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

use macro

Browse source
This commit is contained in:
Abhilash Majumder 2024-02-12 13:08:54 +05:30 • committed by GitHub
parent 60f9508103
commit 28444caf7f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 16 additions and 60 deletions
Download patch file Download diff file Expand all files Collapse all files

76
ggml-sycl.cpp
View file

@ -12423,9 +12423,7 @@ inline void ggml_sycl_op_alibi(const ggml_tensor *src0, const ggml_tensor *src1,
GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32);
const int64_t ne00 = src0->ne[0];
const int64_t ne01 = src0->ne[1];
const int64_t ne02 = src0->ne[2];
GGML_TENSOR_LOCALS_3(int64_t, ne0, src0, ne);
const int64_t nrows = ggml_nrows(src0);
//const int n_past = ((int32_t *) dst->op_params)[0];
@ -12755,15 +12753,9 @@ static void ggml_sycl_op_mul_mat(const ggml_tensor *src0,
ggml_sycl_op_mul_mat_t op,
const bool convert_src1_to_q8_1) try {
const int64_t ne00 = src0->ne[0];
const int64_t ne01 = src0->ne[1];
const int64_t ne02 = src0->ne[2];
const int64_t ne03 = src0->ne[3];
GGML_TENSOR_LOCALS(int64_t, ne0, src0, ne);
const int64_t ne10 = src1->ne[0];
const int64_t ne11 = src1->ne[1];
const int64_t ne12 = src1->ne[2];
const int64_t ne13 = src1->ne[3];
GGML_TENSOR_LOCALS(int64_t, ne1, src1, ne);
const int64_t nrows1 = ggml_nrows(src1);
GGML_ASSERT(ne03 == ne13);
@ -13334,24 +13326,14 @@ static void ggml_sycl_mul_mat_mat_batched_sycl(const ggml_tensor *src0,
GGML_ASSERT(src0->type == GGML_TYPE_F16);
GGML_ASSERT(src1->type == GGML_TYPE_F32);
const int64_t ne00 = src0->ne[0]; GGML_UNUSED(ne00);
const int64_t ne01 = src0->ne[1];
const int64_t ne02 = src0->ne[2];
const int64_t ne03 = src0->ne[3];
GGML_TENSOR_LOCALS(int64_t, ne0, src0, ne);
const int64_t nb01 = src0->nb[1];
const int64_t nb02 = src0->nb[2]; GGML_UNUSED(nb02);
const int64_t nb03 = src0->nb[3]; GGML_UNUSED(nb03);
GGML_TENSOR_LOCALS(int64_t, nb0, src0, nb);
const int64_t ne10 = src1->ne[0];
const int64_t ne11 = src1->ne[1];
const int64_t ne12 = src1->ne[2];
const int64_t ne13 = src1->ne[3];
const int64_t nb11 = src1->nb[1];
const int64_t nb12 = src1->nb[2]; GGML_UNUSED(nb12);
const int64_t nb13 = src1->nb[3]; GGML_UNUSED(nb13);
GGML_TENSOR_LOCALS(int64_t, ne1, src1, ne);
GGML_TENSOR_LOCALS(int64_t, nb1, src1, nb);
const int64_t ne1 = ggml_nelements(src1);
const int64_t ne = ggml_nelements(dst);
@ -13652,24 +13634,16 @@ static void ggml_sycl_mul_mat_id_sycl(ggml_tensor * dst) {
GGML_ASSERT(src00->backend != GGML_BACKEND_GPU_SPLIT);
GGML_ASSERT(src1->type == GGML_TYPE_F32);
const int64_t ne00 = src00->ne[0]; GGML_UNUSED(ne00);
const int64_t ne01 = src00->ne[1];
const int64_t ne02 = src00->ne[2];
const int64_t ne03 = src00->ne[3];
GGML_TENSOR_LOCALS(int64_t, ne0, src00, ne);
//const int64_t nb01 = src00->nb[1];
const int64_t nb02 = src00->nb[2]; GGML_UNUSED(nb02);
const int64_t nb03 = src00->nb[3]; GGML_UNUSED(nb03);
const int64_t ne10 = src1->ne[0];
const int64_t ne11 = src1->ne[1];
const int64_t ne12 = src1->ne[2];
const int64_t ne13 = src1->ne[3];
GGML_TENSOR_LOCALS(int64_t, nb0, src00, nb);
GGML_TENSOR_LOCALS(int64_t, ne1, src1, ne);
GGML_TENSOR_LOCALS(int64_t, nb1, src1, nb);
//const int64_t nb11 = src1->nb[1];
const int64_t nb12 = src1->nb[2]; GGML_UNUSED(nb12);
const int64_t nb13 = src1->nb[3]; GGML_UNUSED(nb13);
const int64_t ne1 = ggml_nelements(src1);
const int64_t ne = ggml_nelements(dst);
@ -13937,25 +13911,7 @@ static void ggml_sycl_cpy(const ggml_tensor *src0, const ggml_tensor *src1,
GGML_ASSERT(ggml_nbytes(src0) <= INT_MAX);
GGML_ASSERT(ggml_nbytes(src1) <= INT_MAX);
const int64_t ne00 = src0->ne[0];
const int64_t ne01 = src0->ne[1];
const int64_t ne02 = src0->ne[2];
const int64_t nb00 = src0->nb[0];
const int64_t nb01 = src0->nb[1];
const int64_t nb02 = src0->nb[2];
const int64_t nb03 = src0->nb[3];
const int64_t ne10 = src1->ne[0];
const int64_t ne11 = src1->ne[1];
const int64_t ne12 = src1->ne[2];
const int64_t nb10 = src1->nb[0];
const int64_t nb11 = src1->nb[1];
const int64_t nb12 = src1->nb[2];
const int64_t nb13 = src1->nb[3];
GGML_TENSOR_BINARY_OP_LOCALS;
SYCL_CHECK(ggml_sycl_set_device(g_main_device));
dpct::queue_ptr main_stream = g_syclStreams[g_main_device_index][0];