[Mesa-dev] [PATCH 7/7] radv: dump LLVM IR when AMD_shader_info is used

Samuel Pitoiset samuel.pitoiset at gmail.com
Thu Mar 15 16:16:02 UTC 2018



On 03/15/2018 05:04 PM, Bas Nieuwenhuizen wrote:
> On Wed, Mar 14, 2018 at 12:27 PM, Samuel Pitoiset
> <samuel.pitoiset at gmail.com> wrote:
>> Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
>> ---
>>   src/amd/vulkan/radv_shader.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/src/amd/vulkan/radv_shader.c b/src/amd/vulkan/radv_shader.c
>> index 327ae9fe3d..1ece13657c 100644
>> --- a/src/amd/vulkan/radv_shader.c
>> +++ b/src/amd/vulkan/radv_shader.c
>> @@ -743,6 +743,7 @@ radv_GetShaderInfoAMD(VkDevice _device,
>>                  buf = _mesa_string_buffer_create(NULL, 1024);
>>
>>                  _mesa_string_buffer_printf(buf, "%s:\n", radv_get_shader_name(variant, stage));
>> +               _mesa_string_buffer_printf(buf, "%s\n\n", variant->llvm_ir_string);
> 
> I don't think this belongs here. We don't dump the nir here either.
> I'd argue we should either be consistent with the other AMD driver and
> only do assembly or be consistent and do the whole thing.

Your call, but it might be useful when debugging with renderdoc. Though, 
I can able it locally if I need it.

> 
>>                  _mesa_string_buffer_printf(buf, "%s\n\n", variant->disasm_string);
>>                  generate_shader_stats(device, variant, stage, buf);
>>
>> --
>> 2.16.2
>>
>> _______________________________________________
>> mesa-dev mailing list
>> mesa-dev at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/mesa-dev


More information about the mesa-dev mailing list