Pull RAS updates from Borislav Petkov:
- Just cleanups and fixes this time around: make threshold_ktype const,
an objtool fix and use proper size for a bitmap
* tag 'ras_core_for_v6.4_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/MCE/AMD: Use an u64 for bank_map
x86/mce: Always inline old MCA stubs
x86/MCE/AMD: Make kobj_type structure constant
Pull EDAC updates from Borislav Petkov:
- skx_edac: Fix overflow when decoding 32G DIMM ranks
- i10nm_edac: Add Sierra Forest support
- amd64_edac: Split driver code between legacy and SMCA systems. The
final goal is adding support for more hw, like GPUs
- The usual minor cleanups and fixes
* tag 'edac_updates_for_v6.4' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras: (25 commits)
EDAC/i10nm: Add Intel Sierra Forest server support
EDAC/amd64: Fix indentation in umc_determine_edac_cap()
EDAC/altera: Remove MODULE_LICENSE in non-module
EDAC: Sanitize MODULE_AUTHOR strings
EDAC/amd81[13]1: Remove trailing newline from MODULE_AUTHOR
EDAC/amd64: Add get_err_info() to pvt->ops
EDAC/amd64: Split dump_misc_regs() into dct/umc functions
EDAC/amd64: Split init_csrows() into dct/umc functions
EDAC/amd64: Split determine_edac_cap() into dct/umc functions
EDAC/amd64: Rename f17h_determine_edac_ctl_cap()
EDAC/amd64: Split setup_mci_misc_attrs() into dct/umc functions
EDAC/amd64: Split ecc_enabled() into dct/umc functions
EDAC/amd64: Split read_mc_regs() into dct/umc functions
EDAC/amd64: Split determine_memory_type() into dct/umc functions
EDAC/amd64: Split read_base_mask() into dct/umc functions
EDAC/amd64: Split prep_chip_selects() into dct/umc functions
EDAC/amd64: Rework hw_info_{get,put}
EDAC/amd64: Merge struct amd64_family_type into struct amd64_pvt
EDAC/amd64: Do not discover ECC symbol size for Family 17h and later
EDAC/amd64: Drop dbam_to_cs() for Family 17h and later
...
Pull m68k updates from Geert Uytterhoeven:
- defconfig updates
- miscellaneous fixes and improvements
* tag 'm68k-for-v6.4-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:
m68k: kexec: Include <linux/reboot.h>
m68k: defconfig: Update defconfigs for v6.3-rc1
m68k: Remove obsolete config NO_KERNEL_MSG
nubus: Drop noop match function
Using this macro makes the code more readable.
It also inits the members of dev_pm_ops in the following manner
without us explicitly needing to:
.suspend = bcm63xx_spi_suspend, \
.resume = bcm63xx_spi_resume, \
.freeze = bcm63xx_spi_suspend, \
.thaw = bcm63xx_spi_resume, \
.poweroff = bcm63xx_spi_suspend, \
.restore = bcm63xx_spi_resume
Signed-off-by: Dhruva Gole <d-gole@ti.com>
Link: https://lore.kernel.org/r/20230424102546.1604484-1-d-gole@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
There is no need to have the functions blkif_get_x86_32_req() and
blkif_get_x86_64_req() in a header file, as they are used in one place
only.
So move them into the using source file and drop the inline qualifier.
While at it fix some style issues, and simplify the code by variable
reusing and using min() instead of open coding it.
