OSDN Git Service
(root)
/
android-x86
/
external-llvm.git
/ history
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
first ⋅ prev ⋅ next
[IR] Make getParamAttributes take argument numbers, not ArgNo+1
[android-x86/external-llvm.git]
/
lib
/
Target
/
AMDGPU
/
AMDGPUTargetTransformInfo.cpp
2017-04-13
Reid Kleckner
[IR] Make getParamAttributes take argument numbers...
blob
|
commitdiff
2017-04-07
Stanislav Mekhanoshin
[AMDGPU] Unroll more to eliminate phis and conditions
blob
|
commitdiff
|
diff to current
2017-03-28
Stanislav Mekhanoshin
[AMDGPU] Boost unroll threshold for loops reading local...
blob
|
commitdiff
|
diff to current
2017-03-27
Yaxun Liu
[AMDGPU] Get address space mapping by target triple...
blob
|
commitdiff
|
diff to current
2017-03-09
Changpeng Fang
AMDGPU/SI: Disable unrolling in the loop vectorizer...
blob
|
commitdiff
|
diff to current
2017-02-23
Matt Arsenault
LoadStoreVectorizer: Split even sized illegal chains...
blob
|
commitdiff
|
diff to current
2017-02-16
Matt Arsenault
AMDGPU: Remove SI_fs_constant and SI_fs_interp intrinsics
blob
|
commitdiff
|
diff to current
2017-02-03
Stanislav Mekhanoshin
[AMDGPU] Bump -amdgpu-unroll-threshold-private to 2000
blob
|
commitdiff
|
diff to current
2017-02-03
Matt Arsenault
AMDGPU: Don't unroll for private with dynamic allocas
blob
|
commitdiff
|
diff to current
2017-02-03
Stanislav Mekhanoshin
[AMDGPU] Unroll preferences improvements
blob
|
commitdiff
|
diff to current
2017-01-30
Matt Arsenault
AMDGPU: Fix atomic_inc/atomic_dec + ds_swizzle not...
blob
|
commitdiff
|
diff to current
2017-01-11
Mohammed Agabaria
[X86] updating TTI costs for arithmetic instructions...
blob
|
commitdiff
|
diff to current
2016-12-12
Nicolai Haehnle
AMDGPU: llvm.amdgcn.interp.mov is a source of divergence
blob
|
commitdiff
|
diff to current
2016-10-03
Volkan Keles
Add new target hooks for LoadStoreVectorizer
blob
|
commitdiff
|
diff to current
2016-07-01
Matt Arsenault
AMDGPU: Implement getLoadStoreVecRegBitWidth
blob
|
commitdiff
|
diff to current
2016-06-17
Matt Arsenault
AMDGPU: Remove llvm.SI.tid intrinsic
blob
|
commitdiff
|
diff to current
2016-05-02
Nicolai Haehnle
AMDGPU: llvm.SI.fs.constant is a source of divergence
blob
|
commitdiff
|
diff to current
2016-04-22
Nicolai Haehnle
AMDGPU/SI: add llvm.amdgcn.ps.live intrinsic
blob
|
commitdiff
|
diff to current
2016-04-06
Nicolai Haehnle
AMDGPU: Add a shader calling convention
blob
|
commitdiff
|
diff to current
2016-03-25
Matt Arsenault
AMDGPU: Cost model for basic integer operations
blob
|
commitdiff
|
diff to current
2016-03-25
Matt Arsenault
AMDGPU: Partially implement getArithmeticInstrCost...
blob
|
commitdiff
|
diff to current
2016-03-25
Matt Arsenault
AMDGPU: TTI: Make insertelement free.
blob
|
commitdiff
|
diff to current
2016-03-18
Nicolai Haehnle
AMDGPU/SI: Add llvm.amdgcn.buffer.atomic.* intrinsics
blob
|
commitdiff
|
diff to current
2016-03-17
Nicolai Haehnle
AMDGPU: mark atomic instructions as sources of divergence
blob
|
commitdiff
|
diff to current
2016-03-14
Nicolai Haehnle
AMDGPU: mark llvm.amdgcn.image.atomic.* as a source...
blob
|
commitdiff
|
diff to current
2016-02-12
Tom Stellard
AMDGPU/SI: Detect uniform branches and emit s_cbranch...
blob
|
commitdiff
|
diff to current
2016-02-11
Matt Arsenault
AMDGPU: Fix not handling new workitem intrinsics in...
blob
|
commitdiff
|
diff to current
2015-12-24
Matt Arsenault
AMDGPU: Fix getRegisterBitWidth for vectors
blob
|
commitdiff
|
diff to current
2015-12-19
Tom Stellard
AMDGPU/SI: Fix implemenation of isSourceOfDivergence...
blob
|
commitdiff
|
diff to current
2015-12-16
Matt Arsenault
AMDGPU: Override getCFInstrCost
blob
|
commitdiff
|
diff to current
2015-12-15
Tom Stellard
AMDGPU/SI: Implement AMDGPUTargetTransformInfo::isSourc...
blob
|
commitdiff
|
diff to current
2015-12-01
Matt Arsenault
AMDGPU: Report extractelement as free in cost model
blob
|
commitdiff
|
diff to current
2015-06-13
Tom Stellard
R600 -> AMDGPU rename
blob
|
commitdiff
|
diff to current