Re: [PATCHv1] sunrpc: fix write space race causing stalls

2016-09-16 Thread Trond Myklebust
> On Sep 16, 2016, at 13:29, David Vrabel wrote: > > On 16/09/16 18:06, Trond Myklebust wrote: >> >>> On Sep 16, 2016, at 12:41, David Vrabel wrote: >>> >>> On 16/09/16 17:01, Trond Myklebust wrote: > On Sep 16, 2016, at 08:28,

Re: [PATCHv1] sunrpc: fix write space race causing stalls

2016-09-16 Thread David Vrabel
On 16/09/16 18:06, Trond Myklebust wrote: > >> On Sep 16, 2016, at 12:41, David Vrabel wrote: >> >> On 16/09/16 17:01, Trond Myklebust wrote: >>> On Sep 16, 2016, at 08:28, David Vrabel wrote: Write space becoming available may

Re: [PATCHv1] sunrpc: fix write space race causing stalls

2016-09-16 Thread Trond Myklebust
> On Sep 16, 2016, at 12:41, David Vrabel wrote: > > On 16/09/16 17:01, Trond Myklebust wrote: >> >>> On Sep 16, 2016, at 08:28, David Vrabel wrote: >>> >>> Write space becoming available may race with putting the task to sleep >>> in

Re: [PATCHv1] sunrpc: fix write space race causing stalls

2016-09-16 Thread David Vrabel
On 16/09/16 17:01, Trond Myklebust wrote: > >> On Sep 16, 2016, at 08:28, David Vrabel wrote: >> >> Write space becoming available may race with putting the task to sleep >> in xprt_wait_for_buffer_space(). The existing mechanism to avoid the >> race does not work. >>

Re: [PATCHv1] sunrpc: fix write space race causing stalls

2016-09-16 Thread Trond Myklebust
> On Sep 16, 2016, at 08:28, David Vrabel wrote: > > Write space becoming available may race with putting the task to sleep > in xprt_wait_for_buffer_space(). The existing mechanism to avoid the > race does not work. > > This (edited) partial trace illustrates the