Instead of using barrier() use READ_ONCE() for avoiding multiple reads
of nr_segments.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
The interface of free_persistent_gnts() can be simplified, as there is
only a single caller of free_persistent_gnts() and the 2nd and 3rd
parameters are easily obtainable via the ring pointer, which is passed
as the first parameter anyway.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
There is no function xen_blkif_purge_persistent(), so remove its
prototype from common.h.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Before this patch, function gfs2_ail_empty_gl called gfs2_log_flush even
in cases where it encountered an error. It should probably skip the log
flush and leave the file system in an inconsistent state, letting a
subsequent withdraw force the journal to be replayed to reestablish
metadata consistency.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Before this patch, function gfs2_ail_empty_gl would silently return an
error to the caller. This would get silently set into sd_log_error which
would cause a withdraw, but there was no indication why the file system
was withdrawn. This patch adds a fs_err to log the appropriate error
message.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Before this patch, function gfs2_make_fs_ro called gfs2_log_flush once to
finalize the log. However, if there's dirty metadata, log flushes tend
to sync the metadata and formulate revokes. Before this patch, those
revokes may not be written out to the journal immediately, which meant
unresolved glocks could still have revokes in their ail lists. When the
glock worker runs, it tries to transition the glock, but the unresolved
revokes in the ail still need to be written, so it tries to start a
transaction. It's impossible to start a transaction because at that
point, the SDF_JOURNAL_LIVE flag has been cleared by gfs2_make_fs_ro.
That causes the glock worker to fail, unable to write the revokes. The
calling sequence looked something like this:
gfs2_make_fs_ro
gfs2_log_flush - with GFS2_LOG_HEAD_FLUSH_SHUTDOWN flag set
if (flags & GFS2_LOG_HEAD_FLUSH_SHUTDOWN)
clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
...meanwhile...
glock_work_func
do_xmote
rgrp_go_sync (or possibly inode_go_sync)
...
gfs2_ail_empty_gl
__gfs2_trans_begin
if (unlikely(!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))) {
...
return -EROFS;
The previous patch in the series ("gfs2: return errors from
gfs2_ail_empty_gl") now causes the transaction error to no longer be
ignored, so it causes a warning from MOST of the xfstests:
WARNING: CPU: 11 PID: X at fs/gfs2/super.c:603 gfs2_put_super [gfs2]
which corresponds to:
WARN_ON(gfs2_withdrawing(sdp));
The withdraw was triggered silently from do_xmote by:
if (unlikely(sdp->sd_log_error && !gfs2_withdrawn(sdp)))
gfs2_withdraw_delayed(sdp);
This patch adds a second log_flush to gfs2_make_fs_ro: one to sync the
data and one to sync any outstanding revokes and finalize the journal.
Note that both of these log flushes need to be "special," in other
words, not GFS2_LOG_HEAD_FLUSH_NORMAL.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Before this patch, function gfs2_ail_empty_gl did not return errors it
encountered from __gfs2_trans_begin. Those errors usually came from the
fact that the file system was made read-only, often due to unmount
(but theoretically could be due to -o remount,ro), which prevented
the transaction from starting.
The inability to start a transaction prevented its revokes from being
properly written to the journal for glocks during unmount (and
transition to ro).
That meant glocks could be unlocked without the metadata properly
revoked in the journal. So other nodes could grab the glock thinking
that their lvb values were correct, but in fact corresponded to the
glock without its revokes properly synced. That presented as lvb
mismatch errors.
This patch allows gfs2_ail_empty_gl to return the error properly to
the caller.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
syzkaller reported [0] memory leaks of an UDP socket and ZEROCOPY
skbs. We can reproduce the problem with these sequences:
sk = socket(AF_INET, SOCK_DGRAM, 0)
sk.setsockopt(SOL_SOCKET, SO_TIMESTAMPING, SOF_TIMESTAMPING_TX_SOFTWARE)
sk.setsockopt(SOL_SOCKET, SO_ZEROCOPY, 1)
sk.sendto(b'', MSG_ZEROCOPY, ('127.0.0.1', 53))
sk.close()
sendmsg() calls msg_zerocopy_alloc(), which allocates a skb, sets
skb->cb->ubuf.refcnt to 1, and calls sock_hold(). Here, struct
ubuf_info_msgzc indirectly holds a refcnt of the socket. When the
skb is sent, __skb_tstamp_tx() clones it and puts the clone into
the socket's error queue with the TX timestamp.
