Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-28 04:47:04 +01:00
Code Issues Packages Projects Releases Wiki Activity

metal : fix the fix

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2024-03-22 09:58:22 +02:00
parent 3966d68127
commit 072c56fcdb
No known key found for this signature in database
GPG Key ID: BF970631944C16B7
2 changed files with 36 additions and 28 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
ggml-metal.m
View File

@ -1392,9 +1392,13 @@ static enum ggml_status ggml_metal_graph_compute(
(ne11 > ne11_mm_min || (ggml_is_quantized(src0t) && ne12 > 1))) { (ne11 > ne11_mm_min || (ggml_is_quantized(src0t) && ne12 > 1))) {
//printf("matrix: ne00 = %6d, ne01 = %6d, ne02 = %6d, ne11 = %6d, ne12 = %6d\n", ne00, ne01, ne02, ne11, ne12); //printf("matrix: ne00 = %6d, ne01 = %6d, ne02 = %6d, ne11 = %6d, ne12 = %6d\n", ne00, ne01, ne02, ne11, ne12);
// some Metal matrix data types require alignment to 16 bytes // some Metal matrix data types require aligned pointers
// ref: https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf (Table 2.5) // ref: https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf (Table 2.5)
GGML_ASSERT(nb01 % 16 == 0); switch (src0->type) {
case GGML_TYPE_F32: GGML_ASSERT(nb01 % 16 == 0); break;
case GGML_TYPE_F16: GGML_ASSERT(nb01 % 8 == 0); break;
default: break;
}
id<MTLComputePipelineState> pipeline = nil; id<MTLComputePipelineState> pipeline = nil;
@ -1710,9 +1714,13 @@ static enum ggml_status ggml_metal_graph_compute(
ne20 % 32 == 0 && ne20 >= 64 && ne20 % 32 == 0 && ne20 >= 64 &&
ne11 > ne11_mm_min) { ne11 > ne11_mm_min) {
// some Metal matrix data types require alignment to 16 bytes // some Metal matrix data types require aligned pointers
// ref: https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf (Table 2.5) // ref: https://developer.apple.com/metal/Metal-Shading-Language-Specification.pdf (Table 2.5)
GGML_ASSERT(nb01 % 16 == 0); switch (src0->type) {
case GGML_TYPE_F32: GGML_ASSERT(nb01 % 16 == 0); break;
case GGML_TYPE_F16: GGML_ASSERT(nb01 % 8 == 0); break;
default: break;
}
id<MTLComputePipelineState> pipeline = nil; id<MTLComputePipelineState> pipeline = nil;