On Fri, Nov 21, 2014 at 11:09:44AM -0800, Matt Turner wrote: > On Fri, Nov 21, 2014 at 10:50 AM, Ben Widawsky > <benjamin.widaw...@intel.com> wrote: > > Fix one of the few cases where we can't reliable touch the destination > > hazard > > bits. I am explicitly doing this patch individually so it is easy to > > backport. I > > was tempted to do this patch before the previous patch which reorganized the > > code, but I believe even doing that first, this is still easy to backport. > > > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86268 > > Signed-off-by: Ben Widawsky <b...@bwidawsk.net> > > --- > > src/mesa/drivers/dri/i965/brw_vec4.cpp | 20 ++++++++++++++++++-- > > src/mesa/drivers/dri/i965/brw_vec4.h | 1 + > > 2 files changed, 19 insertions(+), 2 deletions(-) > > > > diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp > > b/src/mesa/drivers/dri/i965/brw_vec4.cpp > > index 0c2bbe9..d583f27 100644 > > --- a/src/mesa/drivers/dri/i965/brw_vec4.cpp > > +++ b/src/mesa/drivers/dri/i965/brw_vec4.cpp > > @@ -841,9 +841,25 @@ vec4_visitor::move_push_constants_to_pull_constants() > > } > > > > /* Conditions for which we want to avoid setting the dependency control > > bits */ > > -static bool > > -is_dep_ctrl_safe(vec4_instruction *inst) > > +bool > > +vec4_visitor::is_dep_ctrl_safe(vec4_instruction *inst) > > { > > +#define IS_DWORD(reg) \ > > + (reg.type == BRW_REGISTER_TYPE_UD || \ > > + reg.type == BRW_REGISTER_TYPE_D) > > I don't really love the macro, but meh. > > With the comments about #1 addressed, these two are > > Reviewed-by: Matt Turner <matts...@gmail.com> > > Did they fix anything that you can tell?
Yes. It does fix the hang (and make the ES3 test pass). I've asked QA to confirm. They confirmed a more rudimentary form yesterday. -- Ben Widawsky, Intel Open Source Technology Center _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
