vulkan: use ternary expressions to avoid switch lowering for CM2 quant block size (#30164)
* vulkan: specialize quant block size for CM2 matmul * vulkan: use ternary expressions to avoid switch lowering for CM2 quant block size * vulkan: clarify CM2 quant block size comment
This commit is contained in:
1 parent
a05e58c981
commit
08ffba21e5
1 file changed
+6
-11
@@ -49,17 +49,12 @@ layout (constant_id = 6) const uint subgroup_size = 32;
|
||||
#ifdef MULMAT_QUANT
|
||||
|
||||
uint mm_quant_k() {
|
||||
switch (MmTypeA) {
|
||||
case GGML_TYPE_Q4_0: case GGML_TYPE_Q4_1: case GGML_TYPE_Q5_0: case GGML_TYPE_Q5_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
return 32u;
|
||||
case GGML_TYPE_Q1_0:
|
||||
return 128u;
|
||||
case GGML_TYPE_Q2_0:
|
||||
return 64u;
|
||||
default:
|
||||
return 256u;
|
||||
}
|
||||
// Use ternary expressions instead of a switch to keep qk as a specialization-constant expression.
|
||||
return (MmTypeA == GGML_TYPE_Q1_0) ? 128u :
|
||||
(MmTypeA == GGML_TYPE_Q2_0) ? 64u :
|
||||
((MmTypeA == GGML_TYPE_Q4_0 || MmTypeA == GGML_TYPE_Q4_1 ||
|
||||
MmTypeA == GGML_TYPE_Q5_0 || MmTypeA == GGML_TYPE_Q5_1 ||
|
||||
MmTypeA == GGML_TYPE_Q8_0) ? 32u : 256u);
|
||||
}
|
||||
#endif
|
||||
|
||||
|
||||
Reference in new issue
Block a user