smb_grant_oplock() checks the previous oplock holder's o_fp to decide
whether a durable handle should be invalidated when the oplock break
cannot be delivered. prev_opinfo is obtained with opinfo_get_list(),
which pins only the oplock_info. It does not pin the ksmbd_file stored
in opinfo->o_fp.
A concurrent last close can unlink the opinfo from ci->m_op_list under
ci->m_lock and then free the ksmbd_file. The oplock_info can still be
kept alive by the refcount taken by opinfo_get_list(), but o_fp may
already point at freed memory by the time smb_grant_oplock() reads
is_durable, conn, or tcon.
Snapshot the previous holder's durable state while ci->m_lock is held,
then use only the copied values after dropping the lock. This keeps the
o_fp lifetime tied to the inode lock without taking an extra ksmbd_file
reference. Taking such a reference is unsafe here because smb_grant_oplock()
does not necessarily have the previous holder's session work, and dropping
the temporary reference can otherwise become the final putter.
Fixes: 26fa88dc87 ("ksmbd: invalidate durable handles on oplock break")
Reported-by: Gil Portnoy <dddhkts1@gmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
ksmbd_close_disconnected_durable_delete_on_close() collects disconnected
durable handles for a name being superseded by a new delete-on-close
open, drops ci->m_lock, then closes each collected handle directly with
__ksmbd_close_fd().
That bypasses the FP_CLOSED and refcount handoff used by the other close
paths. If a durable reconnect or the durable scavenger already took a
reference to the same fp, the direct __ksmbd_close_fd() can free the
ksmbd_file while that other holder still owns a live reference.
Claim the disconnected durable handle before unlinking it from m_fp_list.
While holding ci->m_lock and global_ft.lock, only take ownership when the
durable lifetime reference is the only remaining reference. Then take a
transient reference, remove the fp from global_ft, mark it FP_CLOSED, and
move it to the local dispose list. If another holder already has a
reference, leave the fp linked and let that holder complete its path.
The dispose loop then drops both references owned by the claim. This keeps
the force-close path in the same refcount handoff model as the durable
scavenger and avoids leaving a live reconnected fp detached from
m_fp_list.
Fixes: 166e4c0702 ("ksmbd: supersede disconnected delete-on-close durable handle")
Signed-off-by: Gil Portnoy <dddhkts1@gmail.com>
Co-developed-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Two concurrent SMB2 durable reconnects (DH2C/DHnC) on the same
persistent_id race the fp->owner.name compare-read in
ksmbd_vfs_compare_durable_owner() against the kfree() in
ksmbd_reopen_durable_fd()'s reopen-success path. fp->owner.name is a
standalone kstrdup() buffer whose lifetime is independent of the fp
refcount, and the two sites share no lock: the compare reads the buffer
while the reopen frees it, so the strcmp() can dereference freed memory.
Commit 7ce4fc4001 ("ksmbd: fix durable reconnect double-bind race in
ksmbd_reopen_durable_fd") made the fp->conn claim atomic under
global_ft.lock (closing the owner.name double-free and the ksmbd_file
write-UAF), but the compare-read versus reopen-free pair was left
unserialized.
BUG: KASAN: slab-use-after-free in strcmp+0x2c/0x80
Read of size 1 by task kworker
strcmp
ksmbd_vfs_compare_durable_owner
smb2_check_durable_oplock
smb2_open
Freed by task kworker:
kfree
ksmbd_reopen_durable_fd
smb2_open
Allocated by task kworker:
kstrdup
session_fd_check
smb2_session_logoff
The buggy address belongs to the cache kmalloc-8
Serialize both sides of the race with fp->f_lock. The global durable
file-table lock still protects the durable reconnect claim, but
fp->owner.name is per-open state and does not need to block unrelated
durable table lookups or reconnects. The teardown is left at its
existing location after the reopen-success point so that an __open_id()
rollback still retains owner.name for a later legitimate reconnect to
verify.
Fixes: 49110a8ce6 ("ksmbd: validate owner of durable handle on reconnect")
Assisted-by: Henry (Claude):claude-opus-4
Signed-off-by: Gil Portnoy <dddhkts1@gmail.com>
Co-developed-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
When a cifs.ko client caches a read-handle (RH) lease via deferred close
and a conflicting open arrives, ksmbd breaks the lease and waits for the
acknowledgment in wait_for_break_ack() for up to OPLOCK_WAIT_TIME (35s).
