Skip to content

CANN: Mask unsupported TRANSPOSE_1D operator #15733

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 3, 2025
Merged

CANN: Mask unsupported TRANSPOSE_1D operator #15733

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 1 addition & 3 deletions ggml/src/ggml-cann/aclnn_ops.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2803,16 +2803,14 @@ void ggml_cann_conv_transpose_1d(ggml_backend_cann_context& ctx, ggml_tensor* ds
aclIntArray *padding = aclCreateIntArray(paddingVal, 1);
int64_t dilationVal[] = {1};
aclIntArray *dilation = aclCreateIntArray(dilationVal, 1);
bool transposed = true;
int64_t groups = 1;
int8_t cubeMathType = 0;

#ifdef ASCEND_310P
cubeMathType = 1;
#endif

GGML_CANN_CALL_ACLNN_OP(ctx, Convolution, acl_input, acl_weight, nullptr, stride,
padding, dilation, transposed, padding, groups, acl_dst, cubeMathType);
padding, dilation, true, padding, 1, acl_dst, cubeMathType);

ggml_cann_release_resources(ctx, acl_weight, acl_dst, stride, padding, dilation);
}
Expand Down
4 changes: 3 additions & 1 deletion ggml/src/ggml-cann/ggml-cann.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2479,12 +2479,14 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev,
case GGML_OP_ARGMAX:
case GGML_OP_COS:
case GGML_OP_SIN:
case GGML_OP_CONV_TRANSPOSE_1D:
case GGML_OP_LOG:
case GGML_OP_MEAN:
case GGML_OP_PAD_REFLECT_1D:
case GGML_OP_COUNT_EQUAL:
return true;
case GGML_OP_CONV_TRANSPOSE_1D:
// TODO: ((weightL - 1) * dilationW - padLeft)=1336 should not be larger than 255.
return (op->src[0]->ne[0] - 1) <= 255;
case GGML_OP_SCALE:
float bias;
memcpy(&bias, (const float *)(op->op_params) + 1, sizeof(float));
Expand Down
Loading