Menglong Dong says:
====================
bpf trampoline support "jmp" mode
For now, the bpf trampoline is called by the "call" instruction. However,
it break the RSB and introduce extra overhead in x86_64 arch.
For example, we hook the function "foo" with fexit, the call and return
logic will be like this:
call foo -> call trampoline -> call foo-body ->
return foo-body -> return foo
As we can see above, there are 3 call, but 2 return, which break the RSB
balance. We can pseudo a "return" here, but it's not the best choice,
as it will still cause once RSB miss:
call foo -> call trampoline -> call foo-body ->
return foo-body -> return dummy -> return foo
The "return dummy" doesn't pair the "call trampoline", which can also
cause the RSB miss.
Therefore, we introduce the "jmp" mode for bpf trampoline, as advised by
Alexei in [1]. And the logic will become this:
call foo -> jmp trampoline -> call foo-body ->
return foo-body -> return foo
As we can see above, the RSB is totally balanced after this series.
In this series, we introduce the FTRACE_OPS_FL_JMP for ftrace to make it
use the "jmp" instruction instead of "call".
And we also do some adjustment to bpf_arch_text_poke() to allow us specify
the old and new poke_type.
For the BPF_TRAMP_F_SHARE_IPMODIFY case, we will fallback to the "call"
mode, as it need to get the function address from the stack, which is not
supported in "jmp" mode.
Before this series, we have the following performance with the bpf
benchmark:
$ cd tools/testing/selftests/bpf
$ ./benchs/run_bench_trigger.sh
usermode-count : 890.171 ± 1.522M/s
kernel-count : 409.184 ± 0.330M/s
syscall-count : 26.792 ± 0.010M/s
fentry : 171.242 ± 0.322M/s
fexit : 80.544 ± 0.045M/s
fmodret : 78.301 ± 0.065M/s
rawtp : 192.906 ± 0.900M/s
tp : 81.883 ± 0.209M/s
kprobe : 52.029 ± 0.113M/s
kprobe-multi : 62.237 ± 0.060M/s
kprobe-multi-all: 4.761 ± 0.014M/s
kretprobe : 23.779 ± 0.046M/s
kretprobe-multi: 29.134 ± 0.012M/s
kretprobe-multi-all: 3.822 ± 0.003M/
And after this series, we have the following performance:
usermode-count : 890.443 ± 0.307M/s
kernel-count : 416.139 ± 0.055M/s
syscall-count : 31.037 ± 0.813M/s
fentry : 169.549 ± 0.519M/s
fexit : 136.540 ± 0.518M/s
fmodret : 159.248 ± 0.188M/s
rawtp : 194.475 ± 0.144M/s
tp : 84.505 ± 0.041M/s
kprobe : 59.951 ± 0.071M/s
kprobe-multi : 63.153 ± 0.177M/s
kprobe-multi-all: 4.699 ± 0.012M/s
kretprobe : 23.740 ± 0.015M/s
kretprobe-multi: 29.301 ± 0.022M/s
kretprobe-multi-all: 3.869 ± 0.005M/s
As we can see above, the performance of fexit increase from 80.544M/s to
136.540M/s, and the "fmodret" increase from 78.301M/s to 159.248M/s.
Link: https://lore.kernel.org/bpf/20251117034906.32036-1-dongml2@chinatelecom.cn/
Changes since v2:
* reject if the addr is already "jmp" in register_ftrace_direct() and
__modify_ftrace_direct() in the 1st patch.
* fix compile error in powerpc in the 5th patch.
* changes in the 6th patch:
- fix the compile error by wrapping the write to tr->fops->flags with
CONFIG_DYNAMIC_FTRACE_WITH_JMP
- reset BPF_TRAMP_F_SKIP_FRAME when the second try of modify_fentry in
bpf_trampoline_update()
Link: https://lore.kernel.org/bpf/20251114092450.172024-1-dongml2@chinatelecom.cn/
Changes since v1:
* change the bool parameter that we add to save_args() to "u32 flags"
* rename bpf_trampoline_need_jmp() to bpf_trampoline_use_jmp()
* add new function parameter to bpf_arch_text_poke instead of introduce
bpf_arch_text_poke_type()
* rename bpf_text_poke to bpf_trampoline_update_fentry
* remove the BPF_TRAMP_F_JMPED and check the current mode with the origin
flags instead.
Link: https://lore.kernel.org/bpf/CAADnVQLX54sVi1oaHrkSiLqjJaJdm3TQjoVrgU-LZimK6iDcSA@mail.gmail.com/[1]
====================
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://patch.msgid.link/20251118123639.688444-1-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Implement the "jmp" mode for the bpf trampoline. For the ftrace_managed
case, we need only to set the FTRACE_OPS_FL_JMP on the tr->fops if "jmp"
is needed.
For the bpf poke case, we will check the origin poke type with the
"origin_flags", and current poke type with "tr->flags". The function
bpf_trampoline_update_fentry() is introduced to do the job.
The "jmp" mode will only be enabled with CONFIG_DYNAMIC_FTRACE_WITH_JMP
enabled and BPF_TRAMP_F_SHARE_IPMODIFY is not set. With
BPF_TRAMP_F_SHARE_IPMODIFY, we need to get the origin call ip from the
stack, so we can't use the "jmp" mode.
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20251118123639.688444-7-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In the origin logic, the bpf_arch_text_poke() assume that the old and new
instructions have the same opcode. However, they can have different opcode
if we want to replace a "call" insn with a "jmp" insn.
Therefore, add the new function parameter "old_t" along with the "new_t",
which are used to indicate the old and new poke type. Meanwhile, adjust
the implement of bpf_arch_text_poke() for all the archs.
"BPF_MOD_NOP" is added to make the code more readable. In
bpf_arch_text_poke(), we still check if the new and old address is NULL to
determine if nop insn should be used, which I think is more safe.
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Link: https://lore.kernel.org/r/20251118123639.688444-6-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In the origin call case, if BPF_TRAMP_F_SKIP_FRAME is not set, it means
that the trampoline is not called, but "jmp".
Introduce the function bpf_trampoline_use_jmp() to check if the trampoline
is in "jmp" mode.
Do some adjustment on the "jmp" mode for the x86_64. The main adjustment
that we make is for the stack parameter passing case, as the stack
alignment logic changes in the "jmp" mode without the "rip". What's more,
the location of the parameters on the stack also changes.
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Link: https://lore.kernel.org/r/20251118123639.688444-5-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Implement the DYNAMIC_FTRACE_WITH_JMP for x86_64. In ftrace_call_replace,
we will use JMP32_INSN_OPCODE instead of CALL_INSN_OPCODE if the address
should use "jmp".
Meanwhile, adjust the direct call in the ftrace_regs_caller. The RSB is
balanced in the "jmp" mode. Take the function "foo" for example:
original_caller:
call foo -> foo:
call fentry -> fentry:
[do ftrace callbacks ]
move tramp_addr to stack
RET -> tramp_addr
tramp_addr:
[..]
call foo_body -> foo_body:
[..]
RET -> back to tramp_addr
[..]
RET -> back to original_caller
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20251118123639.688444-3-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
For now, the "nop" will be replaced with a "call" instruction when a
function is hooked by the ftrace. However, sometimes the "call" can break
the RSB and introduce extra overhead. Therefore, introduce the flag
FTRACE_OPS_FL_JMP, which indicate that the ftrace_ops should be called
with a "jmp" instead of "call". For now, it is only used by the direct
call case.
When a direct ftrace_ops is marked with FTRACE_OPS_FL_JMP, the last bit of
the ops->direct_call will be set to 1. Therefore, we can tell if we should
use "jmp" for the callback in ftrace_call_replace().
Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20251118123639.688444-2-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Puranjay Mohan says:
====================
bpf: Nested rcu critical sections
v1: https://lore.kernel.org/bpf/20250916113622.19540-1-puranjay@kernel.org/
Changes in v1->v2:
- Move the addition of new tests to a separate patch (Alexei)
- Avoid incrementing active_rcu_locks at two places (Eduard)
Support nested rcu critical sections by making the boolean flag
active_rcu_lock a counter and use it to manage rcu critical section
state. bpf_rcu_read_lock() increments this counter and
bpf_rcu_read_unlock() decrements it, MEM_RCU -> PTR_UNTRUSTED transition
happens when active_rcu_locks drops to 0.
====================
Link: https://patch.msgid.link/20251117200411.25563-1-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, nested rcu critical sections are rejected by the verifier and
rcu_lock state is managed by a boolean variable. Add support for nested
rcu critical sections by make active_rcu_locks a counter similar to
active_preempt_locks. bpf_rcu_read_lock() increments this counter and
bpf_rcu_read_unlock() decrements it, MEM_RCU -> PTR_UNTRUSTED transition
happens when active_rcu_locks drops to 0.
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251117200411.25563-2-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Three tests are added:
- invalidate_pkt_pointers_by_tail_call checks that one can use the
packet pointer after a tail call. This was originally possible
and also poses not problems, but was made impossible by 1a4607ffba.
- invalidate_pkt_pointers_by_static_tail_call tests a corner case
found by Eduard Zingerman during the discussion of the original fix,
which was broken in that fix.
- subprog_result_tail_call tests that precision propagation works
correctly across tail calls. This did not work before.
Signed-off-by: Martin Teichmann <martin.teichmann@xfel.eu>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251119160355.1160932-3-martin.teichmann@xfel.eu
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
A successful ebpf tail call does not return to the caller, but to the
caller-of-the-caller, often just finishing the ebpf program altogether.
Any restrictions that the verifier needs to take into account - notably
the fact that the tail call might have modified packet pointers - are to
be checked on the caller-of-the-caller. Checking it on the caller made
the verifier refuse perfectly fine programs that would use the packet
pointers after a tail call, which is no problem as this code is only
executed if the tail call was unsuccessful, i.e. nothing happened.
This patch simulates the behavior of a tail call in the verifier. A
conditional jump to the code after the tail call is added for the case
of an unsucessful tail call, and a return to the caller is simulated for
a successful tail call.
For the successful case we assume that the tail call returns an int,
as tail calls are currently only allowed in functions that return and
int. We always assume that the tail call modified the packet pointers,
as we do not know what the tail call did.
For the unsuccessful case we know nothing happened, so we do not need to
add new constraints.
This approach also allows to check other problems that may occur with
tail calls, namely we are now able to check that precision is properly
propagated into subprograms using tail calls, as well as checking the
live slots in such a subprogram.
Fixes: 1a4607ffba ("bpf: consider that tail calls invalidate packet pointers")
Link: https://lore.kernel.org/bpf/20251029105828.1488347-1-martin.teichmann@xfel.eu/
Signed-off-by: Martin Teichmann <martin.teichmann@xfel.eu>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251119160355.1160932-2-martin.teichmann@xfel.eu
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In [1] Dan Carpenter reported that the following code makes the
Smatch static analyser unhappy:
17904 value = map->ops->map_lookup_elem(map, &i);
17905 if (!value)
17906 return -EINVAL;
--> 17907 items[i - start] = value->xlated_off;
The analyser assumes that the `value` variable may contain an error
and thus it should be properly checked before the dereference.
On practice this will never happen as array maps do not return
error values in map_lookup_elem, but to make the Smatch and other
possible analysers happy this patch adds a formal check.
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/bpf/aR2BN1Ix--8tmVrN@stanley.mountain/ [1]
Fixes: 493d9e0d60 ("bpf, x86: add support for indirect jumps")
Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Link: https://lore.kernel.org/r/20251119112517.1091793-1-a.s.protopopov@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, test_perf_branches_no_hw() relies on the busy loop within
test_perf_branches_common() being slow enough to allow at least one
perf event sample tick to occur before starting to tear down the
backing perf event BPF program. With a relatively small fixed
iteration count of 1,000,000, this is not guaranteed on modern fast
CPUs, resulting in the test run to subsequently fail with the
following:
bpf_testmod.ko is already unloaded.
Loading bpf_testmod.ko...
Successfully loaded bpf_testmod.ko.
test_perf_branches_common:PASS:test_perf_branches_load 0 nsec
test_perf_branches_common:PASS:attach_perf_event 0 nsec
test_perf_branches_common:PASS:set_affinity 0 nsec
check_good_sample:PASS:output not valid 0 nsec
check_good_sample:PASS:read_branches_size 0 nsec
check_good_sample:PASS:read_branches_stack 0 nsec
check_good_sample:PASS:read_branches_stack 0 nsec
check_good_sample:PASS:read_branches_global 0 nsec
check_good_sample:PASS:read_branches_global 0 nsec
check_good_sample:PASS:read_branches_size 0 nsec
test_perf_branches_no_hw:PASS:perf_event_open 0 nsec
test_perf_branches_common:PASS:test_perf_branches_load 0 nsec
test_perf_branches_common:PASS:attach_perf_event 0 nsec
test_perf_branches_common:PASS:set_affinity 0 nsec
check_bad_sample:FAIL:output not valid no valid sample from prog
Summary: 0/1 PASSED, 0 SKIPPED, 1 FAILED
Successfully unloaded bpf_testmod.ko.
