[Mesa-dev] [PATCH] nvir: Always split 64-bit IMAD/IMUL operations

[Karol Herbst]

Hi Dylan,

this patch only matters for debug builds (DEBUG set), but your merge
result looks correct nonetheless.

Thanks

On Fri, Sep 14, 2018 at 6:16 PM, Dylan Baker <dylan at pnwbakers.com> wrote:
> Quoting Pierre Moreau (2017-12-04 15:51:04)
>> Those operations do not map to actual hardware instructions, therefore
>> those should always be lowered to 32-bit instructions.
>>
>> Fixes: 009c54aa7af "nv50/ir: Split 64-bit integer MAD/MUL operations"
>> Signed-off-by: Pierre Moreau <pierre.morrow at free.fr>
>> ---
>>  src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
>> index 61d4e6a2d0..14bdcea2ca 100644
>> --- a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
>> +++ b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
>> @@ -3794,7 +3794,7 @@ Program::optimizeSSA(int level)
>>     RUN_PASS(2, AlgebraicOpt, run);
>>     RUN_PASS(2, ModifierFolding, run); // before load propagation -> less checks
>>     RUN_PASS(1, ConstantFolding, foldAll);
>> -   RUN_PASS(1, Split64BitOpPreRA, run);
>> +   RUN_PASS(0, Split64BitOpPreRA, run);
>>     RUN_PASS(1, LoadPropagation, run);
>>     RUN_PASS(1, IndirectPropagation, run);
>>     RUN_PASS(2, MemoryOpt, run);
>> --
>> 2.15.0
>>
>
> Hi Pierre,
>
> There was a small conflict when applying this to 18.1; I think my resolution is
> correct, but you can see the version of the patch here:
> https://gitlab.freedesktop.org/mesa/mesa/commit/649aff1a8788684c3160ab6001016054de251f39
> please let me know if any changes are needed.
>
> Dylan
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>