Add x86_64 JIT support for BPF functions and kfuncs with more than
5 arguments. The extra arguments are passed through a stack area
addressed by register r11 (BPF_REG_PARAMS) in BPF bytecode,
which the JIT translates to native code.
The JIT follows the x86-64 calling convention for both BPF-to-BPF
and kfunc calls:
- Arg 6 is passed in the R9 register
- Args 7+ are passed on the stack
Incoming arg 6 (BPF r11+8) is translated to a MOV from R9 rather
than a memory load. Incoming args 7+ (BPF r11+16, r11+24, ...) map
directly to [rbp + 16], [rbp + 24], ..., matching the x86-64 stack
layout after CALL + PUSH RBP, so no offset adjustment is needed.
tail_call_reachable is rejected by the verifier and priv_stack is
disabled by the JIT when stack args exist, so R9 is always
available. When BPF bytecode writes to the arg-6 stack slot
(offset -8), the JIT emits a MOV into R9 instead of a memory store.
Outgoing args 7+ are placed at [rsp] in a pre-allocated area below
callee-saved registers, using:
native_off = outgoing_arg_base - outgoing_rsp - bpf_off - 16
The native x86_64 stack layout with stack arguments:
high address
+-------------------------+
| incoming stack arg N | [rbp + 16 + (N-7)*8] (from caller)
| ... |
| incoming stack arg 7 | [rbp + 16]
+-------------------------+
| return address | [rbp + 8]
| saved rbp | [rbp]
+-------------------------+
| BPF program stack | (round_up(stack_depth, 8) bytes)
+-------------------------+
| callee-saved regs | (r12, rbx, r13, r14, r15 as needed)
+-------------------------+
| outgoing arg M | [rsp + (M-7)*8]
| ... |
| outgoing arg 7 | [rsp]
+-------------------------+ rsp
low address
Acked-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045122.2393118-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Extend the stack argument mechanism to kfunc calls, allowing kfuncs
with more than 5 parameters to receive additional arguments via the
r11-based stack arg area.
For kfuncs, the caller is a BPF program and the callee is a kernel
function. The BPF program writes outgoing args at negative r11
offsets, following the same convention as BPF-to-BPF calls:
Outgoing: r11 - 8 (arg6), ..., r11 - N*8 (last arg)
The following is an example:
int foo(int a1, int a2, int a3, int a4, int a5, int a6, int a7) {
...
kfunc1(a1, a2, a3, a4, a5, a6, a7, a8);
...
kfunc2(a1, a2, a3, a4, a5, a6, a7, a8, a9);
...
}
Caller (foo), generated by llvm
===============================
Incoming (positive offsets):
r11+8: [incoming arg 6]
r11+16: [incoming arg 7]
Outgoing for kfunc1 (negative offsets):
r11-8: [outgoing arg 6]
r11-16: [outgoing arg 7]
r11-24: [outgoing arg 8]
Outgoing for kfunc2 (negative offsets):
r11-8: [outgoing arg 6]
r11-16: [outgoing arg 7]
r11-24: [outgoing arg 8]
r11-32: [outgoing arg 9]
Later JIT will marshal outgoing arguments to the native calling
convention for kfunc1() and kfunc2().
For kfunc calls where stack args are used as constant or size
parameters, a mark_stack_arg_precision() helper is used to propagate
precision and do proper backtracking.
There are two places where meta->release_regno needs to keep
regno for later releasing the reference. Also, 'cur_aux(env)->arg_prog = regno'
is also keeping regno for later fixup. Since stack arguments don't have a valid
register number (regno is negative), these three cases are rejected for now
if the argument is on the stack.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045104.2391543-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_REG_PARAMS (r11) is used for stack argument accesses and
the following are only insns with r11 presence:
- load incoming stack arg
- store register to outgoing stack arg
- store immediate to outgoing stack arg
The detailed insn format can be found in is_stack_arg_ldx/st/stx()
helpers. After this patch, stack arg ldx/st/stx insns become valid
for kernel and these insns can be properly checked by verifier.
The LLVM compiler [1] implemented the above BPF_REG_PARAMS insns.
[1] https://github.com/llvm/llvm-project/pull/189060
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045059.2391192-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add bpf_jit_supports_stack_args() as a weak function defaulting to
false. Architectures that implement JIT support for stack arguments
override it to return true.
Reject BPF functions with more than 5 parameters at verification
time if the architecture does not support stack arguments.
Acked-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045054.2390945-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_REG_PARAMS (R11) is at index MAX_BPF_REG, which is beyond the
register tracking arrays in const_fold.c and liveness.c. Handle it
explicitly to avoid out-of-bounds accesses.
Extend the arg tracking dataflow to cover stack arg slots. Otherwise,
pointers passed through stack args are invisible to liveness, causing
the pointed-to stack slots to be incorrectly poisoned.
Extend the at_out tracking array to MAX_AT_TRACK_REGS (registers
plus stack arg slots) so that outgoing stack arg stores are tracked
alongside registers. Add a separate at_stack_arg_entry array in
compute_subprog_args(), passed to arg_track_xfer(), to restore
FP-derived values on incoming stack arg reads.
Extend record_call_access() to check stack arg slots for FP-derived
pointers at kfunc call sites, reusing the record_arg_access() helper
extracted in the previous patch. Pass stack arg state from caller to
callee in analyze_subprog() so that callees can track pointers received
through stack args, hence avoid poisoning.
Skip stack arg instructions in record_load_store_access(). Stack arg
STX uses dst_reg=BPF_REG_PARAMS (index 11), but at[11] is repurposed
to track the value stored in stack arg slot 0. Without the skip, if a
prior stack arg STX stored an FP-derived pointer (e.g., fp-64) into
slot 0, a subsequent stack arg STX would read that FP-derived value as
the base pointer and spuriously mark a regular stack slot (e.g., fp-72
from -64 + -8) as accessed in the liveness bitmap.
Extend arg_track_log() to log state transitions for outgoing stack arg
slots at indices MAX_BPF_REG through MAX_AT_TRACK_REGS-1. Without this,
changes to at_out[11..17] caused by stack arg store instructions are
silently omitted from BPF_LOG_LEVEL2 output. For example, when a
caller passes fp-64 through a stack argument:
subprog#0:
10: (bf) r6 = r10
11: (07) r6 += -64
12: (7b) *(u64 *)(r11 -8) = r6
sa0: none -> fp0-64
13: (85) call pc+5
Without the fix, the "sa0: none -> fp0-64" transition at insn 12
would not appear.
Extend print_subprog_arg_access() to include stack arg slots in the
per-instruction FP-derived state dump. For example:
subprog#0:
12: (7b) *(u64 *)(r11 - 8) = r6 // r6=fp0-64
13: (85) call pc+5 // r6=fp0-64 sa0=fp0-64
Without the fix, the "sa0=fp0-64" annotation at insn 13 would not
appear, making it harder to debug liveness analysis for programs
that pass FP-derived pointers through stack arguments.
Extend has_fp_args() to also check stack arg slots for FP-derived
pointers, so that callees receiving pointers only through stack args
are still recursively analyzed.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045043.2389049-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Replace "frame != ARG_NONE" with arg_is_fp() in has_fp_args().
The function's purpose is to check whether any argument is derived
from a frame pointer, which is exactly what arg_is_fp() tests
(frame >= 0 || frame == ARG_IMPRECISE). Using the dedicated
predicate is clearer and more consistent with the rest of the file.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045035.2388671-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Extract the per-argument FP-derived pointer handling from
record_call_access() into a new record_arg_access() helper.
The existing loop body — checking arg_is_fp, querying stack access
bytes, and calling record_stack_access/record_imprecise — will be
reused for stack argument slots in the next patch. Factoring it out
now avoids duplicating the logic.
No functional change.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045030.2388067-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Extend the precision marking and backtracking infrastructure to
support stack argument slots (r11-based accesses). Without this,
precision demands for scalar values passed through stack arguments
are silently dropped, which could allow the verifier to incorrectly
prune states with different constant values in stack arg slots.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045025.2387526-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Move stack slot index (spi) and frame number out of the flags field
in bpf_jmp_history_entry into dedicated bitfields. This simplifies
the encoding and makes room for new flags.
Previously, spi and frame were packed into the lower 9 bits of the
12-bit flags field (3 bits frame + 6 bits spi), with INSN_F_STACK_ACCESS
at BIT(9) and INSN_F_DST/SRC_REG_STACK at BIT(10)/BIT(11).
But this has no room for an INSN_F_* flag for stack arguments.
To resolve this issue, bpf_jmp_history_entry field idx is narrowed to
20 bits (sufficient for insn indices up to 1M), and the freed bits hold
spi (6 bits) and frame (3 bits) as dedicated struct fields. The flags
enum is simplified accordingly:
INSN_F_STACK_ACCESS -> BIT(0)
INSN_F_DST_REG_STACK -> BIT(1)
INSN_F_SRC_REG_STACK -> BIT(2)
which allows more room for additional INSN_F_* flags.
bpf_push_jmp_history() now takes explicit spi and frame parameters
instead of encoding them into flags. The insn_stack_access_flags(),
insn_stack_access_spi(), and insn_stack_access_frameno() helpers are
removed.
No functional change.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045020.2385962-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently BPF functions (subprogs) are limited to 5 register arguments.
With [1], the compiler can emit code that passes additional arguments
via a dedicated stack area through bpf register BPF_REG_PARAMS (r11),
introduced in an earlier patch ([2]).
The compiler uses positive r11 offsets for incoming (callee-side) args
and negative r11 offsets for outgoing (caller-side) args, following the
x86_64/arm64 calling convention direction. There is an 8-byte gap at
offset 0 separating two regions:
Incoming (callee reads): r11+8 (arg6), r11+16 (arg7), ...
Outgoing (caller writes): r11-8 (arg6), r11-16 (arg7), ...
The following is an example to show how stack arguments are saved
and transferred between caller and callee:
int foo(int a1, int a2, int a3, int a4, int a5, int a6, int a7) {
...
bar(a1, a2, a3, a4, a5, a6, a7, a8);
...
}
Caller (foo) Callee (bar)
============ ============
Incoming (positive offsets): Incoming (positive offsets):
r11+8: [incoming arg 6] r11+8: [incoming arg 6] <-+
r11+16: [incoming arg 7] r11+16: [incoming arg 7] <-|+
r11+24: [incoming arg 8] <-||+
Outgoing (negative offsets): |||
r11-8: [outgoing arg 6 to bar] -------->-------------------------+||
r11-16: [outgoing arg 7 to bar] -------->--------------------------+|
r11-24: [outgoing arg 8 to bar] -------->---------------------------+
If the bpf function has more than one call:
int foo(int a1, int a2, int a3, int a4, int a5, int a6, int a7) {
...
bar1(a1, a2, a3, a4, a5, a6, a7, a8);
...
bar2(a1, a2, a3, a4, a5, a6, a7, a8, a9);
...
}
Caller (foo) Callee (bar2)
============ ==============
Incoming (positive offsets): Incoming (positive offsets):
r11+8: [incoming arg 6] r11+8: [incoming arg 6] <+
r11+16: [incoming arg 7] r11+16: [incoming arg 7] <|+
r11+24: [incoming arg 8] <||+
Outgoing for bar2 (negative offsets): r11+32: [incoming arg 9] <|||+
r11-8: [outgoing arg 6] ---->----------->-------------------------+|||
r11-16: [outgoing arg 7] ---->----------->--------------------------+||
r11-24: [outgoing arg 8] ---->----------->---------------------------+|
r11-32: [outgoing arg 9] ---->----------->----------------------------+
The verifier tracks outgoing stack arguments in stack_arg_regs[] and
out_stack_arg_cnt in bpf_func_state, separately from the regular
r10 stack. The callee does not copy incoming args — it reads them
directly from the caller's outgoing slots at positive r11 offsets.
Similar to stacksafe(), introduce stack_arg_safe() to do pruning
check.
Outgoing stack arg slots are invalidated when the callee returns
(e.g. in prepare_func_exit), not at call time. This allows the callee to
read incoming args from the caller's outgoing slots during
verification. The following are a few examples.