__smb_break_all_levII_oplock() runs that wait while holding ci->m_lock
for read.
cifs.ko reacts to a handle-lease break by closing the deferred handle
rather than sending a lease break acknowledgment. That close path
(close_id_del_oplock() -> opinfo_del()) takes ci->m_lock for write and
is exactly what would wake the waiter, but it blocks on the read lock
held by the waiting thread. The break is then resolved only by the 35s
timeout, so xfstests generic/001 takes ~78s with leases enabled versus
~4s with oplocks only.
Collect the target opinfos (each pinned with a reference) while holding
ci->m_lock, then break them after releasing it, matching how
smb_grant_oplock() already breaks a conflicting lease using only a
reference. The reference keeps the opinfo (and its conn and lease)
alive across the unlocked window, and a close racing the break is
handled by the existing OPLOCK_CLOSING state check. Apply the same fix
to the parent lease break paths.
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Update ksmbd.rst to reflect the current implementation status of SMB
features. Durable handles (v1, v2) and SMB3.1.1 Compression are now
fully supported in ksmbd, so update their status from "Planned for future"
to "Supported".
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
session_fd_check() and ksmbd_reopen_durable_fd() walk ci->m_op_list with
list_for_each_entry_rcu() while holding ci->m_lock for write. That is
the local inode/oplock serializer, but the RCU-list iterator does not
currently tell lockdep about it.
Pass lockdep_is_held(&ci->m_lock) to these iterators so
CONFIG_PROVE_RCU_LIST can see the rwsem protection already in place.
This was found by our static analysis tool and then manually reviewed
against the current tree. The dynamic triage evidence is a
target-matched CONFIG_PROVE_RCU_LIST warning; the change is limited
to documenting the existing protection contract.
This is a lockdep annotation cleanup. It does not change oplock list
lifetime or durable-handle behavior.
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
smb2_validate_credit_charge() adds the request's CreditCharge to
conn->outstanding_credits when an SMB2 PDU is received, and
smb2_set_rsp_credits() subtracts it again when the response is built.
However smb2_set_rsp_credits() only runs on the normal response path:
- __process_request() returning SERVER_HANDLER_ABORT (unimplemented
command, command index out of range, signature check failure, or a
handler that sets send_no_response such as a cancelled blocking
lock) breaks out of the processing loop before set_rsp_credits() is
called;
- smb2_set_rsp_credits() itself returns early with -EINVAL (total
credit overflow or insufficient credits) before the subtraction.
On all of these paths the charge added at receive time is never
returned, so conn->outstanding_credits only grows. Because a client can
repeatedly trigger them (e.g. by sending unimplemented commands or by
issuing and cancelling blocking locks), outstanding_credits eventually
reaches total_credits and smb2_validate_credit_charge() then rejects
every subsequent request, wedging the connection.
Record the charge that was added in work->credit_charge and release any
charge still pending at the single send. exit point of
__handle_ksmbd_work(), which all abort and error paths fall through to.
smb2_set_rsp_credits() clears work->credit_charge once it has returned
the charge so the response path is unchanged and the credit is never
released twice. Paths that never charged a credit (no multi-credit
support, validation failure) leave work->credit_charge at zero and are
unaffected.
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
smb2_validate_credit_charge() computes the credit charge a request is
allowed to consume from the payload size:
CreditCharge = (max(SendPayloadSize, ResponsePayloadSize) - 1)/65536 + 1
For SMB2 QUERY_INFO, the server must validate CreditCharge based on the
*maximum* of InputBufferLength and OutputBufferLength. ksmbd instead
summed the two lengths, which overestimates the required charge.
As a result a single-credit QUERY_INFO whose InputBufferLength and
OutputBufferLength each fit in 64KB but whose sum exceeds 64KB is
rejected with STATUS_INVALID_PARAMETER, even though it is a valid
request. IOCTL already uses max() of the request and response sizes;
make QUERY_INFO consistent by feeding InputBufferLength as the request
length and OutputBufferLength as the expected response length so that
smb2_validate_credit_charge() takes their maximum.
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
smb2_read() allocates the read payload buffer with kvzalloc(), zeroing up
to max_read_size bytes (1MB or more with multichannel) on every read,
only to immediately overwrite the region with file data via kernel_read().
