This is a series of patches to move mutable file-scope variables
into the driver state. This change will help to introduce another
version of the pipe driver (with different state) for the older
host interface or having several instances of this device.
Signed-off-by: Roman Kiryanov <rkir@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is a series of patches to move mutable file-scope variables
into the driver state. This change will help to introduce another
version of the pipe driver (with different state) for the older
host interface or having several instances of this device.
Signed-off-by: Roman Kiryanov <rkir@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add new GSMI commands (GSMI_CMD_LOG_S0IX_SUSPEND = 0xa,
GSMI_CMD_LOG_S0IX_RESUME = 0xb) that allow firmware to log any
information during S0ix suspend/resume paths.
Traditional ACPI suspend S3 involves BIOS both during the suspend and
the resume paths. However, modern suspend type like S0ix does not
involve firmware on either of the paths. This command gives the
firmware an opportunity to log any required information about the
suspend and resume operations e.g. wake sources.
Additionally, this change adds a module parameter to allow platforms
to specifically enable S0ix logging if required. This prevents any
other platforms from unnecessarily making a GSMI call which could have
any side-effects.
Tested by verifying that wake sources are correctly logged in eventlog.
Signed-off-by: Furquan Shaikh <furquan@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Rajat Jain <rajatja@chromium.org>
Signed-off-by: Furquan Shaikh <furquan@google.com>
Tested-by: Furquan Shaikh <furquan@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
[zwisler: update changelog for upstream]
Signed-off-by: Ross Zwisler <zwisler@google.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of selecting EFI and EFI_VARS automatically when GSMI is
enabled let that portion of the driver be conditionally compiled
if EFI and EFI_VARS are enabled.
This allows the rest of the driver (specifically event log) to
be used if EFI_VARS is not enabled.
To test:
1) verify that EFI_VARS is not automatically selected when
CONFIG_GOOGLE_GSMI is enabled
2) verify that the kernel boots on Link and that GSMI event log
is still available and functional
3) specifically boot the kernel on Alex to ensure it does not
try to load efivars and that gsmi also does not load because it
is not in the supported DMI table
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Olof Johansson <olofj@chromium.org>
Signed-off-by: Benson Leung <bleung@chromium.org>
Signed-off-by: Ben Zhang <benzh@chromium.org>
Signed-off-by: Filipe Brandenburger <filbranden@chromium.org>
Signed-off-by: Furquan Shaikh <furquan@google.com>
Tested-by: Furquan Shaikh <furquan@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
[zwisler: update changelog for upstream]
Signed-off-by: Ross Zwisler <zwisler@google.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The sysfs handler should return the number of bytes consumed, which in the
case of a successful write is the entire buffer. Also fix a bug where
param.data_len was being set to (count - (2 * sizeof(u32))) instead of just
(count - sizeof(u32)). The latter is correct because we skip over the
leading u32 which is our param.type, but we were also incorrectly
subtracting sizeof(u32) on the line where we were actually setting
param.data_len:
param.data_len = count - sizeof(u32);
This meant that for our example event.kernel_software_watchdog with total
length 10 bytes, param.data_len was just 2 prior to this change.
To test, successfully append an event to the log with gsmi sysfs.
This sample event is for a "Kernel Software Watchdog"
> xxd -g 1 event.kernel_software_watchdog
0000000: 01 00 00 00 ad de 06 00 00 00
> cat event.kernel_software_watchdog > /sys/firmware/gsmi/append_to_eventlog
> mosys eventlog list | tail -1
14 | 2012-06-25 10:14:14 | Kernl Event | Software Watchdog
Signed-off-by: Duncan Laurie <dlaurie@chromium.org>
Reviewed-by: Vadim Bendebury <vbendeb@chromium.org>
Reviewed-by: Stefan Reinauer <reinauer@chromium.org>
Signed-off-by: Furquan Shaikh <furquan@google.com>
Tested-by: Furquan Shaikh <furquan@chromium.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Justin TerAvest <teravest@chromium.org>
[zwisler: updated changelog for 2nd bug fix and upstream]
Signed-off-by: Ross Zwisler <zwisler@google.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I am one of the main engineers working on ashmem. I have been fixing
bugs in the driver and have been involved in the memfd conversion
discussions and sending patches about that. I also have an understanding
of the binder driver and was involved with some of the development on
finer grained locking. So I would like to be added to the MAINTAINERS
file for android drivers for review and maintenance of ashmem and other
Android drivers.