Example 1:
*(u64 *)(r11 - 8) = r6;
*(u64 *)(r11 - 16) = r7;
call bar1; // arg6 = r6, arg7 = r7
call bar2; // expected with 2 stack arguments, failed
Example 2:
To fix the Example 1:
*(u64 *)(r11 - 8) = r6;
*(u64 *)(r11 - 16) = r7;
call bar1; // arg6 = r6, arg7 = r7
*(u64 *)(r11 - 8) = r8;
*(u64 *)(r11 - 16) = r9;
call bar2; // arg6 = r8, arg7 = r9
Example 3:
The compiler can hoist the shared stack arg stores above the branch:
*(u64 *)(r11 - 16) = r7;
if cond goto else;
*(u64 *)(r11 - 8) = r8;
call bar1; // arg6 = r8, arg7 = r7
goto end;
else:
*(u64 *)(r11 - 8) = r9;
call bar2; // arg6 = r9, arg7 = r7
end:
Example 4:
Within a loop:
loop:
*(u64 *)(r11 - 8) = r6; // arg6, before loop
call bar; // reuses arg6 each iteration
if ... goto loop;
A separate max_out_stack_arg_cnt field in bpf_subprog_info tracks
the deepest outgoing slot actually written. This intends to
reject programs that write to slots beyond what any callee expects.
It is necessary for JIT.
Similar to typical compiler generated code, enforce the following
orderings:
- all stack arg reads must be ahead of any stack arg write
- all stack arg reads must be before any bpf func, kfunc and helpers
This is needed as JIT may emit 'mov' insns for read/write with
the same register and bpf function, kfunc and helper will invalidate
all arguments immediately after the call.
Callback functions with stack arguments need kernel setup parameter
types (including stack parameters) properly and then callback function
can retrieve such information for verification purpose.
Global subprogs and freplace with >5 args are not yet supported.
[1] https://github.com/llvm/llvm-project/pull/189060
[2] https://lore.kernel.org/bpf/20260423033506.2542005-1-yonghong.song@linux.dev/
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045015.2385013-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Move the "sub->arg_cnt = nargs" assignment to immediately after
nargs is computed from btf_type_vlen(), instead of at the end of
btf_prepare_func_args().
btf_prepare_func_args() can return -EINVAL early in several cases,
e.g. when a static function has some non-int/enum arguments.
Since -EINVAL from btf_prepare_func_args() does not immediately
reject verification, arg_cnt remains zero after the early return.
This causes later stack argument based load/store insns to
incorrectly assume the function has no arguments.
Setting arg_cnt right after nargs ensures it is available regardless
of which path btf_prepare_func_args() takes.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260513045010.2384635-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Many BPF_MAP_CREATE validation failures currently return -EINVAL without
any explanation to userspace.
Plumb common syscall log attributes into map_create(), create a verifier
log from bpf_common_attr::log_buf/log_size/log_level, and report
map-creation failure reasons through that buffer.
This improves debuggability by allowing userspace to inspect why map
creation failed and read back log_true_size from common attributes.
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20260512153157.28382-7-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_BTF_LOAD can now take log parameters from both union bpf_attr and
struct bpf_common_attr, with the same merge rules as BPF_PROG_LOAD:
- if both sides provide a complete log tuple (buf/size/level) and they
match, use it;
- if only one side provides log parameters, use that one;
- if both sides provide complete tuples but they differ, return -EINVAL.
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20260512153157.28382-6-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_PROG_LOAD can now take log parameters from both union bpf_attr and
struct bpf_common_attr. The merge rules are:
- if both sides provide a complete log tuple (buf/size/level) and they
match, use it;
- if only one side provides log parameters, use that one;
- if both sides provide complete tuples but they differ, return -EINVAL.
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20260512153157.28382-5-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The next commit will add support for reporting logs via extended common
attributes, including 'log_true_size'.
To prepare for that, refactor the 'log_true_size' reporting logic by
introducing a new struct bpf_log_attr to encapsulate log-related behavior:
* bpf_log_attr_init(): initialize log fields, which will support
extended common attributes in the next commit.
* bpf_log_attr_finalize(): handle log finalization and write back
'log_true_size' to userspace.
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20260512153157.28382-4-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add generic BPF syscall support for passing common attributes.
The initial set of common attributes includes:
1. 'log_buf': User-provided buffer for storing logs.
2. 'log_size': Size of the log buffer.
3. 'log_level': Log verbosity level.
4. 'log_true_size': Actual log size reported by kernel.
The common-attribute pointer and its size are passed as the 4th and 5th
syscall arguments. A new command bit, 'BPF_COMMON_ATTRS' ('1 << 16'),
indicates that common attributes are supplied.
This commit adds syscall and uapi plumbing. Command-specific handling is
added in follow-up patches.
Signed-off-by: Leon Hwang <leon.hwang@linux.dev>
Link: https://lore.kernel.org/r/20260512153157.28382-2-leon.hwang@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_MAP_TYPE_ARENA accepts BPF_PSEUDO_MAP_VALUE offsets at exactly
the end of the arena mapping (off == arena_size). The boundary check
in arena_map_direct_value_addr() uses `>` instead of `>=`, which
incorrectly allows a one-past-end pointer to be accepted.
Change the condition to `>=` to correctly reject offsets that fall
outside the valid arena user_vm range.
Fixes: 317460317a ("bpf: Introduce bpf_arena.")
Signed-off-by: Junyoung Jang <graypanda.inzag@gmail.com>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Link: https://lore.kernel.org/r/20260426172505.1947915-1-graypanda.inzag@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Because subprog 0, the main subprog, is considered a global function,
we end up running the arg-tracking dataflow analysis twice on it. That
results in slightly longer verification but mostly in more verbose
verifier logs. This patch fixes it by keeping only the iteration over
global subprogs.
When running over all of Cilium's programs with BPF_LOG_LEVEL2, this
reduces verbosity by ~20% on average.
