ON(!caifdevs); - caifd = kzalloc(sizeof(*caifd), GFP_ATOMIC); + caifd = kzalloc(sizeof(*caifd), GFP_KERNEL); if (!caifd) return NULL; caifd->pcpu_refcnt = alloc_percpu(int); + if (!caifd->pcpu_refcnt) { + kfree(caifd); + return NULL; + } caifd->netdev = dev; dev_hold(dev); return caifd; From 19c1ea14c930db5e9c0cd7c3c6f4d01457dfcd69 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Sun, 4 Sep 2011 20:24:20 +0000 Subject: [PATCH 305/326] ipv4: Fix fib_info->fib_metrics leak Commit 4670994d(net,rcu: convert call_rcu(fc_rport_free_rcu) to kfree_rcu()) introduced a memory leak. This patch reverts it. Signed-off-by: Zheng Yan Signed-off-by: David S. Miller --- net/ipv4/fib_semantics.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 33e2c35b74b7..80106d89d548 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -142,6 +142,14 @@ const struct fib_prop fib_props[RTN_MAX + 1] = { }; /* Release a nexthop info record */ +static void free_fib_info_rcu(struct rcu_head *head) +{ + struct fib_info *fi = container_of(head, struct fib_info, rcu); + + if (fi->fib_metrics != (u32 *) dst_default_metrics) + kfree(fi->fib_metrics); + kfree(fi); +} void free_fib_info(struct fib_info *fi) { @@ -156,7 +164,7 @@ void free_fib_info(struct fib_info *fi) } endfor_nexthops(fi); fib_info_cnt--; release_net(fi->fib_net); - kfree_rcu(fi, rcu); + call_rcu(&fi->rcu, free_fib_info_rcu); } void fib_release_info(struct fib_info *fi) From 728871bc05afc8ff310b17dba3e57a2472792b13 Mon Sep 17 00:00:00 2001 From: David Ward Date: Mon, 5 Sep 2011 16:47:23 +0000 Subject: [PATCH 306/326] net: Align AF-specific flowi structs to long AF-specific flowi structs are now passed to flow_key_compare, which must also be aligned to a long. Signed-off-by: David Ward Signed-off-by: David S. Miller --- include/net/flow.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/net/flow.h b/include/net/flow.h index 78113daadd63..2ec377d9ab9f 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -68,7 +68,7 @@ struct flowi4 { #define fl4_ipsec_spi uli.spi #define fl4_mh_type uli.mht.type #define fl4_gre_key uli.gre_key -}; +} __attribute__((__aligned__(BITS_PER_LONG/8))); static inline void flowi4_init_output(struct flowi4 *fl4, int oif, __u32 mark, __u8 tos, __u8 scope, @@ -112,7 +112,7 @@ struct flowi6 { #define fl6_ipsec_spi uli.spi #define fl6_mh_type uli.mht.type #define fl6_gre_key uli.gre_key -}; +} __attribute__((__aligned__(BITS_PER_LONG/8))); struct flowidn { struct flowi_common __fl_common; @@ -127,7 +127,7 @@ struct flowidn { union flowi_uli uli; #define fld_sport uli.ports.sport #define fld_dport uli.ports.dport -}; +} __attribute__((__aligned__(BITS_PER_LONG/8))); struct flowi { union { From aa1c366e4febc7f5c2b84958a2dd7cd70e28f9d0 Mon Sep 17 00:00:00 2001 From: dpward Date: Mon, 5 Sep 2011 16:47:24 +0000 Subject: [PATCH 307/326] net: Handle different key sizes between address families in flow cache With the conversion of struct flowi to a union of AF-specific structs, some operations on the flow cache need to account for the exact size of the key. Signed-off-by: David Ward Signed-off-by: David S. Miller --- include/net/flow.h | 19 +++++++++++++++++++ net/core/flow.c | 31 +++++++++++++++++-------------- 2 files changed, 36 insertions(+), 14 deletions(-) diff --git a/include/net/flow.h b/include/net/flow.h index 2ec377d9ab9f..a09447749e2d 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -7,6 +7,7 @@ #ifndef _NET_FLOW_H #define _NET_FLOW_H +#include #include #include @@ -161,6 +162,24 @@ static inline struct flowi *flowidn_to_flowi(struct flowidn *fldn) return container_of(fldn, struct flowi, u.dn); } +typedef unsigned long flow_compare_t; + +static inline size_t flow_key_size(u16 family) +{ + switch (family) { + case AF_INET: + BUILD_BUG_ON(sizeof(struct flowi4) % sizeof(flow_compare_t)); + return sizeof(struct flowi4) / sizeof(flow_compare_t); + case AF_INET6: + BUILD_BUG_ON(sizeof(struct flowi6) % sizeof(flow_compare_t)); + return sizeof(struct flowi6) / sizeof(flow_compare_t); + case AF_DECnet: + BUILD_BUG_ON(sizeof(struct flowidn) % sizeof(flow_compare_t)); + return sizeof(struct flowidn) / sizeof(flow_compare_t); + } + return 0; +} + #define FLOW_DIR_IN 0 #define FLOW_DIR_OUT 1 #define FLOW_DIR_FWD 2 diff --git a/net/core/flow.c b/net/core/flow.c index 47b6d26c2afb..555a456efb07 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -173,29 +173,26 @@ static void flow_new_hash_rnd(struct flow_cache *fc, static u32 flow_hash_code(struct flow_cache *fc, struct flow_cache_percpu *fcp, - const struct flowi *key) + const struct flowi *key, + size_t keysize) { const u32 *k = (const u32 *) key; + const u32 length = keysize * sizeof(flow_compare_t) / sizeof(u32); - return jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd) + return jhash2(k, length, fcp->hash_rnd) & (flow_cache_hash_size(fc) - 1); } -typedef unsigned long flow_compare_t; - /* I hear what you're saying, use memcmp. But memcmp cannot make - * important assumptions that we can here, such as alignment and - * constant size. + * important assumptions that we can here, such as alignment. */ -static int flow_key_compare(const struct flowi *key1, const struct flowi *key2) +static int flow_key_compare(const struct flowi *key1, const struct flowi *key2, + size_t keysize) { const flow_compare_t *k1, *k1_lim, *k2; - const int n_elem = sizeof(struct flowi) / sizeof(flow_compare_t); - - BUILD_BUG_ON(sizeof(struct flowi) % sizeof(flow_compare_t)); k1 = (const flow_compare_t *) key1; - k1_lim = k1 + n_elem; + k1_lim = k1 + keysize; k2 = (const flow_compare_t *) key2; @@ -216,6 +213,7 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, struct flow_cache_entry *fle, *tfle; struct