diff options
author | Peter Zijlstra <peterz@infradead.org> | 2019-03-19 15:18:56 +0300 |
---|---|---|
committer | Ingo Molnar <mingo@kernel.org> | 2019-04-03 12:42:33 +0300 |
commit | a1247d06d01045d7ab2882a9c074fbf21137c690 (patch) | |
tree | de30aa6f8d6a1f35a5bf17932d0ea5fe1c598c62 /arch/x86/lib | |
parent | 5e7a8ca319268a70a6c7c3c1fde5bea38e1e5539 (diff) | |
download | linux-a1247d06d01045d7ab2882a9c074fbf21137c690.tar.xz |
locking/static_key: Fix false positive warnings on concurrent dec/inc
Even though the atomic_dec_and_mutex_lock() in
__static_key_slow_dec_cpuslocked() can never see a negative value in
key->enabled the subsequent sanity check is re-reading key->enabled, which may
have been set to -1 in the meantime by static_key_slow_inc_cpuslocked().
CPU A CPU B
__static_key_slow_dec_cpuslocked(): static_key_slow_inc_cpuslocked():
# enabled = 1
atomic_dec_and_mutex_lock()
# enabled = 0
atomic_read() == 0
atomic_set(-1)
# enabled = -1
val = atomic_read()
# Oops - val == -1!
The test case is TCP's clean_acked_data_enable() / clean_acked_data_disable()
as tickled by KTLS (net/ktls).
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reported-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Tested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: ard.biesheuvel@linaro.org
Cc: oss-drivers@netronome.com
Cc: pbonzini@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'arch/x86/lib')
0 files changed, 0 insertions, 0 deletions