-
-
Notifications
You must be signed in to change notification settings - Fork 6.1k
feat(ROCm): expand is_rdna() to cover RDNA2, RDNA3, RDNA3.5, RDNA4 #4428
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
Changes from 4 commits
d82c013
e713559
a27154a
6c978f3
c727139
c0e6320
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -88,10 +88,22 @@ def is_cdna(): | |
|
|
||
| @functools.lru_cache(1) | ||
| def is_rdna(): | ||
| """Detect ROCm-supported RDNA consumer/workstation GPUs (RDNA3, RDNA4).""" | ||
| """Detect RDNA consumer/workstation GPUs (RDNA2, RDNA3, RDNA3.5, RDNA4).""" | ||
| return is_hip() and triton.runtime.driver.active.get_current_target().arch in ( | ||
| # RDNA2 | ||
| "gfx1030", | ||
| "gfx1031", | ||
| "gfx1032", | ||
| # RDNA3 | ||
|
Comment on lines
+94
to
+97
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
The new Useful? React with 👍 / 👎. |
||
| "gfx1100", | ||
| "gfx1101", | ||
| "gfx1102", | ||
| "gfx1103", | ||
| # RDNA3.5 | ||
| "gfx1150", | ||
| "gfx1151", | ||
| "gfx1152", | ||
| # RDNA4 | ||
| "gfx1200", | ||
| "gfx1201", | ||
| ) | ||
|
|
||
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.
The RDNA2 section in
is_rdna()currently ends atgfx1032, so RDNA2 targetsgfx1033–gfx1036still evaluate as non-RDNA and miss RDNA-gated behavior (for example, the Gemma3 ROCm compile workaround that callsis_rdna()). This leaves part of the stated RDNA2 support path disabled for affected GPUs.Useful? React with 👍 / 👎.