The zero-fill is pure overhead: ksmbd_vfs_read() returns the number of
bytes actually read ('nbytes'), and only those nbytes are ever consumed -
they are pinned into the response iov (ksmbd_iov_pin_rsp_read()), sent
over the RDMA channel (smb2_read_rdma_channel()), or copied by the
compression path (ksmbd_compress_response() uses iov_len == nbytes). The
ALIGN(length, 8) tail padding and any short-read remainder are never read
or transmitted, so they need not be initialized.
Use kvmalloc() instead to skip the redundant zeroing. This reduces CPU
and memory-bandwidth usage on large sequential reads.
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
set_ntacl_dacl() copies each ACE from the attacker-controlled stored
security descriptor verbatim into the response DACL without checking
sid.num_subauth. The ACE bytes (including an unchecked num_subauth)
originate from an authenticated SMB2_SET_INFO(SecInfo=DACL) that is
stored raw via ksmbd_vfs_set_sd_xattr(); parse_dacl() rejects a bad ACE
with `break` rather than an error, so parse_sec_desc() still returns
success and the malformed SD reaches the xattr intact.
On a subsequent SMB2_QUERY_INFO(SecInfo=DACL) for an inode carrying a
POSIX access ACL, build_sec_desc() -> set_ntacl_dacl() ->
set_posix_acl_entries_dacl() walks the copied ACEs and reads
ntace->sid.sub_auth[ntace->sid.num_subauth - 1]
with num_subauth taken straight from the stored SD. Since sub_auth[]
is fixed at SID_MAX_SUB_AUTHORITIES (15), a crafted num_subauth (e.g.
255) drives an out-of-bounds heap read of ~1 KB with an offset fully
controlled by an authenticated client.
The sibling functions already gate this field:
parse_dacl() -- num_subauth == 0 || > SID_MAX_SUB_AUTHORITIES
parse_sid() -- num_subauth > SID_MAX_SUB_AUTHORITIES
smb_copy_sid() -- min_t(u8, num_subauth, SID_MAX_SUB_AUTHORITIES)
set_ntacl_dacl() is the lone inconsistent path that omits the check.
Add the same num_subauth validation in set_ntacl_dacl() before copying
the ACE, matching the gate already enforced by parse_dacl().
Signed-off-by: Haofeng Li <lihaofeng@kylinos.cn>
Reviewed-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Suggested-by: Namjae Jeon <linkinjeon@kernel.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
parse_dacl() limits the attacker-controlled ACE count by comparing it
with the number of minimal ACEs that fit in the DACL size. The DACL size
field is 16 bits, but the expression subtracts sizeof(struct smb_acl).
Because sizeof() is unsigned, a DACL size smaller than the ACL header
underflows to a large size_t.
A malicious client can reach this with:
SMB2_SET_INFO (InfoType=SMB2_O_INFO_SECURITY)
-> smb2_set_info_sec()
-> set_info_sec()
-> parse_sec_desc()
-> parse_dacl()
-> init_acl_state(..., 0xffff)
-> init_acl_state(..., 0xffff)
-> kmalloc_objs(..., 0xffff)
Thus a malformed security descriptor can make num_aces pass the guard
and drive large temporary ACL state and pointer-array allocations.
Reject DACLs smaller than struct smb_acl before doing the subtraction,
so the ACE count check cannot be bypassed by the underflow.
Fixes: e2f34481b2 ("cifsd: add server-side procedures for SMB3")
Signed-off-by: Haofeng Li <lihaofeng@kylinos.cn>
Reviewed-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Free ndr buffer data when ndr_encode_dos_attr() returns error
to avoid memory leak.
Signed-off-by: Qiang Liu <liuqiang@kylinos.cn>
Reviewed-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
1. When ndr_decode_v4_ntacl() fails, the code jumped to free_n_data
which only freed n.data, skipping kfree(acl.sd_buf) and leaking
the buffer. Zero-initialize struct xattr_ntacl acl, reorder error
labels to out_free to release acl.sd_buf on all error paths.
2. if (acl.sd_size < sizeof(struct smb_ntsd)) is true, original code
returned success without freeing sd_buf and left stale *pntsd.
Set rc = -EINVAL before jumping to out_free to return error code and
free buffer.
Signed-off-by: Qiang Liu <liuqiang@kylinos.cn>
Reviewed-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
ndr_encode_v4_ntacl() allocates sd_ndr.data via kzalloc() at entry.
If any subsequent ndr_write_*() call returns error during encoding,
the allocated sd_ndr.data won't be freed and causes memory leak.
Move kfree(sd_ndr.data) into out label to ensure the buffer gets
released on all success and error return paths.
Signed-off-by: Qiang Liu <liuqiang@kylinos.cn>
Reviewed-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
tcf_bpf_act() can run from the tc egress path, which holds only
rcu_read_lock_bh(), but reads prog->filter with rcu_dereference() and
trips lockdep:
WARNING: suspicious RCU usage
net/sched/act_bpf.c:47 suspicious rcu_dereference_check() usage!
1 lock held by syz.2.1588/12756:
#0: (rcu_read_lock_bh){....}-{1:3}, at: __dev_queue_xmit net/core/dev.c:4792
tcf_bpf_act+0x6ae/0x940 net/sched/act_bpf.c:47
tcf_classify+0x6e4/0x1080 net/sched/cls_api.c:1860
sch_handle_egress net/core/dev.c:4545 [inline]
__dev_queue_xmit+0x2185/0x2c00 net/core/dev.c:4808
packet_sendmsg+0x3dfa/0x5120 net/packet/af_packet.c:3114
The other tc actions and cls_bpf already use rcu_dereference_bh() here.
Do the same.
Fixes: 1f211a1b92 ("net, sched: add clsact qdisc")
Signed-off-by: Sechang Lim <rhkrqnwk98@gmail.com>
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Link: https://patch.msgid.link/20260629154112.1164986-1-rhkrqnwk98@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The Kconfig entry for the al3320a is missing a `select REGMAP_I2C`,
causing build failures.
Fixes: 1850e6ae7f ("iio: light: al3320a: Implement regmap support")
Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Reviewed-by: David Heidelberg <david@ixit.cz>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The KConfig entry for the AL3010 is missing a `select REGMAP_I2C`,
causing build failures.
Fixes: 0e5e21e23d ("iio: light: al3010: Implement regmap support")
Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Reviewed-by: David Heidelberg <david@ixit.cz>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The KConfig entry for the al3000a is missing a `select REGMAP_I2C`,
causing build failures.
Fixes: d531b9f789 ("iio: light: Add support for AL3000a illuminance sensor")
Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Reviewed-by: David Heidelberg <david@ixit.cz>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
selinux_genfs_get_sid() allocates memory for a path with __get_free_page().
Such usage does not require a "page" and the size of the buffer should
actually be PATH_MAX which may be less than PAGE_SIZE on some
architectures.
Replace __get_free_page() for allocation of a path buffer with kmalloc()
and make it explicit that the buffer size is PATH_MAX.
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Acked-by: Stephen Smalley <stephen.smalley.work@gmail.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
Reading a word from the stack in a kretprobe crashes a risc-v kernel.
$ cd /sys/kernel/tracing/
$ echo 'r n_tty_write $stack0' > dynamic_events
$ echo 1 > events/kprobes/enable
Unable to handle kernel paging request at virtual address 0000000200000128
...
[<ffffffff80016d16>] regs_get_kernel_stack_nth+0x26/0x38
[<ffffffff80177196>] process_fetch_insn+0x3ee/0x760
[<ffffffff80177836>] kretprobe_trace_func+0x116/0x1f0
[<ffffffff8017795a>] kretprobe_dispatcher+0x4a/0x58
[<ffffffff8013572e>] kretprobe_rethook_handler+0x5e/0x90
[<ffffffff80180838>] rethook_trampoline_handler+0x70/0x108
[<ffffffff8001ba32>] arch_rethook_trampoline_callback+0x12/0x1c
[<ffffffff8001ba84>] arch_rethook_trampoline+0x48/0x94
[<ffffffff8067872a>] tty_write+0x1a/0x30
In regs_get_kernel_stack_nth, regs->sp contains an arbitrary value.
arch_rethook_trampoline saves the registers from the probed function in a
struct pt_regs. sp is not saved. Instead, sp is decremented for
arch_rethook_trampoline's local stack.
Fix this crash and save the original sp along with the other registers.
Use a0 as a temporary register, it is overwritten anyway.
Cc: stable@vger.kernel.org
Fixes: c22b0bcb1d ("riscv: Add kprobes supported")
Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://patch.msgid.link/20260630194010.1824039-1-martin@kaiser.cx
[pjw@kernel.org: added Fixes tag; cc'ed stable]
Signed-off-by: Paul Walmsley <pjw@kernel.org>
query_nic_features() detects which offloads depend on tx-gso-partial
by enabling everything, turning tx-gso-partial off, and seeing which
active features drop out. Enabling all hw features is dangerous:
we may end up enabling rx-fcs and loopback for example. For the
ice driver we end up getting into problems with feature dependencies
so the cleanup isn't successful either, and the test exits with
rx-fcs and loopback enabled.