On a modern CPU (i.e. one with a 3.5 GHz clock rate), executing 1
million increments of a volatile integer can take significantly less
than 1 millisecond. If the spin loop and detachment of the perf event
BPF program elapses before the first 1 ms sampling interval elapses,
the perf event will never end up firing. Fix this by bumping the loop
iteration counter a little within test_perf_branches_common(), along
with ensuring adding another loop termination condition which is
directly influenced by the backing perf event BPF program
executing. Notably, a concious decision was made to not adjust the
sample_freq value as that is just not a reliable way to go about
fixing the problem. It effectively still leaves the race window open.
Fixes: 67306f84ca ("selftests/bpf: Add bpf_read_branch_records() selftest")
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20251119143540.2911424-1-mattbobrowski@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Gracefully skip the test_perf_branches_hw subtest on platforms that
do not support LBR or require specialized perf event attributes
to enable branch sampling.
For example, AMD's Milan (Zen 3) supports BRS rather than traditional
LBR. This requires specific configurations (attr.type = PERF_TYPE_RAW,
attr.config = RETIRED_TAKEN_BRANCH_INSTRUCTIONS) that differ from the
generic setup used within this test. Notably, it also probably doesn't
hold much value to special case perf event configurations for selected
micro architectures.
Fixes: 67306f84ca ("selftests/bpf: Add bpf_read_branch_records() selftest")
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Acked-by: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/r/20251120142059.2836181-1-mattbobrowski@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add support for a new instruction
BPF_JMP|BPF_X|BPF_JA, SRC=0, DST=Rx, off=0, imm=0
which does an indirect jump to a location stored in Rx. The register
Rx should have type PTR_TO_INSN. This new type assures that the Rx
register contains a value (or a range of values) loaded from a
correct jump table – map of type instruction array.
ARM64 JIT supports indirect jumps to all registers through the A64_BR()
macro, use it to implement this new instruction.
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Reviewed-by: Anton Protopopov <a.s.protopopov@gmail.com>
Acked-by: Xu Kuohai <xukuohai@huawei.com>
Link: https://lore.kernel.org/r/20251117130732.11107-3-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add support for the instructions array map type in the arm64 JIT by
calling bpf_prog_update_insn_ptrs() with the offsets that map
xlated_offset to the jited_offset in the final image. arm64 JIT already
has this offset array which was being used for
bpf_prog_fill_jited_linfo() and can be used directly for
bpf_prog_update_insn_ptrs.
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Reviewed-by: Anton Protopopov <a.s.protopopov@gmail.com>
Acked-by: Xu Kuohai <xukuohai@huawei.com>
Link: https://lore.kernel.org/r/20251117130732.11107-2-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Hoyeon Lee says:
====================
selftests/bpf: networking test cleanups
This series finishes the sockaddr_storage migration in the networking
selftests by removing the remaining open-coded IPv4/IPv6 wrappers
(addr_port/tuple in cls_redirect, sa46 in select_reuseport). The tests
now use sockaddr_storage directly. No other custom socket-address
wrappers remain after this series, so the churn stops here and behavior
is unchanged.
====================
Link: https://patch.msgid.link/20251121081332.2309838-1-hoyeon.lee@suse.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
The select_reuseport selftest uses a custom sa46 union to represent
IPv4 and IPv6 addresses. This custom wrapper requires extra manual
handling for address family and field extraction.
Replace sa46 with sockaddr_storage and update the helper functions to
operate on native socket structures. This simplifies the code and
removes unnecessary custom address-handling logic. No functional
changes intended.
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Hoyeon Lee <hoyeon.lee@suse.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251121081332.2309838-3-hoyeon.lee@suse.com
The cls_redirect test uses a custom addr_port/tuple wrapper to represent
IPv4/IPv6 addresses and ports. This custom wrapper requires extra
conversion logic and specific helpers such as fill_addr_port(), which
are no longer necessary when using standard socket address structures.
This commit replaces addr_port/tuple with the standard sockaddr_storage
so test handles address families and ports using native socket types.