Fixes: bf0c571f7f ("bpf: introduce forward arg-tracking dataflow analysis")
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/e4d7b53d4963ef520541a782f5fc8108a168877c.1778176504.git.paul.chaignon@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull timer fix from Ingo Molnar:
"Fix CPU hotplug activation race in the timer migration code, by
Frederic Weisbecker"
* tag 'timers-urgent-2026-05-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
timers/migration: Fix another hotplug activation race
Pull scheduler fixes from Ingo Molnar:
- Fix spurious failures in rseq self-tests (Mark Brown)
- Fix rseq rseq::cpu_id_start ABI regression due to TCMalloc's creative
use of the supposedly read-only field
The fix is to introduce a new ABI variant based on a new (larger)
rseq area registration size, to keep the TCMalloc use of rseq
backwards compatible on new kernels (Thomas Gleixner)
- Fix wakeup_preempt_fair() for not waking up task (Vincent Guittot)
- Fix s64 mult overflow in vruntime_eligible() (Zhan Xusheng)
* tag 'sched-urgent-2026-05-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Fix wakeup_preempt_fair() for not waking up task
sched/fair: Fix overflow in vruntime_eligible()
selftests/rseq: Expand for optimized RSEQ ABI v2
rseq: Reenable performance optimizations conditionally
rseq: Implement read only ABI enforcement for optimized RSEQ V2 mode
selftests/rseq: Validate legacy behavior
selftests/rseq: Make registration flexible for legacy and optimized mode
selftests/rseq: Skip tests if time slice extensions are not available
rseq: Revert to historical performance killing behaviour
rseq: Don't advertise time slice extensions if disabled
rseq: Protect rseq_reset() against interrupts
rseq: Set rseq::cpu_id_start to 0 on unregistration
selftests/rseq: Don't run tests with runner scripts outside of the scripts
Pull perf events fixes from Ingo Molnar:
- Fix deadlock in the perf_mmap() failure path (Peter Zijlstra)
- Intel ACR (Auto Counter Reload) fixes (Dapeng Mi):
- Fix validation and configuration of ACR masks
- Fix ACR rescheduling bug causing stale masks
- Disable the PMI on ACR-enabled hardware
- Enable ACR on Panther Cover uarch too
* tag 'perf-urgent-2026-05-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel: Enable auto counter reload for DMR
perf/x86/intel: Disable PMI for self-reloaded ACR events
perf/x86/intel: Always reprogram ACR events to prevent stale masks
perf/x86/intel: Improve validation and configuration of ACR masks
perf/core: Fix deadlock in perf_mmap() failure path
Make sure to only call pick_next_entity() on an non-empty cfs_rq.
The assumption that p is always enqueued and not delayed, is only true for
wakeup. If p was moved while delayed, pick_next_entity() will dequeue it and
the cfs might become empty. Test if there are still queued tasks before trying
again to determine if p could be the next one to be picked.
There are at least 2 cases:
When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
delayed, they will be dequeued when attached to cfs. attach_tasks() ->
attach_task() -> wakeup_preempt(rq, p, 0);
A misfit task running on cfs A triggers a load balance to be pulled on a better
cpu, the load balance on cfs B starts an active load balance to pulled the
running misfit task. If there is a delayed dequeue task on cfs A, it can be
pulled instead of the previously running misfit task. attach_one_task() ->
attach_task() -> wakeup_preempt(rq, p, 0);
Fixes: ac8e69e693 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260503104503.1732682-1-vincent.guittot@linaro.org
Zhan Xusheng reported running into sporadic a s64 mult overflow in
vruntime_eligible().
When constructing a worst case scenario:
If you have cgroups, then you can have an entity of weight 2 (per
calc_group_shares()), and its vlag should then be bounded by: (slice+TICK_NSEC)
* NICE_0_LOAD, which is around 44 bits as per the comment on entity_key().
The other extreme is 100*NICE_0_LOAD, thus you get:
{key, weight}[] := {
puny: { (slice + TICK_NSEC) * NICE_0_LOAD, 2 },
max: { 0, 100*NICE_0_LOAD },
}
The avg_vruntime() would end up being very close to 0 (which is
zero_vruntime), so no real help making that more accurate.
vruntime_eligible(puny) ends up with:
avg = 2 * puny.key (+ 0)
load = 2 + 100 * NICE_0_LOAD
avg >= puny.key * load
And that is: (slice + TICK_NSEC) * NICE_0_LOAD * NICE_0_LOAD * 100, which will
overflow s64.
Zhan suggested using __builtin_mul_overflow(), however after staring at
compiler output for various architectures using godbolt, it seems that using an
__int128 multiplication often results in better code.
Specifically, a number of architectures already compute the __int128 product to
determine the overflow. Eg. arm64 already has the 'smulh' instruction used. By
explicitly doing an __int128 multiply, it will emit the 'mul; smulh' pattern,
which modern cores can fuse (armv8-a clang-22.1.0). x86_64 has less branches
(no OF handling).
Since Linux has ARCH_SUPPORTS_INT128 to gate __int128 usage, also provide the
__builtin_mul_overflow() variant as a fallback.
[peterz: Changelog and __int128 bits]
Fixes: 556146ce5e ("sched/fair: Avoid overflow in enqueue_entity()")
Reported-by: Zhan Xusheng <zhanxusheng1024@gmail.com>
Closes: https://patch.msgid.link/20260415145742.10359-1-zhanxusheng%40xiaomi.com
Signed-off-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260505103155.GN3102924%40noisy.programming.kicks-ass.net
Due to the incompatibility with TCMalloc the RSEQ optimizations and
extended features (time slice extensions) have been disabled and made
run-time conditional.
The original RSEQ implementation, which TCMalloc depends on, registers a 32
byte region (ORIG_RSEG_SIZE). This region has a 32 byte alignment
requirement.
The extension safe newer variant exposes the kernel RSEQ feature size via
getauxval(AT_RSEQ_FEATURE_SIZE) and the alignment requirement via
getauxval(AT_RSEQ_ALIGN). The alignment requirement is that the registered
RSEQ region is aligned to the next power of two of the feature size. The
kernel currently has a feature size of 33 bytes, which means the alignment
requirement is 64 bytes.
The TCMalloc RSEQ region is embedded into a cache line aligned data
structure starting at offset 32 bytes so that bytes 28-31 and the
cpu_id_start field at bytes 32-35 form a 64-bit little endian pointer with
the top-most bit (63 set) to check whether the kernel has overwritten
cpu_id_start with an actual CPU id value, which is guaranteed to not have
the top most bit set.
As this is part of their performance tuned magic, it's a pretty safe
assumption, that TCMalloc won't use a larger RSEQ size.
This allows the kernel to declare that registrations with a size greater
than the original size of 32 bytes, which is the cases since time slice
extensions got introduced, as RSEQ ABI v2 with the following differences to
the original behaviour:
1) Unconditional updates of the user read only fields (CPU, node, MMCID)
are removed. Those fields are only updated on registration, task
migration and MMCID changes.
2) Unconditional evaluation of the criticial section pointer is
removed. It's only evaluated when user space was interrupted and was
scheduled out or before delivering a signal in the interrupted
context.
3) The read/only requirement of the ID fields is enforced. When the
kernel detects that userspace manipulated the fields, the process is
terminated. This ensures that multiple entities (libraries) can
utilize RSEQ without interfering.
4) Todays extended RSEQ feature (time slice extensions) and future
extensions are only enabled in the v2 enabled mode.
