Re: [PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-18 Thread Keith Packard
Peter Hutterer writes: > f9c58cdf8b51 and 59bba16c have my rev-by and the revised v5 > (f2161ee249c1b6806) as well, this is as close as I'm going to get to review > those anyway ;) Thanks. I've pushed an updated version of the branch. Just one more to review, the

Re: [PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-17 Thread Peter Hutterer
On Mon, May 16, 2016 at 09:37:18PM -0500, Keith Packard wrote: > Peter Hutterer writes: > > > long-term it's better to have separate patches here, it makes archeology > > less mysterious :) > > And we never do any archeology on X server code... > > I've pushed an

Re: [PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-16 Thread Peter Hutterer
On Mon, May 16, 2016 at 01:24:21AM -0500, Keith Packard wrote: > Peter Hutterer writes: > > Thanks for taking a look at these. This one is less mechanical than I'd > like, so some of the changes aren't obvious. It's definitely good to > review them carefully. > > >>

Re: [PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-16 Thread Keith Packard
Peter Hutterer writes: Thanks for taking a look at these. This one is less mechanical than I'd like, so some of the changes aren't obvious. It's definitely good to review them carefully. >> KdNotifyFd(int fd, int ready, void *data) >> { >> int i = (int)

Re: [PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-15 Thread Peter Hutterer
On Wed, May 11, 2016 at 01:54:51PM -0700, Keith Packard wrote: > This removes all of the SIGIO handling support used for input > throughout the X server, preparing the way for using threads for input > handling instead. > > Places calling OsBlockSIGIO and OsReleaseSIGIO are marked with calls > to

[PATCH xserver 2/9] Remove SIGIO support for input [v3]

2016-05-11 Thread Keith Packard
This removes all of the SIGIO handling support used for input throughout the X server, preparing the way for using threads for input handling instead. Places calling OsBlockSIGIO and OsReleaseSIGIO are marked with calls to stub functions input_lock/input_unlock so that we don't lose this