It removes the custom helper, eliminates redundant casts, and simplifies
the setup helpers without functional changes. set_up_conn() and
build_input() now take src/dst sockaddr_storage directly.
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Hoyeon Lee <hoyeon.lee@suse.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251121081332.2309838-2-hoyeon.lee@suse.com
Add missing kernel-doc documentation for the cfi_stubs and owner
fields in struct bpf_struct_ops to fix the following warnings:
Warning: include/linux/bpf.h:1931 struct member 'cfi_stubs' not
described in 'bpf_struct_ops'
Warning: include/linux/bpf.h:1931 struct member 'owner' not
described in 'bpf_struct_ops'
The cfi_stubs field was added in commit 2cd3e3772e ("x86/cfi,bpf:
Fix bpf_struct_ops CFI") to provide CFI stub functions for trampolines,
and the owner field is used for module reference counting.
Signed-off-by: Nirbhay Sharma <nirbhay.lkd@gmail.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251120204620.59571-2-nirbhay.lkd@gmail.com
subtest_kmem_cache_iter_check_slabinfo() fundamentally compares slab
cache names parsed out from /proc/slabinfo against those stored within
struct kmem_cache_result. The current problem is that the slab cache
name within struct kmem_cache_result is stored within a bounded
fixed-length array (sized to SLAB_NAME_MAX(32)), whereas the name
parsed out from /proc/slabinfo is not. Meaning, using ASSERT_STREQ()
can certainly lead to test failures, particularly when dealing with
slab cache names that are longer than SLAB_NAME_MAX(32)
bytes. Notably, kmem_cache_create() allows callers to create slab
caches with somewhat arbitrarily sized names via its __name identifier
argument, so exceeding the SLAB_NAME_MAX(32) limit that is in place
now can certainly happen.
Make subtest_kmem_cache_iter_check_slabinfo() more reliable by only
checking up to sizeof(struct kmem_cache_result.name) - 1 using
ASSERT_STRNEQ().
Fixes: a496d0cdc8 ("selftests/bpf: Add a test for kmem_cache_iter")
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Acked-by: Song Liu <song@kernel.org>
Link: https://patch.msgid.link/20251118073734.4188710-1-mattbobrowski@google.com
Amery Hung says:
====================
Replace BPF memory allocator with kmalloc_nolock() in local storage
This patchset tries to simplify bpf_local_storage.c by adopting
kmalloc_nolock(). This removes memory preallocation and reduces the
dependency of smap in bpf_selem_free() and bpf_local_storage_free().
The later will simplify a future refactor that replaces
local_storage->lock and b->lock [1].
RFC v1 tried to switch to kmalloc_nolock() unconditionally. However,
as there is substantial performance loss in socket local storage due to
1) defer_free() in kfree_nolock() and 2) no kfree_rcu() batching,
replacing kzalloc() is postponed until necessary improvements in mm
land.
Benchmark
./bench -p 1 local-storage-create --storage-type <socket,task> \
--batch-size <16,32,64>
The benchmark is a microbenchmark stress-testing how fast local storage
can be created. For task local storage, switching from BPF memory
allocator to kmalloc_nolock() yields a small amount of improvement. For
socket local storage, it remains roughly the same as nothing has changed.
Socket local storage
memory alloc batch creation speed creation speed diff
--------------- ---- ------------------ ----
kzalloc 16 144.149 ± 0.642k/s 3.10 kmallocs/create
(before) 32 144.379 ± 1.070k/s 3.08 kmallocs/create
64 144.491 ± 0.818k/s 3.13 kmallocs/create
kzalloc 16 146.180 ± 1.403k/s 3.10 kmallocs/create +1.4%
(not changed) 32 146.245 ± 1.272k/s 3.10 kmallocs/create +1.3%
64 145.012 ± 1.545k/s 3.10 kmallocs/create +0.4%
Task local storage
memory alloc batch creation speed creation speed diff
--------------- ---- ------------------ ----
BPF memory 16 24.668 ± 0.121k/s 2.54 kmallocs/create
allocator 32 22.899 ± 0.097k/s 2.67 kmallocs/create
(before) 64 22.559 ± 0.076k/s 2.56 kmallocs/create
kmalloc_nolock 16 25.796 ± 0.059k/s 2.52 kmallocs/create +4.6%
(after) 32 23.412 ± 0.069k/s 2.50 kmallocs/create +2.2%
64 23.717 ± 0.108k/s 2.60 kmallocs/create +5.1%
[1] https://lore.kernel.org/bpf/20251002225356.1505480-1-ameryhung@gmail.com/
v1 -> v2
- Only replace BPF memory allocator with kmalloc_nolock()
Link: https://lore.kernel.org/bpf/20251112175939.2365295-1-ameryhung@gmail.com/
====================
Link: https://patch.msgid.link/20251114201329.3275875-1-ameryhung@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Replace bpf memory allocator with kmalloc_nolock() to reduce memory
wastage due to preallocation.