When the original skb is received locally, skb_copy_ubufs() calls
skb_unclone(), and pskb_expand_head() increments skb->cb->ubuf.refcnt.
This additional count is decremented while freeing the skb, but struct
ubuf_info_msgzc still has a refcnt, so __msg_zerocopy_callback() is
not called.
The last refcnt is not released unless we retrieve the TX timestamped
skb by recvmsg(). Since we clear the error queue in inet_sock_destruct()
after the socket's refcnt reaches 0, there is a circular dependency.
If we close() the socket holding such skbs, we never call sock_put()
and leak the count, sk, and skb.
TCP has the same problem, and commit e0c8bccd40 ("net: stream:
purge sk_error_queue in sk_stream_kill_queues()") tried to fix it
by calling skb_queue_purge() during close(). However, there is a
small chance that skb queued in a qdisc or device could be put
into the error queue after the skb_queue_purge() call.
In __skb_tstamp_tx(), the cloned skb should not have a reference
to the ubuf to remove the circular dependency, but skb_clone() does
not call skb_copy_ubufs() for zerocopy skb. So, we need to call
skb_orphan_frags_rx() for the cloned skb to call skb_copy_ubufs().
[0]:
BUG: memory leak
unreferenced object 0xffff88800c6d2d00 (size 1152):
comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 cd af e8 81 00 00 00 00 ................
02 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00 ...@............
backtrace:
[<0000000055636812>] sk_prot_alloc+0x64/0x2a0 net/core/sock.c:2024
[<0000000054d77b7a>] sk_alloc+0x3b/0x800 net/core/sock.c:2083
[<0000000066f3c7e0>] inet_create net/ipv4/af_inet.c:319 [inline]
[<0000000066f3c7e0>] inet_create+0x31e/0xe40 net/ipv4/af_inet.c:245
[<000000009b83af97>] __sock_create+0x2ab/0x550 net/socket.c:1515
[<00000000b9b11231>] sock_create net/socket.c:1566 [inline]
[<00000000b9b11231>] __sys_socket_create net/socket.c:1603 [inline]
[<00000000b9b11231>] __sys_socket_create net/socket.c:1588 [inline]
[<00000000b9b11231>] __sys_socket+0x138/0x250 net/socket.c:1636
[<000000004fb45142>] __do_sys_socket net/socket.c:1649 [inline]
[<000000004fb45142>] __se_sys_socket net/socket.c:1647 [inline]
[<000000004fb45142>] __x64_sys_socket+0x73/0xb0 net/socket.c:1647
[<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
[<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
BUG: memory leak
unreferenced object 0xffff888017633a00 (size 240):
comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s)
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 2d 6d 0c 80 88 ff ff .........-m.....
backtrace:
[<000000002b1c4368>] __alloc_skb+0x229/0x320 net/core/skbuff.c:497
[<00000000143579a6>] alloc_skb include/linux/skbuff.h:1265 [inline]
[<00000000143579a6>] sock_omalloc+0xaa/0x190 net/core/sock.c:2596
[<00000000be626478>] msg_zerocopy_alloc net/core/skbuff.c:1294 [inline]
[<00000000be626478>] msg_zerocopy_realloc+0x1ce/0x7f0 net/core/skbuff.c:1370
[<00000000cbfc9870>] __ip_append_data+0x2adf/0x3b30 net/ipv4/ip_output.c:1037
[<0000000089869146>] ip_make_skb+0x26c/0x2e0 net/ipv4/ip_output.c:1652
[<00000000098015c2>] udp_sendmsg+0x1bac/0x2390 net/ipv4/udp.c:1253
[<0000000045e0e95e>] inet_sendmsg+0x10a/0x150 net/ipv4/af_inet.c:819
[<000000008d31bfde>] sock_sendmsg_nosec net/socket.c:714 [inline]
[<000000008d31bfde>] sock_sendmsg+0x141/0x190 net/socket.c:734
[<0000000021e21aa4>] __sys_sendto+0x243/0x360 net/socket.c:2117
[<00000000ac0af00c>] __do_sys_sendto net/socket.c:2129 [inline]
[<00000000ac0af00c>] __se_sys_sendto net/socket.c:2125 [inline]
[<00000000ac0af00c>] __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2125
[<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
[<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
Fixes: f214f915e7 ("tcp: enable MSG_ZEROCOPY")
Fixes: b5947e5d1e ("udp: msg_zerocopy")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After failing to verify configuration, it returns directly without
releasing link, which may cause memory leak.
