Gitiles
Code Review
Sign In
gerrit-public.fairphone.software
/
toolchain
/
llvm-project
/
0cab80c9b323ff53f4c775ca7e76b68bf7f0f2c2
/
llvm
/
lib
/
Target
/
AMDGPU
/
AMDGPUTargetMachine.cpp
84db5d9
AMDGPU/SI: Fix read2 merging into a super register.
by Matt Arsenault
· 10 years ago
5010ebf
Make TargetTransformInfo keeping a reference to the Module DataLayout
by Mehdi Amini
· 10 years ago
db7781c
AMDGPU: Run SIInsertWaits as pre-emit pass
by Matt Arsenault
· 10 years ago
45bb48e
R600 -> AMDGPU rename
by Tom Stellard
· 10 years ago
[Renamed (99%) from llvm/lib/Target/R600/AMDGPUTargetMachine.cpp]
3e5de88
Replace string GNU Triples with llvm::Triple in TargetMachine. NFC.
by Daniel Sanders
· 10 years ago
ed64d62
Replace string GNU Triples with llvm::Triple in computeDataLayout(). NFC.
by Daniel Sanders
· 10 years ago
a73f1fd
Replace string GNU Triples with llvm::Triple in MCSubtargetInfo and create*MCSubtargetInfo(). NFC.
by Daniel Sanders
· 10 years ago
28d13a4
R600/SI: add pass to mark CF live ranges as non-spillable
by Tom Stellard
· 10 years ago
8024d03
Grab a subtarget off of an AMDGPUTargetMachine rather than a
by Eric Christopher
· 11 years ago
93e1ea1
Move the DataLayout to the generic TargetMachine, making it mandatory.
by Mehdi Amini
· 11 years ago
30d69c2
[PM] Remove the old 'PassManager.h' header file at the top level of
by Chandler Carruth
· 11 years ago
de5b7b1
R600: Split AMDGPUPassConfig into R600PassConfig and GCNPassConfig
by Tom Stellard
· 11 years ago
c65b360
R600: Create an R600TargetMachine for pre-gcn GPUs
by Tom Stellard
· 11 years ago
8b04c0d
[multiversion] Switch all of the targets over to use the
by Chandler Carruth
· 11 years ago
93dcdc4
[PM] Switch the TargetMachine interface from accepting a pass manager
by Chandler Carruth
· 11 years ago
705b185
[PM] Change the core design of the TTI analysis to use a polymorphic
by Chandler Carruth
· 11 years ago
40ce8af
R600: Move DataLayout to AMDGPUTargetMachine
by Tom Stellard
· 11 years ago
42fb60e
R600/SI: Spill VGPRs to scratch space for compute shaders
by Tom Stellard
· 11 years ago
49f8bfd
R600/SI: Add a stub GCNTargetMachine
by Tom Stellard
· 11 years ago
7e37a5f
[CodeGen] Add print and verify pass after each MachineFunctionPass by default
by Matthias Braun
· 11 years ago
01c7361
This reverts commit r224043 and r224042.
by Rafael Espindola
· 11 years ago
a7c82a9
[CodeGen] Add print and verify pass after each MachineFunctionPass by default
by Matthias Braun
· 11 years ago
05cd445
R600/SI: Move SIInsertWaits into AMDGPUPassConfig::addPreSched2()
by Tom Stellard
· 11 years ago
92105e8
R600/SI: Don't run SI passes on R600 subtargets
by Tom Stellard
· 11 years ago
691ae3d
R600/SI: Fix running SILowerI1Copies a second time
by Matt Arsenault
· 11 years ago
6596ba7
R600/SI: Add SIFoldOperands pass
by Tom Stellard
· 11 years ago
162c101
R600/SI: Move SIFixSGPRCopies to inst selector passes
by Matt Arsenault
· 11 years ago
a271932
This patch changes the ownership of TLOF from TargetLoweringBase to TargetMachine so that different subtargets could share the TLOF effectively
by Aditya Nandakumar
· 11 years ago
5cbb53c
Reapply: R600: Make sure to inline all internal functions
by Tom Stellard
· 11 years ago
9abe268
Revert "R600: Make sure to inline all internal functions"
by Reid Kleckner
· 11 years ago
aa73831
R600: Make sure to inline all internal functions
by Tom Stellard
· 11 years ago
4103328
R600/SI: Add load / store machine optimizer pass.
