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

Change MAX_MEMORY_BANDWIDTH device query to uint64 #2653

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

Conversation

omarahmed1111
Copy link
Contributor

@omarahmed1111 omarahmed1111 commented Jan 31, 2025

The max value that could be assigned to MemoryBandwidth variable is (250 * 3200000 * 256) which will overflow the uint32 max capacity. so we should change that to be uint64 to fit the maximum value from this computation.

@omarahmed1111 omarahmed1111 requested a review from a team as a code owner January 31, 2025 17:40
@github-actions github-actions bot added the cuda CUDA adapter specific issues label Jan 31, 2025
Copy link
Contributor

@jchlanda jchlanda left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

source/adapters/cuda/device.cpp Outdated Show resolved Hide resolved
@omarahmed1111 omarahmed1111 force-pushed the change-MAX_MEMORY_BANDWIDTH-device-query-to-int64 branch 3 times, most recently from f3775e3 to 3c22a26 Compare February 3, 2025 13:28
@omarahmed1111 omarahmed1111 requested a review from a team as a code owner February 3, 2025 13:28
@omarahmed1111 omarahmed1111 force-pushed the change-MAX_MEMORY_BANDWIDTH-device-query-to-int64 branch from 3c22a26 to 8922e51 Compare February 3, 2025 13:28
@github-actions github-actions bot added conformance Conformance test suite issues. specification Changes or additions to the specification labels Feb 3, 2025
@omarahmed1111 omarahmed1111 changed the title [CUDA] Change MAX_MEMORY_BANDWIDTH device query to uint64 Change MAX_MEMORY_BANDWIDTH device query to uint64 Feb 3, 2025
@@ -393,7 +393,7 @@ etors:
- name: GPU_HW_THREADS_PER_EU
desc: "[uint32_t][optional-query] return Intel GPU number of threads per EU"
- name: MAX_MEMORY_BANDWIDTH
desc: "[uint32_t][optional-query] return max memory bandwidth in Mb/s"
desc: "[uint64_t][optional-query] return max memory bandwidth in Mb/s"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The spec for this says it should be in Mb/s, which I'd be surprised if we're overflowing (uint32 max mb would be around 4000 terabytes/second). Is the sycl RT expecting mb/s? Should we update it to b/s?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
conformance Conformance test suite issues. cuda CUDA adapter specific issues specification Changes or additions to the specification
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants