-
Notifications
You must be signed in to change notification settings - Fork 1.4k
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
[BACKEND] Add support to convert INT8 MMAV2 accumulator layout to dot_operand layout #3595
base: main
Are you sure you want to change the base?
Changes from 9 commits
aa9d90e
e464114
4a3f982
8567a02
06b19ce
ac9a6d4
1dd0dcf
8920609
2007a11
0a56c9a
4470f03
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 |
---|---|---|
|
@@ -186,8 +186,6 @@ module attributes {"triton_gpu.compute-capability" = 90 : i32, "triton_gpu.num-c | |
#shared = #triton_gpu.shared<{vec = 8, perPhase = 1, maxPhase = 8, order = [0, 1], CTAsPerCGA = [1, 1], CTASplitNum = [1, 1], CTAOrder = [0, 1], hasLeadingOffset = true}> | ||
module attributes {"triton_gpu.num-ctas" = 1 : i32, "triton_gpu.num-warps" = 4 : i32} { | ||
// CHECK-LABEL: cvt_mma_to_dot_fp8 | ||
// CHECK: prmt.b32 | ||
// CHECK: prmt.b32 | ||
Comment on lines
-189
to
-190
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. Why are those removed? 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. These two corresponding to the |
||
// CHECK: nvvm.shfl.sync | ||
// CHECK: nvvm.shfl.sync | ||
// CHECK: prmt.b32 | ||
|
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.
that doesn't make sense. That will change behavior of existing tests. If we want to tests i32 dtype it can be set in the config?
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.
Output type for INT8 MMA is always INT32 (https://github.com/openai/triton/blob/main/python/triton/language/semantic.py#L1358), and
out_dtype
is simply ignored, as opposed to FP MMA (https://github.com/openai/triton/blob/main/python/triton/language/semantic.py#L1367)I changed it to
tl.int32
here just for correctness. If you feel necessary, I can add a check thatout_dtype
must betl.int32
fortl.int8
inputs, in this or a separate PR.