by Matt Arsenault
· 11 years ago
60024a0
R600/SI: Fix the FixSGPRLiveRanges pass
by Tom Stellard
· 11 years ago
8c90fd7
Add override to overriden virtual methods, remove virtual keywords.
by Benjamin Kramer
· 11 years ago
34aaf97
Move the R600 intrinsic support back to the target machine - there's
by Eric Christopher
· 11 years ago
ac4b69e
Move R600 subtarget dependent variables onto the subtarget.
by Eric Christopher
· 11 years ago
1aaad69
R600/SI: Add instruction shrinking pass
by Tom Stellard
· 11 years ago
b02094e
R600/SI: Use scratch memory for large private arrays
by Tom Stellard
· 11 years ago
d9a23ab
R600: Add option to disable promote alloca
by Matt Arsenault
· 11 years ago
b2de94e
R600/SI: Adjsut SGPR live ranges before register allocation
by Tom Stellard
· 11 years ago
880a80a
R600: Use LDS and vectors for private memory
by Tom Stellard
· 11 years ago
bc5b537
R600: Remove AMDIL instruction and register definitions
by Tom Stellard
· 11 years ago
2e59a45
R600: Move AMDGPUInstrInfo from AMDGPUTargetMachine into AMDGPUSubtarget
by Tom Stellard
· 11 years ago
46b51b7
R600: Add definition for flat address space ID.
by Matt Arsenault
· 11 years ago
1bd8072
R600/SI: Use VALU instructions for copying i1 values
by Tom Stellard
· 12 years ago
5656db4
[C++11] Add 'override' keywords and remove 'virtual'. Additionally add 'final' and leave 'virtual' on some methods that are marked virtual without overriding anything and have no obvious overrides themselves. R600 edition
by Craig Topper
· 12 years ago
062a2ba
[C++] Use 'nullptr'. Target edition.
by Craig Topper
· 12 years ago
422b93d
Use unique_ptr to manage objects owned by the ScheduleDAGMI.
by David Blaikie
· 12 years ago
1583409
R600/SI: Handle MUBUF instructions in SIInstrInfo::moveToVALU()
by Tom Stellard
· 12 years ago
d0ce2bd
R600: Make check clearer.
by Matt Arsenault
· 12 years ago
5ad5f15
[cleanup] Move the Dominators.h and Verifier.h headers into the IR
by Chandler Carruth
· 12 years ago
d7f890e
Factor MI-Sched in preparation for post-ra scheduling support.
by Andrew Trick
· 12 years ago
4fa7975
Small simplification, p0 is the same as p.
by Rafael Espindola
· 12 years ago
a98cd6a
R600/SI: Make private pointers be 32-bit.
by Matt Arsenault
· 12 years ago
e89b414
One last cleanup of LLVM's DataLayout strings.
by Rafael Espindola
· 12 years ago
0eb1ebe
Structure R600's computeDataLayout more like every other target.
by Rafael Espindola
· 12 years ago
bccb9d4
The preferred alignment defaults to the abi alignment. Omit if it is the same.
by Rafael Espindola
· 12 years ago
f057093
Don't duplicate the DataLayout defaults for integer, floats and vectors.
by Rafael Espindola
· 12 years ago
8afbb28
On DataLayout, omit the default of p:64:64:64.
by Rafael Espindola
· 12 years ago
ceb0c49
Turn AMDGPUSubtarget::getDataLayout into a static function.
by Rafael Espindola
· 12 years ago
f2ba972
R600: Register AMDGPUCFGStructurizer pass
by Tom Stellard
· 12 years ago
1de5582
R600: Register R600EmitClauseMarkers pass
by Tom Stellard
· 12 years ago
92b0a64
Add a RequireStructuredCFG Field to TargetMachine.
by Vincent Lejeune
· 12 years ago
66df8a2
R600: Enable the IR structurizer by default
by Tom Stellard
· 12 years ago
783893a
R600: Add a SubtargetFeatture for disabling the ifcvt pass.
by Tom Stellard
· 12 years ago
af77543
R600: Fix handling of vector kernel arguments
by Tom Stellard
· 12 years ago
26a3b67
R600: Simplify handling of private address space
by Tom Stellard
· 12 years ago
4ee6dd6
R600/SI: Add SinkingPass before ISel
by Vincent Lejeune
· 12 years ago
ed0ceec
R600: Use StructurizeCFGPass for non SI targets
by Tom Stellard
· 12 years ago
a4da6fb
R600: add a pass that merges clauses.
by Vincent Lejeune
· 12 years ago
978674b
Allow subtarget selection of the default MachineScheduler and document the interface.
by Andrew Trick
· 12 years ago
9fa1791
R600/SI: Convert v16i8 resource descriptors to i128
by Tom Stellard
· 12 years ago
2f7cdda
R600/SI: Use VSrc_* register classes as the default classes for types
by Tom Stellard
· 12 years ago
aa664d9
Factor FlattenCFG out from SimplifyCFG
by Tom Stellard
· 12 years ago
8b1e021
SimplifyCFG: Use parallel-and and parallel-or mode to consolidate branch conditions
by Tom Stellard
· 12 years ago
960a622
R600: Simplify AMDILCFGStructurize by removing templates and assuming single exit
by Vincent Lejeune
· 12 years ago
ce49974
R600: Do not predicated basic block with multiple alu clause
by Vincent Lejeune
· 12 years ago
d46fce1
Move StructurizeCFG out of R600 to generic Transforms.
by Matt Arsenault
· 12 years ago
a6c6e1b
R600: Rework subtarget info and remove AMDILDevice classes
by Tom Stellard
· 12 years ago
dec1875
R600: Add a pass that merge Vector Register
by Vincent Lejeune
· 12 years ago
beef23f
Revert "R600: Add a pass that merge Vector Register"
by Rafael Espindola
· 12 years ago
a45aafa
R600: Add a pass that merge Vector Register
by Vincent Lejeune
· 12 years ago
39aca62
Fix a leak on the r600 backend.
by Rafael Espindola
· 12 years ago
d3eed66
R600: Improve texture handling
by Vincent Lejeune
· 12 years ago
227144c
Remove the MachineMove class.
by Rafael Espindola
· 12 years ago
2b971eb
R600: Remove AMDILPeeopholeOptimizer and replace optimizations with tablegen patterns
by Tom Stellard
· 12 years ago
147700b
R600: Packetize instructions
by Vincent Lejeune
· 13 years ago
bfaa63a6
R600: Add support for native control flow
by Vincent Lejeune
· 13 years ago
f43bc57
R600: Emit CF_ALU and use true kcache register.
by Vincent Lejeune
· 13 years ago
99ee0f4
R600/SI: rework input interpolation v2
by Christian Konig
· 13 years ago
68b6b6d
R600: initial scheduler code
by Vincent Lejeune
· 13 years ago
0b72f10
R600: Remove LowerConstCopyPass and lower CONST_COPY right after ISel.
by Vincent Lejeune
· 13 years ago
c756cb99
R600/SI: cleanup literal handling v3
by Christian Konig
· 13 years ago
f3b2a1e
R600: Support for indirect addressing v4
by Tom Stellard
· 13 years ago
df063e6
R600: Fold remaining CONST_COPY after expand pseudo inst
by Tom Stellard
· 13 years ago
365366f
R600: rework handling of the constants
by Tom Stellard
· 13 years ago
c4cabef
R600: Proper insert S_WAITCNT instructions
by Tom Stellard
· 13 years ago
f879435
R600: New control flow for SI v2
by Tom Stellard
· 13 years ago
75aadc2
Add R600 backend
by Tom Stellard
· 13 years ago