| Age | Commit message (Collapse) | Author |
|---|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: 7e10b6554ff2 ("tcp: add dsack blocks received stats")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-10-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: fb31c9b9f6c8 ("tcp: add data bytes retransmitted stats")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-9-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: ba113c3aa79a ("tcp: add data bytes sent stats")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-8-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: 87ecc95d81d9 ("tcp: add send queue size stat in SCM_TIMESTAMPING_OPT_STATS")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: feb5f2ec6464 ("tcp: export packets delivery info")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: 7156d194a077 ("tcp: add snd_ssthresh stat in SCM_TIMESTAMPING_OPT_STATS")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE(), WRITE_ONCE() data_race() annotations to keep KCSAN happy.
Fixes: bb7c19f96012 ("tcp: add related fields into SCM_TIMESTAMPING_OPT_STATS")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() intentionally runs lockless, we must
add READ_ONCE() and WRITE_ONCE() annotations to keep KCSAN happy.
Fixes: 7e98102f4897 ("tcp: record pkts sent and retransmistted")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_get_timestamping_opt_stats() does not own the socket lock,
this is intentional.
It calls tcp_get_info_chrono_stats() while other threads could
change chrono fields in tcp_chrono_set().
I do not think we need coherent TCP socket state snapshot
in tcp_get_timestamping_opt_stats(), I chose to only
add annotations to keep KCSAN happy.
Fixes: 1c885808e456 ("tcp: SOF_TIMESTAMPING_OPT_STATS option for SO_TIMESTAMPING")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260416200319.3608680-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add field for missing lock around the hrtimer. Add 'Required' where
the core checks for valid entries.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
|
|
Fail gracefully if there is no channel array attached to the mailbox
controller. Otherwise the later dereference will cause an OOPS which
might not be seen because mailbox controllers might instantiate very
early. Remove the comment explaining the obvious while here.
Fixes: 2b6d83e2b8b7 ("mailbox: Introduce framework for mailbox")
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
|
|
smb2_get_ea() applies 4-byte alignment padding via memset() after
writing each EA entry. The bounds check on buf_free_len is performed
before the value memcpy, but the alignment memset fires unconditionally
afterward with no check on remaining space.
When the EA value exactly fills the remaining buffer (buf_free_len == 0
after value subtraction), the alignment memset writes 1-3 NUL bytes
past the buf_free_len boundary. In compound requests where the response
buffer is shared across commands, the first command (e.g., READ) can
consume most of the buffer, leaving a tight remainder for the QUERY_INFO
EA response. The alignment memset then overwrites past the physical
kvmalloc allocation into adjacent kernel heap memory.
Add a bounds check before the alignment memset to ensure buf_free_len
can accommodate the padding bytes.
This is the same bug pattern fixed by commit beef2634f81f ("ksmbd: fix
potencial OOB in get_file_all_info() for compound requests") and
commit fda9522ed6af ("ksmbd: fix OOB write in QUERY_INFO for compound
requests"), both of which added bounds checks before unconditional
writes in QUERY_INFO response handlers.
Cc: stable@vger.kernel.org
Fixes: e2b76ab8b5c9 ("ksmbd: add support for read compound")
Signed-off-by: Tristan Madani <tristan@talencesecurity.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
set_posix_acl_entries_dacl() and set_ntacl_dacl() accumulate ACE sizes
in u16 variables. When a file has many POSIX ACL entries, the
accumulated size can wrap past 65535, causing the pointer arithmetic
(char *)pndace + *size to land within already-written ACEs. Subsequent
writes then overwrite earlier entries, and pndacl->size gets a
truncated value.
Use check_add_overflow() at each accumulation point to detect the
wrap before it corrupts the buffer, consistent with existing
check_mul_overflow() usage elsewhere in smbacl.c.
Cc: stable@vger.kernel.org
Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3")
Signed-off-by: Tristan Madani <tristan@talencesecurity.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
In smb2_open, the call to ksmbd_put_durable_fd(fp) drops the reference
to the durable file descriptor early during the durable reconnect
process. If an error occurs subsequently (eg, ksmbd_iov_pin_rsp fails)
or a scavenger accesses the file, it leads to a use-after-free when
accessing fp properties (eg fp->create_time).
Move the single put to the end of the function below err_out2 so fp
stays valid until smb2_open returns.
Fixes: c8efcc786146 ("ksmbd: add support for durable handles v1/v2")
Signed-off-by: Akif <akif.sait111@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
smb_inherit_dacl() trusts the on-disk num_aces value from the parent
directory's DACL xattr and uses it to size a heap allocation:
aces_base = kmalloc(sizeof(struct smb_ace) * num_aces * 2, ...);
num_aces is a u16 read from le16_to_cpu(parent_pdacl->num_aces)
without checking that it is consistent with the declared pdacl_size.
An authenticated client whose parent directory's security.NTACL is
tampered (e.g. via offline xattr corruption or a concurrent path that
bypasses parse_dacl()) can present num_aces = 65535 with minimal
actual ACE data. This causes a ~8 MB allocation (not kzalloc, so
uninitialized) that the subsequent loop only partially populates, and
may also overflow the three-way size_t multiply on 32-bit kernels.
Additionally, the ACE walk loop uses the weaker
offsetof(struct smb_ace, access_req) minimum size check rather than
the minimum valid on-wire ACE size, and does not reject ACEs whose
declared size is below the minimum.
Reproduced on UML + KASAN + LOCKDEP against the real ksmbd code path.
