vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

vulkan: only use M-sized matmul on Apple GPUs

Browse source 
L-sized and S-sized matmuls are broken on Apple GPUs, force using
M-size with this vendor.

Signed-off-by: Sergio Lopez <slp@redhat.com>
This commit is contained in:
Sergio Lopez 2024-02-11 11:30:16 +01:00
parent f79cef94ae
commit 3a5a7e3718
1 changed files with 19 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

19
ggml-vulkan.cpp
View file

@ -27,6 +27,7 @@
#define CEIL_DIV(M, N) (((M) + (N)-1) / (N))
#define VK_VENDOR_ID_AMD 0x1002
#define VK_VENDOR_ID_APPLE 0x106b
#define VK_VENDOR_ID_INTEL 0x8086
#define VK_VENDOR_ID_NVIDIA 0x10de
@ -2073,6 +2074,22 @@ static vk_pipeline* ggml_vk_guess_matmul_pipeline_amd(ggml_backend_vk_context *
return aligned ? &ctx->pipeline_matmul_f32_aligned_m : &ctx->pipeline_matmul_f32_m;
}
static vk_pipeline* ggml_vk_guess_matmul_pipeline_apple(ggml_backend_vk_context * ctx, bool bit16_x, bool bit16_y, bool aligned) {
#ifdef GGML_VULKAN_DEBUG
std::cerr << " M" << std::endl;
#endif
if (bit16_x && bit16_y) {
return aligned ? &ctx->pipeline_matmul_f16_aligned_m : &ctx->pipeline_matmul_f16_m;
}
if (bit16_x && !bit16_y) {
return aligned ? &ctx->pipeline_matmul_f16_f32_aligned_m : &ctx->pipeline_matmul_f16_f32_m;
}
if (!bit16_x && bit16_y) {
GGML_ASSERT(false);
}
return aligned ? &ctx->pipeline_matmul_f32_aligned_m : &ctx->pipeline_matmul_f32_m;
}
static vk_pipeline* ggml_vk_guess_matmul_pipeline_intel(ggml_backend_vk_context * ctx, bool bit16_x, bool bit16_y, bool aligned) {
#ifdef GGML_VULKAN_DEBUG
std::cerr << " S" << std::endl;
@ -2096,6 +2113,8 @@ static vk_pipeline* ggml_vk_guess_matmul_pipeline(ggml_backend_vk_context * ctx,
switch (ctx->device.lock()->vendor_id) {
case VK_VENDOR_ID_AMD:
return ggml_vk_guess_matmul_pipeline_amd(ctx, bit16_x, bit16_y, m, n, aligned);
case VK_VENDOR_ID_APPLE:
return ggml_vk_guess_matmul_pipeline_apple(ctx, bit16_x, bit16_y, aligned);
case VK_VENDOR_ID_INTEL:
return ggml_vk_guess_matmul_pipeline_intel(ctx, bit16_x, bit16_y, aligned);
}