hlist_node *entry; struct flow_cache_object *flo; + size_t keysize; unsigned int hash; local_bh_disable(); @@ -223,6 +221,11 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, fle = NULL; flo = NULL; + + keysize = flow_key_size(family); + if (!keysize) + goto nocache; + /* Packet really early in init? Making flow_cache_init a * pre-smp initcall would solve this. --RR */ if (!fcp->hash_table) @@ -231,12 +234,12 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, if (fcp->hash_rnd_recalc) flow_new_hash_rnd(fc, fcp); - hash = flow_hash_code(fc, fcp, key); + hash = flow_hash_code(fc, fcp, key, keysize); hlist_for_each_entry(tfle, entry, &fcp->hash_table[hash], u.hlist) { if (tfle->net == net && tfle->family == family && tfle->dir == dir && - flow_key_compare(key, &tfle->key) == 0) { + flow_key_compare(key, &tfle->key, keysize) == 0) { fle = tfle; break; } @@ -251,7 +254,7 @@ flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir, fle->net = net; fle->family = family; fle->dir = dir; - memcpy(&fle->key, key, sizeof(*key)); + memcpy(&fle->key, key, keysize * sizeof(flow_compare_t)); fle->object = NULL; hlist_add_head(&fle->u.hlist, &fcp->hash_table[hash]); fcp->hash_count++; From 9566042ef84fd2a282d00d3163074ec9b3f93a70 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 16 Sep 2011 09:09:50 +0000 Subject: [PATCH 308/326] IRDA: Fix global type conflicts in net/irda/irsysctl.c v2 The externs here didn't agree with the declarations in qos.c. Better would be probably to move this into a header, but since it's common practice to have naked externs with sysctls I left it for now. Cc: samuel@sortiz.org Signed-off-by: Andi Kleen Signed-off-by: David S. Miller --- net/irda/irsysctl.c | 6 +++--- net/irda/qos.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/net/irda/irsysctl.c b/net/irda/irsysctl.c index d0b70dadf73b..2615ffc8e785 100644 --- a/net/irda/irsysctl.c +++ b/net/irda/irsysctl.c @@ -40,9 +40,9 @@ extern int sysctl_slot_timeout; extern int sysctl_fast_poll_increase; extern char sysctl_devname[]; extern int sysctl_max_baud_rate; -extern int sysctl_min_tx_turn_time; -extern int sysctl_max_tx_data_size; -extern int sysctl_max_tx_window; +extern unsigned int sysctl_min_tx_turn_time; +extern unsigned int sysctl_max_tx_data_size; +extern unsigned int sysctl_max_tx_window; extern int sysctl_max_noreply_time; extern int sysctl_warn_noreply_time; extern int sysctl_lap_keepalive_time; diff --git a/net/irda/qos.c b/net/irda/qos.c index 1b51bcf42394..4369f7f41bcb 100644 --- a/net/irda/qos.c +++ b/net/irda/qos.c @@ -60,7 +60,7 @@ int sysctl_max_noreply_time = 12; * Default is 10us which means using the unmodified value given by the * peer except if it's 0 (0 is likely a bug in the other stack). */ -unsigned sysctl_min_tx_turn_time = 10; +unsigned int sysctl_min_tx_turn_time = 10; /* * Maximum data size to be used in transmission in payload of LAP frame. * There is a bit of confusion in the IrDA spec : @@ -75,13 +75,13 @@ unsigned sysctl_min_tx_turn_time = 10; * bytes frames or all negotiated frame sizes, but you can use the sysctl * to play with this value anyway. * Jean II */ -unsigned sysctl_max_tx_data_size = 2042; +unsigned int sysctl_max_tx_data_size = 2042; /* * Maximum transmit window, i.e. number of LAP frames between turn-around. * This allow to override what the peer told us. Some peers are buggy and * don't always support what they tell us. * Jean II */ -unsigned sysctl_max_tx_window = 7; +unsigned int sysctl_max_tx_window = 7; static int irlap_param_baud_rate(void *instance, irda_param_t *param, int get); static int irlap_param_link_disconnect(void *instance, irda_param_t *parm, From 34b8686d278f00fb16234e74be44c253d6d6b676 Mon Sep 17 00:00:00 2001 From: Daniel Mack Date: Fri, 16 Sep 2011 07:57:43 +0000 Subject: [PATCH 309/326] can: ti_hecc: include linux/io.h This fixes a build breakage for OMAP3 boards. Signed-off-by: Daniel Mack Cc: Wolfgang Grandegger Cc: netdev@vger.kernel.org Acked-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/ti_hecc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index a81249246ece..2adc294f512a 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include From 40b054970afcf067896d62cd6f7e617c62665304 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Fri, 16 Sep 2011 16:55:47 -0700 Subject: [PATCH 310/326] iscsi-target: Fix sendpage breakage with proper padding+DataDigest iovec offsets This patch fixes a bug in the iscsit_fe_sendpage_sg() transmit codepath that was originally introduced with the v3.1 iscsi-target merge that incorrectly uses hardcoded cmd->iov_data_count values to determine cmd->iov_data[] offsets for extra outgoing padding and DataDigest payload vectors. This code is obviously incorrect for the DataDigest enabled case with sendpage offload, and this fix ensures correct operation for padding + DataDigest, padding only, and DataDigest only cases. The bug was introduced during a pre-merge change in iscsit_fe_sendpage_sg() to natively use struct scatterlist instead of the legacy v3.0 struct se_mem logic. Cc: Andy Grover Cc: Christoph Hellwig Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target_util.