A legitimate mount.cifs client creates a parent directory over SMB
(ksmbd writes a valid security.NTACL xattr), then the NTACL blob on
the backing filesystem is rewritten to set num_aces = 0xFFFF while
keeping the posix_acl_hash bytes intact so ksmbd_vfs_get_sd_xattr()'s
hash check still passes. A subsequent SMB2 CREATE of a child under
that parent drives smb2_open() into smb_inherit_dacl() (share has
"vfs objects = acl_xattr" set), which fails the page allocator:
WARNING: mm/page_alloc.c:5226 at __alloc_frozen_pages_noprof+0x46c/0x9c0
Workqueue: ksmbd-io handle_ksmbd_work
__alloc_frozen_pages_noprof+0x46c/0x9c0
___kmalloc_large_node+0x68/0x130
__kmalloc_large_node_noprof+0x24/0x70
__kmalloc_noprof+0x4c9/0x690
smb_inherit_dacl+0x394/0x2430
smb2_open+0x595d/0xabe0
handle_ksmbd_work+0x3d3/0x1140
With the patch applied the added guard rejects the tampered value
with -EINVAL before any large allocation runs, smb2_open() falls back
to smb2_create_sd_buffer(), and the child is created with a default
SD. No warning, no splat.
Fix by:
1. Validating num_aces against pdacl_size using the same formula
applied in parse_dacl().
2. Replacing the raw kmalloc(sizeof * num_aces * 2) with
kmalloc_array(num_aces * 2, sizeof(...)) for overflow-safe
allocation.
3. Tightening the per-ACE loop guard to require the minimum valid
ACE size (offsetof(smb_ace, sid) + CIFS_SID_BASE_SIZE) and
rejecting under-sized ACEs, matching the hardening in
smb_check_perm_dacl() and parse_dacl().
v1 -> v2:
- Replace the synthetic test-module splat in the changelog with a
real-path UML + KASAN reproduction driven through mount.cifs and
SMB2 CREATE; Namjae flagged the kcifs3_test_inherit_dacl_old name
in v1 since it does not exist in ksmbd.
- Drop the commit-hash citation from the code comment per Namjae's
review; keep the parse_dacl() pointer.
Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-6
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The global max_connections check in ksmbd's TCP accept path counts
the newly accepted connection with atomic_inc_return(), but then
rejects the connection when the result is greater than or equal to
server_conf.max_connections.
That makes the effective limit one smaller than configured. For
example:
- max_connections=1 rejects the first connection
- max_connections=2 allows only one connection
The per-IP limit in the same function uses <= correctly because it
counts only pre-existing connections. The global limit instead checks
the post-increment total, so it should reject only when that total
exceeds the configured maximum.
Fix this by changing the comparison from >= to >, so exactly
max_connections simultaneous connections are allowed and the next one
is rejected. This matches the documented meaning of max_connections
in fs/smb/server/ksmbd_netlink.h as the "Number of maximum simultaneous
connections".
Fixes: 0d0d4680db22 ("ksmbd: add max connections parameter")
Cc: stable@vger.kernel.org
Signed-off-by: DaeMyung Kang <charsyam@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Both ACE-walk loops in smb_check_perm_dacl() only guard against an
under-sized remaining buffer, not against an ACE whose declared
`ace->size` is smaller than the struct it claims to describe:
if (offsetof(struct smb_ace, access_req) > aces_size)
break;
ace_size = le16_to_cpu(ace->size);
if (ace_size > aces_size)
break;
The first check only requires the 4-byte ACE header to be in bounds;
it does not require access_req (4 bytes at offset 4) to be readable.
An attacker who has set a crafted DACL on a file they own can declare
ace->size == 4 with aces_size == 4, pass both checks, and then
granted |= le32_to_cpu(ace->access_req); /* upper loop */
compare_sids(&sid, &ace->sid); /* lower loop */
reads access_req at offset 4 (OOB by up to 4 bytes) and ace->sid at
offset 8 (OOB by up to CIFS_SID_BASE_SIZE + SID_MAX_SUB_AUTHORITIES
* 4 bytes).
Tighten both loops to require
ace_size >= offsetof(struct smb_ace, sid) + CIFS_SID_BASE_SIZE
which is the smallest valid on-wire ACE layout (4-byte header +
4-byte access_req + 8-byte sid base with zero sub-auths). Also
reject ACEs whose sid.num_subauth exceeds SID_MAX_SUB_AUTHORITIES
before letting compare_sids() dereference sub_auth[] entries.
parse_sec_desc() already enforces an equivalent check (lines 441-448);
smb_check_perm_dacl() simply grew weaker validation over time.
Reachability: authenticated SMB client with permission to set an ACL
on a file. On a subsequent CREATE against that file, the kernel
walks the stored DACL via smb_check_perm_dacl() and triggers the
OOB read. Not pre-auth, and the OOB read is not reflected to the
attacker, but KASAN reports and kernel state corruption are
possible.
Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-6
Assisted-by: Codex:gpt-5-4
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
ipc_validate_msg() computes the expected message size for each
response type by adding (or multiplying) attacker-controlled fields
from the daemon response to a fixed struct size in unsigned int
arithmetic. Three cases can overflow:
KSMBD_EVENT_RPC_REQUEST:
msg_sz = sizeof(struct ksmbd_rpc_command) + resp->payload_sz;
KSMBD_EVENT_SHARE_CONFIG_REQUEST:
msg_sz = sizeof(struct ksmbd_share_config_response) +
resp->payload_sz;
KSMBD_EVENT_LOGIN_REQUEST_EXT:
msg_sz = sizeof(struct ksmbd_login_response_ext) +
resp->ngroups * sizeof(gid_t);
resp->payload_sz is __u32 and resp->ngroups is __s32. Each addition
can wrap in unsigned int; the multiplication by sizeof(gid_t) mixes
signed and size_t, so a negative ngroups is converted to SIZE_MAX
before the multiply. A wrapped value of msg_sz that happens to
equal entry->msg_sz bypasses the size check on the next line, and
downstream consumers (smb2pdu.c:6742 memcpy using rpc_resp->payload_sz,
kmemdup in ksmbd_alloc_user using resp_ext->ngroups) then trust the
unverified length.
Use check_add_overflow() on the RPC_REQUEST and SHARE_CONFIG_REQUEST
paths to detect integer overflow without constraining functional
payload size; userspace ksmbd-tools grows NDR responses in 4096-byte
chunks for calls like NetShareEnumAll, so a hard transport cap is
unworkable on the response side. For LOGIN_REQUEST_EXT, reject
resp->ngroups outside the signed [0, NGROUPS_MAX] range up front and
report the error from ipc_validate_msg() so it fires at the IPC
boundary; with that bound the subsequent multiplication and addition
stay well below UINT_MAX. The now-redundant ngroups check and
pr_err in ksmbd_alloc_user() are removed.
