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

fix supported ops for kompute backend

Browse source
This commit is contained in:
Jared Van Bortel 2024-01-18 15:32:55 -05:00
parent 7addf2b878
commit 610394fff8
2 changed files with 30 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

41
ggml-kompute.cpp
View file

@ -1316,27 +1316,13 @@ static bool ggml_kompute_supports_op(const struct ggml_tensor * op) {
case GGML_OP_VIEW:
case GGML_OP_TRANSPOSE:
case GGML_OP_PERMUTE:
case GGML_OP_CONCAT:
case GGML_OP_ADD:
case GGML_OP_ACC:
case GGML_OP_MUL:
case GGML_OP_DIV:
case GGML_OP_SCALE:
case GGML_OP_SQR:
case GGML_OP_SUM_ROWS:
case GGML_OP_SOFT_MAX:
case GGML_OP_RMS_NORM:
case GGML_OP_GROUP_NORM:
case GGML_OP_NORM:
case GGML_OP_ALIBI:
case GGML_OP_ROPE:
case GGML_OP_IM2COL:
case GGML_OP_UPSCALE:
case GGML_OP_PAD:
case GGML_OP_ARGSORT:
case GGML_OP_LEAKY_RELU:
case GGML_OP_MUL_MAT:
case GGML_OP_MUL_MAT_ID:
return true;
case GGML_OP_DUP:
case GGML_OP_CPY:
@ -1357,8 +1343,33 @@ static bool ggml_kompute_supports_op(const struct ggml_tensor * op) {
}
return true;
case GGML_OP_DIAG_MASK_INF:
case GGML_OP_GET_ROWS:
return op->ne[3] == 1;
case GGML_OP_GET_ROWS:
switch (op->src[0]->type) {
case GGML_TYPE_F16:
case GGML_TYPE_Q4_0:
case GGML_TYPE_Q4_1:
case GGML_TYPE_Q6_K:
return op->ne[3] == 1;
default:
;
}
return false;
case GGML_OP_MUL_MAT:
if (op->src[1]->type != GGML_TYPE_F32 || ggml_is_transposed(op->src[0]) || ggml_is_transposed(op->src[1]))
return false;
switch (op->src[0]->type) {
case GGML_TYPE_F32:
case GGML_TYPE_F16:
case GGML_TYPE_Q8_0:
case GGML_TYPE_Q4_0:
case GGML_TYPE_Q4_1:
case GGML_TYPE_Q6_K:
return true;
default:
;
}
default:
;
}