linux_dsm_epyc7002/lockdep_states.h at d018031f562b9c2eff038969ab1955a370c52d8f - linux_dsm_epyc7002 - Cephas TEA

AuxXxilium/linux_dsm_epyc7002

mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-11-30 12:16:45 +07:00

Peter Zijlstra d92a8cfcb3 locking/lockdep: Rework FS_RECLAIM annotation

A while ago someone, and I cannot find the email just now, asked if we
could not implement the RECLAIM_FS inversion stuff with a 'fake' lock
like we use for other things like workqueues etc. I think this should
be possible which allows reducing the 'irq' states and will reduce the
amount of __bfs() lookups we do.

Removing the 1 IRQ state results in 4 less __bfs() walks per
dependency, improving lockdep performance. And by moving this
annotation out of the lockdep code it becomes easier for the mm people
to extend.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Byungchul Park <byungchul.park@lge.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nikolay Borisov <nborisov@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akpm@linux-foundation.org
Cc: boqun.feng@gmail.com
Cc: iamjoonsoo.kim@lge.com
Cc: kernel-team@lge.com
Cc: kirill@shutemov.name
Cc: npiggin@gmail.com
Cc: walken@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

2017-08-10 12:29:03 +02:00

9 lines

207 B

C

Raw Blame History

 /*
  * Lockdep states,
  *
  * please update XXX_LOCK_USAGE_STATES in include/linux/lockdep.h whenever
  * you add one, or come up with a nice dynamic solution.
  */
 LOCKDEP_STATE(HARDIRQ)
 LOCKDEP_STATE(SOFTIRQ)