Scope the feature probing just to segmentation bits.
Fixes: 266b835e5e ("selftests: drv-net: tso: enable test cases based on hw_features")
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Daniel Zahka <daniel.zahka@gmail.com>
Link: https://patch.msgid.link/20260629233923.2151144-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Depending on the value of chip_version, the correct decode set is selected.
However, the subsequent matching with the t4 encoding type in the if-else
block results in a reassignment, which leads to the loss of support for
t6_decode as well as reinitializing of values t4_decode and t5_decode.
The component history shows that the if-else block previously used for
this purpose, as well as the execution order, was not affected by the
change.
Furthermore, it is suggested by the execution order that the scenario with
overwriting and loss of support will be implemented.
Delete the if-else block.
Fixes: 6df397539c ("cxgb4: Update correct encoding of SGE Ingress DMA States for T6 adapter")
Signed-off-by: Gleb Markov <markov.gi@npc-ksb.ru>
Reviewed-by: Potnuri Bharat Teja <bharat@chelsio.com>
Link: https://patch.msgid.link/20260629130856.1168-1-markov.gi@npc-ksb.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Adjust the indentation and use tabs between function parameters and
their descriptions to align with the convention used in FPU code.
Signed-off-by: Chao Gao <chao.gao@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://patch.msgid.link/20250620052152.490414-1-chao.gao@intel.com
Breno Leitao says:
====================
net: do not warn on best-effort skb allocation failures
Both netconsole and netpoll keep a small preallocated pool of skbs
(skb_pool) so they can still get a buffer under memory pressure.
On the hot path they first attempt a normal GFP_ATOMIC allocation and only
fall back to the pool when that fails, keeping the pool as a last resort.
This is where the problem happens. If alloc_skb() fails, we now have
more than 100 message coming from the page=0 failure, which consumes
the scarce pool of skb, making the real issue disappear.
So the noise (memory allocation failure) deplets the SKB buffer and
crowds out the real message we were trying to deliver.
This is happening on the Meta fleet. The stack trace looks like:
pr/netcon_ext0: page allocation failure: order:0, mode:0x40820(GFP_ATOMIC|__GFP_COMP), nodemask=(null),cpuset=/,mems_allowed=0
...
dump_stack_lvl
warn_alloc
__alloc_pages_slowpath
__alloc_frozen_pages_noprof
alloc_pages_mpol
alloc_slab_page
allocate_slab
kmem_cache_alloc_node_noprof
__alloc_skb
send_udp
netconsole_write
nbcon_emit_next_record
nbcon_emit_one
nbcon_kthread_func
kthread
Solution: Do not warn if netconsole/netpoll fails to allocate these SKBs.
Pass __GFP_NOWARN on these best-effort allocations -- both the hot-path
attempt in netconsole's find_skb() and the pool refill in netpoll's
refill_skbs() -- and let the existing fallback paths do their job
quietly. The allocation will happen on SKB refill workqueue.
Given I am touching this code, if alloc_skb() fails, reschedule the
workqueue to try later.
====================
Link: https://patch.msgid.link/20260629-netpoll_no_warn-v1-0-f380f0b2cd0c@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
refill_skbs() tops up the per-netpoll skb pool with GFP_ATOMIC and
simply stops on the first allocation failure, leaving the pool partially
filled; a later refill tops it up once memory frees up. The allocation
failing is therefore an expected and fully handled condition, but
without __GFP_NOWARN the page allocator emits a warn_alloc() splat with
a full stack trace on every miss.
Pass __GFP_NOWARN so the best-effort refill stays quiet, mirroring the
same change in netconsole's find_skb().
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20260629-netpoll_no_warn-v1-2-f380f0b2cd0c@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
find_skb() allocates the skb with GFP_ATOMIC as a best-effort attempt:
on failure it falls back to the preallocated skb pool and, failing that,
polls the device and retries. The allocation failing is therefore an
expected and fully handled condition, but without __GFP_NOWARN the page
allocator still emits a warn_alloc() splat with a full stack trace on
every miss, which then consumes the whole SKB pool, that would be useful
printing the real issue rather than the memory failure.