In bpf_selem_free(), an selem now needs to wait for a RCU grace period
before being freed when reuse_now == true. Therefore, rcu_barrier()
should be always be called in bpf_local_storage_map_free().
In bpf_local_storage_free(), since smap->storage_ma is no longer needed
to return the memory, the function is now independent from smap.
Remove the outdated comment in bpf_local_storage_alloc(). We already
free selem after an RCU grace period in bpf_local_storage_update() when
bpf_local_storage_alloc() failed the cmpxchg since commit c0d63f3091
("bpf: Add bpf_selem_free()").
Signed-off-by: Amery Hung <ameryhung@gmail.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20251114201329.3275875-5-ameryhung@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Save the memory allocation method used for bpf_local_storage in the
struct explicitly so that we don't need to go through the hassle to
find out the info. When a later patch replaces BPF memory allocator
with kmalloc_noloc(), bpf_local_storage_free() will no longer need
smap->storage_ma to return the memory and completely remove the
dependency on smap in bpf_local_storage_free().
Signed-off-by: Amery Hung <ameryhung@gmail.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20251114201329.3275875-4-ameryhung@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Since selem already saves a pointer to smap, use it instead of an
additional argument in bpf_selem_free(). This requires moving the
SDATA(selem)->smap assignment from bpf_selem_link_map() to
bpf_selem_alloc() since bpf_selem_free() may be called without the
selem being linked to smap in bpf_local_storage_update().
Signed-off-by: Amery Hung <ameryhung@gmail.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20251114201329.3275875-3-ameryhung@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The connect4_prog and bpf_iter_setsockopt tests duplicate the same
open-coded TCP congestion control string comparison logic. Since
bpf_strncmp() provides the same functionality, use it instead to
avoid repeated open-coded loops.
This change applies only to functional BPF tests and does not affect
the verifier performance benchmarks (veristat.cfg). No functional
changes intended.
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Hoyeon Lee <hoyeon.lee@suse.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251115225550.1086693-5-hoyeon.lee@suse.com
Some BPF selftests contain identical copies of the min(), max(),
before(), and after() helpers. These repeated snippets are the same
across the tests and do not need to be defined separately.
Move these helpers into bpf_tracing_net.h so they can be shared by
TCP related BPF programs. This removes repeated code and keeps the
helpers in a single place.
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Hoyeon Lee <hoyeon.lee@suse.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251115225550.1086693-4-hoyeon.lee@suse.com
Syzkaller triggers an invalid memory access issue following fault
injection in update_effective_progs. The issue can be described as
follows:
__cgroup_bpf_detach
update_effective_progs
compute_effective_progs
bpf_prog_array_alloc <-- fault inject
purge_effective_progs
/* change to dummy_bpf_prog */
array->items[index] = &dummy_bpf_prog.prog
---softirq start---
__do_softirq
...
__cgroup_bpf_run_filter_skb
__bpf_prog_run_save_cb
bpf_prog_run
stats = this_cpu_ptr(prog->stats)
/* invalid memory access */
flags = u64_stats_update_begin_irqsave(&stats->syncp)
---softirq end---
static_branch_dec(&cgroup_bpf_enabled_key[atype])
The reason is that fault injection caused update_effective_progs to fail
and then changed the original prog into dummy_bpf_prog.prog in
purge_effective_progs. Then a softirq came, and accessing the members of
dummy_bpf_prog.prog in the softirq triggers invalid mem access.
To fix it, skip updating stats when stats is NULL.
Fixes: 492ecee892 ("bpf: enable program stats")
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Link: https://lore.kernel.org/r/20251115102343.2200727-1-pulehui@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add a test to check that bpf_skb_check_mtu(BPF_MTU_CHK_SEGS) is
rejected (-EINVAL) if skb->transport_header is not set. The test
needs to lower the MTU of the loopback device. Thus, take this
opportunity to run the test in a netns by adding "ns_" to the test
name. The "serial_" prefix can then be removed.
