-
Notifications
You must be signed in to change notification settings - Fork 447
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
Porting TPC Data compression decoding to GPU #12616
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Approving to start the CI, please do not merge yet!
The code-formatting check fails. Could you please run git-clang-format, to fix the formatting. Please also squash all test commits, and leave only meaningfull commits in. Alternatively, we can squash the full PR when merging.
Error while checking build/O2/fullCI for f0249ef at 2024-01-30 11:34:
Full log here. |
c49f074
to
166b6b8
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks pretty good so far, though I didn't fully check the functionality. I have mostly a couple of cosmetic comments.
f920549
to
ce91114
Compare
46aab55
to
60b937e
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
approved to start the CI
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, let's see what the CI says. I have only 2 cosmetic comments below, which can also be fixed afterwards.
Error while checking build/O2/fullCI for 60b937e at 2024-02-29 08:55:
Full log here. |
@ktf : It seems framework core ctest is failing regularly in the CI today, this seems to be a new problem? |
I am checking. |
Should be fixed by #12781 . |
…sed clusters for decompression
…ang formats Cosmetic fixes
…ssion and change mTmpNativeClusters to MEMORY_SCRATCH
…emory, fix synchronization bug
Error while checking build/O2/fullCI for d3ce442 at 2024-03-01 18:19:
Full log here. |
…changed memory handling in case of CPU code
Error while checking build/O2/fullCI for 7ee82f7 at 2024-03-02 00:54:
Full log here. |
Updates for TPC GPU Decoding
…s for Decompression Kernels and added memory scratch flag to decompression context
Error while checking build/O2/fullCI for 579d623 at 2024-03-03 20:55:
Full log here. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can be merged now if the FullCI becomes green
Full CI again failed randomly. Tried it locally and it worked, and also works as part of #12799, so should be fine, merging. |
Average of 300 clusters over 53175874 not correctly decoded, maybe fast-math problem. Still need to incorporate GPU decoding in the offline chain.