Fully unroll the scan in jnp.searchsorted, when method 'scan_unrolled' is specified. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.
#17509
Conversation
This file contains 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
![copybara-service[bot]](https://avatars.githubusercontent.com/in/44061?s=60&v=4){kind=small}
copybara-service
bot
changed the title
jnp.searchsorted. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.scan in the default jnp.searchsorted implementation. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.
copybara-service
bot
force-pushed
the
test_563615727
branch
from
3d508e8
to
b03c1b8
Compare
copybara-service
bot
changed the title
scan in the default jnp.searchsorted implementation. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.scan in the default jnp.searchsorted implementation, when method 'scan_unrolled' is specified. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.
copybara-service
bot
force-pushed
the
test_563615727
branch
from
b03c1b8
to
711fb4e
Compare
copybara-service
bot
changed the title
scan in the default jnp.searchsorted implementation, when method 'scan_unrolled' is specified. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.scan in jnp.searchsorted, when method 'scan_unrolled' is specified. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.
copybara-service
bot
force-pushed
the
test_563615727
branch
7 times, most recently
from
cf2d72b
to
54530bb
Compare
copybara-service
bot
force-pushed
the
test_563615727
branch
from
54530bb
to
ed955ea
Compare
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.
Fully unroll the
scaninjnp.searchsorted, when method 'scan_unrolled' is specified. On GPU, XLA's 'scan' (fori_loop) implementation launches multiple calls to the body_fun GPU kernel, whereas a fully unrolled scan can be fused into a single kernel launch.Since we only require log-many steps, this is often quite practical, and can be a nice speedup. (from 4.5ms down to 1.5ms in my scenario.)