Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

metal : avoid divisions in bin kernel (#20426)

Browse Source 
* metal : avoid modulus in bin kernel when not broadcasting

* metal : fix capture_started flag
This commit is contained in:
Georgi Gerganov
2026-03-12 09:42:40 +02:00
committed by GitHub
parent 4cc6eb158c
commit e4cff0956b
4 changed files with 14 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-metal/ggml-metal.metal
+7 -4
View File
@@ -1111,6 +1111,7 @@ template [[host_name("kernel_unary_f16_f16_4")]] kernel kernel_unary_t kernel_un
constant short FC_bin_op [[function_constant(FC_BIN + 0)]];
constant short FC_bin_f [[function_constant(FC_BIN + 1)]];
constant bool FC_bin_rb [[function_constant(FC_BIN + 2)]];
constant bool FC_bin_cb [[function_constant(FC_BIN + 3)]];
template <typename T0, typename T1, typename T>
kernel void kernel_bin_fuse_impl(
@@ -1124,11 +1125,12 @@ kernel void kernel_bin_fuse_impl(
#define FC_OP FC_bin_op
#define FC_F FC_bin_f
#define FC_RB FC_bin_rb
#define FC_CB FC_bin_cb
if (FC_RB) {
// row broadcast
const uint i0 = tgpig.x;
const uint i1 = i0%args.ne10;
const uint i0 = tgpig.y*args.ne00 + tgpig.x;
const uint i1 = FC_CB ? tgpig.x%args.ne10 : tgpig.x;
device const T0 * src0_row = (device const T0 *) (src0);
device T * dst_row = (device T *) (dst);
@@ -1200,7 +1202,7 @@ kernel void kernel_bin_fuse_impl(
device const T1 * src1_ptr = (device const T1 *) (src1 + args.o1[0] + i13*args.nb13 + i12*args.nb12 + i11*args.nb11);
for (int i0 = tpitg.x; i0 < args.ne0; i0 += ntg.x) {
const int i10 = i0%args.ne10;
const int i10 = FC_CB ? i0%args.ne10 : i0;
if (FC_OP == 0) {
dst_ptr[i0] = src0_ptr[i0] + src1_ptr[i10];
@@ -1225,7 +1227,7 @@ kernel void kernel_bin_fuse_impl(
}
for (int i0 = tpitg.x; i0 < args.ne0; i0 += ntg.x) {
const int i10 = i0%args.ne10;
const int i10 = FC_CB ? i0%args.ne10 : i0;
T res = src0_ptr[i0];
@@ -1261,6 +1263,7 @@ kernel void kernel_bin_fuse_impl(
#undef FC_OP
#undef FC_F
#undef FC_RB
#undef FC_CB
}
typedef decltype(kernel_bin_fuse_impl<float, float, float>) kernel_bin_fuse_t;