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target_util.c b/drivers/target/iscsi/iscsi_target_util.c index 1d1b4fe33e43..f00137f377b2 100644 --- a/drivers/target/iscsi/iscsi_target_util.c +++ b/drivers/target/iscsi/iscsi_target_util.c @@ -1256,7 +1256,7 @@ int iscsit_fe_sendpage_sg( struct kvec iov; u32 tx_hdr_size, data_len; u32 offset = cmd->first_data_sg_off; - int tx_sent; + int tx_sent, iov_off; send_hdr: tx_hdr_size = ISCSI_HDR_LEN; @@ -1276,9 +1276,19 @@ int iscsit_fe_sendpage_sg( } data_len = cmd->tx_size - tx_hdr_size - cmd->padding; - if (conn->conn_ops->DataDigest) + /* + * Set iov_off used by padding and data digest tx_data() calls below + * in order to determine proper offset into cmd->iov_data[] + */ + if (conn->conn_ops->DataDigest) { data_len -= ISCSI_CRC_LEN; - + if (cmd->padding) + iov_off = (cmd->iov_data_count - 2); + else + iov_off = (cmd->iov_data_count - 1); + } else { + iov_off = (cmd->iov_data_count - 1); + } /* * Perform sendpage() for each page in the scatterlist */ @@ -1307,8 +1317,7 @@ int iscsit_fe_sendpage_sg( send_padding: if (cmd->padding) { - struct kvec *iov_p = - &cmd->iov_data[cmd->iov_data_count-1]; + struct kvec *iov_p = &cmd->iov_data[iov_off++]; tx_sent = tx_data(conn, iov_p, 1, cmd->padding); if (cmd->padding != tx_sent) { @@ -1322,8 +1331,7 @@ int iscsit_fe_sendpage_sg( send_datacrc: if (conn->conn_ops->DataDigest) { - struct kvec *iov_d = - &cmd->iov_data[cmd->iov_data_count]; + struct kvec *iov_d = &cmd->iov_data[iov_off]; tx_sent = tx_data(conn, iov_d, 1, ISCSI_CRC_LEN); if (ISCSI_CRC_LEN != tx_sent) { From 8e2ec639173f325977818c45011ee176ef2b11f6 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Mon, 5 Sep 2011 21:34:30 +0000 Subject: [PATCH 311/326] ipv6: don't use inetpeer to store metrics for routes. Current IPv6 implementation uses inetpeer to store metrics for routes. The problem of inetpeer is that it doesn't take subnet prefix length in to consideration. If two routes have the same address but different prefix length, they share same inetpeer. So changing metrics of one route also affects the other. The fix is to allocate separate metrics storage for each route. Signed-off-by: Zheng Yan Signed-off-by: David S. Miller --- net/ipv6/route.c | 33 ++++++++++++++++++++++----------- 1 file changed, 22 insertions(+), 11 deletions(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 9e69eb0ec6dd..1250f9020670 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -104,6 +104,9 @@ static u32 *ipv6_cow_metrics(struct dst_entry *dst, unsigned long old) struct inet_peer *peer; u32 *p = NULL; + if (!(rt->dst.flags & DST_HOST)) + return NULL; + if (!rt->rt6i_peer) rt6_bind_peer(rt, 1); @@ -252,6 +255,9 @@ static void ip6_dst_destroy(struct dst_entry *dst) struct inet6_dev *idev = rt->rt6i_idev; struct inet_peer *peer = rt->rt6i_peer; + if (!(rt->dst.flags & DST_HOST)) + dst_destroy_metrics_generic(dst); + if (idev != NULL) { rt->rt6i_idev = NULL; in6_dev_put(idev); @@ -723,9 +729,7 @@ static struct rt6_info *rt6_alloc_cow(const struct rt6_info *ort, ipv6_addr_copy(&rt->rt6i_gateway, daddr); } - rt->rt6i_dst.plen = 128; rt->rt6i_flags |= RTF_CACHE; - rt->dst.flags |= DST_HOST; #ifdef CONFIG_IPV6_SUBTREES if (rt->rt6i_src.plen && saddr) { @@ -775,9 +779,7 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct rt6_info *rt = ip6_rt_copy(ort, daddr); if (rt) { - rt->rt6i_dst.plen = 128; rt->rt6i_flags |= RTF_CACHE; - rt->dst.flags |= DST_HOST; dst_set_neighbour(&rt->dst, neigh_clone(dst_get_neighbour_raw(&ort->dst))); } return rt; @@ -1078,12 +1080,15 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, neigh = NULL; } - rt->rt6i_idev = idev; + rt->dst.flags |= DST_HOST; + rt->dst.output = ip6_output; dst_set_neighbour(&rt->dst, neigh); atomic_set(&rt->dst.__refcnt, 1); - ipv6_addr_copy(&rt->rt6i_dst.addr, addr); dst_metric_set(&rt->dst, RTAX_HOPLIMIT, 255); - rt->dst.output = ip6_output; + + ipv6_addr_copy(&rt->rt6i_dst.addr, addr); + rt->rt6i_dst.plen = 128; + rt->rt6i_idev = idev; spin_lock_bh(&icmp6_dst_lock); rt->dst.next = icmp6_dst_gc_list; @@ -1261,6 +1266,14 @@ int ip6_route_add(struct fib6_config *cfg) if (rt->rt6i_dst.plen == 128) rt->dst.flags |= DST_HOST; + if (!(rt->dst.flags & DST_HOST) && cfg->fc_mx) { + u32 *metrics = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL); + if (!metrics) { + err = -ENOMEM; + goto out; + } + dst_init_metrics(&rt->dst, metrics, 0); + } #ifdef CONFIG_IPV6_SUBTREES ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len); rt->rt6i_src.plen = cfg->fc_src_len; @@ -1607,9 +1620,6 @@ void rt6_redirect(const struct in6_addr *dest, const struct in6_addr *src, if (on_link) nrt->rt6i_flags &= ~RTF_GATEWAY; - nrt->rt6i_dst.plen = 128; - nrt->dst.flags |= DST_HOST; - ipv6_addr_copy(&nrt->rt6i_gateway, (struct in6_addr*)neigh->primary_key); dst_set_neighbour(&nrt->dst, neigh_clone(neigh)); @@ -1754,9 +1764,10 @@ static struct rt6_info *ip6_rt_copy(const struct rt6_info *ort, if (rt) { rt->dst.input = ort->dst.input; rt->dst.output = ort->dst.output; + rt->dst.flags |= DST_HOST; ipv6_addr_copy(&rt->rt6i_dst.addr, dest); - rt->rt6i_dst.plen = ort->rt6i_dst.plen; + rt->rt6i_dst.plen = 128; dst_copy_metrics(&rt->dst, &ort->dst); rt->dst.error = ort->dst.error; rt->rt6i_idev = ort->rt6i_idev; From 3765fefaee2da83f10829fa64a74e6b7360350cb Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Sun, 18 Sep 2011 10:20:46 -0400 Subject: [PATCH 312/326] btrfs: fix d_off in the first dirent Since the d_off in the first dirent for "." (that originates from the 4th argument "offset" of filldir() for the 2nd dirent for "..") is wrongly assigned in btrfs_real_readdir(), telldir returns same offset for different locations. | # mkfs.btrfs /dev/sdb1 | # mount /dev/sdb1 fs0 | # cd fs0 | # touch file0 file1 | # ../test | telldir: 0 | readdir: d_off = 2, d_name = "." | telldir: 2 | readdir: d_off = 2, d_name = ".." | telldir: 2 | readdir: d_off = 3, d_name = "file0" | telldir: 3 | readdir: d_off = 2147483647, d_name = "file1" | telldir: 2147483647 To fix this problem, pass filp->f_pos (which is loff_t) instead. | # ../test | telldir: 0 | readdir: d_off = 1, d_name = "." | telldir: 1 | readdir: d_off = 2, d_name = ".." | telldir: 2 | readdir: d_off = 3, d_name = "file0" : At the moment the "offset" for "." is unused because there is no preceding dirent, however it is better to pass filp->f_pos to follow grammatical usage. Signed-off-by: Hidetoshi Seto Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index b94c0da3b43f..ba951764b005 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4117,7 +4117,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, /* special case for "." */ if (filp->f_pos == 0) { - over = filldir(dirent, ".", 1, 1, btrfs_ino(inode), DT_DIR); + over = filldir(dirent, ".", 1, + filp->f_pos, btrfs_ino(inode), DT_DIR); if (over) return 0; filp->f_pos = 1; @@ -4126,7 +4127,7 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, if (filp->f_pos == 1) { u64 pino = parent_ino(filp->f_path.dentry); over = filldir(dirent, "..", 2, - 2, pino, DT_DIR); + filp->f_pos, pino, DT_DIR); if (over) return 0; filp->f_pos = 2; From 71ef07861080418d125dcf454af41baafa409a2c Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sun, 18 Sep 2011 10:20:46 -0400 Subject: [PATCH 313/326] Btrfs: fix pages truncation in btrfs_ioctl_clone() It's a bug in commit f81c9cdc567cd3160ff9e64868d9a1a7ee226480 (Btrfs: truncate pages from clone ioctl target range) We should pass the dest range to the truncate function, but not the src range. Also move the function before locking extent state. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 63b4de1626d2..8bfb514b26c9 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2234,6 +2234,10 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, goto out_unlock; } + /* truncate page cache pages from target inode range */ + truncate_inode_pages_range(&inode->i_data, destoff, + PAGE_CACHE_ALIGN(destoff + len) - 1); + /* do any pending delalloc/csum calc on src, one way or another, and lock file content */ while (1) { @@ -2250,10 +2254,6 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, btrfs_wait_ordered_range(src, off, len); } - /* truncate page cache pages from target inode range */ - truncate_inode_pages_range(&inode->i_data, off, - ALIGN(off + len, PAGE_CACHE_SIZE) - 1); - /* clone data */ key.objectid = btrfs_ino(src); key.type = BTRFS_EXTENT_DATA_KEY; From 0e7b824c4ef9f5bcf5e48cdce164a7b349dde969 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sun, 18 Sep 2011 10:20:46 -0400 Subject: [PATCH 314/326] Btrfs: don't make a file partly checksummed through file clone To reproduce the bug: # mount /dev/sda7 /mnt # dd if=/dev/zero of=/mnt/src bs=4K count=1 # umount /mnt # mount -o nodatasum /dev/sda7 /mnt # dd if=/dev/zero of=/mnt/dst bs=4K count=1 # clone_range -s 4K -l 4K /mnt/src /mnt/dst # echo 3 > /proc/sys/vm/drop_caches # cat /mnt/dst # dmesg ... btrfs no csum found for inode 258 start 0 btrfs csum failed ino 258 off 0 csum 2566472073 private 0 It's because part of the file is checksummed and the other part is not, and then btrfs will complain checksum is not found when we read the file. Disallow file clone if src and dst file have different checksum flag, so we ensure a file is completely checksummed or unchecksummed. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 8bfb514b26c9..1e766e86f334 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2185,6 +2185,11 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, if (!(src_file->f_mode & FMODE_READ)) goto out_fput; + /* don't make the dst file partly checksummed */ + if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != + (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) + goto out_fput; + ret = -EISDIR; if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode)) goto out_fput; From dde820fbf7176b64daddc1856597d9c61dac19e2 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sun, 18 Sep 2011 10:20:46 -0400 Subject: [PATCH 315/326] Btrfs: don't change inode flag of the dest clone file The dst file will have the same inode flags with dst file after file clone, and I think it's unexpected. For example, the dst file will suddenly become immutable after getting some share of data with src file, if the src is immutable. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 1e766e86f334..9947a0ac7bd5 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2455,7 +2455,6 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, if (endoff > inode->i_size) btrfs_i_size_write(inode, endoff); - BTRFS_I(inode)->flags = BTRFS_I(src)->flags; ret = btrfs_update_inode(trans, root, inode); BUG_ON(ret); btrfs_end_transaction(trans, root); From 48802c8ae2a9d618ec734a61283d645ad527e06c Mon Sep 17 00:00:00 2001 From: Jeff Liu Date: Sun, 18 Sep 2011 10:34:02 -0400 Subject: [PATCH 316/326] BTRFS: Fix lseek return value for error The recent reworking of btrfs' lseek lead to incorrect values being returned. This adds checks for seeking beyond EOF in SEEK_HOLE and makes sure the error values come back correct. Andi Kleen also sent in similar patches. Signed-off-by: Jie Liu Reported-by: Andi Kleen Signed-off-by: Chris Mason --- fs/btrfs/file.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 3c3abff731a7..a381cd22f518 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1817,6 +1817,11 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int origin) goto out; case SEEK_DATA: case SEEK_HOLE: + if (offset >= i_size_read(inode)) { + mutex_unlock(&inode->i_mutex); + return -ENXIO; + } + ret = find_desired_extent(inode, &offset, origin); if (ret) { mutex_unlock(&inode->i_mutex); @@ -1825,11 +1830,11 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int origin) } if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) { - ret = -EINVAL; + offset = -EINVAL; goto out; } if (offset > inode->i_sb->s_maxbytes) { - ret = -EINVAL; + offset = -EINVAL; goto out; } From a66e7cc626f42de6c745963fe0d807518fa49d39 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Sun, 18 Sep 2011 10:34:03 -0400 Subject: [PATCH 317/326] Btrfs: only clear the need lookup flag after the dentry is setup We can race with readdir and the RCU path walking stuff. This is because we clear the need lookup flag before actually instantiating the inode. This will lead the RCU path walk stuff to find a dentry it thinks is valid without a d_inode attached. So instead unhash the dentry when we first start the lookup, and then clear the flag after we've instantiated the dentry so we're garunteed to either try the slow lookup, or have the d_inode set properly. Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 377e9bb0974f..b2d004ad66a0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4018,7 +4018,8 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) memcpy(&location, dentry->d_fsdata, sizeof(struct btrfs_key)); kfree(dentry->d_fsdata); dentry->d_fsdata = NULL; - d_clear_need_lookup(dentry); + /* This thing is hashed, drop it for now */ + d_drop(dentry); } else { ret = btrfs_inode_by_name(dir, dentry, &location); } @@ -4085,7 +4086,15 @@ static void btrfs_dentry_release(struct dentry *dentry) static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { - return d_splice_alias(btrfs_lookup_dentry(dir, dentry), dentry); + struct dentry *ret; + + ret = d_splice_alias(btrfs_lookup_dentry(dir, dentry), dentry); + if (unlikely(d_need_lookup(dentry))) { + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_NEED_LOOKUP; + spin_unlock(&dentry->d_lock); + } + return ret; } unsigned char btrfs_filetype_table[] = { From 18b4fada275dd2b6dd9db904ddf70fe39e272222 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 16 Sep 2011 12:04:07 -0400 Subject: [PATCH 318/326] drm/radeon/kms: fix typo in r100_blit_copy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cur_pages is the number of pages per loop iteration. Signed-off-by: Alex Deucher Reviewed-by: Michel Dänzer Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/r100.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 11e44a3479e3..a536505342d8 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -773,8 +773,8 @@ int r100_copy_blit(struct radeon_device *rdev, radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16)); radeon_ring_write(rdev, 0); radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16)); - radeon_ring_write(rdev, num_pages); - radeon_ring_write(rdev, num_pages); + radeon_ring_write(rdev, cur_pages); + radeon_ring_write(rdev, cur_pages); radeon_ring_write(rdev, cur_pages | (stride_pixels << 16)); } radeon_ring_write(rdev, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0)); From 003cefe0c238e683a29d2207dba945b508cd45b7 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 16 Sep 2011 12:04:08 -0400 Subject: [PATCH 319/326] drm/radeon/kms: Make GPU/CPU page size handling consistent in blit code (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The BO blit code inconsistenly handled the page size. This wasn't an issue on system with 4k pages since the GPU's page size is 4k as well. Switch the driver blit callbacks to take num pages in GPU page units. Fixes lemote mipsel systems using AMD rs780/rs880 chipsets. v2: incorporate suggestions from Michel. Signed-off-by: Alex Deucher Reviewed-by: Michel Dänzer Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen.c | 10 ++++++---- drivers/gpu/drm/radeon/r100.c | 12 ++++++------ drivers/gpu/drm/radeon/r200.c | 4 ++-- drivers/gpu/drm/radeon/r600.c | 10 ++++++---- drivers/gpu/drm/radeon/radeon.h | 7 ++++--- drivers/gpu/drm/radeon/radeon_asic.h | 8 ++++---- drivers/gpu/drm/radeon/radeon_ttm.c | 7 ++++++- 7 files changed, 34 insertions(+), 24 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index f10d1c1c2554..e8a746712b5b 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -3171,21 +3171,23 @@ int evergreen_suspend(struct radeon_device *rdev) } int evergreen_copy_blit(struct radeon_device *rdev, - uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, struct radeon_fence *fence) + uint64_t src_offset, + uint64_t dst_offset, + unsigned num_gpu_pages, + struct radeon_fence *fence) { int r; mutex_lock(&rdev->r600_blit.mutex); rdev->r600_blit.vb_ib = NULL; - r = evergreen_blit_prepare_copy(rdev, num_pages * RADEON_GPU_PAGE_SIZE); + r = evergreen_blit_prepare_copy(rdev, num_gpu_pages * RADEON_GPU_PAGE_SIZE); if (r) { if (rdev->r600_blit.vb_ib) radeon_ib_free(rdev, &rdev->r600_blit.vb_ib); mutex_unlock(&rdev->r600_blit.mutex); return r; } - evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_pages * RADEON_GPU_PAGE_SIZE); + evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages * RADEON_GPU_PAGE_SIZE); evergreen_blit_done_copy(rdev, fence); mutex_unlock(&rdev->r600_blit.mutex); return 0; diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index a536505342d8..5b1837b4aacf 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -721,11 +721,11 @@ void r100_fence_ring_emit(struct radeon_device *rdev, int r100_copy_blit(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence) { uint32_t cur_pages; - uint32_t stride_bytes = PAGE_SIZE; + uint32_t stride_bytes = RADEON_GPU_PAGE_SIZE; uint32_t pitch; uint32_t stride_pixels; unsigned ndw; @@ -737,7 +737,7 @@ int r100_copy_blit(struct radeon_device *rdev, /* radeon pitch is /64 */ pitch = stride_bytes / 64; stride_pixels = stride_bytes / 4; - num_loops = DIV_ROUND_UP(num_pages, 8191); + num_loops = DIV_ROUND_UP(num_gpu_pages, 8191); /* Ask for enough room for blit + flush + fence */ ndw = 64 + (10 * num_loops); @@ -746,12 +746,12 @@ int r100_copy_blit(struct radeon_device *rdev, DRM_ERROR("radeon: moving bo (%d) asking for %u dw.