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20251112232331.1566074-2-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The bpf_skb_check_mtu helper needs to use skb->transport_header when
the BPF_MTU_CHK_SEGS flag is used:
bpf_skb_check_mtu(skb, ifindex, &mtu_len, 0, BPF_MTU_CHK_SEGS)
The transport_header is not always set. There is a WARN_ON_ONCE
report when CONFIG_DEBUG_NET is enabled + skb->gso_size is set +
bpf_prog_test_run is used:
WARNING: CPU: 1 PID: 2216 at ./include/linux/skbuff.h:3071
skb_gso_validate_network_len
bpf_skb_check_mtu
bpf_prog_3920e25740a41171_tc_chk_segs_flag # A test in the next patch
bpf_test_run
bpf_prog_test_run_skb
For a normal ingress skb (not test_run), skb_reset_transport_header
is performed but there is plan to avoid setting it as described in
commit 2170a1f091 ("net: no longer reset transport_header in __netif_receive_skb_core()").
This patch fixes the bpf helper by checking
skb_transport_header_was_set(). The check is done just before
skb->transport_header is used, to avoid breaking the existing bpf prog.
The WARN_ON_ONCE is limited to bpf_prog_test_run, so targeting bpf-next.
Fixes: 34b2021cc6 ("bpf: Add BPF-helper for MTU checking")
Cc: Jesper Dangaard Brouer <hawk@kernel.org>
Reported-by: Kaiyan Mei <M202472210@hust.edu.cn>
Reported-by: Yinhao Hu <dddddd@hust.edu.cn>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20251112232331.1566074-1-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Metadata about a kfunc call is added to the kfunc_tab in
add_kfunc_call() but the call instruction itself could get removed by
opt_remove_dead_code() later if it is not reachable.
If the call instruction is removed, specialize_kfunc() is never called
for it and the desc->imm in the kfunc_tab is never initialized for this
kfunc call. In this case, sort_kfunc_descs_by_imm_off(env->prog); in
do_misc_fixups() doesn't sort the table correctly.
This is a problem for s390 as its JIT uses this table to find the
addresses for kfuncs, and if this table is not sorted properly, JIT may
fail to find addresses for valid kfunc calls.
This was exposed by:
commit d869d56ca8 ("bpf: verifier: refactor kfunc specialization")
as before this commit, desc->imm was initialised in add_kfunc_call()
which happens before dead code elimination.
Move desc->imm setup down to sort_kfunc_descs_by_imm_off(), this fixes
the problem and also saves us from having the same logic in
add_kfunc_call() and specialize_kfunc().
Suggested-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251114154023.12801-1-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf_task_work_schedule_resume() and bpf_task_work_schedule_signal() have
been renamed in bpf tree to bpf_task_work_schedule_resume_impl() and
bpf_task_work_schedule_signal_impl() accordingly.
There are few uses of these kfuncs in selftests that are not in bpf
tree, so that when we port [1] into bpf-next, those BPF programs will
not compile.
This patch aligns those remaining callsites with the kfunc renaming.
It should go on top of [1] when applying on bpf-next.
1: https://lore.kernel.org/all/20251104-implv2-v3-0-4772b9ae0e06@meta.com/
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Link: https://lore.kernel.org/r/20251105132105.597344-1-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Paul Houssel says:
====================
libbpf: fix BTF dedup to support recursive typedef
Pahole fails to encode BTF for some Go projects (e.g. Kubernetes and
Podman) due to recursive type definitions that create reference loops
not representable in C. These recursive typedefs trigger a failure in
the BTF deduplication algorithm.
This patch extends btf_dedup_struct_types() to properly handle potential
recursion for BTF_KIND_TYPEDEF, similar to how recursion is already
handled for BTF_KIND_STRUCT. This allows pahole to successfully
generate BTF for Go binaries using recursive types without impacting
existing C-based workflows.
Changes in v4: fix typo found by Claude-based CI
Changes in v3:
1. Patch 1: Adjusted the comment of btf_dedup_ref_type() to refer to
typedef as well.
2. Patch 2: Update of the "dedup: recursive typedef" test to include a
duplicated version of the types to make sure deduplication still happens
in this case.