Paolo Abeni thinks that the whole code of this driver is quite
"suboptimal" and looks unmainatained since at least ~15y, so he
suggests that we could simply remove the whole driver, please
take it into consideration.
Simon Horman suggests that the fix label should be set to
"Linux-2.6.12-rc2" considering that the problem has existed
since the driver was introduced and the commit above doesn't
seem to exist in net/net-next.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Gan Gecen <gangecen@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
devm_clk_get() will return either an error or NULL, which the driver
handles, continuing to use the clock of reading the value of the
clock-frequency property.
However, the value of ov5670->xvclk is left as-is and the other clock
framework functions aren't capable of handling error values.
Use devm_clk_get_optional() to obtain NULL instead of -ENOENT.
Fixes: 8004c91e20 ("media: i2c: ov5670: Use common clock framework")
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Pull trivial nios2 cleanup from Al Viro.
* tag 'pull-nios2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
nios2: _TIF_ALLWORK_MASK is unused
Pull misc vfs pile from Al Viro.
Random minor cleanups.
* tag 'pull-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fs: Fix description of vfs_tmpfile()
sysv: switch to put_and_unmap_page()
fs/sysv: Don't round down address for kunmap_flush_on_unmap()
Pull legacy dio cleanup from Al Viro.
* tag 'pull-old-dio' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
__blockdev_direct_IO(): get rid of submit_io callback
Pull vfs write_one_page removal from Al Viro:
"write_one_page series"
* tag 'pull-write-one-page' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
mm,jfs: move write_one_page/folio_write_one to jfs
ocfs2: don't use write_one_page in ocfs2_duplicate_clusters_by_page
ufs: don't flush page immediately for DIRSYNC directories
Pull vfs fget updates from Al Viro:
"fget() to fdget() conversions"
* tag 'pull-fd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fuse_dev_ioctl(): switch to fdget()
cgroup_get_from_fd(): switch to fdget_raw()
bpf: switch to fdget_raw()
build_mount_idmapped(): switch to fdget()
kill the last remaining user of proc_ns_fget()
SVM-SEV: convert the rest of fget() uses to fdget() in there
convert sgx_set_attribute() to fdget()/fdput()
convert setns(2) to fdget()/fdput()
When the action of an xdp program was XDP_TX, lan966x was creating
a xdp_frame and use this one to send the frame back. But it is also
possible to send back the frame without needing a xdp_frame, because
it is possible to send it back using the page.
And then once the frame is transmitted is possible to use directly
page_pool_recycle_direct as lan966x is using page pools.
This would save some CPU usage on this path, which results in higher
number of transmitted frames. Bellow are the statistics:
Frame size: Improvement:
64 ~8%
256 ~11%
512 ~8%
1000 ~0%
1500 ~0%
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Link: https://lore.kernel.org/r/20230422142344.3630602-1-horatiu.vultur@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Alexei Starovoitov says:
====================
pull-request: bpf-next 2023-04-24
We've added 5 non-merge commits during the last 3 day(s) which contain
a total of 7 files changed, 87 insertions(+), 44 deletions(-).
The main changes are:
1) Workaround for bpf iter selftest due to lack of subprog support
in precision tracking, from Andrii.
2) Disable bpf_refcount_acquire kfunc until races are fixed, from Dave.
3) One more test_verifier test converted from asm macro to asm in C,
from Eduard.