Registrations with the original size of 32 bytes operate in backwards
compatible legacy mode without performance improvements and extended
features.
Unfortunately that also affects users of older GLIBC versions which
register the original size of 32 bytes and do not evaluate the kernel
required size in the auxiliary vector AT_RSEQ_FEATURE_SIZE.
That's the result of the lack of enforcement in the original implementation
and the unwillingness of a single entity to cooperate with the larger
ecosystem for many years.
Implement the required registration changes by restructuring the spaghetti
code and adding the size/version check. Also add documentation about the
differences of legacy and optimized RSEQ V2 mode.
Thanks to Mathieu for pointing out the ORIG_RSEQ_SIZE constraints!
Fixes: d6200245c7 ("rseq: Allow registering RSEQ with slice extension")
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Link: https://patch.msgid.link/20260428224427.927160119%40kernel.org
Cc: stable@vger.kernel.org
The optimized RSEQ V2 mode requires that user space adheres to the ABI
specification and does not modify the read-only fields cpu_id_start,
cpu_id, node_id and mm_cid behind the kernel's back.
While the kernel does not rely on these fields, the adherence to this is a
fundamental prerequisite to allow multiple entities, e.g. libraries, in an
application to utilize the full potential of RSEQ without stepping on each
other toes.
Validate this adherence on every update of these fields. If the kernel
detects that user space modified the fields, the application is force
terminated.
Fixes: d6200245c7 ("rseq: Allow registering RSEQ with slice extension")
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Link: https://patch.msgid.link/20260428224427.845230956%40kernel.org
Cc: stable@vger.kernel.org
The hotplug control CPU is assumed to be active in the hierarchy but
that doesn't imply that the root is active. If the current CPU is not
the one that activated the current hierarchy, and the CPU performing
this duty is still halfway through the tree, the root may still be
observed inactive. And this can break the activation of a new root as in
the following scenario:
1) Initially, the whole system has 64 CPUs and only CPU 63 is awake.
[GRP1:0]
active
/ | \
/ | \
[GRP0:0] [...] [GRP0:7]
idle idle active
/ | \ |
CPU 0 CPU 1 ... CPU 63
idle idle active
2) CPU 63 goes idle _but_ due to a #VMEXIT it hasn't yet reached the
[GRP1:0]->parent dereference (that would be NULL and stop the walk)
in __walk_groups_from().
[GRP1:0]
idle
/ | \
/ | \
[GRP0:0] [...] [GRP0:7]
idle idle idle
/ | \ |
CPU 0 CPU 1 ... CPU 63
idle idle idle
3) CPU 1 wakes up, activates GRP0:0 but didn't yet manage to propagate
up to GRP1:0 due to yet another #VMEXIT.
[GRP1:0]
idle
/ | \
/ | \
[GRP0:0] [...] [GRP0:7]
active idle idle
/ | \ |
CPU 0 CPU 1 ... CPU 63
idle active idle
3) CPU 0 wakes up and doesn't need to walk above GRP0:0 as it's CPU 1
role.
[GRP1:0]
idle
/ | \
/ | \
[GRP0:0] [...] [GRP0:7]
active idle idle
/ | \ |
CPU 0 CPU 1 ... CPU 63
active active idle
4) CPU 0 boots CPU 64. It creates a new root for it.
[GRP2:0]
idle
/ \
/ \
[GRP1:0] [GRP1:1]
idle idle
/ | \ \
/ | \ \
[GRP0:0] [...] [GRP0:7] [GRP0:8]
active idle idle idle
/ | \ | |
CPU 0 CPU 1 ... CPU 63 CPU 64
active active idle offline
5) CPU 0 activates the new root, but note that GRP1:0 is still idle,
waiting for CPU 1 to resume from #VMEXIT and activate it.
[GRP2:0]
active
/ \
/ \
[GRP1:0] [GRP1:1]
idle idle
/ | \ \
/ | \ \
[GRP0:0] [...] [GRP0:7] [GRP0:8]
active idle idle idle
/ | \ | |
CPU 0 CPU 1 ... CPU 63 CPU 64
active active idle offline
6) CPU 63 resumes after #VMEXIT and sees the new GRP1:0 parent.
Therefore it propagates the stale inactive state of GRP1:0 up to
GRP2:0.
[GRP2:0]
idle
/ \
/ \
[GRP1:0] [GRP1:1]
idle idle
/ | \ \
/ | \ \
[GRP0:0] [...] [GRP0:7] [GRP0:8]
active idle idle idle
/ | \ | |
CPU 0 CPU 1 ... CPU 63 CPU 64
active active idle offline
7) CPU 1 resumes after #VMEXIT and finally activates GRP1:0. But it
doesn't observe its parent link because no ordering enforced that.
Therefore GRP2:0 is spuriously left idle.
[GRP2:0]
idle
/ \
/ \
[GRP1:0] [GRP1:1]
active idle
/ | \ \
/ | \ \
[GRP0:0] [...] [GRP0:7] [GRP0:8]
active idle idle idle
/ | \ | |
CPU 0 CPU 1 ... CPU 63 CPU 64
active active idle offline
Such races are highly theoretical and the problem would solve itself
once the old root ever becomes idle again. But it still leaves a taste
of discomfort.
Fix it with enforcing a fully ordered atomic read of the old root state
before propagating the activate state up to the new root. It has a two
directions ordering effect:
* Acquire + release of the latest old root state: If the hotplug control
CPU is not the one that woke up the old root, make sure to acquire its
active state and propagate it upwards through the ordered chain of
activation (the acquire pairs with the cmpxchg() in tmigr_active_up()
and subsequent releases will pair with atomic_read_acquire() and
smp_mb__after_atomic() in tmigr_inactive_up()).
* Release: If the hotplug control CPU is not the one that must wake up
the old root, but the CPU covering that is lagging behind its duty,
publish the links from the old root to the new parents. This way the
lagging CPU will propagate the active state itself.
Fixes: 7ee9887703 ("timers: Implement the hierarchical pull model")
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260423165354.95152-2-frederic@kernel.org
Pull workqueue fixes from Tejun Heo:
- Fix devm_alloc_workqueue() passing a va_list as a positional arg to
the variadic alloc_workqueue() macro, which garbled wq->name and
skipped lockdep init on the devm path. Fold both noprof entry points
onto a va_list helper.
Also, annotate it using __printf(1, 0)
* tag 'wq-for-7.1-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: Annotate alloc_workqueue_va() with __printf(1, 0)
workqueue: fix devm_alloc_workqueue() va_list misuse
Pull cgroup fixes from Tejun Heo:
- During v6.19, cgroup task unlink was moved from do_exit() to after the
final task switch to satisfy a controller invariant. That left the kernel
seeing tasks past exit_signals() longer than userspace expected, and
several v7.0 follow-ups tried to bridge the gap by making rmdir wait for
the kernel side. None held up.
The latest is an A-A deadlock when rmdir is invoked by the reaper of
zombies whose pidns teardown the rmdir itself is waiting on, which
points at the synchronizing approach being fundamentally wrong.
Take a different approach: drop the wait, leave rmdir's user-visible
side returning as soon as cgroup.procs is empty, and defer the css
percpu_ref kill that drives ->css_offline() until the cgroup is fully
depopulated.
Tagged for stable. Somewhat invasive but contained. The hope is that
fixing forward sticks. If not, the fallback is to revert the entire
chain and rework on the development branch.
Note that this doesn't plug a pre-existing analogous race in
cgroup_apply_control_disable() (controller disable via
subtree_control). Not a regression. The development branch will do
the more invasive restructuring needed for that.
- Documentation update for cgroup-v1 charge-commit section that still
referenced functions removed when the memcg hugetlb try-commit-cancel
protocol was retired.
* tag 'cgroup-for-7.1-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
docs: cgroup-v1: Update charge-commit section
cgroup: Defer css percpu_ref kill on rmdir until cgroup is depopulated
Pull sched_ext fixes from Tejun Heo:
- Fix idle CPU selection returning prev_cpu outside the task's cpus_ptr
when the BPF caller's allowed mask was wider. Stable backport.
- Two opposite-direction gaps in scx_task_iter's cgroup-scoped mode
versus the global mode:
- Tasks past exit_signals() are filtered by the cgroup walk but kept
by global. Sub-scheduler enable abort leaked __scx_init_task()
state. Add a CSS_TASK_ITER_WITH_DEAD flag to cgroup's task
iterator (scx_task_iter is its only user) and use it.
- Tasks past sched_ext_dead() are still returned, tripping
WARN_ON_ONCE() in callers or making them touch torn-down state.
Mark and skip under the per-task rq lock.
* tag 'sched_ext-for-7.1-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: idle: Recheck prev_cpu after narrowing allowed mask
sched_ext: Skip past-sched_ext_dead() tasks in scx_task_iter_next_locked()
cgroup, sched_ext: Include exiting tasks in cgroup iter
The recent RSEQ optimization work broke the TCMalloc abuse of the RSEQ ABI
as it not longer unconditionally updates the CPU, node, mm_cid fields,
which are documented as read only for user space. Due to the observed
behavior of the kernel it was possible for TCMalloc to overwrite the
cpu_id_start field for their own purposes and rely on the kernel to update
it unconditionally after each context switch and before signal delivery.
The RSEQ ABI only guarantees that these fields are updated when the data
changes, i.e. the task is migrated or the MMCID of the task changes due to
switching from or to per CPU ownership mode.
The optimization work eliminated the unconditional updates and reduced them
to the documented ABI guarantees, which results in a massive performance
win for syscall, scheduling heavy work loads, which in turn breaks the
TCMalloc expectations.
There have been several options discussed to restore the TCMalloc
functionality while preserving the optimization benefits. They all end up
in a series of hard to maintain workarounds, which in the worst case
introduce overhead for everyone, e.g. in the scheduler.
The requirements of TCMalloc and the optimization work are diametral and
the required work arounds are a maintainence burden. They end up as fragile
constructs, which are blocking further optimization work and are pretty
much guaranteed to cause more subtle issues down the road.
The optimization work heavily depends on the generic entry code, which is
not used by all architectures yet. So the rework preserved the original
mechanism moslty unmodified to keep the support for architectures, which
handle rseq in their own exit to user space loop. That code is currently
optimized out by the compiler on architectures which use the generic entry
code.
This allows to revert back to the original behaviour by replacing the
compile time constant conditions with a runtime condition where required,
which disables the optimization and the dependend time slice extension
feature until the run-time condition can be enabled in the RSEQ
registration code on a per task basis again.
The following changes are required to restore the original behavior, which
makes TCMalloc work again:
1) Replace the compile time constant conditionals with runtime
conditionals where appropriate to prevent the compiler from optimizing
the legacy mode out
2) Enforce unconditional update of IDs on context switch for the
non-optimized v1 mode
3) Enforce update of IDs in the pre signal delivery path for the
non-optimized v1 mode
4) Enforce update of IDs in the membarrier(RSEQ) IPI for the
non-optimized v1 mode
5) Make time slice and future extensions depend on optimized v2 mode
This brings back the full performance problems, but preserves the v2
optimization code and for generic entry code using architectures also the
TIF_RSEQ optimization which avoids a full evaluation of the exit to user
mode loop in many cases.
Fixes: 566d8015f7 ("rseq: Avoid CPU/MM CID updates when no event pending")
Reported-by: Mathias Stearn <mathias@mongodb.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Closes: https://lore.kernel.org/CAHnCjA25b+nO2n5CeifknSKHssJpPrjnf+dtr7UgzRw4Zgu=oA@mail.gmail.com
Link: https://patch.msgid.link/20260428224427.517051752%40kernel.org
Cc: stable@vger.kernel.org
Ian noted that commit 77de62ad3d ("perf/core: Fix refcount bug and
potential UAF in perf_mmap") would cause a deadlock due to
event->mmap_mutex recursion.
This happens because we're now calling perf_mmap_close() under
mmap_mutex, while that function itself can also take mmap_mutex.
Solve this by noting that perf_mmap_close() is far more complicated
than we need at this particular point, since it deals with scenarios
that cannot happen in this particular case.
Replace the call to perf_mmap_close() with a very narrow undo for the
case of first-exposure. If this is not the first mmap(), there is no
race and it is fine to drop the lock and call perf_mmap_close() to
handle to more complicated scenarios.
Note: move the rb->mmap_user (namespace) handling into the rb
init/free code such that it does not complicate the mmap handling.