Pass __GFP_NOWARN so the best-effort allocation stays quiet and lets the
existing fallback path do its job.
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20260629-netpoll_no_warn-v1-1-f380f0b2cd0c@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When a linear skb crosses a 16 KiB boundary, ioc3_start_xmit()
splits it into two buffers of lengths s1 and s2. The descriptor
advertises those lengths through B1CNT and B2CNT.
The first buffer is mapped with s1, but the second buffer is also
mapped with s1 even though the device is told to fetch s2 bytes from
it. When the lengths differ, the DMA mapping does not cover the same
region as the second descriptor buffer, which can result in incorrect
cache maintenance or a DMA fault on implementations that enforce the
mapped range.
There is a separate mismatch in the error path. If mapping the second
buffer fails, only d1 needs to be unmapped. d1 was mapped for s1 bytes,
but the driver unmaps it using the full packet length. Streaming DMA
mappings must be unmapped with the same size used for the corresponding
map operation.
Map the second buffer with s2 and unmap the first buffer with s1 when
the second mapping fails.
Cc: <stable+noautosel@kernel.org> # untested fix for ancient HW
Signed-off-by: Xu Rao <raoxu@uniontech.com>
Reviewed-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Link: https://patch.msgid.link/4E1486BC4536407E+20260629080623.908426-1-raoxu@uniontech.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ioc3eth_remove() frees the coherent RX and TX descriptor rings before
unregistering the netdev. If the interface is running,
unregister_netdev() invokes ioc3_close() through ndo_stop.
ioc3_close() stops the device and then calls ioc3_free_rx_bufs() and
ioc3_clean_tx_ring(). Both cleanup functions access descriptors in the
rings, so the current ordering causes CPU accesses to freed coherent
memory. Until ioc3_stop() disables RX and TX DMA, the device may also
continue using the freed ring addresses.
Unregister the netdev before releasing the rings. This lets the core
close a running interface and quiesce the device while the rings are
still valid. Keep the explicit timer deletion because ndo_stop is not
called when the interface is already down.
Cc: <stable+noautosel@kernel.org> # untested fix for ancient HW
Signed-off-by: Xu Rao <raoxu@uniontech.com>
Reviewed-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Link: https://patch.msgid.link/40CD736C4911C181+20260629085053.964383-1-raoxu@uniontech.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The JUnit XML output correctly captures the stderr/stdout output of failed
tests inside the <failure> element. However, for skipped tests, the output
was completely discarded and the XML only received a self-closing <skipped
message="reason"/> tag.
This expands the <skipped> element to include the test's err_output when
available, which is extremely helpful for debugging why a test was skipped
(e.g. diagnosing missing prerequisites or unexpected environment states
that triggered the skip) directly from CI systems parsing the XML report.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Optimize the execution of the metric validation and metric listing shell
test suites:
1. `stat_metrics_values.sh`:
The Python metric validator runs the `perf bench futex hash` workload
for each validated metric relationship. Reduce the benchmark runtime
limit from `-r 2` (2 seconds) to `-r 1` (1 second). This cuts the
workload duration in half while still generating sufficient PMU events
to satisfy non-zero threshold metric validations.
2. `stat_all_metrics.sh`:
The metric checking test runs `perf stat` sequentially across all
433+ listed metrics. Change the default workload for system-wide runs
from `sleep 0.01` to `true`. This avoids the 10ms sleep delay on each
sequential metric invocation, saving over 4 seconds of total wall
time during full test suite runs.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The lock contention analysis test suite (`lock_contention.sh`) performs a
series of 13 separate profiling checks to verify various aggregation and
filtering parameters of `perf lock contention`. Each of these checks runs
the `perf bench sched messaging` messaging benchmark as its workload.
By default, `sched messaging` runs 10 groups of 40 processes (400
processes total) generating substantial task scheduling, context
switching, and IPC message passing. When traced system-wide for lock
events, the tracing overhead (handling millions of lock acquisitions and
releases) slows execution down significantly, causing the test suite to
take over 80 seconds.
Optimize this by introducing a scaled-down messaging benchmark workload:
`perf bench sched messaging -g 1 -p`. Running 1 group (40 processes) takes
only 0.01 seconds natively (instead of 0.08 seconds), drastically reduces
the sheer volume of lock acquire/release trace events, and reduces CPU
context switching during tracing while still generating sufficient lock
events to fully exercise the BPF/record filters.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The off-cpu profiling test suite runs multiple recording commands with a
default workload of `sleep 1` to test the off-cpu threshold configurations
(specifically, above 999ms and below 1200ms). This adds a mandatory 3.0
seconds of sleep overhead.