4) Fix build with NETFILTER=y INET=n config, from Florian.
5) Add __rcu_read_{lock,unlock} into deny list, from Yafang.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
selftests/bpf: avoid mark_all_scalars_precise() trigger in one of iter tests
bpf: Add __rcu_read_{lock,unlock} into btf id deny list
bpf: Disable bpf_refcount_acquire kfunc calls until race conditions are fixed
selftests/bpf: verifier/prevent_map_lookup converted to inline assembly
bpf: fix link failure with NETFILTER=y INET=n
====================
Link: https://lore.kernel.org/r/20230425005648.86714-1-alexei.starovoitov@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Gerhard Engleder says:
====================
tsnep: XDP socket zero-copy support
Implement XDP socket zero-copy support for tsnep driver. I tried to
follow existing drivers like igc as far as possible. But one main
difference is that tsnep does not need any reconfiguration for XDP BPF
program setup. So I decided to keep this behavior no matter if a XSK
pool is used or not. As a result, tsnep starts using the XSK pool even
if no XDP BPF program is available.
Another difference is that I tried to prevent potentially failing
allocations during XSK pool setup. E.g. both memory models for page pool
and XSK pool are registered all the time. Thus, XSK pool setup cannot
end up with not working queues.
Some prework is done to reduce the last two XSK commits to actual XSK
changes.
====================
Link: https://lore.kernel.org/r/20230421194656.48063-1-gerhard@engleder-embedded.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Send and complete XSK pool frames within TX NAPI context. NAPI context
is triggered by ndo_xsk_wakeup.
Test results with A53 1.2GHz:
xdpsock txonly copy mode, 64 byte frames:
pps pkts 1.00
tx 284,409 11,398,144
Two CPUs with 100% and 10% utilization.
xdpsock txonly zero-copy mode, 64 byte frames:
pps pkts 1.00
tx 511,929 5,890,368
Two CPUs with 100% and 1% utilization.
xdpsock l2fwd copy mode, 64 byte frames:
pps pkts 1.00
rx 248,985 7,315,885
tx 248,921 7,315,885
Two CPUs with 100% and 10% utilization.
xdpsock l2fwd zero-copy mode, 64 byte frames:
pps pkts 1.00
rx 254,735 3,039,456
tx 254,735 3,039,456
Two CPUs with 100% and 4% utilization.
Packet rate increases and CPU utilization is reduced in both cases.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add support for XSK zero-copy to RX path. The setup of the XSK pool can
be done at runtime. If the netdev is running, then the queue must be
disabled and enabled during reconfiguration. This can be done easily
with functions introduced in previous commits.
A more important property is that, if the netdev is running, then the
setup of the XSK pool shall not stop the netdev in case of errors. A
broken netdev after a failed XSK pool setup is bad behavior. Therefore,
the allocation and setup of resources during XSK pool setup is done only
before any queue is disabled. Additionally, freeing and later allocation
of resources is eliminated in some cases. Page pool entries are kept for
later use. Two memory models are registered in parallel. As a result,
the XSK pool setup cannot fail during queue reconfiguration.
In contrast to other drivers, XSK pool setup and XDP BPF program setup
are separate actions. XSK pool setup can be done without any XDP BPF
program. The XDP BPF program can be added, removed or changed without
any reconfiguration of the XSK pool.
Test results with A53 1.2GHz:
xdpsock rxdrop copy mode, 64 byte frames:
pps pkts 1.00
rx 856,054 10,625,775
Two CPUs with both 100% utilization.
xdpsock rxdrop zero-copy mode, 64 byte frames:
pps pkts 1.00
rx 889,388 4,615,284
Two CPUs with 100% and 20% utilization.
Packet rate increases and CPU utilization is reduced.
100% CPU load seems to the base load. This load is consumed by ksoftirqd
just for dropping the generated packets without xdpsock running.