Fixes: 77de62ad3d ("perf/core: Fix refcount bug and potential UAF in perf_mmap")
Reported-by: Ian Rogers <irogers@google.com>
Closes: https://patch.msgid.link/CAP-5%3DfVJyVMZw%3DDqP53Kxg58nUmJ_0bxoaeOKAbC03BVc11HaA%40mail.gmail.com
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260326112821.GK3738786@noisy.programming.kicks-ass.net
scx_select_cpu_dfl() narrows @allowed to @cpus_allowed & @p->cpus_ptr
when the BPF caller supplies a @cpus_allowed that differs from
@p->cpus_ptr and @p doesn't have full affinity. However,
@is_prev_allowed was computed against the original (wider)
@cpus_allowed, so the prev_cpu fast paths could pick a @prev_cpu that
is in @cpus_allowed but not in @p->cpus_ptr, violating the intended
invariant that the returned CPU is always usable by @p. The kernel
masks this via the SCX_EV_SELECT_CPU_FALLBACK fallback, but the
behavior contradicts the documented contract.
Move the @is_prev_allowed evaluation past the narrowing block so it
tests against the final @allowed mask.
Fixes: ee9a4e9279 ("sched_ext: idle: Properly handle invalid prev_cpu during idle selection")
Cc: stable@vger.kernel.org # v6.16+
Assisted-by: Claude <noreply@anthropic.com>
Signed-off-by: David Carlier <devnexen@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_task_iter's cgroup-scoped mode can return tasks whose
sched_ext_dead() has already completed: cgroup_task_dead() removes
from cset->tasks after sched_ext_dead() in finish_task_switch() and is
irq-work deferred on PREEMPT_RT. The global mode is fine -
sched_ext_dead() removes from scx_tasks via list_del_init() first.
Callers (sub-sched enable prep/abort/apply, scx_sub_disable(),
scx_fail_parent()) assume returned tasks are still on @sch and trip
WARN_ON_ONCE() or operate on torn-down state otherwise.
Set %SCX_TASK_OFF_TASKS in sched_ext_dead() under @p's rq lock and
have scx_task_iter_next_locked() skip flagged tasks under the same
lock. Setter and reader serialize on the per-task rq lock - no race.
Signed-off-by: Tejun Heo <tj@kernel.org>
a72f73c4dd ("cgroup: Don't expose dead tasks in cgroup") made
css_task_iter_advance() skip exiting tasks so cgroup.procs stays consistent
with waitpid() visibility. Unfortunately, this broke scx_task_iter.
scx_task_iter walks either scx_tasks (global) or a cgroup subtree via
css_task_iter() and the two modes are expected to cover the same set of
tasks. After the above change the cgroup-scoped mode silently skips tasks
past exit_signals() that are still on scx_tasks.
scx_sub_enable_workfn()'s abort path is one of the symptoms: an exiting
SCX_TASK_SUB_INIT task can race past the cgroup iter leaking
__scx_init_task() state. Other iterations share the same gap.
Add CSS_TASK_ITER_WITH_DEAD to opt out of the skip and use it from
scx_task_iter().
Fixes: b0e4c2f8a0 ("sched_ext: Implement cgroup subtree iteration for scx_task_iter")
Reported-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
A chain of commits going back to v7.0 reworked rmdir to satisfy the
controller invariant that a subsystem's ->css_offline() must not run while
tasks are still doing kernel-side work in the cgroup.
[1] d245698d72 ("cgroup: Defer task cgroup unlink until after the task is done switching out")
[2] a72f73c4dd ("cgroup: Don't expose dead tasks in cgroup")
[3] 1b164b876c ("cgroup: Wait for dying tasks to leave on rmdir")
[4] 4c56a8ac68 ("cgroup: Fix cgroup_drain_dying() testing the wrong condition")
[5] 13e786b64b ("cgroup: Increment nr_dying_subsys_* from rmdir context")
[1] moved task cset unlink from do_exit() to finish_task_switch() so a
task's cset link drops only after the task has fully stopped scheduling.
That made tasks past exit_signals() linger on cset->tasks until their final
context switch, which led to a series of problems as what userspace expected
to see after rmdir diverged from what the kernel needs to wait for. [2]-[5]
tried to bridge that divergence: [2] filtered the exiting tasks from
cgroup.procs; [3] had rmdir(2) sleep in TASK_UNINTERRUPTIBLE for them; [4]
fixed the wait's condition; [5] made nr_dying_subsys_* visible
synchronously.
The cgroup_drain_dying() wait in [3] turned out to be a dead end. When the
rmdir caller is also the reaper of a zombie that pins a pidns teardown (e.g.
host PID 1 systemd reaping orphan pids that were re-parented to it during
the same teardown), rmdir blocks in TASK_UNINTERRUPTIBLE waiting for those
pids to free, the pids can't free because PID 1 is the reaper and it's stuck
in rmdir, and the system A-A deadlocks. No internal lock ordering breaks
this; the wait itself is the bug.
The css killing side that drove the original reorder, however, can be made
cleanly asynchronous: ->css_offline() is already async, run from
css_killed_work_fn() driven by percpu_ref_kill_and_confirm(). The fix is to
make that chain start only after all tasks have left the cgroup. rmdir's
user-visible side then returns as soon as cgroup.procs and friends are
empty, while ->css_offline() still runs only after the cgroup is fully
drained.
Verified by the original reproducer (pidns teardown + zombie reaper, runs
under vng) which hangs vanilla and succeeds here, and by per-commit
deterministic repros for [2], [3], [4], [5] with a boot parameter that
widens the post-exit_signals() window so each state is reliably reachable.
Some stress tests on top of that.
cgroup_apply_control_disable() has the same shape of pre-existing race:
when a controller is disabled via subtree_control, kill_css() ran
synchronously while tasks past exit_signals() could still be linked to
the cgroup's csets, and ->css_offline() could fire before they drained.
This patch preserves the existing synchronous behavior at that call site
(kill_css_sync() + kill_css_finish() back-to-back) and a follow-up patch
will defer kill_css_finish() there using a per-css trigger.
This seems like the right approach and I don't see problems with it. The
changes are somewhat invasive but not excessively so, so backporting to
-stable should be okay. If something does turn out to be wrong, the fallback
is to revert the entire chain ([1]-[5]) and rework in the development branch
instead.
v2: Pin cgrp across the deferred destroy work with explicit
cgroup_get()/cgroup_put() around queue_work() and the work_fn. v1
wasn't actually broken (ordered cgroup_offline_wq + queue_work order
in cgroup_task_dead() saved it) but the explicit ref removes the
dependency on those non-obvious invariants. Also note the
pre-existing cgroup_apply_control_disable() race in the description;
a follow-up will defer kill_css_finish() there.
