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

Fix stats #231

Merged
merged 9 commits into from
Jul 9, 2024
Merged

Fix stats #231

merged 9 commits into from
Jul 9, 2024

Conversation

barnes88
Copy link
Contributor

Adds microbenchmark / correlation data for shared memory bank conflicts counter. Relies on accel-sim/gpgpu-sim_distribution#58 and accel-sim/gpu-app-collection#16 part of #226

For the ubench with bank conflicts:
HW: 260593 bank conflicts, Simulator: 253952 (97%)

For the ubench without bank conflicts:
HW: 2484, Simulator 0

It's odd that for some reason nsight compute shows bank conflicts on the example that has none.

@barnes88
Copy link
Contributor Author

barnes88 commented Jul 5, 2023

this PR can't be merged until we add the traces for the new benchmark to common/accel-sim/traces/volta-tesla-v100/latest/. Right now only Tim has write access

@JRPan JRPan linked an issue Feb 20, 2024 that may be closed by this pull request
@JRPan JRPan mentioned this pull request Jul 7, 2024
32 tasks
@JRPan JRPan added this to the v2.0 milestone Jul 7, 2024
@barnes88 barnes88 added this pull request to the merge queue Jul 9, 2024
Merged via the queue into accel-sim:dev with commit 174a78c Jul 9, 2024
3 checks passed
JRPan added a commit to JRPan/accel-sim-framework that referenced this pull request Sep 3, 2024
* fix shared memory and l1d cache bankconflict stat and add shmem benchmark

* this change requires dev branch of app collection

---------

Co-authored-by: JRPan <25518778+JRPan@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Fix broken output Statistics
3 participants