Skip to content

Conversation

@whitneywhtsang
Copy link
Contributor

@whitneywhtsang whitneywhtsang commented Jan 8, 2025

This PR change the Triton base from 67dc627 to 37817d7 (Jan 2).
Pass rate: 99.88%->99.87%

Please do not squash and merge this PR.

peterbell10 and others added 9 commits January 2, 2025 12:59
It's a bit annoying to type `ninja -C python/build/cmake-.../` to run
build commands. This adds a top-level makefile that just forwards
commands down to the ninja build system so you can just type `make`.

Building and running the tests from a fresh install should now be as
simple as `make dev-install && make test`.

I also add `make golden-samples` to regenerate the golden sample tests,
which should make it fairly painless to accept a failing sample.
…s (#5517)

this simplifies our JIT a bunch and will make it quite a bit easier to
add support for namedtuples in the future
Since tl.debug_barrier is not chained to any other ops, it cannot be
determined where to schedule to keep appropriate order w.r.t memory ops.
This PR fixes case filtering and adds AMD specific layout pairs.
Add address sanitizer pass to LLVM pass pipeline. 

Known limitations for this PR:
- Currently only the AMD backend is supported
- Source code line support not implemented here,
  coming in follow up patch
@whitneywhtsang whitneywhtsang self-assigned this Jan 8, 2025
@whitneywhtsang whitneywhtsang marked this pull request as ready for review January 9, 2025 00:17
@whitneywhtsang whitneywhtsang changed the title Merge OpenAI Triton commit 5524b7c Merge OpenAI Triton commit 37817d7 Jan 9, 2025
@whitneywhtsang whitneywhtsang merged commit a643415 into main Jan 9, 2025
5 of 6 checks passed
@whitneywhtsang whitneywhtsang deleted the whitneywhtsang/merge branch January 9, 2025 00:17
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.

7 participants