Besides my comment on one of the patches, the series is:
Reviewed-by: Marek Olšák <marek.ol...@amd.com>
Marek
On Sat, May 4, 2019 at 9:39 AM Nicolai Hähnle <nhaeh...@gmail.com> wrote:
> From: Nicolai Hähnle <nicolai.haeh...@amd.com>
>
> This implies that the memory will always be at address 0, which allows
> LLVM to generate slightly better code.
> ---
> src/gallium/drivers/radeonsi/si_shader.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/gallium/drivers/radeonsi/si_shader.c
> b/src/gallium/drivers/radeonsi/si_shader.c
> index 0cf4d01a36f..91f4c177bd0 100644
> --- a/src/gallium/drivers/radeonsi/si_shader.c
> +++ b/src/gallium/drivers/radeonsi/si_shader.c
> @@ -2201,21 +2201,21 @@ void si_declare_compute_memory(struct
> si_shader_context *ctx)
>
> LLVMTypeRef i8p = LLVMPointerType(ctx->i8, AC_ADDR_SPACE_LDS);
> LLVMValueRef var;
>
> assert(!ctx->ac.lds);
>
> var = LLVMAddGlobalInAddressSpace(ctx->ac.module,
> LLVMArrayType(ctx->i8, lds_size),
> "compute_lds",
> AC_ADDR_SPACE_LDS);
> - LLVMSetAlignment(var, 4);
> + LLVMSetAlignment(var, 64 * 1024);
>
> ctx->ac.lds = LLVMBuildBitCast(ctx->ac.builder, var, i8p, "");
> }
>
> void si_tgsi_declare_compute_memory(struct si_shader_context *ctx,
> const struct tgsi_full_declaration
> *decl)
> {
> assert(decl->Declaration.MemType == TGSI_MEMORY_TYPE_SHARED);
> assert(decl->Range.First == decl->Range.Last);
>
> --
> 2.20.1
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