Optimize this by scaling down the thresholds and workload durations by a
factor of 10:
- Use `sleep 0.1` as the workload duration.
- Change the above-threshold test to use `--off-cpu-thresh 50` and `sleep
0.1`.
- Change the below-threshold test to use `--off-cpu-thresh 500` and `sleep
0.1`.
- Update the awk period check in the above-threshold test to look for a
period greater than 50,000,000 ns (50ms) instead of 999,000,000 ns
(999ms).
This reduces raw test sleep overhead from 3.0s down to 0.3s, yielding a
~2.7 second speedup for this test.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The branch stack sampling test (test 130) runs short iteration-based
workloads to verify syscall, kernel, and trap branch stack sampling.
Specifically, `test_syscall()` and `test_kernel_branches()` run `perf
bench syscall basic` with loop counts of 8000 and 1000, and
`test_trap_eret_branches()` runs `traploop` with 1000 iterations.
Because these loop limits are extremely small, the total benchmark
runtimes last only a few milliseconds (or less). Under high load,
virtualization, or coarse sampling conditions, PMU cycle sampling fails to
capture enough samples inside the brief benchmark loops. This leads to
false negatives where the script output lacks the expected syscall,
kernel, or trap branch entries (e.g. "ERROR: Branches missing getppid[^
]*/SYSCALL/").
Fix this by increasing the workload loop counts to 100,000 across all
three test sections. Running 100,000 loops still finishes virtually
instantaneously (less than 0.1 seconds), but generates enough iterations
to guarantee robust branch stack capture.
Fixes: b55878c90a ("perf test: Add test for branch stack sampling")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The `perf stat --bpf-counters test` fails intermittently on hybrid
architectures or systems with dynamic frequency scaling (DVFS). This
happens because the test workload (`sqrtloop`) runs for a fixed 1-second
duration, and the CPU frequency can scale dynamically between idle and
maximum frequency. As the first run runs on a cold CPU and the second run
runs on a warmed-up CPU (or vice versa), the number of instructions
executed in 1 second differs by up to 2.2x, violating the comparison
tolerance.
Also, when running as root, BPF tracepoints and scheduling programs
trigger frequently. Since standard `perf stat -e instructions` measures
both user and kernel space instructions, it counts BPF helper and program
execution overheads, whereas the BPF counters themselves do not self-
measure. This introduces a large kernel-space instruction count
discrepancy between standard and BPF counters.
Fix these issues by:
1. Switching the workload to a strictly deterministic, iteration-based
workload: `awk 'BEGIN { for (i=0; i<10000000; i++) sum+=i }'`. We pin
the
workload to a single random allowed CPU using `taskset -c $CPU` via a
bash array.
2. Restricting the counted event to user-space only (`instructions:u` or
`/u`).
3. Tightening the comparison tolerance from 20% to 15%.
These modifications isolate the measurements to user-space instructions of
the deterministic loop, which executes a virtually identical number of
instructions on both runs (with less than 0.001% variation), eliminating
Dynamic Frequency Scaling (DVFS), kernel scheduling noise, and BPF helper
self-measurement overheads.
Fixes: 2c0cb9f560 ("perf test: Add a shell test for 'perf stat --bpf-counters' new option")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The `perf trace record and replay` test fails intermittently on slow or
virtualized hosts because the default recording workload (`sleep 1`)
occasionally completes without scheduling the target `nanosleep` or
`clock_nanosleep` system calls inside the recorded sample window,
resulting in the error: `Failed: cannot find *nanosleep syscall`.
Generalize the `perf_record_with_retry` helper in
`tests/shell/lib/perf_record.sh` to support a custom record command prefix
via the `PERF_RECORD_CMD` environment variable (defaulting to "perf
record").
Update `trace_record_replay.sh` to use this robust retry loop running with
`PERF_RECORD_CMD="perf trace record"` and a base workload of `sleep`. The
test will automatically retry with scaled sleep durations (from 0.01s up
to 2.0s) until the required `nanosleep` event is successfully captured.
Fixes: 15bcfb96d0 ("perf test: Add trace record and replay test")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The `python profiling with jitdump` test failed due to:
1. Target PID extraction resolving to duplicate space-separated values,
which broke the buildid-cache loops.
