ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

vulkan: Add bfloat16 support (#12554)

Browse source 
* vulkan: Add bfloat16 support

This adds bfloat16 matrix multiply support based on VK_KHR_shader_bfloat16.
The extension is required for coopmat multiply support, but matrix-vector
multiply trivially promotes bf16 to fp32 and doesn't require the extension.
The copy/get_rows shaders also don't require the extension.

It's probably possible to fall back to non-coopmat and promote to fp32 when
the extension isn't supported, but this change doesn't do that.

The coopmat support also requires a glslc that supports the extension, which
currently requires a custom build.

* vulkan: Support bf16 tensors without the bf16 extension or coopmat support

Compile a variant of the scalar mul_mm shader that will promote the bf16
values to float, and use that when either the bf16 extension or the coopmat
extensions aren't available.

* vulkan: bfloat16 fixes (really works without bfloat16 support now)

* vulkan: fix spirv-val failure and reenable -O
This commit is contained in:
Jeff Bolz 2025-05-01 13:49:39 -05:00 • committed by GitHub
parent fc727bcdd5
commit 79f26e9e12
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
13 changed files with 368 additions and 67 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vec.comp
View file

@ -6,7 +6,7 @@
layout(local_size_x_id = 0, local_size_y = 1, local_size_z = 1) in;
#if !defined(DATA_A_F32) && !defined(DATA_A_F16)
#if !defined(DATA_A_F32) && !defined(DATA_A_F16) && !defined(DATA_A_BF16)
#define K_PER_ITER 8
#else
#define K_PER_ITER 2