Repository navigation
[tpums] specs: Use the bfloat16 peak for float32 compute rooflines - #165
Merged
Merged
Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
copybara-service
Bot
force-pushed
the
test_1022617738
branch
3 times, most recently
from
November 18, 2026 01:45
f93befe to
0c15d31
Compare
float32 GEMM reported roofline efficiency above 100% because the float32 peak was set to half the bfloat16 peak. MXU multiplies take bfloat16 inputs (https://docs.cloud.google.com/tpu/docs/system-architecture-tpu-vm) and the benchmarks run at JAX's default matmul precision, so a float32 matmul is a single bfloat16 pass. Set the float32 peak to the bfloat16 peak for tpu7x and v6e. PiperOrigin-RevId: 1024209837
copybara-service
Bot
force-pushed
the
test_1022617738
branch
from
November 18, 2026 01:59
0c15d31 to
d961958
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
[tpums] specs: Use the bfloat16 peak for float32 compute rooflines
float32 GEMM reported roofline efficiency above 100% because the float32 peak was set to half the bfloat16 peak. MXU multiplies take bfloat16 inputs (https://docs.cloud.google.com/tpu/docs/system-architecture-tpu-vm) and the benchmarks run at JAX's default matmul precision, so a float32 matmul is a single bfloat16 pass. Set the float32 peak to the bfloat16 peak for tpu7x and v6e.