This is the response-side analogue of aab98e2dbd64 ("ksmbd: fix
integer overflows on 32 bit systems"), which hardened the request
side.
Fixes: 0626e6641f6b ("cifsd: add server handler for central processing and tranport layers")
Fixes: a77e0e02af1c ("ksmbd: add support for supplementary groups")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-6
Assisted-by: Codex:gpt-5-4
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Commit 77ffbcac4e56 ("smb: server: fix leak of active_num_conn in
ksmbd_tcp_new_connection()") addressed the kthread_run() failure
path. The earlier alloc_transport() == NULL path in the same
function has the same leak, is reachable pre-authentication via any
TCP connect to port 445, and was empirically reproduced on UML
(ARCH=um, v7.0-rc7): a small number of forced allocation failures
were sufficient to put ksmbd into a state where every subsequent
connection attempt was rejected for the remainder of the boot.
ksmbd_kthread_fn() increments active_num_conn before calling
ksmbd_tcp_new_connection() and discards the return value, so when
alloc_transport() returns NULL the socket is released and -ENOMEM
returned without decrementing the counter. Each such failure
permanently consumes one slot from the max_connections pool; once
cumulative failures reach the cap, atomic_inc_return() hits the
threshold on every subsequent accept and every new connection is
rejected. The counter is only reset by module reload.
An unauthenticated remote attacker can drive the server toward the
memory pressure that makes alloc_transport() fail by holding open
connections with large RFC1002 lengths up to MAX_STREAM_PROT_LEN
(0x00FFFFFF); natural transient allocation failures on a loaded
host produce the same drift more slowly.
Mirror the existing rollback pattern in ksmbd_kthread_fn(): on the
alloc_transport() failure path, decrement active_num_conn gated on
server_conf.max_connections.
Repro details: with the patch reverted, forced alloc_transport()
NULL returns leaked counter slots and subsequent connection
attempts -- including legitimate connects issued after the
forced-fail window had closed -- were all rejected with "Limit the
maximum number of connections". With this patch applied, the same
connect sequence produces no rejections and the counter cycles
cleanly between zero and one on every accept.
Fixes: 0d0d4680db22 ("ksmbd: add max connections parameter")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-6
Assisted-by: Codex:gpt-5-4
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c updates from Wolfram Sang:
"The biggest news in this pull request is that it will start the last
cycle of me handling the I2C subsystem. From 7.2. on, I will pass
maintainership to Andi Shyti who has been maintaining the I2C drivers
for a while now and who has done a great job in doing so.
We will use this cycle for a hopefully smooth transition. Thanks must
go to Andi for stepping up! I will still be around for guidance.
Updates:
- generic cleanups in npcm7xx, qcom-cci, xiic and designware DT
bindings
- atr: use kzalloc_flex for alias pool allocation
- ixp4xx: convert bindings to DT schema
- ocores: use read_poll_timeout_atomic() for polling waits
- qcom-geni: skip extra TX DMA TRE for single read messages
- s3c24xx: validate SMBus block length before using it
- spacemit: refactor xfer path and add K1 PIO support
- tegra: identify DVC and VI with SoC data variants
- tegra: support SoC-specific register offsets
- xiic: switch to devres and generic fw properties
- xiic: skip input clock setup on non-OF systems
- various minor improvements in other drivers
rtl9300:
- add per-SoC callbacks and clock support for RTL9607C
- add support for new 50 kHz and 2.5 MHz bus speeds
- general refactoring in preparation for RTL9607C support
New support:
- DesignWare GOOG5000 (ACPI HID)
- Intel Nova Lake (ACPI ID)
- Realtek RTL9607C
- SpacemiT K3 binding
- Tegra410 register layout support"
* tag 'i2c-for-7.1-rc1-part1' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux: (40 commits)
i2c: usbio: Add ACPI device-id for NVL platforms
i2c: qcom-geni: Avoid extra TX DMA TRE for single read message in GPI mode
i2c: atr: use kzalloc_flex
i2c: spacemit: introduce pio for k1
i2c: spacemit: move i2c_xfer_msg()
i2c: xiic: skip input clock setup on non-OF systems
i2c: xiic: use numbered adapter registration
i2c: xiic: cosmetic: use resource format specifier in debug log
i2c: xiic: cosmetic cleanup
i2c: xiic: switch to generic device property accessors
i2c: xiic: remove duplicate error message
i2c: xiic: switch to devres managed APIs
i2c: rtl9300: add RTL9607C i2c controller support
i2c: rtl9300: introduce new function properties to driver data
i2c: rtl9300: introduce clk struct for upcoming rtl9607 support
dt-bindings: i2c: realtek,rtl9301-i2c: extend for clocks and RTL9607C support
i2c: rtl9300: introduce a property for 8 bit width reg address
i2c: rtl9300: introduce F_BUSY to the reg_fields struct
i2c: rtl9300: introduce max length property to driver data
i2c: rtl9300: split data_reg into read and write reg
...
|
|
Pull ipmi updates from Corey Minyard:
"Small updates and fixes (mostly to the BMC software):
- Fix one issue in the host side driver where a kthread can be left
running on a specific memory allocation failre at probe time
- Replace system_wq with system_percpu_wq so system_wq can eventually
go away"
* tag 'for-linus-7.1-1' of https://github.com/cminyard/linux-ipmi:
ipmi:ssif: Clean up kthread on errors
ipmi:ssif: Remove unnecessary indention
ipmi: ssif_bmc: Fix KUnit test link failure when KUNIT=m
ipmi: ssif_bmc: add unit test for state machine
ipmi: ssif_bmc: change log level to dbg in irq callback
ipmi: ssif_bmc: fix message desynchronization after truncated response
ipmi: ssif_bmc: fix missing check for copy_to_user() partial failure
ipmi: ssif_bmc: cancel response timer on remove
ipmi: Replace use of system_wq with system_percpu_wq
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools
Pull perf tools updates from Namhyung Kim:
"perf report:
- Add 'comm_nodigit' sort key to combine similar threads that only
have different numbers in the comm. In the following example, the
'comm_nodigit' will have samples from all threads starting with
"bpfrb/" into an entry "bpfrb/<N>".