\n", r, ndw); return -EINVAL; } - while (num_pages > 0) { - cur_pages = num_pages; + while (num_gpu_pages > 0) { + cur_pages = num_gpu_pages; if (cur_pages > 8191) { cur_pages = 8191; } - num_pages -= cur_pages; + num_gpu_pages -= cur_pages; /* pages are in Y direction - height page width in X direction - width */ diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c index f24058300413..a1f3ba063c2d 100644 --- a/drivers/gpu/drm/radeon/r200.c +++ b/drivers/gpu/drm/radeon/r200.c @@ -84,7 +84,7 @@ static int r200_get_vtx_size_0(uint32_t vtx_fmt_0) int r200_copy_dma(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence) { uint32_t size; @@ -93,7 +93,7 @@ int r200_copy_dma(struct radeon_device *rdev, int r = 0; /* radeon pitch is /64 */ - size = num_pages << PAGE_SHIFT; + size = num_gpu_pages << RADEON_GPU_PAGE_SHIFT; num_loops = DIV_ROUND_UP(size, 0x1FFFFF); r = radeon_ring_lock(rdev, num_loops * 4 + 64); if (r) { diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index c68427612e3b..720dd99163f8 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2353,21 +2353,23 @@ void r600_fence_ring_emit(struct radeon_device *rdev, } int r600_copy_blit(struct radeon_device *rdev, - uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, struct radeon_fence *fence) + uint64_t src_offset, + uint64_t dst_offset, + unsigned num_gpu_pages, + struct radeon_fence *fence) { int r; mutex_lock(&rdev->r600_blit.mutex); rdev->r600_blit.vb_ib = NULL; - r = r600_blit_prepare_copy(rdev, num_pages * RADEON_GPU_PAGE_SIZE); + r = r600_blit_prepare_copy(rdev, num_gpu_pages * RADEON_GPU_PAGE_SIZE); if (r) { if (rdev->r600_blit.vb_ib) radeon_ib_free(rdev, &rdev->r600_blit.vb_ib); mutex_unlock(&rdev->r600_blit.mutex); return r; } - r600_kms_blit_copy(rdev, src_offset, dst_offset, num_pages * RADEON_GPU_PAGE_SIZE); + r600_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages * RADEON_GPU_PAGE_SIZE); r600_blit_done_copy(rdev, fence); mutex_unlock(&rdev->r600_blit.mutex); return 0; diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 32807baf55e2..c1e056b35b29 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -322,6 +322,7 @@ union radeon_gart_table { #define RADEON_GPU_PAGE_SIZE 4096 #define RADEON_GPU_PAGE_MASK (RADEON_GPU_PAGE_SIZE - 1) +#define RADEON_GPU_PAGE_SHIFT 12 struct radeon_gart { dma_addr_t table_addr; @@ -914,17 +915,17 @@ struct radeon_asic { int (*copy_blit)(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence); int (*copy_dma)(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence); int (*copy)(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence); uint32_t (*get_engine_clock)(struct radeon_device *rdev); void (*set_engine_clock)(struct radeon_device *rdev, uint32_t eng_clock); diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 3d7a0d7c6a9a..3dedaa07aac1 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -75,7 +75,7 @@ uint32_t r100_pll_rreg(struct radeon_device *rdev, uint32_t reg); int r100_copy_blit(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence); int r100_set_surface_reg(struct radeon_device *rdev, int reg, uint32_t tiling_flags, uint32_t pitch, @@ -143,7 +143,7 @@ extern void r100_post_page_flip(struct radeon_device *rdev, int crtc); extern int r200_copy_dma(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, + unsigned num_gpu_pages, struct radeon_fence *fence); void r200_set_safe_registers(struct radeon_device *rdev); @@ -311,7 +311,7 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib); int r600_ring_test(struct radeon_device *rdev); int r600_copy_blit(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, struct radeon_fence *fence); + unsigned num_gpu_pages, struct radeon_fence *fence); void r600_hpd_init(struct radeon_device *rdev); void r600_hpd_fini(struct radeon_device *rdev); bool r600_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd); @@ -403,7 +403,7 @@ void evergreen_bandwidth_update(struct radeon_device *rdev); void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib); int evergreen_copy_blit(struct radeon_device *rdev, uint64_t src_offset, uint64_t dst_offset, - unsigned num_pages, struct radeon_fence *fence); + unsigned num_gpu_pages, struct radeon_fence *fence); void evergreen_hpd_init(struct radeon_device *rdev); void evergreen_hpd_fini(struct radeon_device *rdev); bool evergreen_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd); diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 9b86fb0e4122..0b5468bfaf54 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -277,7 +277,12 @@ static int radeon_move_blit(struct ttm_buffer_object *bo, DRM_ERROR("Trying to move memory with CP turned off.\n"); return -EINVAL; } - r = radeon_copy(rdev, old_start, new_start, new_mem->num_pages, fence); + + BUILD_BUG_ON((PAGE_SIZE % RADEON_GPU_PAGE_SIZE) != 0); + + r = radeon_copy(rdev, old_start, new_start, + new_mem->num_pages * (PAGE_SIZE / RADEON_GPU_PAGE_SIZE), /* GPU pages */ + fence); /* FIXME: handle copy error */ r = ttm_bo_move_accel_cleanup(bo, (void *)fence, NULL, evict, no_wait_reserve, no_wait_gpu, new_mem); From e05c82d3666119075615fdbf6abca0266344f27b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 18 Sep 2011 21:02:55 -0400 Subject: [PATCH 320/326] tcp: fix build error if !CONFIG_SYN_COOKIES commit 946cedccbd7387 (tcp: Change possible SYN flooding messages) added a build error if CONFIG_SYN_COOKIES=n Reported-by: Markus Trippelsdorf Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index e9b48b094683..acc620a4a45f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -431,17 +431,34 @@ extern int tcp_disconnect(struct sock *sk, int flags); extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS]; extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt); +#ifdef