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Acked-by: Todd Kjos <tkjos@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
nvmem_find_cell_by_index() is only called from inside an #ifdef,
so we get a build warning without CONFIG_OF:
drivers/nvmem/core.c:496:1: error: 'nvmem_find_cell_by_index' defined but not used [-Werror=unused-function]
Move it into the same #ifdef as the caller to avoid the warning.
Fixes: e888d445ac ("nvmem: resolve cells from DT at registration time")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We check if the pointer returned by __nvmem_device_get() is not NULL
while we should actually check if it is not IS_ERR(nvmem). Fix it.
While we're at it: fix the next error path where we should assign an
error value to cell before returning.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The check for ret < 0 is redundant as any places prior to this point
where ret is set to an error value the code will exit out of the loop
to the error exit label 'err'. Remove this redundant dead code.
Detected by CoverityScan, CID#1339528 ("Logically dead code")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This adds a helper to paste 2 pointer arrays together, useful for merging
various types of attribute arrays. There are a few places in the kernel
tree where this is open coded, and I just added one more in the STM class.
The naming is inspired by memset_p() and memcat(), and partial credit for
it goes to Andy Shevchenko.
This patch adds the function wrapped in a type-enforcing macro and a test
module.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The rules and order of identification of trace sources against the
"stp-policy" have changed; update the documentation to reflect that.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This adds support for CLOCKSYNC SyS-T packets, that establish correlation
between the transport clock (STP timestamps) and SyS-T timestamps. These
packets are sent periodically to allow the decoder to keep both time
sources in sync.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This adds support for MIPI SyS-T protocol as specified in an open
standard [1]. In addition to marking message boundaries, it also
supports tagging messages with the source UUID, to provide better
distinction between trace sources, including payload length and
timestamp in the message's metadata.
This driver adds attributes to STP policy nodes to control/configure
these metadata features.
[1] https://www.mipi.org/specifications/sys-t
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Now that the default framing protocol is factored out into its own driver,
switch over to using the driver for writing data. To that end, make the
policy code require a valid protocol name (or absence thereof, which is
equivalent to "p_basic").
Also, to make transition easier, make stm class request "p_basic" module
at initialization time.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The STP framing pattern that the stm class implicitly applies to the
data payload is, in fact, a protocol. This patch moves the relevant code
out of the stm core into its own driver module.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add a helper to write a sequence of bytes as STP data packets. This
is used by protocol drivers to output their metadata, as well as the
actual data payload.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
At the moment, the stm class applies a certain STP framing pattern to
the data as it is written to the underlying STM device. In order to
allow different framing patterns (aka protocols), this patch introduces
the concept of STP protocol drivers, defines data structures and APIs
for the protocol drivers to use.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, if no matching policy node can be found for a trace source,
we'll try to use "default" policy node, then, if that doesn't exist,
we'll pick the first node, in order of creation. If that also fails,
we'll allocate M/C range from the beginning of the device's M/C range.
This makes it difficult to know which node (if any) was used in any
particular case.
In order to make things more deterministic, the new order is as follows:
* if they supply ID string, use that and nothing else,
* if they are a task, use their task name (comm),
* use "default", if it exists,
* return failure, to let them know there is no suitable rule.
This should provide enough convenience with the "default" catch-all node,
while not leaving *everything* to chance. As a side effect, this relaxes
the requirement of using ioctl() for identification with the possibility of
using task names as policy nodes.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Tested-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I wrote:
"Char/Misc fixes for 4.19-rc7
Here are 8 small fixes for some char/misc driver issues
Included here are:
- fpga driver fixes
- thunderbolt bugfixes
- firmware core revert/fix
- hv core fix
- hv tool fix
All of these have been in linux-next with no reported issues."
* tag 'char-misc-4.19-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
thunderbolt: Initialize after IOMMUs
thunderbolt: Do not handle ICM events after domain is stopped
firmware: Always initialize the fw_priv list object
docs: fpga: document fpga manager flags
fpga: bridge: fix obvious function documentation error
tools: hv: fcopy: set 'error' in case an unknown operation was requested
fpga: do not access region struct after fpga_region_unregister
Drivers: hv: vmbus: Use get/put_cpu() in vmbus_connect()
I wrote:
"Serial driver fixes for 4.19-rc7
Here are 3 small serial driver fixes for 4.19-rc7
- 2 sh-sci bugfixes for reported issues
- a revert of the PM handling for the 8250_dw code
All of these have been in linux-next with no reported issues."