$ perf report -s comm_nodigit,comm -H
...
#
# Overhead CommandNoDigit / Command
# ........... ........................
#
20.30% swapper
20.30% swapper
13.37% chrome
13.37% chrome
10.07% bpfrb/<N>
7.47% bpfrb/0
0.70% bpfrb/1
0.47% bpfrb/3
0.46% bpfrb/2
0.25% bpfrb/4
0.23% bpfrb/5
0.20% bpfrb/6
0.14% bpfrb/10
0.07% bpfrb/7
- Support flat layout for symfs. The --symfs option is to specify the
location of debugging symbol files. The default 'hierarchy' layout
would search the symbol file using the same path of the original
file under the symfs root. The new 'flat' layout would search only
in the root directory.
- Update 'simd' sort key for ARM SIMD flags to cover ASE/SME and more
predicate flags.
perf stat:
- Add --pmu-filter option to select specific PMUs. This would be
useful when you measure metrics from multiple instance of uncore
PMUs with similar names.
# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':
19,417,779,115 hisi_sicl0_cpa0/cpa_cycles/ # 0.00 cpa_p0_avg_bw
0 hisi_sicl0_cpa0/cpa_p0_wr_dat/
0 hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
0 hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
19,417,751,103 hisi_sicl10_cpa0/cpa_cycles/ # 0.00 cpa_p0_avg_bw
0 hisi_sicl10_cpa0/cpa_p0_wr_dat/
0 hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
0 hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
19,417,730,679 hisi_sicl2_cpa0/cpa_cycles/ # 0.31 cpa_p0_avg_bw
75,635,749 hisi_sicl2_cpa0/cpa_p0_wr_dat/
18,520,640 hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
0 hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
19,417,674,227 hisi_sicl8_cpa0/cpa_cycles/ # 0.00 cpa_p0_avg_bw
0 hisi_sicl8_cpa0/cpa_p0_wr_dat/
0 hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
0 hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
19.417734480 seconds time elapsed
With --pmu-filter, users can select only hisi_sicl2_cpa0 PMU.
# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':
6,234,093,559 cpa_cycles # 0.60 cpa_p0_avg_bw
50,548,465 cpa_p0_wr_dat
7,552,182 cpa_p0_rd_dat_64b
0 cpa_p0_rd_dat_32b
6.234139320 seconds time elapsed
Data type profiling:
- Quality improvements by tracking register state more precisely
- Ensure array members to get the type
- Handle more cases for global variables
Vendor event/metric updates:
- Update various Intel events and metrics
- Add NVIDIA Tegra 410 Olympus events
Internal changes:
- Verify perf.data header for maliciously crafted files
- Update perf test to cover more usages and make them robust
- Move a couple of copied kernel headers not to annoy objtool build
- Fix a bug in map sorting in name order
- Remove some unused codes
Misc:
- Fix module symbol resolution with non-zero text address
- Add -t/--threads option to `perf bench mem mmap`
- Track duration of exit*() syscall by `perf trace -s`
- Add core.addr2line-timeout and core.addr2line-disable-warn config
items"
* tag 'perf-tools-for-v7.1-2026-04-17' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (131 commits)
perf loongarch: Fix build failure with CONFIG_LIBDW_DWARF_UNWIND
perf annotate: Use jump__delete when freeing LoongArch jumps
perf test: Fixes for check branch stack sampling
perf test: Fix inet_pton probe failure and unroll call graph
perf build: fix "argument list too long" in second location
perf header: Add sanity checks to HEADER_BPF_BTF processing
perf header: Sanity check HEADER_BPF_PROG_INFO
perf header: Sanity check HEADER_PMU_CAPS
perf header: Sanity check HEADER_HYBRID_TOPOLOGY
perf header: Sanity check HEADER_CACHE
perf header: Sanity check HEADER_GROUP_DESC
perf header: Sanity check HEADER_PMU_MAPPINGS
perf header: Sanity check HEADER_MEM_TOPOLOGY
perf header: Sanity check HEADER_NUMA_TOPOLOGY
perf header: Sanity check HEADER_CPU_TOPOLOGY
perf header: Sanity check HEADER_NRCPUS and HEADER_CPU_DOMAIN_INFO
perf header: Bump up the max number of command line args allowed
perf header: Validate nr_domains when reading HEADER_CPU_DOMAIN_INFO
perf sample: Fix documentation typo
perf arm_spe: Improve SIMD flags setting
...
|
|
CONFIG_FIRMWARE_EDID=y depends on X86 or EFI_GENERIC_STUB. Neither
is true here, so drop the lines from the defconfig files.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
|
|
AT_SYSINFO_EHDR defines the auxvector index representing the vDSO
entrypoint. Its value or presence does not depend on whether a vDSO
is actually provided by the kernel.
The definition of AT_SYSINFO_EHDR was gated between CONFIG_VSYSCALL to
avoid a default gate VMA to be created. However that default gate VMA
was removed entirely in commit a6c19dfe3994
("arm64,ia64,ppc,s390,sh,tile,um,x86,mm: remove default gate area").
Remove the now unnecessary conditional.
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Reviewed-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
|
|
Both platform_early.c and platform_early.h have an extra dash in
their SPDX-License-Identifier lines. Use the correct (single-dash)
syntax for these lines.
Signed-off-by: Tim Bird <tim.bird@sony.com>
Reviewed-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
|
|
Include <linux/io.h> to avoid depending on <linux/backlight.h>
for including it. Declares __raw_readb() and __raw_writeb().