CONFIG_SYN_COOKIES extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mss); +#else +static inline __u32 cookie_v4_init_sequence(struct sock *sk, + struct sk_buff *skb, + __u16 *mss) +{ + return 0; +} +#endif extern __u32 cookie_init_timestamp(struct request_sock *req); extern bool cookie_check_timestamp(struct tcp_options_received *opt, bool *); /* From net/ipv6/syncookies.c */ extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb); +#ifdef CONFIG_SYN_COOKIES extern __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mss); - +#else +static inline __u32 cookie_v6_init_sequence(struct sock *sk, + struct sk_buff *skb, + __u16 *mss) +{ + return 0; +} +#endif /* tcp_output.c */ extern void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, From f779b2d60ab95c17f1e025778ed0df3ec2f05d75 Mon Sep 17 00:00:00 2001 From: Zheng Yan Date: Sun, 18 Sep 2011 22:37:34 -0400 Subject: [PATCH 321/326] tcp: fix validation of D-SACK D-SACK is allowed to reside below snd_una. But the corresponding check in tcp_is_sackblock_valid() is the exact opposite. It looks like a typo. Signed-off-by: Zheng Yan Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index ea0d2183df4b..21fab3edb92c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1124,7 +1124,7 @@ static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack, return 0; /* ...Then it's D-SACK, and must reside below snd_una completely */ - if (!after(end_seq, tp->snd_una)) + if (after(end_seq, tp->snd_una)) return 0; if (!before(start_seq, tp->undo_marker)) From 1a51410abe7d0ee4b1d112780f46df87d3621043 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2011 17:04:37 -0700 Subject: [PATCH 322/326] Make TASKSTATS require root access Ok, this isn't optimal, since it means that 'iotop' needs admin capabilities, and we may have to work on this some more. But at the same time it is very much not acceptable to let anybody just read anybody elses IO statistics quite at this level. Use of the GENL_ADMIN_PERM suggested by Johannes Berg as an alternative to checking the capabilities by hand. Reported-by: Vasiliy Kulikov Cc: Johannes Berg Acked-by: Balbir Singh Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index e19ce1454ee1..e66046456f4f 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -655,6 +655,7 @@ static struct genl_ops taskstats_ops = { .cmd = TASKSTATS_CMD_GET, .doit = taskstats_user_cmd, .policy = taskstats_cmd_get_policy, + .flags = GENL_ADMIN_PERM, }; static struct genl_ops cgroupstats_ops = { From 58c3c3aa01b455ecb99d61ce73f1444274af696b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2011 17:10:57 -0700 Subject: [PATCH 323/326] Make taskstats round statistics down to nearest 1k bytes/events Even with just the interface limited to admin, there really is little to reason to give byte-per-byte counts for taskstats. So round it down to something less intrusive. Acked-by: Balbir Singh Signed-off-by: Linus Torvalds --- kernel/tsacct.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 24dc60d9fa1f..5bbfac85866e 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -78,6 +78,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) #define KB 1024 #define MB (1024*KB) +#define KB_MASK (~(KB-1)) /* * fill in extended accounting fields */ @@ -95,14 +96,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) stats->hiwater_vm = get_mm_hiwater_vm(mm) * PAGE_SIZE / KB; mmput(mm); } - stats->read_char = p->ioac.rchar; - stats->write_char = p->ioac.wchar; - stats->read_syscalls = p->ioac.syscr; - stats->write_syscalls = p->ioac.syscw; + stats->read_char = p->ioac.rchar & KB_MASK; + stats->write_char = p->ioac.wchar & KB_MASK; + stats->read_syscalls = p->ioac.syscr & KB_MASK; + stats->write_syscalls = p->ioac.syscw & KB_MASK; #ifdef CONFIG_TASK_IO_ACCOUNTING - stats->read_bytes = p->ioac.read_bytes; - stats->write_bytes = p->ioac.write_bytes; - stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes; + stats->read_bytes = p->ioac.read_bytes & KB_MASK; + stats->write_bytes = p->ioac.write_bytes & KB_MASK; + stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes & KB_MASK; #else stats->read_bytes = 0; stats->write_bytes = 0; From c19cc78efe922e86da7ba694dbfc4be066dd7eb4 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 19 Sep 2011 16:05:10 -0700 Subject: [PATCH 324/326] staging: fix comedi build when ISA_DMA_API is enabled but COMEDI_PCI is not enabled Fix build when CONFIG_ISA_DMA_API is enabled but CONFIG_COMEDI_PCI[_DRIVERS] is not enabled. Fixes these build errors: drivers/staging/comedi/drivers/ni_labpc.c: In function 'labpc_ai_cmd': drivers/staging/comedi/drivers/ni_labpc.c:1351: error: implicit declaration of function 'labpc_suggest_transfer_size' drivers/staging/comedi/drivers/ni_labpc.c: At top level: drivers/staging/comedi/drivers/ni_labpc.c:1802: error: conflicting types for 'labpc_suggest_transfer_size' drivers/staging/comedi/drivers/ni_labpc.c:1351: note: previous implicit declaration of 'labpc_suggest_transfer_size' was here Signed-off-by: Randy Dunlap Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- drivers/staging/comedi/drivers/ni_labpc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/staging/comedi/drivers/ni_labpc.c b/drivers/staging/comedi/drivers/ni_labpc.c index 6859af0778cf..7611def97d06 100644 --- a/drivers/staging/comedi/drivers/ni_labpc.c +++ b/drivers/staging/comedi/drivers/ni_labpc.c @@ -241,8 +241,10 @@ static int labpc_eeprom_write_insn(struct comedi_device *dev, struct comedi_insn *insn, unsigned int *data); static void labpc_adc_timing(struct comedi_device *dev, struct comedi_cmd *cmd); -#ifdef CONFIG_COMEDI_PCI +#ifdef CONFIG_ISA_DMA_API static