[RFC] Add f8E4M3 and f8E3M4 types support #2486
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
apivovarov
force-pushed
the
rfc_f8E4M3_f8E3M4
branch
3 times, most recently
from
e4fddfa
to
78b7485
Compare
apivovarov
force-pushed
the
rfc_f8E4M3_f8E3M4
branch
from
78b7485
to
389e72c
Compare
|
Signal boosted this to XLA devs, and generally positive feedback. Will give this a week comment period but general consensus is LGTM. Summarizing some early feedback I've heard:
Also want to note that I'll continue socializing / signal boosting this, will report back additional feedback as I hear it / request feedback be left on the PR! |
|
Hello! Yes sorry waiting on feedback from one more person from xla who said they wanted to look into this. Will follow up first thing tomorrow. |
GleasonK
approved these changes
Sep 3, 2024
|
RFC LGTM. I wasn't able to get a hold of that last dev that wanted to chime in, but found a few proxy approvals internally who all agree that given that this is in IEEE and LLVM it should be good to go. Thanks for the contribution and apologies for the delay! |
|
Great news! Thank you, Kevin, for you help and support! |
GleasonK
pushed a commit
that referenced
this pull request
Sep 4, 2024
This PR adds f8E4M3 and f8E3M4 types support.
f8E4M3 and f8E3M4 types follow IEEE 754 convention.
```c
f8E4M3 (IEEE 754)
- Exponent bias: 7
- Maximum stored exponent value: 14 (binary 1110)
- Maximum unbiased exponent value: 14 - 7 = 7
- Minimum stored exponent value: 1 (binary 0001)
- Minimum unbiased exponent value: 1 − 7 = −6
- Precision specifies the total number of bits used for the significand (mantisa),
including implicit leading integer bit = 3 + 1 = 4
- Follows IEEE 754 conventions for representation of special values
- Has Positive and Negative zero
- Has Positive and Negative infinity
- Has NaNs
Additional details:
- Max exp (unbiased): 7
- Min exp (unbiased): -6
- Infinities (+/-): S.1111.000
- Zeros (+/-): S.0000.000
- NaNs: S.1111.{001, 010, 011, 100, 101, 110, 111}
- Max normal number: S.1110.111 = +/-2^(7) x (1 + 0.875) = +/-240
- Min normal number: S.0001.000 = +/-2^(-6)
- Max subnormal number: S.0000.111 = +/-2^(-6) x 0.875 = +/-2^(-9) x 7
- Min subnormal number: S.0000.001 = +/-2^(-6) x 0.125 = +/-2^(-9)
```
```c
f8E3M4 (IEEE 754)
- Exponent bias: 3
- Maximum stored exponent value: 6 (binary 110)
- Maximum unbiased exponent value: 6 - 3 = 3
- Minimum stored exponent value: 1 (binary 001)
- Minimum unbiased exponent value: 1 − 3 = −2
- Precision specifies the total number of bits used for the significand (mantissa),
including implicit leading integer bit = 4 + 1 = 5
- Follows IEEE 754 conventions for representation of special values
- Has Positive and Negative zero
- Has Positive and Negative infinity
- Has NaNs
Additional details:
- Max exp (unbiased): 3
- Min exp (unbiased): -2
- Infinities (+/-): S.111.0000
- Zeros (+/-): S.000.0000
- NaNs: S.111.{0,1}⁴ except S.111.0000
- Max normal number: S.110.1111 = +/-2^(6-3) x (1 + 15/16) = +/-2^3 x 31 x 2^(-4) = +/-15.5
- Min normal number: S.001.0000 = +/-2^(1-3) x (1 + 0) = +/-2^(-2)
- Max subnormal number: S.000.1111 = +/-2^(-2) x 15/16 = +/-2^(-2) x 15 x 2^(-4) = +/-15 x 2^(-6)
- Min subnormal number: S.000.0001 = +/-2^(-2) x 1/16 = +/-2^(-2) x 2^(-4) = +/-2^(-6)
```
Related PRs:
- LLVM [PR-97179](llvm/llvm-project#97179)
[APFloat] Add support for f8E4M3 IEEE 754 type (Merged)
- LLVM [PR-97118](llvm/llvm-project#97118)
[MLIR] Add f8E4M3 IEEE 754 type (Merged)
- LLVM [PR-99698](llvm/llvm-project#99698)
[APFloat] Add support for f8E3M4 IEEE 754 type (Merged)
- LLVM [PR-101230](llvm/llvm-project#101230)
[MLIR] Add f8E3M4 IEEE 754 type (Merged)
- StableHLO [PR-2486](#2486)
[RFC] Add f8E4M3 and f8E3M4 types support
- ml_dtypes [PR-161](jax-ml/ml_dtypes#161) Add
float8_e4m3 (Merged)
- ml_dtypes [PR-171](jax-ml/ml_dtypes#171) Add
float8_e3m4 (Merged)
- XLA [PR-16585](openxla/xla#16585) Add support
for float8_e4m3
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.
Summary
This is a proposal to add
Float8E4M3andFloat8E3M4floating point types to StableHLO.Feedback welcome, see RFC: Float8E4M3 and Float8E3M4 for more details.
References and Links