Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings

Commit a53f7f7

Browse filesBrowse files
lslusarczyklslusarczyk
authored
fixed compilation warnings in ggml-sycl (ggml-org#12424)
1 parent 7dfad38 commit a53f7f7
Copy full SHA for a53f7f7

File tree

Expand file treeCollapse file tree

9 files changed

+101
-106
lines changed
Filter options
Expand file treeCollapse file tree

9 files changed

+101
-106
lines changed

‎ggml/src/ggml-sycl/convert.cpp

Copy file name to clipboardExpand all lines: ggml/src/ggml-sycl/convert.cpp
+1-1Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -138,7 +138,7 @@ static void dequantize_row_q4_0_sycl_reorder(const void *vx, dst_t *y, const int
138138
stream->parallel_for(sycl::nd_range<3>(sycl::range<3>(1, 1, n_warp) *
139139
sycl::range<3>(1, 1, WARP_SIZE),
140140
sycl::range<3>(1, 1, WARP_SIZE)),
141-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]]{
141+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]]{
142142
dequantize_block_q4_0_reorder(vx, y, k, item_ct1);
143143
});
144144

‎ggml/src/ggml-sycl/dmmv.cpp

Copy file name to clipboardExpand all lines: ggml/src/ggml-sycl/dmmv.cpp
+12-13Lines changed: 12 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -210,7 +210,7 @@ static void convert_mul_mat_vec_f16_sycl(const void *vx, const dfloat *y,
210210

211211
stream->parallel_for(
212212
sycl::nd_range<3>(block_nums * block_dims, block_dims),
213-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
213+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
214214
dequantize_mul_mat_vec<1, 1, convert_f16>(vx, y, dst, ncols,
215215
nrows, item_ct1);
216216
});
@@ -879,7 +879,7 @@ static void dequantize_mul_mat_vec_q4_0_sycl_reorder(const void *vx, const dfloa
879879

880880
stream->parallel_for(
881881
sycl::nd_range<3>(block_nums * block_dims, block_dims),
882-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
882+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
883883
dequantize_mul_mat_vec_reorder<QK4_0, QR4_0, dequantize_q4_0_reorder>(
884884
vx, y, dst, ncols, nrows, item_ct1);
885885
});
@@ -902,7 +902,7 @@ static void dequantize_mul_mat_vec_q4_0_sycl(const void *vx, const dfloat *y,
902902

903903
stream->parallel_for(
904904
sycl::nd_range<3>(block_nums * block_dims, block_dims),
905-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
905+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
906906
dequantize_mul_mat_vec<QK4_0, QR4_0, dequantize_q4_0>(
907907
vx, y, dst, ncols, nrows, item_ct1);
908908
});
@@ -923,7 +923,7 @@ static void dequantize_mul_mat_vec_q4_1_sycl(const void *vx, const dfloat *y,
923923

924924
stream->parallel_for(
925925
sycl::nd_range<3>(block_nums * block_dims, block_dims),
926-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
926+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
927927
dequantize_mul_mat_vec<QK4_1, QR4_1, dequantize_q4_1>(
928928
vx, y, dst, ncols, nrows, item_ct1);
929929
});
@@ -944,7 +944,7 @@ static void dequantize_mul_mat_vec_q5_0_sycl(const void *vx, const dfloat *y,
944944

945945
stream->parallel_for(
946946
sycl::nd_range<3>(block_nums * block_dims, block_dims),
947-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
947+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
948948
dequantize_mul_mat_vec<QK5_0, QR5_0, dequantize_q5_0>(
949949
vx, y, dst, ncols, nrows, item_ct1);
950950
});
@@ -965,7 +965,7 @@ static void dequantize_mul_mat_vec_q5_1_sycl(const void *vx, const dfloat *y,
965965

966966
stream->parallel_for(
967967
sycl::nd_range<3>(block_nums * block_dims, block_dims),
968-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
968+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
969969
dequantize_mul_mat_vec<QK5_1, QR5_1, dequantize_q5_1>(
970970
vx, y, dst, ncols, nrows, item_ct1);
971971
});
@@ -986,7 +986,7 @@ static void dequantize_mul_mat_vec_q8_0_sycl(const void *vx, const dfloat *y,
986986