unsigned int labpc_suggest_transfer_size(struct comedi_cmd cmd); +#endif +#ifdef CONFIG_COMEDI_PCI static int labpc_find_device(struct comedi_device *dev, int bus, int slot); #endif static int labpc_dio_mem_callback(int dir, int port, int data, From 44f4c3ed60fb21e1d2dd98304390ac121e6c7c6d Mon Sep 17 00:00:00 2001 From: Greg KH Date: Mon, 19 Sep 2011 16:05:11 -0700 Subject: [PATCH 325/326] USB: xhci: Set change bit when warm reset change is set. Sometimes, when a USB 3.0 device is disconnected, the Intel Panther Point xHCI host controller will report a link state change with the state set to "SS.Inactive". This causes the xHCI host controller to issue a warm port reset, which doesn't finish before the USB core times out while waiting for it to complete. When the warm port reset does complete, and the xHC gives back a port status change event, the xHCI driver kicks khubd. However, it fails to set the bit indicating there is a change event for that port because the logic in xhci-hub.c doesn't check for the warm port reset bit. After that, the warm port status change bit is never cleared by the USB core, and the xHC stops reporting port status change bits. (The xHCI spec says it shouldn't report more port events until all change bits are cleared.) This means any port changes when a new device is connected will never be reported, and the port will seem "dead" until the xHCI driver is unloaded and reloaded, or the computer is rebooted. Fix this by making the xHCI driver set the port change bit when a warm port reset change bit is set. A better solution would be to make the USB core handle warm port reset in differently, merging the current code with the standard port reset code that does an incremental backoff on the timeout, and tries to complete the port reset two more times before giving up. That more complicated fix will be merged next window, and this fix will be backported to stable. This should be backported to kernels as old as 3.0, since that was the first kernel with commit a11496ebf375 ("xHCI: warm reset support"). Signed-off-by: Sarah Sharp Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- drivers/usb/host/xhci-hub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index 1e96d1f1fe6b..723f8231193d 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -761,7 +761,7 @@ int xhci_hub_status_data(struct usb_hcd *hcd, char *buf) memset(buf, 0, retval); status = 0; - mask = PORT_CSC | PORT_PEC | PORT_OCC | PORT_PLC; + mask = PORT_CSC | PORT_PEC | PORT_OCC | PORT_PLC | PORT_WRC; spin_lock_irqsave(&xhci->lock, flags); /* For each port, did anything change? If so, set that bit in buf. */ From c2d7b49f42f50d7fc5cbfd195b785a128723fdf4 Mon Sep 17 00:00:00 2001 From: Andiry Xu Date: Mon, 19 Sep 2011 16:05:12 -0700 Subject: [PATCH 326/326] USB: xHCI: prevent infinite loop when processing MSE event When a xHC host is unable to handle isochronous transfer in the interval, it reports a Missed Service Error event and skips some tds. Currently xhci driver handles MSE event in the following ways: 1. When encounter a MSE event, set ep->skip flag, update event ring dequeue pointer and return. 2. When encounter the next event on this ep, the driver will run the do-while loop, fetch td from ep's td_list to find the td corresponding to this event. All tds missed are marked as short transfer(-EXDEV). The do-while loop will end in two ways: 1. If the td pointed by the event trb is found; 2. If the ep ring's td_list is empty. However, if a buggy HW reports some unpredicted event (for example, an overrun event following a MSE event while the ep ring is actually not empty), the driver will never find the td, and it will loop until the td_list is empty. Unfortunately, the spinlock is dropped when give back a urb in the do-while loop. During the spinlock released period, the class driver may still submit urbs and add tds to the td_list. This may cause disaster, since the td_list will never be empty and the loop never ends, and the system hangs. To fix this, count the number of TDs on the ep ring before skipping TDs, and quit the loop when skipped that number of tds. This guarantees the do-while loop will end after certain number of cycles, and driver will not be trapped in an infinite loop. Signed-off-by: Andiry Xu Signed-off-by: Sarah Sharp Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- drivers/usb/host/xhci-ring.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 54139a2f06ce..952e2ded61af 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1934,8 +1934,10 @@ static int handle_tx_event(struct xhci_hcd *xhci, int status = -EINPROGRESS; struct urb_priv *urb_priv; struct xhci_ep_ctx *ep_ctx; + struct list_head *tmp; u32 trb_comp_code; int ret = 0; + int td_num = 0; slot_id = TRB_TO_SLOT_ID(le32_to_cpu(event->flags)); xdev = xhci->devs[slot_id]; @@ -1957,6 +1959,12 @@ static int handle_tx_event(struct xhci_hcd *xhci, return -ENODEV; } + /* Count current td numbers if ep->skip is set */ + if (ep->skip) { + list_for_each(tmp, &ep_ring->td_list) + td_num++; + } + event_dma = le64_to_cpu(event->buffer); trb_comp_code = GET_COMP_CODE(le32_to_cpu(event->transfer_len)); /* Look for common error cases */ @@ -2068,7 +2076,18 @@ static int handle_tx_event(struct xhci_hcd *xhci, goto cleanup; } + /* We've skipped all the TDs on the ep ring when ep->skip set */ + if (ep->skip && td_num == 0) { + ep->skip = false; + xhci_dbg(xhci, "All tds on the ep_ring skipped. " + "Clear skip flag.\n"); + ret = 0; + goto cleanup; + } + td = list_entry(ep_ring->td_list.next, struct xhci_td, td_list); + if (ep->skip) + td_num--; /* Is this a TRB in the currently executing TD? */ event_seg = trb_in_td(ep_ring->deq_seg, ep_ring->dequeue,