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202510282206.wI0HrqcK-lkp@intel.com/
Fixes: 243ce64b2b37 ("backlight: Do not include <linux/fb.h> in header file")
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Daniel Thompson (RISCstar) <danielt@kernel.org>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Cc: Lee Jones <lee@kernel.org>
Cc: Daniel Thompson <danielt@kernel.org>
Cc: Jingoo Han <jingoohan1@gmail.com>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Reviewed-by: Daniel Thompson (RISCstar) <danielt@kernel.org>
Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
|
|
The EL2 code defines ranges of host hypercalls that are either
enabled at boot-time only, used by [nh]VHE KVM, or reserved to pKVM.
The way these ranges are delineated is error prone, as the enum symbols
defining the limits are expressed in terms of actual function symbols.
This means that should a new function be added, special care must be
taken to also update the limit symbol.
Improve this by reusing the mechanism introduced for the vcpu_sysreg
enum, which uses a MARKER() macro and some extra trickery to make
the limit symbol standalone. Crucially, the limit symbol has the
same value as the *following* symbol.
The handle_host_hcall() function is then updated to make use of
the new limit definitions and get rid of the brittle default
upper limit. This allows for some more strict checks at build
time, and the removal of an comparison at run time.
Tested-by: Fuad Tabba <tabba@google.com>
Reviewed-by: Fuad Tabba <tabba@google.com>
Link: https://patch.msgid.link/20260414160528.2218858-1-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
|
|
Add myself as a page cache reviewer since I tend to review changes in
these areas anyway.
[akpm@linux-foundation.org: add linux-mm@kvack.org]
Link: https://lore.kernel.org/20260415174039.13016-2-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Acked-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Lorenzo Stoakes <ljs@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When the -fsanitize=bounds sanitizer is enabled, gcc-16 sometimes runs
into a corner case in the read_ctrl_pos() pos function, where it sees
possible undefined behavior from the 'tier' index overflowing, presumably
in the case that this was called with a negative tier:
In function 'get_tier_idx',
inlined from 'isolate_folios' at mm/vmscan.c:4671:14:
mm/vmscan.c: In function 'isolate_folios':
mm/vmscan.c:4645:29: error: 'pv.refaulted' is used uninitialized [-Werror=uninitialized]
Part of the problem seems to be that read_ctrl_pos() has unusual calling
conventions since commit 37a260870f2c ("mm/mglru: rework type selection")
where passing MAX_NR_TIERS makes it accumulate all tiers but passing a
smaller positive number makes it read a single tier instead.
Shut up the warning by adding a fake initialization to the two instances
of this variable that can run into that corner case.
Link: https://lore.kernel.org/all/CAJHvVcjtFW86o5FoQC8MMEXCHAC0FviggaQsd5EmiCHP+1fBpg@mail.gmail.com/
Link: https://lore.kernel.org/20260414065206.3236176-1-arnd@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Barry Song <baohua@kernel.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Kairui Song <kasong@tencent.com>
Cc: Koichiro Den <koichiro.den@canonical.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Wei Xu <weixugc@google.com>
Cc: Yuanchu Xie <yuanchu@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Use my personal email address due to the Red Hat work will stop soon
Link: https://lore.kernel.org/ad8GFhh3SI1wb7IC@darkstar.users.ipa.redhat.com
Signed-off-by: Dave Young <ruirui.yang@linux.dev>
Acked-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The directory does not exist any more.
Link: https://lore.kernel.org/20260414121752.1912847-4-pratyush@kernel.org
Signed-off-by: Pratyush Yadav (Google) <pratyush@kernel.org>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: SeongJae Park <sj@kernel.org>
Cc: Alexander Graf <graf@amazon.com>
Cc: Baoquan He <bhe@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The KHO entry already includes include/linux/kho. Listing its
subdirectory is redundant.
Link: https://lore.kernel.org/20260414121752.1912847-3-pratyush@kernel.org
Signed-off-by: Pratyush Yadav (Google) <pratyush@kernel.org>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: SeongJae Park <sj@kernel.org>
Cc: Alexander Graf <graf@amazon.com>
Cc: Baoquan He <bhe@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "MAINTAINERS: update KHO and LIVE UPDATE entries".
This series contains some updates for the Kexec Handover (KHO) and Live
update entries. Patch 1 updates the maintainers list and adds the
liveupdate tree. Patches 2 and 3 clean up stale files in the list.
This patch (of 3):
I have been helping out with reviewing and developing KHO. I would also
like to help maintain it. Change my entry from R to M for KHO and live
update. Alex has been inactive for a while, so to avoid over-crowding the
KHO entry and to keep the information up-to-date, move his entry from M to
R.
We also now have a tree for KHO and live update at liveupdate/linux.git
where we plan to start maintaining those subsystems and start queuing the
patches. List that in the entries as well.
Link: https://lore.kernel.org/20260414121752.1912847-1-pratyush@kernel.org
Link: https://lore.kernel.org/20260414121752.1912847-2-pratyush@kernel.org
Signed-off-by: Pratyush Yadav (Google) <pratyush@kernel.org>
Reviewed-by: Alexander Graf <graf@amazon.com>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Baoquan He <bhe@redhat.com>
Cc: David Hildenbrand <david@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Update KEXEC and KDUMP maintainer entries by adding the live update group
maintainers. Remove Vivek Goyal due to inactivity to keep the MAINTAINERS
file up-to-date, and add Vivek to the CREDITS file to recognize their
contributions.
Link: https://lore.kernel.org/20260413121146.49215-1-pasha.tatashin@soleen.com
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: Pratyush Yadav <pratyush@kernel.org>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Diego Viola <diego.viola@gmail.com>
Cc: Jakub Kacinski <kuba@kernel.org>
Cc: Magnus Karlsson <magnus.karlsson@intel.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Martin Kepplinger <martink@posteo.de>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
migrate_vma_collect_huge_pmd()
The softleaf_is_migration() check is unreachable as entries that are not
device_private are filtered out. Similarly, the PTE-level equivalent in
migrate_vma_collect_pmd() skips migration entries.
This dead branch also contained a double spin_unlock(ptl) bug.