* tag 'tty-4.19-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
Revert "serial: sh-sci: Allow for compressed SCIF address"
Revert "serial: sh-sci: Remove SCIx_RZ_SCIFA_REGTYPE"
Revert "serial: 8250_dw: Fix runtime PM handling"
I wrote:
"USB fixes for 4.19-rc7
Here are some small USB fixes for 4.19-rc7
These include:
- the usual xhci bugfixes for reported issues
- some new serial driver device ids
- bugfix for the option serial driver for some devices
- bugfix for the cdc_acm driver that has been there for a long time.
All of these have been in linux-next for a while with no reported
issues."
* tag 'usb-4.19-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
usb: xhci-mtk: resume USB3 roothub first
xhci: Add missing CAS workaround for Intel Sunrise Point xHCI
usb: cdc_acm: Do not leak URB buffers
USB: serial: simple: add Motorola Tetra MTP6550 id
USB: serial: option: add two-endpoints device-id flag
USB: serial: option: improve Quectel EP06 detection
Wolfram writes:
"i2c for 4.19
I2C has three driver bugfixes and a fix for a typo for you."
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: designware: Call i2c_dw_clk_rate() only when calculating timings
i2c: i2c-scmi: fix for i2c_smbus_write_block_data
i2c: i2c-isch: fix spelling mistake "unitialized" -> "uninitialized"
i2c: i2c-qcom-geni: Properly handle DMA safe buffers
James writes:
"SCSI fixes on 20181006
Small fix for an unititialized mutex in the qedi driver."
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: qedi: Initialize the stats mutex lock
Michael writes:
"powerpc fixes for 4.19 #4
Four regression fixes.
A fix for a change to lib/xz which broke our zImage loader when
building with XZ compression. OK'ed by Herbert who merged the
original patch.
The recent fix we did to avoid patching __init text broke some 32-bit
machines, fix that.
Our show_user_instructions() could be tricked into printing kernel
memory, add a check to avoid that.
And a fix for a change to our NUMA initialisation logic, which causes
crashes in some kdump configurations.
Thanks to:
Christophe Leroy, Hari Bathini, Jann Horn, Joel Stanley, Meelis
Roos, Murilo Opsfelder Araujo, Srikar Dronamraju."
* tag 'powerpc-4.19-4' of https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/numa: Skip onlining a offline node in kdump path
powerpc: Don't print kernel instructions in show_user_instructions()
powerpc/lib: fix book3s/32 boot failure due to code patching
lib/xz: Put CRC32_POLY_LE in xz_private.h
Dave writes:
"Networking fixes:
1) Fix truncation of 32-bit right shift in bpf, from Jann Horn.
2) Fix memory leak in wireless wext compat, from Stefan Seyfried.
3) Use after free in cfg80211's reg_process_hint(), from Yu Zhao.
4) Need to cancel pending work when unbinding in smsc75xx otherwise
we oops, also from Yu Zhao.
5) Don't allow enslaving a team device to itself, from Ido Schimmel.
6) Fix backwards compat with older userspace for rtnetlink FDB dumps.
From Mauricio Faria.
7) Add validation of tc policy netlink attributes, from David Ahern.
8) Fix RCU locking in rawv6_send_hdrinc(), from Wei Wang."
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (26 commits)
net: mvpp2: Extract the correct ethtype from the skb for tx csum offload
ipv6: take rcu lock in rawv6_send_hdrinc()
net: sched: Add policy validation for tc attributes
rtnetlink: fix rtnl_fdb_dump() for ndmsg header
yam: fix a missing-check bug
net: bpfilter: Fix type cast and pointer warnings
net: cxgb3_main: fix a missing-check bug
bpf: 32-bit RSH verification must truncate input before the ALU op
net: phy: phylink: fix SFP interface autodetection
be2net: don't flip hw_features when VXLANs are added/deleted
net/packet: fix packet drop as of virtio gso
net: dsa: b53: Keep CPU port as tagged in all VLANs
openvswitch: load NAT helper
bnxt_en: get the reduced max_irqs by the ones used by RDMA
bnxt_en: free hwrm resources, if driver probe fails.
bnxt_en: Fix enables field in HWRM_QUEUE_COS2BW_CFG request
bnxt_en: Fix VNIC reservations on the PF.
team: Forbid enslaving team device to itself
net/usb: cancel pending work when unbinding smsc75xx
mlxsw: spectrum: Delete RIF when VLAN device is removed
...