Using batch API reduced CPU utilization slightly, but measurements are
not stable enough to provide meaningful numbers.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The function tsnep_rx_poll() is already pretty long and the skb receive
action can be reused for XSK zero-copy support. Move page based skb
receive to separate function.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Move queue enable and disable code to separate functions. This way the
activation and deactivation of the queues are defined actions, which can
be used in future execution paths.
This functions will be used for the queue reconfiguration at runtime,
which is necessary for XSK zero-copy support.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Make initialization of TX and RX queues less dynamic by moving some
initialization from netdev open/close to device probing.
Additionally, move some initialization code to separate functions to
enable future use in other execution paths.
This is done as preparation for queue reconfigure at runtime, which is
necessary for XSK zero-copy support.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
TX/RX ring size is static and power of 2 to enable compiler to optimize
modulo operation to mask operation. Make this optimization already in
the code and don't rely on the compiler.
CPU utilisation during high packet rate has not changed. So no
performance improvement has been measured. But it is best practice to
prevent modulo operations.
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
SET_COALESCE may change operation mode and parameters in one call.
Changing operation mode may cause the driver to reset the parameter
values to what is a reasonable default for new operation mode.
Since driver does not know which parameters come from user and which
are echoed back from ->get, driver may ignore the parameters when
switching operation modes.
This used to be inevitable for ioctl() but in netlink we know which
parameters are actually specified by the user.
We could inform which parameters were set by the user but this would
lead to a lot of code duplication in the drivers. Instead try to call
the drivers twice if both mode and params are changed. The set method
already checks if any params need updating so in case the driver did
the right thing the first time around - there will be no second call
to it's ->set method (only an extra call to ->get()).
For mlx5 for example before this patch we'd see:
# ethtool -C eth0 adaptive-rx on adaptive-tx on
# ethtool -C eth0 adaptive-rx off adaptive-tx off \
tx-usecs 123 rx-usecs 123
Adaptive RX: off TX: off
rx-usecs: 3
rx-frames: 32
tx-usecs: 16
tx-frames: 32
[...]
After the change:
# ethtool -C eth0 adaptive-rx on adaptive-tx on
# ethtool -C eth0 adaptive-rx off adaptive-tx off \
tx-usecs 123 rx-usecs 123
Adaptive RX: off TX: off
rx-usecs: 123
rx-frames: 32
tx-usecs: 123
tx-frames: 32
[...]
This only works for netlink, so it's a small discrepancy between
netlink and ioctl(). Since we anticipate most users to move to
netlink I believe it's worth making their lives easier.
Link: https://lore.kernel.org/r/20230420233302.944382-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Haiyang Zhang says:
====================
Update coding style and check alloc_frag
Follow up patches for the jumbo frame support.
As suggested by Jakub Kicinski, update coding style, and check napi_alloc_frag
for possible fallback to single pages.
====================
Link: https://lore.kernel.org/r/1682096818-30056-1-git-send-email-haiyangz@microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
netdev/napi_alloc_frag() may fall back to single page which is smaller
than the requested size.
Add error checking to avoid memory overwritten.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Rename mana_refill_rxoob for naming consistency.
And remove some empty lines between function call and error
checking.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Lorenzo Bianconi says:
====================
add page_pool support for page recycling in veth driver
Introduce page_pool support in veth driver in order to recycle pages in
veth_convert_skb_to_xdp_buff routine and avoid reallocating the skb through
the page allocator when we run a xdp program on the device and we receive
skbs from the stack.
====================
Link: https://lore.kernel.org/r/cover.1682188837.git.lorenzo@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Introduce page_pool stats support to report info about local page_pool
through ethtool
Tested-by: Maryam Tahhan <mtahhan@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Introduce page_pool support in veth driver in order to recycle pages
in veth_convert_skb_to_xdp_buff routine and avoid reallocating the skb
through the page allocator.
