OSchip
/
llvm-project
mirror of https://github.com/THU-DSP-LAB/llvm-project.git
13
0
Fork 6
Code Issues Packages Projects Releases Wiki Activity
433,172 Commits 31 Branches 0 Tags 2.2 GiB
Commit Graph

11 Commits

Author SHA1 Message Date
Simon Pilgrim 4178e33470 [CostModel] Update RUN -passes=* to double quotes to appease update scripts on windows 
DOS really doesn't like `` quotes to be used in command lines

Some prep work as I'm intending to resurrect D79483 soon
 2022-08-10 17:54:06 +01:00
Piotr Sobczak bd675af2a2 [AMDGPU] Make v16i16/v16f16 legal 
There are upcoming intrinsics to use the new types.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D128865
 2022-06-30 23:08:40 +02:00
Arthur Eubanks 15ba588d6d [test] Migrate '-analyze -cost-model' to '-passes=print<cost-model>' 2022-02-09 15:42:16 -08:00
Stanislav Mekhanoshin bb1fe36977 [AMDGPU] Make v8i16/v8f16 legal 
Differential Revision: https://reviews.llvm.org/D117721
 2022-01-24 11:51:08 -08:00
Daniil Fukalov e5c64b45be [CostModel][AMDGPU] Fix intrinsics costs estimations. 
1. Fixed costs inconsistency for llvm.fma.vXf16 instinsiscs.
2. Added tests for llvm.sadd.sat, llvm.ssub.sat, llvm.uadd.sat, llvm.usub.sat
   intrisics since they have special processing in cost model.
3. Minor intrisics' costs tests updat and refinement.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D115385
 2021-12-13 17:17:34 +03:00
Daniil Fukalov cf362ff4ca [NFC][AMDGPU] Improve cost model tests coverage. 2021-09-30 18:13:17 +03:00
Daniil Fukalov 1f73f0c19d [NFC][AMDGPU] Update cost model tests: 
1. Convert to generated tests.
2. Added code-size case in few places.
 2021-09-27 19:26:02 +03:00
Daniil Fukalov 4f28a2eb03 [NFC] Refactor tests to improve readability. 2021-09-24 01:57:30 +03:00
Stanislav Mekhanoshin a8d9d50762 [AMDGPU] gfx90a support 
Differential Revision: https://reviews.llvm.org/D96906
 2021-02-17 16:01:32 -08:00
dfukalov 9068c20965 [AMDGPU][CostModel] Refine cost model for half- and quarter-rate instructions. 
1. Throughput and codesize costs estimations was separated and updated.
2. Updated fdiv cost estimation for different cases.
3. Added scalarization processing for types that are treated as !isSimple() to
improve codesize estimation in getArithmeticInstrCost() and
getArithmeticInstrCost(). The code was borrowed from TCK_RecipThroughput path
of base implementation.

Next step is unify scalarization part in base class that is currently works for
TCK_RecipThroughput path only.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D89973
 2020-10-24 19:53:08 +03:00
Stanislav Mekhanoshin 58578f7056 [AMDGPU] Implemented fma cost analysis 
Differential Revision: https://reviews.llvm.org/D71676
 2019-12-18 23:54:20 -08:00