Logo
Explore Help
Sign In
shylie/llvm-project
1
0
Fork 0
You've already forked llvm-project
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/mlir/test/Dialect/Math
History
Hyunsung Lee 3a3377579f
[mlir][math]Update convertPowfOp ExpandPatterns.cpp (#124402)
The current implementation of `convertPowfOp` requires a calculation of
`a * a` but, max\<fp16\> ~= 65,504, and if `a` is about 16, it will
overflow so get INF in fp8 or fp16 easily.


Remove support when `a < 0`. Overhead of handling negative value of `a`
is large and easy to overflow;

- related issue in iree:
https://github.com/iree-org/iree/issues/15936
2025-01-28 22:56:43 -05:00
..
algebraic-simplification.mlir
[mlir][Transforms] GreedyPatternRewriteDriver: Do not CSE constants during iterations (#75897)
2024-01-05 09:22:18 +01:00
canonicalize_ipowi.mlir
[mlir][vector] add result type to vector.extract assembly format (#66499)
2023-09-28 11:11:16 +01:00
canonicalize.mlir
[mlir][ub] Add poison support to CommonFolders.h
2023-09-07 12:30:29 +02:00
depends-on-arith.mlir
…
expand-math.mlir
[mlir][math]Update convertPowfOp ExpandPatterns.cpp (#124402)
2025-01-28 22:56:43 -05:00
extend-to-supported-types-f16.mlir
[mlir] Refactor LegalizeToF32 to specify extra supported float types and target type as arguments (#108815)
2024-09-27 10:02:16 -05:00
extend-to-supported-types.mlir
[mlir] Refactor LegalizeToF32 to specify extra supported float types and target type as arguments (#108815)
2024-09-27 10:02:16 -05:00
ops.mlir
…
polynomial-approximation.mlir
[mlir][Math] Fix 0-rank support for PolynomialApproximation (#114826)
2024-11-05 09:40:52 +00:00
uplift-to-fma.mlir
…
Powered by Gitea Version: 1.23.1 Page: 1026ms Template: 3ms
English
Bahasa Indonesia Deutsch English Español Français Gaeilge Italiano Latviešu Magyar nyelv Nederlands Polski Português de Portugal Português do Brasil Suomi Svenska Türkçe Čeština Ελληνικά Български Русский Українська فارسی മലയാളം 日本語 简体中文 繁體中文(台灣) 繁體中文(香港) 한국어
Licenses API