* akpm:
mm: madvise(MADV_DODUMP): allow hugetlbfs pages
ocfs2: fix locking for res->tracking and dlm->tracking_list
mm/vmscan.c: fix int overflow in callers of do_shrink_slab()
mm/vmstat.c: skip NR_TLB_REMOTE_FLUSH* properly
mm/vmstat.c: fix outdated vmstat_text
proc: restrict kernel stack dumps to root
mm/hugetlb: add mmap() encodings for 32MB and 512MB page sizes
mm/migrate.c: split only transparent huge pages when allocation fails
ipc/shm.c: use ERR_CAST() for shm_lock() error return
mm/gup_benchmark: fix unsigned comparison to zero in __gup_benchmark_ioctl
mm, thp: fix mlocking THP page with migration enabled
ocfs2: fix crash in ocfs2_duplicate_clusters_by_page()
hugetlb: take PMD sharing into account when flushing tlb/caches
mm: migration: fix migration of huge PMD shared pages
Reproducer, assuming 2M of hugetlbfs available:
Hugetlbfs mounted, size=2M and option user=testuser
# mount | grep ^hugetlbfs
hugetlbfs on /dev/hugepages type hugetlbfs (rw,pagesize=2M,user=dan)
# sysctl vm.nr_hugepages=1
vm.nr_hugepages = 1
# grep Huge /proc/meminfo
AnonHugePages: 0 kB
ShmemHugePages: 0 kB
HugePages_Total: 1
HugePages_Free: 1
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 2048 kB
Code:
#include <sys/mman.h>
#include <stddef.h>
#define SIZE 2*1024*1024
int main()
{
void *ptr;
ptr = mmap(NULL, SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_HUGETLB | MAP_ANONYMOUS, -1, 0);
madvise(ptr, SIZE, MADV_DONTDUMP);
madvise(ptr, SIZE, MADV_DODUMP);
}
Compile and strace:
mmap(NULL, 2097152, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB, -1, 0) = 0x7ff7c9200000
madvise(0x7ff7c9200000, 2097152, MADV_DONTDUMP) = 0
madvise(0x7ff7c9200000, 2097152, MADV_DODUMP) = -1 EINVAL (Invalid argument)
hugetlbfs pages have VM_DONTEXPAND in the VmFlags driver pages based on
author testing with analysis from Florian Weimer[1].
The inclusion of VM_DONTEXPAND into the VM_SPECIAL defination was a
consequence of the large useage of VM_DONTEXPAND in device drivers.
A consequence of [2] is that VM_DONTEXPAND marked pages are unable to be
marked DODUMP.
A user could quite legitimately madvise(MADV_DONTDUMP) their hugetlbfs
memory for a while and later request that madvise(MADV_DODUMP) on the same
memory. We correct this omission by allowing madvice(MADV_DODUMP) on
hugetlbfs pages.
[1] https://stackoverflow.com/questions/52548260/madvisedodump-on-the-same-ptr-size-as-a-successful-madvisedontdump-fails-wit
[2] commit 0103bd16fb ("mm: prepare VM_DONTDUMP for using in drivers")
Link: http://lkml.kernel.org/r/20180930054629.29150-1-daniel@linux.ibm.com
Link: https://lists.launchpad.net/maria-discuss/msg05245.html
Fixes: 0103bd16fb ("mm: prepare VM_DONTDUMP for using in drivers")
Reported-by: Kenneth Penza <kpenza@gmail.com>
Signed-off-by: Daniel Black <daniel@linux.ibm.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, you can use /proc/self/task/*/stack to cause a stack walk on
a task you control while it is running on another CPU. That means that
the stack can change under the stack walker. The stack walker does
have guards against going completely off the rails and into random
kernel memory, but it can interpret random data from your kernel stack
as instruction pointers and stack pointers. This can cause exposure of
kernel stack contents to userspace.
