[Pallas GPU] Enable Pallas OpsExtraTest in 64-bit mode
#23798
Merged
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
force-pushed
the
test_676838304
branch
9 times, most recently
from
a5d03e5
to
e8216c4
Compare
This is a follow-up of #23747, which enables Pallas `OpsTest` in 64-bit mode. In order to enable Pallas `OpsExtraTest` in 64-bit mode, some of the code in the tests need to be modified. There are three kinds of modifications: 1. Most of the modifications are just changing `jnp.int32` to `intx` and `jnp.float32` to `floatx`, which uses the same approach as the previous PR #23747. `intx` and `floatx` are conventions used in Pallas tests to refer to 64-bit types in 64-bit mode and their 32-bit counterparts in 32-bit mode. 2. For the test `test_array_reduce`, the original code uses a simple approach to determine `out_dtype` from `dtype`, which no longer works in 64-bit mode. Therefore, I modified the code to deduct `out_dtype` by executing the operation on a single element first. 3. For the test `test_masked_load_store`, the `idx` variable is expected to be an `int32` array, which is calculated based on `pl.program_id()` and `block_size`. In 64-bit mode, the computation will give out an `int64` array instead. Since `pl.program_id()` always returns an `int32` result, I modified the computation to produce `int32` result. I also modified the `pl.program_id()` docstring to document the behaviour that `pl.program_id()` always returns an `int32` result. PiperOrigin-RevId: 677007613
copybara-service
bot
force-pushed
the
test_676838304
branch
from
e8216c4
to
d63afd8
Compare
copybara-service bot
pushed a commit
that referenced
this pull request
Sep 22, 2024
…t mode These tests are failing on GPU in 64-bit mode. This fixes test failures introduced by #23798 PiperOrigin-RevId: 677483206
copybara-service bot
pushed a commit
that referenced
this pull request
Sep 22, 2024
…t mode These tests are failing on GPU in 64-bit mode. This fixes test failures introduced by #23798 PiperOrigin-RevId: 677483206
copybara-service bot
pushed a commit
that referenced
this pull request
Sep 23, 2024
…t mode These tests are failing on GPU in 64-bit mode. This fixes test failures introduced by #23798 PiperOrigin-RevId: 677483206
copybara-service bot
pushed a commit
that referenced
this pull request
Sep 23, 2024
…t mode These tests are failing on GPU in 64-bit mode. This fixes test failures introduced by #23798 PiperOrigin-RevId: 677583606
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.
[Pallas GPU] Enable Pallas
OpsExtraTestin 64-bit modeThis is a follow-up of #23747, which enables Pallas
OpsTestin 64-bit mode.In order to enable Pallas
OpsExtraTestin 64-bit mode, some of the code in the tests need to be modified. There are three kinds of modifications:jnp.int32tointxandjnp.float32tofloatx, which uses the same approach as the previous PR Enable Pallasops_teston GPU in 64-bit mode. #23747.intxandfloatxare conventions used in Pallas tests to refer to 64-bit types in 64-bit mode and their 32-bit counterparts in 32-bit mode.test_array_reduce, the original code uses a simple approach to determineout_dtypefromdtype, which no longer works in 64-bit mode. Therefore, I modified the code to deductout_dtypeby executing the operation on a single element first.test_masked_load_store, theidxvariable is expected to be anint32array, which is calculated based onpl.program_id()andblock_size. In 64-bit mode, the computation will give out anint64array instead. Sincepl.program_id()always returns anint32result, I modified the computation to produceint32result. I also modified thepl.program_id()docstring to document the behaviour thatpl.program_id()always returns anint32result.