987987
stream->parallel_for(
988988
sycl::nd_range<3>(block_nums * block_dims, block_dims),
989-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(WARP_SIZE)]] {
989+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
990990
dequantize_mul_mat_vec<QK8_0, QR8_0, dequantize_q8_0>(
991991
vx, y, dst, ncols, nrows, item_ct1);
992992
});
@@ -1004,7 +1004,7 @@ static void dequantize_mul_mat_vec_q2_K_sycl(const void *vx, const float *y,
10041004
const sycl::range<3> block_dims(1, ny, QK_WARP_SIZE);
10051005
stream->parallel_for(
10061006
sycl::nd_range<3>(block_nums * block_dims, block_dims),
1007-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(QK_WARP_SIZE)]] {
1007+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(QK_WARP_SIZE)]] {
10081008
dequantize_mul_mat_vec_q2_k(vx, y, dst, ncols, nrows, item_ct1);
10091009
});
10101010
}
@@ -1020,7 +1020,7 @@ static void dequantize_mul_mat_vec_q3_K_sycl(const void *vx, const float *y,
10201020
const sycl::range<3> block_dims(1, ny, QK_WARP_SIZE);
10211021
stream->parallel_for(
10221022
sycl::nd_range<3>(block_nums * block_dims, block_dims),
1023-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(QK_WARP_SIZE)]] {
1023+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(QK_WARP_SIZE)]] {
10241024
dequantize_mul_mat_vec_q3_k(vx, y, dst, ncols, nrows, item_ct1);
10251025
});
10261026
}
@@ -1036,7 +1036,7 @@ static void dequantize_mul_mat_vec_q4_K_sycl(const void *vx, const float *y,
10361036
const sycl::range<3> block_dims(1, ny, QK_WARP_SIZE);
10371037
stream->parallel_for(
10381038
sycl::nd_range<3>(block_nums * block_dims, block_dims),
1039-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(QK_WARP_SIZE)]] {
1039+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(QK_WARP_SIZE)]] {
10401040
dequantize_mul_mat_vec_q4_k(vx, y, dst, ncols, nrows, item_ct1);
10411041
});
10421042
}
@@ -1049,7 +1049,7 @@ static void dequantize_mul_mat_vec_q5_K_sycl(const void *vx, const float *y,
10491049
const sycl::range<3> block_dims(1, 1, QK_WARP_SIZE);
10501050
stream->parallel_for(
10511051
sycl::nd_range<3>(sycl::range<3>(1, 1, nrows) * block_dims, block_dims),
1052-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(QK_WARP_SIZE)]] {
1052+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(QK_WARP_SIZE)]] {
10531053
dequantize_mul_mat_vec_q5_k(vx, y, dst, ncols, item_ct1);
10541054
});
10551055
}
@@ -1065,7 +1065,7 @@ static void dequantize_mul_mat_vec_q6_K_sycl(const void *vx, const float *y,
10651065
const sycl::range<3> block_dims(1, ny, QK_WARP_SIZE);
10661066
stream->parallel_for(
10671067
sycl::nd_range<3>(block_nums * block_dims, block_dims),
1068-
[=](sycl::nd_item<3> item_ct1) [[intel::reqd_sub_group_size(QK_WARP_SIZE)]] {
1068+
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(QK_WARP_SIZE)]] {
10691069
dequantize_mul_mat_vec_q6_k(vx, y, dst, ncols, nrows, item_ct1);
10701070
});
10711071
}
@@ -1143,7 +1143,6 @@ void ggml_sycl_op_dequantize_mul_mat_vec(
11431143
default:
11441144
printf("ggml_sycl_op_dequantize_mul_mat_vec unsupported GGML_TYPE %d\n", src0->type);
11451145
GGML_ABORT("fatal error");
1146-
break;
11471146
}
11481147

11491148
GGML_UNUSED(src1);

0 commit comments

Comments
0 (0)
Morty Proxy This is a proxified and sanitized view of the page, visit original site.