Changes in v2:
1. Patch 1: Refactored code to prevent copying existing logic. Instead of
adding a new function we modify the existing btf_dedup_struct_type()
function to handle the BTF_KIND_TYPEDEF case. Calls to btf_hash_struct()
and btf_shallow_equal_struct() are replaced with calls to functions that
select btf_hash_struct() / btf_hash_typedef() based on the type.
2. Patch 2: Added tests
v3: https://lore.kernel.org/lkml/cover.1763024337.git.paul.houssel@orange.com/
v2: https://lore.kernel.org/lkml/cover.1762956564.git.paul.houssel@orange.com/
v1: https://lore.kernel.org/lkml/20251107153408.159342-1-paulhoussel2@gmail.com/
====================
Link: https://patch.msgid.link/cover.1763037045.git.paul.houssel@orange.com
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
When test_send_signal_kern__open_and_load() fails parent closes the
pipe which cases ASSERT_EQ(read(pipe_p2c...)) to fail, but child
continues and enters infinite loop, while parent is stuck in wait(NULL).
Other error paths have similar issue, so kill the child before waiting on it.
The bug was discovered while compiling all of selftests with -O1 instead of -O2
which caused progs/test_send_signal_kern.c to fail to load.
Fixes: ab8b7f0cb3 ("tools/bpf: Add self tests for bpf_send_signal_thread()")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20251113171153.2583-1-alexei.starovoitov@gmail.com
Pull pci fixes from Bjorn Helgaas:
- Cache the ASPM L0s/L1 Supported bits early so quirks can override
them if necessary (Bjorn Helgaas)
- Add quirks for PA Semi and Freescale Root Ports and a HiSilicon Wi-Fi
device that are reported to have broken L0s and L1 (Shawn Lin, Bjorn
Helgaas)
* tag 'pci-v6.18-fixes-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI/ASPM: Avoid L0s and L1 on Hi1105 [19e5:1105] Wi-Fi
PCI/ASPM: Avoid L0s and L1 on PA Semi [1959:a002] Root Ports
PCI/ASPM: Avoid L0s and L1 on Freescale [1957:0451] Root Ports
PCI/ASPM: Convert quirks to override advertised link states
PCI/ASPM: Add pcie_aspm_remove_cap() to override advertised link states
PCI/ASPM: Cache L0s/L1 Supported so advertised link states can be overridden
Pull bpf fixes from Alexei Starovoitov:
- Fix interaction between livepatch and BPF fexit programs (Song Liu)
With Steven and Masami acks.
- Fix stack ORC unwind from BPF kprobe_multi (Jiri Olsa)
With Steven and Masami acks.
- Fix out of bounds access in widen_imprecise_scalars() in the verifier
(Eduard Zingerman)
- Fix conflicts between MPTCP and BPF sockmap (Jiayuan Chen)
- Fix net_sched storage collision with BPF data_meta/data_end (Eric
Dumazet)
- Add _impl suffix to BPF kfuncs with implicit args to avoid breaking
them in bpf-next when KF_IMPLICIT_ARGS is added (Mykyta Yatsenko)
* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Test widen_imprecise_scalars() with different stack depth
bpf: account for current allocated stack depth in widen_imprecise_scalars()
bpf: Add bpf_prog_run_data_pointers()
selftests/bpf: Add mptcp test with sockmap
mptcp: Fix proto fallback detection with BPF
mptcp: Disallow MPTCP subflows from sockmap
selftests/bpf: Add stacktrace ips test for raw_tp
selftests/bpf: Add stacktrace ips test for kprobe_multi/kretprobe_multi
x86/fgraph,bpf: Fix stack ORC unwind from kprobe_multi return probe
Revert "perf/x86: Always store regs->ip in perf_callchain_kernel()"
bpf: add _impl suffix for bpf_stream_vprintk() kfunc
bpf:add _impl suffix for bpf_task_work_schedule* kfuncs
selftests/bpf: Add tests for livepatch + bpf trampoline
ftrace: bpf: Fix IPMODIFY + DIRECT in modify_ftrace_direct()
ftrace: Fix BPF fexit with livepatch
Pull Rust fix from Miguel Ojeda:
- Fix a Rust 1.91.0 build issue due to 'bindings.o' not containing
DWARF debug information anymore by teaching gendwarfksyms to skip
object files without exports
* tag 'rust-fixes-6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux:
gendwarfksyms: Skip files with no exports