Fixes: 1b164b876c ("cgroup: Wait for dying tasks to leave on rmdir")
Cc: stable@vger.kernel.org # v7.0+
Reported-and-tested-by: Martin Pitt <martin@piware.de>
Link: https://lore.kernel.org/all/afHNg2VX2jy9bW7y@piware.de/
Link: https://lore.kernel.org/all/35e0670adb4abeab13da2c321582af9f@kernel.org/
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Pull locking fix from Ingo Molnar:
"Fix lockup in requeue-PI during signal/timeout wakeups, by Sebastian
Andrzej Siewior"
* tag 'locking-urgent-2026-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Prevent lockup in requeue-PI during signal/ timeout wakeup
Pull scheduler fixes from Ingo Molnar:
- Fix the delayed dequeue negative lag increase fix in the
fair scheduler (Peter Zijlstra)
- Fix wakeup_preempt_fair() to do proper delayed dequeue
(Vincent Guittot)
- Clear sched_entity::rel_deadline when initializing
forked entities, which bug can cause all tasks to be
EEVDF-ineligible, causing a NULL pointer dereference
crash in pick_next_entity() (Zicheng Qu)
* tag 'sched-urgent-2026-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Clear rel_deadline when initializing forked entities
sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue
sched/fair: Fix the negative lag increase fix
Currently need_futex_hash_allocate_default() depends on strict pthread
semantics, abusing CLONE_THREAD. This breaks the non-concurrency
assumptions when doing the mm->futex_ref pcpu allocations, leading to
bugs[0] when sharing the mm in other ways; ie:
BUG: KASAN: slab-use-after-free in futex_hash_put
... where the +1 bias can end up on a percpu counter that mm->futex_ref
no longer points at.
Loosen the check to cover any CLONE_VM clone, except vfork(). Excluding
vfork keeps the existing paths untouched (no overhead), and we can't
race in the first place: either the parent is suspended and the child
runs alone, or mm->futex_ref is already allocated from an earlier
CLONE_VM.
Link: https://lore.kernel.org/all/CAL_bE8LsmCQ-FAtYDuwbJhOkt9p2wwYQwAbMh=PifC=VsiBM6A@mail.gmail.com/ [0]
Fixes: d9b05321e2 ("futex: Move futex_hash_free() back to __mmput()")
Reported-by: Yiming Qian <yimingqian591@gmail.com>
Signed-off-by: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull MM fixes from Andrew Morton:
"20 hotfixes. All are for MM (and for MMish maintainers). 9 are
cc:stable and the remainder are for post-7.0 issues or aren't deemed
suitable for backporting.
There are two DAMON series from SeongJae Park which address races
which could lead to use-after-free errors, and avoid the possibility
of presenting stale parameter values to users"
* tag 'mm-hotfixes-stable-2026-04-30-15-39' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mm: memcontrol: fix rcu unbalance in get_non_dying_memcg_end()
mm/userfaultfd: detect VMA type change after copy retry in mfill_copy_folio_retry()
MAINTAINERS: remove stale kdump project URL
mm/damon/stat: detect and use fresh enabled value
mm/damon/lru_sort: detect and use fresh enabled and kdamond_pid values
mm/damon/reclaim: detect and use fresh enabled and kdamond_pid values
selftests/mm: specify requirement for PROC_MEM_ALWAYS_FORCE=y
mm/damon/sysfs-schemes: protect path kfree() with damon_sysfs_lock
mm/damon/sysfs-schemes: protect memcg_path kfree() with damon_sysfs_lock
MAINTAINERS: update Li Wang's email address
MAINTAINERS, mailmap: update email address for Qi Zheng
MAINTAINERS: update Liam's email address
mm/hugetlb_cma: round up per_node before logging it
MAINTAINERS: fix regex pattern in CORE MM category
mm/vma: do not try to unmap a VMA if mmap_prepare() invoked from mmap()
mm: start background writeback based on per-wb threshold for strictlimit BDIs
kho: fix error handling in kho_add_subtree()
liveupdate: fix return value on session allocation failure
mailmap: update entry for Dan Carpenter
vmalloc: fix buffer overflow in vrealloc_node_align()
When using global functions (i.e. subprogs), the verifier performs
function-by-function verification. In that case, the sum of the
instructions processed in each global function and in the main program
counts towards the 1 million instructions limit. Only that sum is
reported in the verifier logs.
While starting to use global functions in Cilium (finally!), we found it
can be useful to have the breakdown per global function, to understand
exactly where the budget is currently spent. This patch implements this
breakdown, under BPF_LOG_STATS, as done for the stack depths.
When iterating over subprogs, we need to skip the hidden subprogs at the
end because they don't have a corresponding func_info_aux entry and
calling bpf_subprog_is_global() would result in an OOB access.
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Link: https://lore.kernel.org/bpf/5590f9c67e614ec9054d0c7e74e87cc690a52c56.1777538384.git.paul.chaignon@gmail.com
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Pull tracing fixes from Steven Rostedt:
- Fix inverted check of registering the stats for branch tracing
When calling register_stat_tracer() which returns zero on success and
negative on error, the callers were checking the return of zero as an
error and printing a warning message. Because this was just a normal
printk() message and not a WARN(), it wasn't caught in any testing.
Fix the check to print the warning message when an error actually
happens.
- Fix a typo in a comment in tracepoint.h
- Limit the size of event probes to 3K in size
It is possible to create a dynamic event probe via the tracefs system
that is greater than the max size of an event that the ring buffer
can hold. This basically causes the event to become useless.
Limit the size of an event probe to be 3K as that should be large
enough to handle any dynamic events being created, and fits within
the PAGE_SIZE sub-buffers of the ring buffer.
* tag 'trace-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing/probes: Limit size of event probe to 3K
tracepoint: Fix typo in tracepoint.h comment
tracing: branch: Fix inverted check on stat tracer registration
There currently isn't a max limit an event probe can be. One could make an
event greater than PAGE_SIZE, which makes the event useless because if
it's bigger than the max event that can be recorded into the ring buffer,
then it will never be recorded.
A event probe should never need to be greater than 3K, so make that the
max size. As long as the max is less than the max that can be recorded
onto the ring buffer, it should be fine.
Cc: stable@vger.kernel.org
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Fixes: 93ccae7a22 ("tracing/kprobes: Support basic types on dynamic events")
Link: https://patch.msgid.link/20260428122302.706610ba@gandalf.local.home
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