Link: https://lore.kernel.org/20260212014611.416695-1-dave@stgolabs.net
Fixes: a30b48bf1b244 ("mm/migrate_device: implement THP migration of zone device pages")
Signed-off-by: Davidlohr Bueso <dave@stgolabs.net>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Alistair Popple <apopple@nvidia.com>
Acked-by: Balbir Singh <balbirs@nvidia.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Cc: Byungchul Park <byungchul@sk.com>
Cc: Gregory Price <gourry@gourry.net>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Joshua Hahn <joshua.hahnjy@gmail.com>
Cc: Mathew Brost <matthew.brost@intel.com>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Ying Huang <ying.huang@linux.alibaba.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
charge_reserved_hugetlb.sh tears down background writers with killall from
psmisc. Minimal Ubuntu images do not always provide that tool, so the
selftest fails in cleanup for an environment reason rather than for the
hugetlb behavior it is trying to cover.
Skip the test when killall is unavailable, similar to the existing root
check, so these environments report the dependency clearly instead of
failing the test.
Link: https://lore.kernel.org/20260410044139.67480-1-create0818@163.com
Signed-off-by: Cao Ruichuang <create0818@163.com>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The current implementation of validate_range() in fs/userfaultfd.c
performs a hard check against mmap_min_addr. This is redundant because
UFFDIO_REGISTER operates on memory ranges that must already be backed by a
VMA.
Enforcing mmap_min_addr or capability checks again in userfaultfd is
unnecessary and prevents applications like binary compilers from using
UFFD for valid memory regions mapped by application.
Remove the redundant check for mmap_min_addr.
We started using UFFD instead of the classic mprotect approach in the
binary translator to track application writes. During development, we
encountered this bug. The translator cannot control where the translated
application chooses to map its memory and if the app requires a
low-address area, UFFD fails, whereas mprotect would work just fine. I
believe this is a genuine logic bug rather than an improvement, and I
would appreciate including the fix in stable.
Link: https://lore.kernel.org/20260409103345.15044-1-komlomal@gmail.com
Fixes: 86039bd3b4e6 ("userfaultfd: add new syscall to provide memory externalization")
Signed-off-by: Denis M. Karpov <komlomal@gmail.com>
Reviewed-by: Lorenzo Stoakes <ljs@kernel.org>
Acked-by: Harry Yoo (Oracle) <harry@kernel.org>
Reviewed-by: Pedro Falcato <pfalcato@suse.de>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
vmstat_shepherd uses delayed_work_pending() to check whether vmstat_update
is already scheduled for a given CPU before queuing it. However,
delayed_work_pending() only tests WORK_STRUCT_PENDING_BIT, which is
cleared the moment a worker thread picks up the work to execute it.
This means that while vmstat_update is actively running on a CPU,
delayed_work_pending() returns false. If need_update() also returns true
at that point (per-cpu counters not yet zeroed mid-flush), the shepherd
queues a second invocation with delay=0, causing vmstat_update to run
again immediately after finishing.
On a 72-CPU system this race is readily observable: before the fix, many
CPUs show invocation gaps well below 500 jiffies (the minimum
round_jiffies_relative() can produce), with the most extreme cases
reaching 0 jiffies—vmstat_update called twice within the same jiffy.
Fix this by replacing delayed_work_pending() with work_busy(), which
returns non-zero for both WORK_BUSY_PENDING (timer armed or work queued)
and WORK_BUSY_RUNNING (work currently executing). The shepherd now
correctly skips a CPU in all busy states.
After the fix, all sub-jiffy and most sub-100-jiffie gaps disappear. The
remaining early invocations have gaps in the 700–999 jiffie range,
attributable to round_jiffies_relative() aligning to a nearer
jiffie-second boundary rather than to this race.
Each spurious vmstat_update invocation has a measurable side effect:
refresh_cpu_vm_stats() calls decay_pcp_high() for every zone, which drains
idle per-CPU pages back to the buddy allocator via free_pcppages_bulk(),
taking the zone spinlock each time. Eliminating the double-scheduling
therefore reduces zone lock contention directly. On a 72-CPU stress-ng
workload measured with perf lock contention:
free_pcppages_bulk contention count: ~55% reduction
free_pcppages_bulk total wait time: ~57% reduction
free_pcppages_bulk max wait time: ~47% reduction
Note: work_busy() is inherently racy—between the check and the
subsequent queue_delayed_work_on() call, vmstat_update can finish
execution, leaving the work neither pending nor running. In that narrow
window the shepherd can still queue a second invocation. After the fix,
this residual race is rare and produces only occasional small gaps, a
significant improvement over the systematic double-scheduling seen with
delayed_work_pending().
Link: https://lore.kernel.org/20260409-vmstat-v2-1-e9d9a6db08ad@debian.org
Fixes: 7b8da4c7f07774 ("vmstat: get rid of the ugly cpu_stat_off variable")
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Dmitry Ilvokhin <d@ilvokhin.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Hildenbrand <david@kernel.org>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
If hugepages, hugepagesz, or default_hugepagesz are specified on the
kernel command line without the '=' separator, early parameter parsing
passes NULL to hugetlb_add_param(), which dereferences it in strlen() and
can crash the system during early boot.
Reject NULL values in hugetlb_add_param() and return -EINVAL instead.
Link: https://lore.kernel.org/20260409105437.108686-4-thorsten.blum@linux.dev
Fixes: 5b47c02967ab ("mm/hugetlb: convert cmdline parameters from setup to early")
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Cc: David Hildenbrand <david@kernel.org>
Cc: Frank van der Linden <fvdl@google.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
recompress_store() parses the type= parameter with three if statements
checking for "idle", "huge", and "huge_idle". An unrecognized value
silently falls through with mode left at 0, causing the recompression pass
to run with no slot filter — processing all slots instead of the
intended subset.
Add a !mode check after the type parsing block to return -EINVAL for
unrecognized values, consistent with the function's other parameter
validation.