The patch has been tested sending tcp traffic to a veth pair where the
remote peer is running a simple xdp program just returning xdp_pass:
veth upstream codebase:
MTU 1500B: ~ 8Gbps
MTU 8000B: ~ 13.9Gbps
veth upstream codebase + pp support:
MTU 1500B: ~ 9.2Gbps
MTU 8000B: ~ 16.2Gbps
Tested-by: Maryam Tahhan <mtahhan@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Brad Spencer provided a detailed report [0] that when calling getsockopt()
for AF_NETLINK, some SOL_NETLINK options set only 1 byte even though such
options require at least sizeof(int) as length.
The options return a flag value that fits into 1 byte, but such behaviour
confuses users who do not initialise the variable before calling
getsockopt() and do not strictly check the returned value as char.
Currently, netlink_getsockopt() uses put_user() to copy data to optlen and
optval, but put_user() casts the data based on the pointer, char *optval.
As a result, only 1 byte is set to optval.
To avoid this behaviour, we need to use copy_to_user() or cast optval for
put_user().
Note that this changes the behaviour on big-endian systems, but we document
that the size of optval is int in the man page.
$ man 7 netlink
...
Socket options
To set or get a netlink socket option, call getsockopt(2) to read
or setsockopt(2) to write the option with the option level argument
set to SOL_NETLINK. Unless otherwise noted, optval is a pointer to
an int.
Fixes: 9a4595bc7e ("[NETLINK]: Add set/getsockopt options to support more than 32 groups")
Fixes: be0c22a46c ("netlink: add NETLINK_BROADCAST_ERROR socket option")
Fixes: 38938bfe34 ("netlink: add NETLINK_NO_ENOBUFS socket flag")
Fixes: 0a6a3a23ea ("netlink: add NETLINK_CAP_ACK socket option")
Fixes: 2d4bc93368 ("netlink: extended ACK reporting")
Fixes: 89d35528d1 ("netlink: Add new socket option to enable strict checking on dumps")
Reported-by: Brad Spencer <bspencer@blackberry.com>
Link: https://lore.kernel.org/netdev/ZD7VkNWFfp22kTDt@datsun.rim.net/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Link: https://lore.kernel.org/r/20230421185255.94606-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
iter_pass_iter_ptr_to_subprog subtest is relying on actual array size
being passed as subprog parameter. This combined with recent fixes to
precision tracking in conditional jumps ([0]) is now causing verifier to
backtrack all the way to the point where sum() and fill() subprogs are
called, at which point precision backtrack bails out and forces all the
states to have precise SCALAR registers. This in turn causes each
possible value of i within fill() and sum() subprogs to cause
a different non-equivalent state, preventing iterator code to converge.
For now, change the test to assume fixed size of passed in array. Once
BPF verifier supports precision tracking across subprogram calls, these
changes will be reverted as unnecessary.
[0] 71b547f561 ("bpf: Fix incorrect verifier pruning due to missing register precision taints")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230424235128.1941726-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
1) Reduce jumpstack footprint: Stash chain in last rule marker in blob for
tracing. Remove last rule and chain from jumpstack. From Florian Westphal.
2) nf_tables validates all tables before committing the new rules.
Unfortunately, this has two drawbacks:
- Since addition of the transaction mutex pernet state gets written to
outside of the locked section from the cleanup callback, this is
wrong so do this cleanup directly after table has passed all checks.
- Revalidate tables that saw no changes. This can be avoided by
keeping the validation state per table, not per netns.
From Florian Westphal.
3) Get rid of a few redundant pointers in the traceinfo structure.
The three removed pointers are used in the expression evaluation loop,
so gcc keeps them in registers. Passing them to the (inlined) helpers
thus doesn't increase nft_do_chain text size, while stack is reduced
by another 24 bytes on 64bit arches. From Florian Westphal.
4) IPVS cleanups in several ways without implementing any functional
changes, aside from removing some debugging output:
- Update width of source for ip_vs_sync_conn_options
The operation is safe, use an annotation to describe it properly.
- Consistently use array_size() in ip_vs_conn_init()
It seems better to use helpers consistently.
