extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-08-24 15:06:26 +02:00
Code Issues Packages Projects Releases Wiki Activity

CANN: GGML_OP_CPY optimization (llama/15070)

Browse Source 
Signed-off-by: noemotiovon <757486878@qq.com>
This commit is contained in:
Chenguang Li
2025-08-12 16:12:13 +08:00
committed by Georgi Gerganov
parent 8e2ddfec31
commit 0effaad964
2 changed files with 35 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files

82
ggml/src/ggml-cann/aclnn_ops.cpp
View File

@@ -753,69 +753,55 @@ static void cann_copy(ggml_backend_cann_context& ctx, aclTensor* acl_src,
void ggml_cann_dup(ggml_backend_cann_context& ctx, ggml_tensor* dst) { void ggml_cann_dup(ggml_backend_cann_context& ctx, ggml_tensor* dst) {
ggml_tensor* src0 = dst->src[0]; ggml_tensor* src0 = dst->src[0];
aclTensor* acl_src = ggml_cann_create_tensor(src0);
aclTensor* acl_dst = ggml_cann_create_tensor(dst);
if (ggml_are_same_shape(src0, dst)) { if (ggml_are_same_shape(src0, dst)) {
aclTensor* acl_src = ggml_cann_create_tensor(src0);
aclTensor* acl_dst = ggml_cann_create_tensor(dst);
if (dst->type == src0->type) { if (dst->type == src0->type) {
cann_copy(ctx, acl_src, acl_dst); cann_copy(ctx, acl_src, acl_dst);
} else { } else {
aclnn_cast(ctx, acl_src, acl_dst, ggml_cann_type_mapping(dst->type)); aclnn_cast(ctx, acl_src, acl_dst, ggml_cann_type_mapping(dst->type));
} }
ggml_cann_release_resources(ctx, acl_src, acl_dst);
} else { } else {
if (ggml_is_contiguous(src0) && ggml_is_contiguous(dst)) { void* src_trans_buffer = src0->data;
if (dst->type == src0->type) { ggml_cann_pool_alloc src_buffer_allocator;
size_t cpy_size = ggml_nbytes(dst); if (!ggml_is_contiguous(src0)) {
ggml_cann_async_memcpy(ctx, dst->data, src0->data, cpy_size, aclTensor* acl_src = ggml_cann_create_tensor(src0);
ACL_MEMCPY_DEVICE_TO_DEVICE); src_buffer_allocator.alloc(ctx.pool(),
return; ggml_nelements(src0) * ggml_type_size(src0->type));
} else { src_trans_buffer = src_buffer_allocator.get();
ggml_cann_pool_alloc src_buffer_allocator(
ctx.pool(),
ggml_nelements(dst) * ggml_type_size(dst->type));
void* src_trans_buffer = src_buffer_allocator.get();
size_t src_trans_nb[GGML_MAX_DIMS];
src_trans_nb[0] = ggml_type_size(dst->type);
for (int i = 1; i < GGML_MAX_DIMS; i++) {
src_trans_nb[i] = src_trans_nb[i - 1] * src0->ne[i - 1];
}
aclTensor* src_trans_tensor = ggml_cann_create_tensor(
src_trans_buffer, ggml_cann_type_mapping(dst->type),
ggml_type_size(dst->type), src0->ne, src_trans_nb,
GGML_MAX_DIMS);
aclnn_cast(ctx, acl_src, src_trans_tensor, ggml_cann_type_mapping(dst->type));
size_t cpy_size = ggml_nbytes(dst);
ggml_cann_async_memcpy(ctx, dst->data, src_trans_buffer, cpy_size,
ACL_MEMCPY_DEVICE_TO_DEVICE);
ggml_cann_release_resources(ctx, src_trans_tensor);
return;
}
} else if (ggml_is_contiguous(dst)) {
ggml_cann_pool_alloc src_buffer_allocator(
ctx.pool(), ggml_nelements(dst) * ggml_type_size(dst->type));
void* src_trans_buffer = src_buffer_allocator.get();
size_t src_trans_nb[GGML_MAX_DIMS]; size_t src_trans_nb[GGML_MAX_DIMS];
src_trans_nb[0] = ggml_type_size(dst->type); src_trans_nb[0] = ggml_type_size(src0->type);
for (int i = 1; i < GGML_MAX_DIMS; i++) { for (int i = 1; i < GGML_MAX_DIMS; i++) {
src_trans_nb[i] = src_trans_nb[i - 1] * src0->ne[i - 1]; src_trans_nb[i] = src_trans_nb[i - 1] * src0->ne[i - 1];
} }
aclTensor* src_trans_tensor = ggml_cann_create_tensor( aclTensor* src_trans_tensor = ggml_cann_create_tensor(
src_trans_buffer, ggml_cann_type_mapping(dst->type), src_trans_buffer, ggml_cann_type_mapping(src0->type),
ggml_type_size(dst->type), src0->ne, src_trans_nb, ggml_type_size(src0->type), src0->ne, src_trans_nb,
GGML_MAX_DIMS); GGML_MAX_DIMS);
cann_copy(ctx, acl_src, src_trans_tensor);
aclnn_cast(ctx, acl_src, src_trans_tensor, ggml_cann_type_mapping(dst->type)); ggml_cann_release_resources(ctx, acl_src, src_trans_tensor);
size_t cpy_size = ggml_nbytes(dst);
ggml_cann_async_memcpy(ctx, dst->data, src_trans_buffer, cpy_size,
ACL_MEMCPY_DEVICE_TO_DEVICE);
ggml_cann_release_resources(ctx, src_trans_tensor);
return;
} else {
GGML_ABORT("Unsupport dst is not contiguous.");
} }
size_t src_reshape_nb[GGML_MAX_DIMS];
src_reshape_nb[0] = ggml_type_size(src0->type);
for (int i = 1; i < GGML_MAX_DIMS; i++) {
src_reshape_nb[i] = src_reshape_nb[i - 1] * dst->ne[i - 1];
}
aclTensor* trans_acl_src = ggml_cann_create_tensor(src_trans_buffer,
ggml_cann_type_mapping(src0->type),ggml_type_size(src0->type),
dst->ne, src_reshape_nb, GGML_MAX_DIMS, ACL_FORMAT_ND);
aclTensor* acl_dst = ggml_cann_create_tensor(dst);
if (dst->type == src0->type) {
cann_copy(ctx, trans_acl_src, acl_dst);
} else {
aclnn_cast(ctx, trans_acl_src, acl_dst, ggml_cann_type_mapping(dst->type));
}
ggml_cann_release_resources(ctx, trans_acl_src, acl_dst);
} }
ggml_cann_release_resources(ctx, acl_src, acl_dst); return;
} }
/** /**

3
ggml/src/ggml-cann/ggml-cann.cpp
View File

@@ -2391,7 +2391,7 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev,
// only support F32 and F16. // only support F32 and F16.
return false; return false;
} }
return ggml_is_contiguous(op); return true;
} break; } break;
case GGML_OP_CONT: { case GGML_OP_CONT: {
// TODO: support GGML_TYPE_BF16 // TODO: support GGML_TYPE_BF16
@@ -2457,7 +2457,6 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev,
return (p0 <= (k0 / 2)) && (p1 <= (k1 / 2)); return (p0 <= (k0 / 2)) && (p1 <= (k1 / 2));
} }
case GGML_OP_DUP: case GGML_OP_DUP:
return ggml_is_contiguous(op);
case GGML_OP_SUM: case GGML_OP_SUM:
case GGML_OP_IM2COL: case GGML_OP_IM2COL:
case GGML_OP_CONCAT: case GGML_OP_CONCAT: