thanks
On Mon, May 25, 2015 at 5:38 PM, Marek Olšák wrote:
> Hi Rob,
>
> I've sent a patch that adds the CAP.
>
> Marek
>
> On Mon, May 25, 2015 at 3:17 PM, Rob Clark wrote:
>> Ignoring the compiler for a moment, I think this would probably break
>> my varying linking (where I match up VS out an
Hi Rob,
I've sent a patch that adds the CAP.
Marek
On Mon, May 25, 2015 at 3:17 PM, Rob Clark wrote:
> Ignoring the compiler for a moment, I think this would probably break
> my varying linking (where I match up VS out and FS in).. (and I
> wouldn't be surprised if somewhere between tgsi_to_ni
Overall, this looks like a good idea to me. Probably needs some changes
in our internal code but the idea is quite sane.
Roland
Am 24.05.2015 um 13:19 schrieb Marek Olšák:
> Hi,
>
> The reason I add this is that TGSI doesn't allow indirect indexing of inputs
> and outputs. Consider this:
>
> M
Ignoring the compiler for a moment, I think this would probably break
my varying linking (where I match up VS out and FS in).. (and I
wouldn't be surprised if somewhere between tgsi_to_nir and my
compiler, it also caused breakage)
Unless you have a setup where you can test/fix all the drivers, I
Testing is really easy. Just run all piglit variable-indexing tests.
Also, drivers don't have to do anything for outputs yet, because those
are always moved to temps by lower_output_reads.
Marek
On Sun, May 24, 2015 at 6:22 PM, Ilia Mirkin wrote:
> Right, but you're changing what the driver rece
Right, but you're changing what the driver receives, so we should make
sure they all handle it correctly. Or let the major driver authors
know what's going on so they can test it out and fix their driver
accordingly. Perhaps find a couple of piglit tests that exercise the
functionality. On that not
Drivers that only use tgsi_shader_info won't break.
Drivers that process tgsi_full_declaration manually and interpret
Range.First .. Range.Last correctly won't break either.
A driver can only break if it doesn't handle Range.Last correctly. If
that's the case, the driver should be fixed, because
While I'm all for doing this, won't this break every driver if it no
longer has all the decl's? It'll take special logic to convert
DECL IN[0..5], GENERIC[0]
into
DECL IN[0], GENERIC[0]
DECL IN[1], GENERIC[1]
etc
Perhaps this should be guarded by a cap? Or an audit of all drivers
should be done
Hi,
The reason I add this is that TGSI doesn't allow indirect indexing of inputs
and outputs. Consider this:
MOV OUT[ADDR[0]-1000], IMM[0]
There is no way to know where the output array starts here. It could be for
example OUT[6]=GENERIC4 or anything else. The problem is some outputs are
phys
