[PATCH 1/2] OS support: fix writeable client vs IgnoreClient behavior

[Keith Packard]

On Mon, 28 Jun 2010 16:59:03 -0700, Jesse Barnes <jbarnes at virtuousgeek.org> wrote:
> On Mon, 28 Jun 2010 16:55:30 -0700
> Keith Packard <keithp at keithp.com> wrote:
> 
> > On Mon, 28 Jun 2010 15:58:46 -0700, Jesse Barnes <jbarnes at virtuousgeek.org> wrote:
> > 
> > > When ResetCurrentRequest is called, or IgnoreClient is called when a
> > > client has input pending, IgnoredClientsWithInput will be set.  However,
> > > a subsequent IgnoreClient request will clear the client fd from that fd
> > > set, potentially causing the client to hang.
> > 
> > Sounds like we need a counter for IgnoreClient that leaves the client
> > ignored until it gets back to zero.
> 
> Yeah, that was another approach I coded up, it worked just as well.
> Either way I suppose their could be code depending on the current
> behavior, that's what I'm most worried about.

A counter would make nesting Ignore/Attend client calls work better,
right? Your plan would wake the client at the first AttendClient call,
rather than waiting for all of the suspending conditions to clear.

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg-devel/attachments/20100628/75f6d3d5/attachment-0001.pgp>