Restrict the ability to inspect kernel stacks of arbitrary tasks to root
in order to prevent a local attacker from exploiting racy stack unwinding
to leak kernel task stack contents. See the added comment for a longer
rationale.
There don't seem to be any users of this userspace API that can't
gracefully bail out if reading from the file fails. Therefore, I believe
that this change is unlikely to break things. In the case that this patch
does end up needing a revert, the next-best solution might be to fake a
single-entry stack based on wchan.
Link: http://lkml.kernel.org/r/20180927153316.200286-1-jannh@google.com
Fixes: 2ec220e27f ("proc: add /proc/*/stack")
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Kees Cook <keescook@chromium.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Ken Chen <kenchen@google.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H . Peter Anvin" <hpa@zytor.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
split_huge_page_to_list() fails on HugeTLB pages. I was experimenting
with moving 32MB contig HugeTLB pages on arm64 (with a debug patch
applied) and hit the following stack trace when the kernel crashed.
[ 3732.462797] Call trace:
[ 3732.462835] split_huge_page_to_list+0x3b0/0x858
[ 3732.462913] migrate_pages+0x728/0xc20
[ 3732.462999] soft_offline_page+0x448/0x8b0
[ 3732.463097] __arm64_sys_madvise+0x724/0x850
[ 3732.463197] el0_svc_handler+0x74/0x110
[ 3732.463297] el0_svc+0x8/0xc
[ 3732.463347] Code: d1000400 f90b0e60 f2fbd5a2 a94982a1 (f9000420)
When unmap_and_move[_huge_page]() fails due to lack of memory, the
splitting should happen only for transparent huge pages not for HugeTLB
pages. PageTransHuge() returns true for both THP and HugeTLB pages.
Hence the conditonal check should test PagesHuge() flag to make sure that
given pages is not a HugeTLB one.
Link: http://lkml.kernel.org/r/1537798495-4996-1-git-send-email-anshuman.khandual@arm.com
Fixes: 94723aafb9 ("mm: unclutter THP migration")
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Zi Yan <zi.yan@cs.rutgers.edu>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A transparent huge page is represented by a single entry on an LRU list.
Therefore, we can only make unevictable an entire compound page, not
individual subpages.
If a user tries to mlock() part of a huge page, we want the rest of the
page to be reclaimable.
We handle this by keeping PTE-mapped huge pages on normal LRU lists: the
PMD on border of VM_LOCKED VMA will be split into PTE table.
Introduction of THP migration breaks[1] the rules around mlocking THP
pages. If we had a single PMD mapping of the page in mlocked VMA, the
page will get mlocked, regardless of PTE mappings of the page.
For tmpfs/shmem it's easy to fix by checking PageDoubleMap() in
remove_migration_pmd().
Anon THP pages can only be shared between processes via fork(). Mlocked
page can only be shared if parent mlocked it before forking, otherwise CoW
will be triggered on mlock().
For Anon-THP, we can fix the issue by munlocking the page on removing PTE
migration entry for the page. PTEs for the page will always come after
mlocked PMD: rmap walks VMAs from oldest to newest.
Test-case:
#include <unistd.h>
#include <sys/mman.h>
#include <sys/wait.h>
#include <linux/mempolicy.h>
#include <numaif.h>
int main(void)
{
unsigned long nodemask = 4;
void *addr;
addr = mmap((void *)0x20000000UL, 2UL << 20, PROT_READ | PROT_WRITE,
MAP_PRIVATE | MAP_ANONYMOUS | MAP_LOCKED, -1, 0);
if (fork()) {
wait(NULL);
return 0;
}
mlock(addr, 4UL << 10);
mbind(addr, 2UL << 20, MPOL_PREFERRED | MPOL_F_RELATIVE_NODES,
&nodemask, 4, MPOL_MF_MOVE);
return 0;
}
[1] https://lkml.kernel.org/r/CAOMGZ=G52R-30rZvhGxEbkTw7rLLwBGadVYeo--iizcD3upL3A@mail.gmail.com
Link: http://lkml.kernel.org/r/20180917133816.43995-1-kirill.shutemov@linux.intel.com
Fixes: 616b837153 ("mm: thp: enable thp migration in generic path")
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: Vegard Nossum <vegard.nossum@oracle.com>
Reviewed-by: Zi Yan <zi.yan@cs.rutgers.edu>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: <stable@vger.kernel.org> [4.14+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