- Remove {Enter,Leave}Function. These seem to be well past their
use-by date.
- Correct spelling in comments.
From Simon Horman.
5) Extended netlink error report for netdevice in flowtables and
netdev/chains. Allow for incrementally add/delete devices to netdev
basechain. Allow to create netdev chain without device.
* tag 'nf-next-23-04-22' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
netfilter: nf_tables: allow to create netdev chain without device
netfilter: nf_tables: support for deleting devices in an existing netdev chain
netfilter: nf_tables: support for adding new devices to an existing netdev chain
netfilter: nf_tables: rename function to destroy hook list
netfilter: nf_tables: do not send complete notification of deletions
netfilter: nf_tables: extended netlink error reporting for netdevice
ipvs: Correct spelling in comments
ipvs: Remove {Enter,Leave}Function
ipvs: Consistently use array_size() in ip_vs_conn_init()
ipvs: Update width of source for ip_vs_sync_conn_options
netfilter: nf_tables: do not store rule in traceinfo structure
netfilter: nf_tables: do not store verdict in traceinfo structure
netfilter: nf_tables: do not store pktinfo in traceinfo structure
netfilter: nf_tables: remove unneeded conditional
netfilter: nf_tables: make validation state per table
netfilter: nf_tables: don't write table validation state without mutex
netfilter: nf_tables: don't store chain address on jump
netfilter: nf_tables: don't store address of last rule on jump
netfilter: nf_tables: merge nft_rules_old structure and end of ruleblob marker
====================
Link: https://lore.kernel.org/r/20230421235021.216950-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull erofs updates from Gao Xiang:
"In this cycle, sub-page block support for uncompressed files is
available. It's mainly used to enable original signing ('golden')
4k-block images on arm64 with 16/64k pages. In addition, end users
could also use this feature to build a manifest to directly refer to
golden tar data.
Besides, long xattr name prefix support is also introduced in this
cycle to avoid too many xattrs with the same prefix (e.g. overlayfs
xattrs). It's useful for erofs + overlayfs combination (like Composefs
model): the image size is reduced by ~14% and runtime performance is
also slightly improved.
Others are random fixes and cleanups as usual.
Summary:
- Add sub-page block size support for uncompressed files
- Support flattened block device for multi-blob images to be attached
into virtual machines (including cloud servers) and bare metals
- Support long xattr name prefixes to optimize images with common
xattr namespaces (e.g. files with overlayfs xattrs) use cases
- Various minor cleanups & fixes"
* tag 'erofs-for-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
erofs: cleanup i_format-related stuffs
erofs: sunset erofs_dbg()
erofs: fix potential overflow calculating xattr_isize
erofs: get rid of z_erofs_fill_inode()
erofs: enable long extended attribute name prefixes
erofs: handle long xattr name prefixes properly
erofs: add helpers to load long xattr name prefixes
erofs: introduce on-disk format for long xattr name prefixes
erofs: move packed inode out of the compression part
erofs: keep meta inode into erofs_buf
erofs: initialize packed inode after root inode is assigned
erofs: stop parsing non-compact HEAD index if clusterofs is invalid
erofs: don't warn ztailpacking feature anymore
erofs: simplify erofs_xattr_generic_get()
erofs: rename init_inode_xattrs with erofs_ prefix
erofs: move several xattr helpers into xattr.c
erofs: tidy up EROFS on-disk naming
erofs: support flattened block device for multi-blob images
erofs: set block size to the on-disk block size
erofs: avoid hardcoded blocksize for subpage block support
The tracing recursion prevention mechanism must be protected by rcu, that
leaves __rcu_read_{lock,unlock} unprotected by this mechanism. If we trace
them, the recursion will happen. Let's add them into the btf id deny list.
When CONFIG_PREEMPT_RCU is enabled, it can be reproduced with a simple bpf
program as such:
SEC("fentry/__rcu_read_lock")
int fentry_run()
{
return 0;
}
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230424161104.3737-2-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
