summaryrefslogtreecommitdiff
path: root/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
diff options
context:
space:
mode:
authorKumar Kartikeya Dwivedi <memxor@gmail.com>2025-10-07 22:03:48 +0000
committerAlexei Starovoitov <ast@kernel.org>2025-10-10 10:04:51 -0700
commitf233d4855918547f19c5bff95223706d1c836b7c (patch)
treed07c443f9b34b88458d049b975dcb43e07dc0f4b /tools/testing/selftests/bpf/prog_tests/tracing_struct.c
parent469d638d1520a9332cd0d034690e75e845610a51 (diff)
bpf: Refactor storage_get_func_atomic to generic non_sleepable flag
Rename the storage_get_func_atomic flag to a more generic non_sleepable flag that tracks whether a helper or kfunc may be called from a non-sleepable context. This makes the flag more broadly applicable beyond just storage_get helpers. See [0] for more context. The flag is now set unconditionally for all helpers and kfuncs when: - RCU critical section is active. - Preemption is disabled. - IRQs are disabled. - In a non-sleepable context within a sleepable program (e.g., timer callbacks), which is indicated by !in_sleepable(). Previously, the flag was only set for storage_get helpers in these contexts. With this change, it can be used by any code that needs to differentiate between sleepable and non-sleepable contexts at the per-instruction level. The existing usage in do_misc_fixups() for storage_get helpers is preserved by checking is_storage_get_function() before using the flag. [0]: https://lore.kernel.org/bpf/CAP01T76cbaNi4p-y8E0sjE2NXSra2S=Uja8G4hSQDu_SbXxREQ@mail.gmail.com Cc: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Acked-by: Eduard Zingerman <eddyz87@gmail.com> Acked-by: Mykyta Yatsenko <mykyta.yatsenko5@gmail.com> Link: https://lore.kernel.org/r/20251007220349.3852807-3-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Diffstat (limited to 'tools/testing/selftests/bpf/prog_tests/tracing_struct.c')
0 files changed, 0 insertions, 0 deletions