root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 10:54:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

metal : fix ggml_metal_supports_op for BF16 (#8021)

Browse Source 
Currently the Metal backend does not support BF16. `ggml_metal_supports_op` was returning true in these cases, leading to a crash with models converted with `--leave-output-tensor`. This commit checks if the first few sources types are BF16 and returns false if that's the case.
This commit is contained in:
Michael de Gans 2024-06-19 22:32:01 -07:00 committed by GitHub
parent ba58993152
commit 2075a66a96
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ggml-metal.m
View File

@ -735,6 +735,12 @@ static id<MTLBuffer> ggml_metal_get_buffer(struct ggml_tensor * t, size_t * offs
} }
static bool ggml_metal_supports_op(const struct ggml_metal_context * ctx, const struct ggml_tensor * op) { static bool ggml_metal_supports_op(const struct ggml_metal_context * ctx, const struct ggml_tensor * op) {
for (size_t i = 0, n = 3; i < n; ++i) {
if (op->src[i] != NULL && op->src[i]->type == GGML_TYPE_BF16) {
return false;
}
}
switch (op->op) { switch (op->op) {
case GGML_OP_UNARY: case GGML_OP_UNARY:
switch (ggml_get_unary_op(op)) { switch (ggml_get_unary_op(op)) {