[lkml]   [2012]   [Feb]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 7/9] block: implement bio_associate_current()
On Mon, Feb 20, 2012 at 09:01:28AM -0800, Tejun Heo wrote:
> On Mon, Feb 20, 2012 at 09:36:22AM -0500, Vivek Goyal wrote:
> > In copy_io(), we don't share the io_context if "current" does not have
> > task->io_context set. Does that mean if even if I specify CLONE_IO, two
> > threads might not share io context depending on when clone() happened? Or
> > I am reading the code wrong.
> Yeah, AFAICS, if the cloning task hasn't issued IO before, CLONE_IO is
> ignored.

Will it make sense to try to allocate and attach io_context and then
share it in copy_io()?

Well, you are planning to kill CLONE_IO altogether, so it does not
make a difference.


 \ /
  Last update: 2012-02-20 20:19    [W:0.046 / U:7.540 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site