2. The default workload duration being too short to capture JIT stack
trampoline samples, resulting in 0 matching JIT symbols.
Fix the PID parsing by sorting and retrieving a unique single-line value.
Implement a robust retry loop starting at 1M python loop iterations and
scaling up to 100M iterations until JIT symbols are successfully captured
and verified.
Fixes: c9cd0c7e52 ("perf test: Add python JIT dump test")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The metrics value validation test requires system-wide recording (`-a`),
which can fail on systems without root permissions or where paranoid
levels restrict tracing. Add a check to skip the test if `-a` is not
supported.
Also fix false negatives during validation by updating parse error string
patterns and resolving issues in metric list generation.
Fixes: 3ad7092f51 ("perf test: Add metric value validation test")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Introduce `perf_record_with_retry` and `perf_record_cleanup` in a shared
library `tests/shell/lib/perf_record.sh` to prevent record test failures
caused by transient recording or workload delays.
Update `record.sh`, `record_lbr.sh`, `pipe_test.sh`, `kvm.sh`, and
`stat_all_pfm.sh` to use this robust record retry logic. These tests now
start with very short durations (e.g. 0.01 seconds) and scale up if the
initial recording failed to capture samples, significantly improving test
execution speed on success while remaining resilient to slow systems.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Currently, the noploop and thloop workloads only support sleep durations
in integer seconds because they parse the argument using atoi() and use
alarm() for timer signaling.
To support much shorter execution times in tests (speeding up test suites
and allowing faster retries), change the input parsing to use atof() for
double floating-point seconds. Use ualarm() for fractional durations less
than 1.0 seconds, and fall back to alarm() for durations of 1.0 second or
more.
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The parallel test harness uses the carriage return delete escape sequence
`PERF_COLOR_DELETE_LINE` ("\033[A\33[2K\r") to erase and update the
"Running (X active)" progress lines.
However, if a test description is longer than the terminal width, the line
wraps around. When this happens, the cursor up escape sequence `\033[A`
only moves the cursor to the last wrapped row, leaving the top half of the
description printed on the previous line. This leads to name duplication
and output corruption spilling over multiple rows on consoles narrower
than the maximum description length (e.g., 101 columns wide).
Fix this by dynamically querying the terminal width using
`get_term_dimensions` and truncating the printed test descriptions using
the `%-*.*s` printf format. We reserve 35 characters for prefix, status,
and spacing metrics to guarantee the progress line never wraps.
Fixes: 0e036dcad4 ("perf test: Display number of active running tests")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Commit b1c5efbfd9 ("perf parse-events: Remove hard coded legacy hardware
and cache parsing") introduced a bypass to PMU filtering to prevent uncore
PMUs from being filtered out during event parsing, which was required for
resolving `duration_time` and `uncore_freq` when running with `--cputype`.
However, this bypass was active whenever `pmu_filter` was set, which also
incorrectly bypassed filtering for the `--pmu-filter` option.
Introduce a `cputype_filter` boolean flag in `parse_events_state` and
`parse_events_option_args` to distinguish filtering initiated by
`--cputype` from that initiated by `--pmu-filter`. Restrict the core-only
check in `parse_events__filter_pmu()` to when `cputype_filter` is true.
Fixes: b1c5efbfd9 ("perf parse-events: Remove hard coded legacy hardware and cache parsing")
Assisted-by: Antigravity:gemini-3.1-pro
Signed-off-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The second loop for per_pidcommd is meaningful only when it doesn't have
from and to tasks. Also make sure c->Y is set before copying the comm
string otherwise it will be overwritten by next one.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The command line options can be passed without copy as their lifetime is
limited within the same function.
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The pattern of strdup() and free() is found, and I think it just can
use the original backtrace directly.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The backtrace was used by sched-switch and sched-wakeup only. No need
to call cat_backtrace() in the process_sample_event(). Let's pass NULL
backtrace and generate it from the sched events.
As it needs a pointer to the 'machine', let's save the session in the
timechart struct and use the host machine of the session instead of
passing the pointer to all handlers. It should be fine to assume the
host machine as timechart command doesn't deal with guest machines and
there's no way to get tracepoints from the guest events.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The cat_backtrace() is only called from process_sample_event() which
means the event type is always PERF_RECORD_SAMPLE. We don't need to
pass the event just because to print already known info.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
