generic: sycl: Inner Product FWD #2214
Conversation
AD2605
changed the title
github-actions
bot
added
platform:gpu-nvidia
|
@AD2605 you should probably update https://github.com/oneapi-src/oneDNN/blob/main/src/gpu/generic/sycl/README.md to mention inner product is supported :) |
|
@sgeor255 , updated the documentation |
github-actions
bot
added
the
documentation
AD2605
force-pushed
the
atharva/inner_product_fwd
branch
from
3bd9d00 to
c5d66f1
Compare
AD2605
force-pushed
the
atharva/inner_product_fwd
branch
from
a3b91d6 to
9c63d68
Compare
|
@oneapi-src/onednn-arch @oneapi-src/onednn-doc gentle nudge for reviewing the PR |
ranukund
left a comment
ranukund
left a comment
There was a problem hiding this comment.
One minor comment added. Please incorporate as you see fit.
AD2605
force-pushed
the
atharva/inner_product_fwd
branch
from
5899fe4 to
6989cb2
Compare
| && attr()->has_default_values(sm::scales_runtime | ||
| | sm::zero_points_runtime | sm::post_ops | ||
| | sm::dropout | sm::scales_runtime_data_type | ||
| | sm::zero_points_runtime_data_type) |
There was a problem hiding this comment.
why not defer attribute and datatype checking to init_matmul?
| | sm::zero_points_runtime_data_type) | ||
| && memory_desc_wrapper(src_md()).is_plain() | ||
| && memory_desc_wrapper(dst_md()) | ||
| .is_plain(); // Blocked memory formats are not supported |
There was a problem hiding this comment.
I think is_plain is not enough to flatten the dimensions.
Here you need to check:
- that strides are dense (see is_dense).
- that the order of flattened dimensions for src and weights match.
- there is a missing check for weights_md()
There was a problem hiding this comment.
That's a good point, thanks. I had to create a new PR (#2248) to address the comments as I don't have permissions to push to this one. I think we can close this one.
7 participants
Description
Implements the forward pass of the inner product for the generic sycl backend with support for different post-ops and data types