Link: https://lore.kernel.org/20260407153027.42425-1-astellman@stellman-greene.com
Signed-off-by: Andrew Stellman <astellman@stellman-greene.com>
Suggested-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The ProtectionKey entry was added in v4.9; back then it was x86-specific,
but it now lives in generic code and applies to all architectures
supporting pkeys (currently x86, power, arm64).
Time to document it: add a paragraph to proc.rst about the ProtectionKey
entry.
[akpm@linux-foundation.org: s/system/hardware/, per review discussion]
[akpm@linux-foundation.org: s/hardware/CPU/]
Link: https://lore.kernel.org/20260407125133.564182-1-kevin.brodsky@arm.com
Signed-off-by: Kevin Brodsky <kevin.brodsky@arm.com>
Reported-by: Yury Khrustalev <yury.khrustalev@arm.com>
Acked-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Reviewed-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Lorenzo Stoakes <ljs@kernel.org>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Kevin Brodsky <kevin.brodsky@arm.com>
Cc: Marc Rutland <mark.rutland@arm.com>
Cc: Shuah Khan <skhan@linuxfoundation.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The common order-0 case is important enough to want its own branch, and
avoids the hairy, large loop logic that the CPU does not seem to handle
particularly well.
While at it, encourage the compiler to inline batch PTE logic and resolve
constant branches by adding __always_inline strategically.
Link: https://lore.kernel.org/20260402141628.3367596-3-pfalcato@suse.de
Signed-off-by: Pedro Falcato <pfalcato@suse.de>
Suggested-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Tested-by: Luke Yang <luyang@redhat.com>
Reviewed-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jiri Hladky <jhladky@redhat.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "mm/mprotect: micro-optimization work", v3.
Micro-optimize the change_protection functionality and the
change_pte_range() routine. This set of functions works in an incredibly
tight loop, and even small inefficiencies are incredibly evident when spun
hundreds, thousands or hundreds of thousands of times.
There was an attempt to keep the batching functionality as much as
possible, which introduced some part of the slowness, but not all of it.
Removing it for !arm64 architectures would speed mprotect() up even
further, but could easily pessimize cases where large folios are mapped
(which is not as rare as it seems, particularly when it comes to the page
cache these days).
The micro-benchmark used for the tests was [0] (usable using
google/benchmark and g++ -O2 -lbenchmark repro.cpp)
This resulted in the following (first entry is baseline):
---------------------------------------------------------
Benchmark Time CPU Iterations
---------------------------------------------------------
mprotect_bench 85967 ns 85967 ns 6935
mprotect_bench 70684 ns 70684 ns 9887
After the patchset we can observe an ~18% speedup in mprotect. Wonderful
for the elusive mprotect-based workloads!
Testing & more ideas welcome. I suspect there is plenty of improvement
possible but it would require more time than what I have on my hands right
now. The entire inlined function (which inlines into change_protection())
is gigantic - I'm not surprised this is so finnicky.
Note: per my profiling, the next _big_ bottleneck here is
modify_prot_start_ptes, exactly on the xchg() done by x86.
ptep_get_and_clear() is _expensive_. I don't think there's a properly
safe way to go about it since we do depend on the D bit quite a lot. This
might not be such an issue on other architectures.
Luke Yang reported [1]:
: On average, we see improvements ranging from a minimum of 5% to a
: maximum of 55%, with most improvements showing around a 25% speed up in
: the libmicro/mprot_tw4m micro benchmark.
This patch (of 2):
Move softleaf change_pte_range code into a separate function. This makes
the change_pte_range() function a good bit smaller, and lessens cognitive
load when reading through the function.
Link: https://lore.kernel.org/20260402141628.3367596-1-pfalcato@suse.de
Link: https://lore.kernel.org/20260402141628.3367596-2-pfalcato@suse.de
Link: https://lore.kernel.org/all/aY8-XuFZ7zCvXulB@luyang-thinkpadp1gen7.toromso.csb/
Link: https://gist.github.com/heatd/1450d273005aba91fa5744f44dfcd933 [0]
Link: https://lore.kernel.org/CAL2CeBxT4jtJ+LxYb6=BNxNMGinpgD_HYH5gGxOP-45Q2OncqQ@mail.gmail.com [1]
Signed-off-by: Pedro Falcato <pfalcato@suse.de>
Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Tested-by: Luke Yang <luyang@redhat.com>
Reviewed-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jiri Hladky <jhladky@redhat.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Android systems usually use memory.reclaim interface to implement user
space memory management which expects that the requested reclaim target
and actually reclaimed amount memory are not diverging by too much. With
the current MGRLU implementation there is, however, no bail out when the
reclaim target is reached and this could lead to an excessive reclaim
that scales with the reclaim hierarchy size.For example, we can get a
nr_reclaimed=394/nr_to_reclaim=32 proactive reclaim under a common 1-N
cgroup hierarchy.
This defect arose from the goal of keeping fairness among memcgs that is,
for try_to_free_mem_cgroup_pages -> shrink_node_memcgs -> shrink_lruvec ->
lru_gen_shrink_lruvec -> try_to_shrink_lruvec, the !root_reclaim(sc) check
was there for reclaim fairness, which was necessary before commit
b82b530740b9 ("mm: vmscan: restore incremental cgroup iteration") because
the fairness depended on attempted proportional reclaim from every memcg
under the target memcg. However after commit b82b530740b9 there is no
longer a need to visit every memcg to ensure fairness. Let's have
try_to_shrink_lruvec bail out when the nr_reclaimed achieved.
Link: https://lore.kernel.org/20260318011558.1696310-1-zhaoyang.huang@unisoc.com
Link: https://lore.kernel.org/20260212032111.408865-1-zhaoyang.huang@unisoc.com
Signed-off-by: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
Suggested-by: T.J.Mercier <tjmercier@google.com>
Reviewed-by: T.J. Mercier <tjmercier@google.com>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Qi Zheng <qi.zheng@linux.dev>
Reviewed-by: Barry Song <baohua@kernel.org>
Reviewed-by: Kairui Song <kasong@tencent.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Yu Zhao <yuzhao@google.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Yuanchu Xie <yuanchu@google.com>
Cc: Wei Xu <weixugc@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The comment in mmzone.h currently details exhaustive per-architecture
bit-width lists and explains alignment using min(PAGE_SHIFT,
PFN_SECTION_SHIFT). Such details risk falling out of date over time and
may inadvertently be left un-updated.
We always expect a single section to cover full pages. Therefore, we can
safely assume that PFN_SECTION_SHIFT is large enough to accommodate
SECTION_MAP_LAST_BIT. We use BUILD_BUG_ON() to ensure this.
Update the comment to accurately reflect this consensus, making it clear
that we rely on a single section covering full pages.
Link: https://lore.kernel.org/20260402102320.3617578-1-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Petr Tesarik <ptesarik@suse.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
sio_read_complete() uses sio->pages to account global PSWPIN vm events,
but sio->pages tracks the number of bvec entries (folios), not base pages.
While large folios cannot currently reach this path (SWP_FS_OPS and
SWP_SYNCHRONOUS_IO are mutually exclusive, and mTHP swap-in allocation is
gated on SWP_SYNCHRONOUS_IO), the accounting is semantically inconsistent
with the per-memcg path which correctly uses folio_nr_pages().
Use sio->len >> PAGE_SHIFT instead, which gives the correct base page
count since sio->len is accumulated via folio_size(folio).
Link: https://lore.kernel.org/20260402061408.36119-1-devnexen@gmail.com
Signed-off-by: David Carlier <devnexen@gmail.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Cc: Baoquan He <bhe@redhat.com>
Cc: Chris Li <chrisl@kernel.org>
Cc: Kairui Song <kasong@tencent.com>
Cc: Kemeng Shi <shikemeng@huaweicloud.com>
Cc: NeilBrown <neil@brown.name>
Cc: Nhat Pham <nphamcs@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The test requires thp, skip the test when thp is not available to avoid
false positive.
Tested with thp disabled kernel.
Before the fix:
# --------------------------------
# running ./transhuge-stress -d 20
# --------------------------------
# TAP version 13
# 1..1
# transhuge-stress: allocate 1453 transhuge pages, using 2907 MiB virtual memory and 11 MiB of ram
# Bail out! MADV_HUGEPAGE# Planned tests != run tests (1 != 0)
# # Totals: pass:0 fail:0 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 60 transhuge-stress -d 20 # exit=1
After the fix:
# --------------------------------
# running ./transhuge-stress -d 20
# --------------------------------
# TAP version 13
# 1..0 # SKIP Transparent Hugepages not available
# [SKIP]
ok 5 transhuge-stress -d 20 # SKIP
Link: https://lore.kernel.org/20260402014543.1671131-7-chuhu@redhat.com
Signed-off-by: Chunyu Hu <chuhu@redhat.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Li Wang <liwang@redhat.com>
Cc: Nico Pache <npache@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
When thp is not enabled on some kernel config such as realtime kernel, the
test will report failure. Fix the false positive by skipping the test
directly when thp is not enabled.
Tested with thp disabled kernel:
Before The fix:
# --------------------------------------------------
# running ./split_huge_page_test /tmp/xfs_dir_Ywup9p
# --------------------------------------------------
# TAP version 13
# Bail out! Reading PMD pagesize failed
# # Totals: pass:0 fail:0 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 61 split_huge_page_test /tmp/xfs_dir_Ywup9p # exit=1
After the fix:
# --------------------------------------------------
# running ./split_huge_page_test /tmp/xfs_dir_YHPUPl
# --------------------------------------------------
# TAP version 13
# 1..0 # SKIP Transparent Hugepages not available
# [SKIP]
ok 6 split_huge_page_test /tmp/xfs_dir_YHPUPl # SKIP
Link: https://lore.kernel.org/20260402014543.1671131-6-chuhu@redhat.com
Signed-off-by: Chunyu Hu <chuhu@redhat.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Li Wang <liwang@redhat.com>
Cc: Nico Pache <npache@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Add three more checks for buflen and numwritten. The buflen should be at
least two, that means at least one char and the null-end. The error case
check is added by checking numwriten < 0 instead of numwritten < 1. And
the truncate case is checked. The test will exit if any of these
conditions aren't met.
Additionally, add more print information when a write failure occurs or a
truncated write happens, providing clearer diagnostics.
Link: https://lore.kernel.org/20260402014543.1671131-5-chuhu@redhat.com
Signed-off-by: Chunyu Hu <chuhu@redhat.com>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Lorenzo Stoakes <ljs@kernel.org>
Cc: Nico Pache <npache@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
thp_settings provides write_file() helper for safely writing to a file and
exit when write failure happens. It's a very low level helper and many
sub tests need such a helper, not only thp tests.
split_huge_page_test also defines a write_file locally. The two have
minior differences in return type and used exit api. And there would be
conflicts if split_huge_page_test wanted to include thp_settings.h because
of different prototype, making it less convenient.
It's possisble to merge the two, although some tests don't use the
kselftest infrastrucutre for testing. It would also work when using the
ksft_exit_msg() to exit in my test, as the counters are all zero. Output
will be like:
TAP version 13
1..62
Bail out! /proc/sys/vm/drop_caches1 open failed: No such file or directory
# Totals: pass:0 fail:0 xfail:0 xpass:0 skip:0 error:0
So here we just keep the version in split_huge_page_test, and move it into
the vm_util. This makes it easy to maitain and user could just include
one vm_util.h when they don't need thp setting helpers. Keep the
prototype of void return as the function will exit on any error, return
value is not necessary, and will simply the callers like write_num() and
write_string().
Link: https://lore.kernel.org/20260402014543.1671131-4-chuhu@redhat.com
Signed-off-by: Chunyu Hu <chuhu@redhat.com>
Reviewed-by: Lorenzo Stoakes (Oracle) <ljs@kernel.org>
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Acked-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Suggested-by: Mike Rapoport <rppt@kernel.org>
Cc: Nico Pache <npache@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
