On Mit, 2013-07-10 at 08:15 -0700, Tom Stellard wrote:
> On Wed, Jul 10, 2013 at 12:32:25PM +0200, Michel Dänzer wrote:
> > On Fre, 2013-06-28 at 14:37 -0700, Tom Stellard wrote:
> > > On Wed, Jun 19, 2013 at 06:28:21PM +0200, Michel Dänzer wrote:
> > > >
> > > > These patches implement enough of
On Wed, Jul 10, 2013 at 12:32:25PM +0200, Michel Dänzer wrote:
> On Fre, 2013-06-28 at 14:37 -0700, Tom Stellard wrote:
> > On Wed, Jun 19, 2013 at 06:28:21PM +0200, Michel Dänzer wrote:
> > >
> > > These patches implement enough of local memory support to allow radeonsi
> > > to use that for comp
On Fre, 2013-06-28 at 14:37 -0700, Tom Stellard wrote:
> On Wed, Jun 19, 2013 at 06:28:21PM +0200, Michel Dänzer wrote:
> >
> > These patches implement enough of local memory support to allow radeonsi
> > to use that for computing derivatives, as suggested by Tom.
> >
> > They also almost allow t
On Wed, Jun 19, 2013 at 06:28:21PM +0200, Michel Dänzer wrote:
>
> These patches implement enough of local memory support to allow radeonsi
> to use that for computing derivatives, as suggested by Tom.
>
> They also almost allow test/CodeGen/R600/local-memory.ll to generate
> code for SI. Right n
These patches implement enough of local memory support to allow radeonsi
to use that for computing derivatives, as suggested by Tom.
They also almost allow test/CodeGen/R600/local-memory.ll to generate
code for SI. Right now it still fails because it tries to copy a VGPR to
an SGPR, which is not
