Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1719168
| From | Peter Zijlstra <peterz@infradead.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH] rwsem: fix missed wakeup due to reordering of load |
| Date | 2017-08-24 13:30 +0200 |
| Message-ID | <uhYuK-5Me-5@gated-at.bofh.it> (permalink) |
| References | <uhC1c-8iX-13@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
WTH did you not Cc the people that commented on your patch last time? On Wed, Aug 23, 2017 at 04:58:55PM +0530, Prateek Sood wrote: > If a spinner is present, there is a chance that the load of > rwsem_has_spinner() in rwsem_wake() can be reordered with > respect to decrement of rwsem count in __up_write() leading > to wakeup being missed. > spinning writer up_write caller > --------------- ----------------------- > [S] osq_unlock() [L] osq > spin_lock(wait_lock) > sem->count=0xFFFFFFFF00000001 > +0xFFFFFFFF00000000 > count=sem->count > MB > sem->count=0xFFFFFFFE00000001 > -0xFFFFFFFF00000001 > RMB This doesn't make sense, it appears to order a STORE against something else. > spin_trylock(wait_lock) > return > rwsem_try_write_lock(count) > spin_unlock(wait_lock) > schedule()
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH] rwsem: fix missed wakeup due to reordering of load Prateek Sood <prsood@codeaurora.org> - 2017-08-23 13:30 +0200
Re: [PATCH] rwsem: fix missed wakeup due to reordering of load Peter Zijlstra <peterz@infradead.org> - 2017-08-24 13:30 +0200
Re: [PATCH] rwsem: fix missed wakeup due to reordering of load Peter Zijlstra <peterz@infradead.org> - 2017-08-24 14:40 +0200
Re: [PATCH] rwsem: fix missed wakeup due to reordering of load Peter Zijlstra <peterz@infradead.org> - 2017-08-24 15:00 +0200
csiph-web