Skip to content
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.

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

[CUDA][Bindless] Address USM normalized type image creation failure and functionality #2056

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

Seanst98
Copy link
Contributor

@Seanst98 Seanst98 commented Sep 5, 2024

The specialised formats, CU_AD_FORMAT_{U|S}NORM_INT{8|16}X{1|2|4}, used to specify a format that is normalized, cannot be applied to images backed by USM memory, causing a failure. This commit addresses that issue by using the standard integer formats and making sure the CU_TRSF_READ_AS_INTEGER flag is set appropriately.

Another issue fixed here is that users with CUDA toolkits pre 11.5 are unable to enable normalized types due to the specialised formats being introduced in CUDA 11.5.

Associated DPC++ PR which tests for this: intel/llvm#15299

@Seanst98 Seanst98 requested a review from a team as a code owner September 5, 2024 10:13
@github-actions github-actions bot added images UR images cuda CUDA adapter specific issues labels Sep 5, 2024
@Seanst98 Seanst98 force-pushed the sean/usm-normalized-fix branch 3 times, most recently from b2ac8dc to 94cb7b0 Compare September 11, 2024 12:16
@Seanst98 Seanst98 added the ready to merge Added to PR's which are ready to merge label Sep 11, 2024
@Seanst98 Seanst98 force-pushed the sean/usm-normalized-fix branch 2 times, most recently from e354592 to 1e40614 Compare September 17, 2024 11:17
@Seanst98 Seanst98 force-pushed the sean/usm-normalized-fix branch 3 times, most recently from 9bedc9f to cbb0735 Compare September 24, 2024 14:53
…nd functionality

The specialised formats, CU_AD_FORMAT_{U|S}NORM_INT{8|16}X{1|2|4},
used to specify a format that is normalized, cannot be applied to
images backed by USM memory, causing a failure. This commit
addresses that issue by using the standard integer formats and
making sure the CU_TRSF_READ_AS_INTEGER flag is set appropriately.

Another issue fixed here is that users with CUDA toolkits pre 11.5
are unable to enable normalized types due to the specialised
formats being introduced in CUDA 11.5.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cuda CUDA adapter specific issues images UR images ready to merge Added to PR's which are ready to merge
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants