00aeb119db
R600/SI: Expand and of v2i32/v4i32 for SI
...
Also add lit test for both cases on SI, and v2i32 for evergreen.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 184837
2013-06-25 13:55:23 +00:00
0125f2a6e4
R600/SI: Report unaligned memory accesses as legal for > 32-bit types
...
In reality, some unaligned memory accesses are legal for 32-bit types and
smaller too, but it all depends on the address space. Allowing
unaligned loads/stores for > 32-bit types is mainly to prevent the
legalizer from splitting one load into multiple loads of smaller types.
https://bugs.freedesktop.org/show_bug.cgi?id=65873
llvm-svn: 184822
2013-06-25 02:39:35 +00:00
9810ec613c
R600: Add support for i32 loads from the constant address space on Cayman
...
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 184821
2013-06-25 02:39:30 +00:00
b06f3fc1be
R600/SI: Add support for v4i32 and v4f32 kernel args
...
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 184820
2013-06-25 02:39:25 +00:00
9d2e1500b4
R600: Fix typo in R600Schedule.td
...
This should only make a difference in programs that use a lot of the
vector ALU instructions like BFI_INT and BIT_ALIGN. There is a slight
improvement in the phatk bitcoin mining kernel with this patch on
Evergreen (vector size == 1):
Before:
1173 Instruction Groups / 9520 dwords
After:
1167 Instruction Groups / 9510 dwords
Reviewed-by: Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 184819
2013-06-25 02:39:20 +00:00
52a72c926c
R600: Fix spelling error in comment
...
our -> or
llvm-svn: 184756
2013-06-24 16:57:57 +00:00
96d38760fc
R600/SI: Expand sub for v2i32 and v4i32 for SI
...
Also add a v2i32 test to the existing v4i32 test.
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry<awatry@gmail.com>
llvm-svn: 184482
2013-06-20 21:55:37 +00:00
043795e818
R600/SI: Expand add for v2i32 and v4i32
...
Also add SI tests to existing file and a v2i32 test for both
R600 and SI.
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
llvm-svn: 184481
2013-06-20 21:55:30 +00:00
6ec9e8043c
R600: Expand v2i32 load/store instead of custom lowering
...
The custom lowering causes llc to crash with a segfault.
Ideally, the custom lowering can be fixed, but this allows
programs which load/store v2i32 to work without crashing.
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry<awatry@gmail.com>
llvm-svn: 184480
2013-06-20 21:55:23 +00:00
a3cd350249
Access the TargetLoweringInfo from the TargetMachine object instead of caching it. The TLI may change between functions. No functionality change.
...
llvm-svn: 184360
2013-06-19 21:36:55 +00:00
d46fce1141
Move StructurizeCFG out of R600 to generic Transforms.
...
Register it with PassManager
llvm-svn: 184343
2013-06-19 20:18:24 +00:00
2aabb06175
Use GetUnderlyingObject instead of custom function
...
llvm-svn: 184261
2013-06-18 23:37:58 +00:00
b7b1681157
Remove dead prototype.
...
llvm-svn: 184173
2013-06-18 06:24:14 +00:00
41d4cf26b4
R600: PV stores Reg id, not index
...
llvm-svn: 184117
2013-06-17 20:16:40 +00:00
8bd10421ec
R600: Properly set COUNT_3 bit in TEX clause initiating inst for pre EG gen.
...
Fixes rv7x0 bug in Heaven reported here:
https://bugs.freedesktop.org/show_bug.cgi?id=64257
llvm-svn: 184116
2013-06-17 20:16:26 +00:00
371573448c
R600: Add SI load support for v[24]i32 and store for v2i32
...
Also add a seperate vector lit test file, since r600 doesn't seem to handle
v2i32 load/store yet, but we can test both for SI.
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
llvm-svn: 184021
2013-06-15 00:09:31 +00:00
ecf9d86404
R600: Use correct encoding for Vertex Fetch instructions on Cayman
...
Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 184016
2013-06-14 22:12:30 +00:00
6aa0d5578d
R600: Use EXPORT_RAT_INST_STORE_DWORD for stores on Cayman
...
We were using RAT_INST_STORE_RAW, which seemed to work, but the docs
say this instruction doesn't exist for Cayman, so it's probably safer
to use a documented instruction instead.
Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 184015
2013-06-14 22:12:24 +00:00
d99b7932ae
R600: Factor the instruction encoding out the RAT_WRITE_CACHELESS_eg class
...
Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 184014
2013-06-14 22:12:19 +00:00
3d0823f1cd
R600: Move instruction encoding definitions into a separate .td file
...
Reviewed-by: Vincent Lejeune<vljn at ovi.com>
llvm-svn: 184013
2013-06-14 22:12:09 +00:00
adba083bc2
R600: Don't try to fix reg class when copying IMPLICIT_DEF to a register
...
The test case for this is way too complex to be useful as a lit test,
and I was unable to reduce it.
https://bugs.freedesktop.org/show_bug.cgi?id=65438
llvm-svn: 183937
2013-06-13 20:14:00 +00:00
193960c822
R600: Make helper functions static.
...
llvm-svn: 183744
2013-06-11 13:32:25 +00:00
d1a9d18120
R600: Use a refined heuristic to choose when switching clause
...
This is using a hint from AMD APP OpenCL Programming Guide with
empirically tweaked parameters.
I used Unigine Heaven 3.0 to determine best parameters on my system
(i7 2600/Radeon 6950/Kernel 3.9.4) the benchmark :
it went from 38.8 average fps to 39.6, which is ~3% gain.
(Lightmark 2008.2 gain is much more marginal: from 537 to 539)
There is no lit test provided as the parameter were determined
empirically and it it would be nearly impossiblet to find a test
program that check for optimal behavior.
llvm-svn: 183593
2013-06-07 23:30:34 +00:00
4d143328df
R600: Anti dep better handled in tex clause
...
llvm-svn: 183592
2013-06-07 23:30:26 +00:00
d74583777f
R600: Fix calculation of stack offset in AMDGPUFrameLowering
...
We weren't computing structure size correctly and we were relying on
the original alloca instruction to compute the offset, which isn't
always reliable.
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
llvm-svn: 183568
2013-06-07 20:52:05 +00:00
a6c6e1bfc2
R600: Rework subtarget info and remove AMDILDevice classes
...
This should simplify the subtarget definitions and make it easier to
add new ones.
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
llvm-svn: 183566
2013-06-07 20:37:48 +00:00
37e9adb091
Don't cache the instruction and register info from the TargetMachine, because
...
the internals of TargetMachine could change.
No functionality change intended.
llvm-svn: 183561
2013-06-07 20:28:55 +00:00
3498e4ff1d
R600: Fix the fetch limits for R600 generation GPUs
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
https://bugs.freedesktop.org/show_bug.cgi?id=64257
llvm-svn: 183560
2013-06-07 20:28:55 +00:00
99792774a4
R600: Move Subtarget feature definitions into AMDGPU.td
...
This is the convention used by the other targets.
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
llvm-svn: 183559
2013-06-07 20:28:49 +00:00
b0804ec2ad
R600: Remove unnecessary include
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
llvm-svn: 183558
2013-06-07 20:28:43 +00:00
705d841bb6
R600: Don't compare iterators of different maps.
...
Found be libstdc's debug mode.
llvm-svn: 183549
2013-06-07 19:59:34 +00:00
ebe0be9ca4
Vincent says the element is at most once in the vector, so we don't need a full std::remove.
...
llvm-svn: 183541
2013-06-07 18:18:12 +00:00
a857fe115b
R600: Fix a potential iterator invalidation issue.
...
As a bonus this reduces the loop from O(n^2) to O(n).
llvm-svn: 183532
2013-06-07 16:13:49 +00:00
931bb768fd
R600: Remove an extra break in R600OptimizeVectorRegisters.cpp
...
llvm-svn: 183528
2013-06-07 15:44:53 +00:00
0030362ed9
R600: Rewrite an awkward loop in R600MachineScheduler
...
llvm-svn: 183458
2013-06-06 23:08:32 +00:00
54476a1503
R600: Remove leftover code in R600MachineScheduler.cpp
...
Spotted by Benjamin Kramer.
llvm-svn: 183413
2013-06-06 14:18:29 +00:00
b91216817f
Cast to the correct type. Pointer, not reference.
...
llvm-svn: 183385
2013-06-06 05:39:29 +00:00
4a8f079371
R600OptimizeVectorRegisters.cpp: Tweak a warning. [-Wsometimes-uninitialized]
...
FIXME: Is it false alarm?
llvm-svn: 183371
2013-06-06 02:15:12 +00:00
e5555fc238
R600OptimizeVectorRegisters.cpp: Suppress a warning. [-Wunused-variable]
...
llvm-svn: 183370
2013-06-06 02:15:06 +00:00
372574d447
Trailing linefeed.
...
llvm-svn: 183369
2013-06-06 02:15:00 +00:00
e410576865
Cast to the proper type.
...
llvm-svn: 183365
2013-06-06 01:04:21 +00:00
acec99c948
R600: Replace predicate loop with predicate function
...
llvm-svn: 183351
2013-06-05 23:39:50 +00:00
dec1875207
R600: Add a pass that merge Vector Register
...
Previously commited @183279 but tests were failing, reverted @183286
It was broken because @183336 was missing, now it's there.
llvm-svn: 183343
2013-06-05 21:38:04 +00:00
4b5b849753
R600: Schedule copy from phys register at beginning of block
...
It allows regalloc pass to remove them by trivially assigning associated reg
llvm-svn: 183336
2013-06-05 20:27:35 +00:00
aad5376fb6
R600: Make sure to schedule AR register uses and defs in the same clause
...
Reviewed-by: vljn at ovi.com
llvm-svn: 183294
2013-06-05 03:43:06 +00:00
beef23fe21
Revert "R600: Add a pass that merge Vector Register"
...
This reverts commit r183279. CodeGen/R600/texture-input-merge.ll was failing.
llvm-svn: 183286
2013-06-05 01:48:30 +00:00
a45aafabfe
R600: Add a pass that merge Vector Register
...
llvm-svn: 183279
2013-06-04 23:17:26 +00:00
c689679173
R600: Const/Neg/Abs can be folded to dot4
...
llvm-svn: 183278
2013-06-04 23:17:15 +00:00
276ceb8d5f
R600: Swizzle texture/export instructions
...
llvm-svn: 183229
2013-06-04 15:04:53 +00:00
19978553d4
Silencing an MSVC warning about mixing bool and unsigned int.
...
llvm-svn: 183176
2013-06-04 01:03:03 +00:00
94593ee8c3
R600/SI: Add support for work item and work group intrinsics
...
llvm-svn: 183138
2013-06-03 17:40:18 +00:00
ed882c2f1b
R600/SI: Add a calling convention for compute shaders
...
llvm-svn: 183137
2013-06-03 17:40:11 +00:00
046039e81b
R600/SI: Custom lower i64 sign_extend
...
llvm-svn: 183136
2013-06-03 17:40:03 +00:00
0518ff89ba
R600/SI: Adjust some instructions' out register class after ISel
...
This is necessary to avoid generating VGPR to SGPR copies in some
cases.
llvm-svn: 183135
2013-06-03 17:39:58 +00:00
bad1f59212
R600/SI: Handle REG_SEQUENCE in fitsRegClass()
...
llvm-svn: 183134
2013-06-03 17:39:54 +00:00
b5a97004fb
R600/SI: Handle nodes with glue results correctly SITargetLowering::foldOperands()
...
llvm-svn: 183133
2013-06-03 17:39:50 +00:00
2183b70523
R600/SI: Fixup CopyToReg register class in PostprocessISelDAG()
...
The CopyToReg nodes will sometimes try to copy a value from a VGPR to an
SGPR. This kind of copy is not possible, so we need to detect
VGPR->SGPR copies and do something else. The current strategy is to
replace these copies with VGPR->VGPR copies and hope that all the users
of CopyToReg can accept VGPRs as arguments.
llvm-svn: 183132
2013-06-03 17:39:46 +00:00
07a10a3d3f
R600/SI: Add support for global loads
...
llvm-svn: 183131
2013-06-03 17:39:43 +00:00
556d9aa841
R600/SI: Rework MUBUF store instructions
...
The lowering of stores is now mostly handled in the tablegen files. No
more BUFFER_STORE nodes I generated during legalization.
llvm-svn: 183130
2013-06-03 17:39:37 +00:00
91a942b93e
R600: 3 op instructions have no write bit but the result are store in PV
...
llvm-svn: 183111
2013-06-03 15:56:12 +00:00
eabf83e0a2
R600: CALL_FS consumes a stack size entry
...
llvm-svn: 183108
2013-06-03 15:44:42 +00:00
f83df1f1cb
R600: use capital letter for PV channel
...
llvm-svn: 183107
2013-06-03 15:44:35 +00:00
a09873dda7
R600: Constraints input regs of interp_xy,_zw
...
llvm-svn: 183106
2013-06-03 15:44:16 +00:00
b1a4d9da3b
Make SubRegIndex size mandatory, following r183020.
...
This also makes TableGen able to compute sizes/offsets of synthesized
indices representing tuples.
llvm-svn: 183061
2013-05-31 23:45:26 +00:00
ae8faf2e9a
Temporary fix to get rid of gcc warning.
...
llvm-svn: 182832
2013-05-29 07:32:08 +00:00
ef9de2a739
Track IR ordering of SelectionDAG nodes 2/4.
...
Change SelectionDAG::getXXXNode() interfaces as well as call sites of
these functions to pass in SDLoc instead of DebugLoc.
llvm-svn: 182703
2013-05-25 02:42:55 +00:00
1b086cbcb8
R600: Fix R600ControlFlowFinalizer not considering VTX_READ 128 bit dst reg
...
Patch by: Vincent Lejeune
https://bugs.freedesktop.org/show_bug.cgi?id=64877
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 182600
2013-05-23 18:26:42 +00:00
d78bb468bd
Move passes from namespace llvm into anonymous namespaces. Sort includes while there.
...
llvm-svn: 182594
2013-05-23 17:10:37 +00:00
635e368e33
R600: Hide symbols of implementation details.
...
Also removes an unused function.
llvm-svn: 182587
2013-05-23 15:43:05 +00:00
15f193a1a3
Setting the default value (fixes CRT assertions about uninitialized variable use when doing debug MSVC builds), and fixing coding style.
...
llvm-svn: 182585
2013-05-23 14:55:00 +00:00
00345fa97b
Fix 32 bit build in c++11 mode.
...
The error was:
error: non-constant-expression cannot be narrowed from type 'long long' to 'long' in initializer list [-Wc++11-narrowing]
MI.getOperand(6).getImm() & 0x1F,
llvm-svn: 182584
2013-05-23 13:22:30 +00:00
39aca620db
Fix a leak on the r600 backend.
...
This should bring the valgrind bot back to life.
llvm-svn: 182561
2013-05-23 03:31:47 +00:00
bd6847fbea
clang-format this file.
...
llvm-svn: 182560
2013-05-23 03:28:39 +00:00
e3d83fb8c3
Fix use after free (pr16103).
...
llvm-svn: 182482
2013-05-22 15:31:11 +00:00
ebd8e38849
Check that a function starts with llvm. before using GET_FUNCTION_RECOGNIZER.
...
Fixes a use of uninitialized memory found by asan and valgind.
llvm-svn: 182480
2013-05-22 14:57:42 +00:00
4f328e1c2f
R600ISelLowering.cpp: Avoid "using namespace Intrinsic;" to appease MSC. Specify namespaces explicitly here.
...
MSC is confused about "memcpy" between <cstring> and llvm::Intrinsic::memcpy, when llvm::Intrinsic were exposed.
llvm-svn: 182452
2013-05-22 06:37:31 +00:00
18ca09c1cc
R600: Whitespace and untabify.
...
llvm-svn: 182451
2013-05-22 06:37:25 +00:00
616852848a
Create an FPOW SDNode opcode def in the target independent .td file rather than in a specific backend.
...
llvm-svn: 182450
2013-05-22 06:36:09 +00:00
21ea01d132
Attempt to fix the mingw32 bot.
...
This should hopefully fix
http://lab.llvm.org:8011/builders/clang-x86_64-darwin11-self-mingw32
llvm-svn: 182446
2013-05-22 02:30:47 +00:00
525cf28652
s/u_int32_t/uint32_t/
...
llvm-svn: 182444
2013-05-22 01:36:19 +00:00
f568827654
Fix warning in non-assert build.
...
llvm-svn: 182443
2013-05-22 01:29:38 +00:00
927ca942ce
R600: Fix bug detected by GCC warning.
...
R600TextureIntrinsicsReplacer.cpp:232: warning: the address of ‘ArgsType’ will always evaluate as ‘true’
This doesn't have any effect on the output as a vararg intrinsic behaves the
same way as a non-vararg one.
llvm-svn: 182293
2013-05-20 15:58:43 +00:00
f1ee716446
R600/SI: Use a multiclass for MUBUF_Load_Helper
...
This will simplify the instructions and also the pattern definitions.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182288
2013-05-20 15:02:31 +00:00
b8458f88d6
R600/SI: Add a pattern for S_LOAD_DWORDX2_* instructions
...
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182287
2013-05-20 15:02:28 +00:00
d2eebf001e
R600/SI: Add pattern for rotr
...
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182286
2013-05-20 15:02:24 +00:00
5643c4ac72
R600: Swap the legality of rotl and rotr
...
The hardware supports rotr and not rotl.
llvm-svn: 182285
2013-05-20 15:02:19 +00:00
1cfd7a50bb
R600/SI: Add patterns for 64-bit shift operations
...
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182284
2013-05-20 15:02:12 +00:00
459a79a81c
R600/SI: Use the same names for VOP3 operands and encoding fields
...
This makes it possible to reorder the operands without breaking the
encoding.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182283
2013-05-20 15:02:08 +00:00
b35efba4d9
R600/SI: Make fitsRegClass() operands const
...
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182282
2013-05-20 15:02:01 +00:00
75865923c9
Add LLVMContext argument to getSetCCResultType
...
llvm-svn: 182180
2013-05-18 00:21:46 +00:00
5986ce0e5d
Fix the build in c++11 mode.
...
The errors were:
non-constant-expression cannot be narrowed from type 'int64_t' (aka 'long') to 'uint32_t' (aka 'unsigned int') in initializer list
and
non-constant-expression cannot be narrowed from type 'long' to 'uint32_t' (aka 'unsigned int') in initializer list
llvm-svn: 182168
2013-05-17 22:45:52 +00:00
d3fcb5016c
R600: Lower int_load_input to copyFromReg instead of Register node
...
It solves a bug uncovered by dot4 patch where the register class of
int_load_input use was ignored.
llvm-svn: 182130
2013-05-17 16:51:06 +00:00
3d5118ca40
R600: Use bottom up scheduling algorithm
...
llvm-svn: 182129
2013-05-17 16:50:56 +00:00
4c81d4da6f
R600: Use depth first scheduling algorithm
...
It should increase PV substitution opportunities and lower gpr
usage (pending computations path are "flushed" sooner)
llvm-svn: 182128
2013-05-17 16:50:44 +00:00
e958c8e0d8
R600: Replace big texture opcode switch in scheduler by usesTC/usesVC
...
llvm-svn: 182127
2013-05-17 16:50:37 +00:00
519f21eed3
R600: Relax some vector constraints on Dot4.
...
Dot4 now uses 8 scalar operands instead of 2 vectors one which allows register
coalescer to remove some unneeded COPY.
This patch also defines some structures/functions that can be used to handle
every vector instructions (CUBE, Cayman special instructions...) in a similar
fashion.
llvm-svn: 182126
2013-05-17 16:50:32 +00:00
d3eed66e8c
R600: Improve texture handling
...
llvm-svn: 182125
2013-05-17 16:50:20 +00:00
4ebef18ab5
R600: Rename 128 bit registers.
...
Almost all instructions that takes a 128 bits reg as input (fetch, export...)
have the abilities to swizzle their argument and output. Instead of printing
default swizzle for each 128 bits reg, rename T*.XYZW to T* and let instructions
print potentially optimized swizzles themselves.
llvm-svn: 182124
2013-05-17 16:50:09 +00:00
0fca91d52e
R600: Some factorization
...
llvm-svn: 182123
2013-05-17 16:50:02 +00:00
f9f4e1e7db
R600: Factorize Fetch size limit inside AMDGPUSubTarget
...
llvm-svn: 182122
2013-05-17 16:49:55 +00:00
709e01688d
R600: prettier dump of clamp
...
llvm-svn: 182121
2013-05-17 16:49:49 +00:00
ecc2ad1cd4
R600: Fix encoding for R600 family GPUs
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
https://bugs.freedesktop.org/show_bug.cgi?id=64193
https://bugs.freedesktop.org/show_bug.cgi?id=64257
https://bugs.freedesktop.org/show_bug.cgi?id=64320
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 182113
2013-05-17 15:23:21 +00:00
edade94bbc
R600: Pass MCSubtargetInfo reference to R600CodeEmitter
...
llvm-svn: 182112
2013-05-17 15:23:12 +00:00
b7be72df5b
R600/SI: return undef instead of null for skipped arguments
...
This is a candidate for the stable branch.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=64694
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 182084
2013-05-17 09:46:48 +00:00
1e21b53020
R600/SI: Add processor type for Hainan asic
...
Patch by: Alex Deucher
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181792
2013-05-14 14:42:56 +00:00
b84cde5219
Remove unused fields and arguments.
...
llvm-svn: 181706
2013-05-13 14:34:48 +00:00
227144c23c
Remove the MachineMove class.
...
It was just a less powerful and more confusing version of
MCCFIInstruction. A side effect is that, since MCCFIInstruction uses
dwarf register numbers, calls to getDwarfRegNum are pushed out, which
should allow further simplifications.
I left the MachineModuleInfo::addFrameMove interface unchanged since
this patch was already fairly big.
llvm-svn: 181680
2013-05-13 01:16:13 +00:00
86067ad6a9
Fix the R600 build.
...
llvm-svn: 181621
2013-05-10 18:31:42 +00:00
2b971eb0d0
R600: Remove AMDILPeeopholeOptimizer and replace optimizations with tablegen patterns
...
The BFE optimization was the only one we were actually using, and it was
emitting an intrinsic that we don't support.
https://bugs.freedesktop.org/show_bug.cgi?id=64201
Reviewed-by: Christian König <christian.koenig@amd.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181580
2013-05-10 02:09:45 +00:00
3a7c34c778
R600: Expand SUB for v2i32/v4i32
...
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181579
2013-05-10 02:09:39 +00:00
3deddc5079
R600: Expand MUL for v4i32/v2i32
...
Fixes piglit test for OpenCL builtin mul24, and allows mad24 to run.
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181578
2013-05-10 02:09:34 +00:00
7fb3963498
R600: Expand SRA for v4i32/v2i32
...
v2: Add v4i32 test
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181577
2013-05-10 02:09:29 +00:00
a99c6ae47a
R600: Expand vselect for v4i32 and v2i32
...
v2: Add vselect v4i32 test
Patch by: Aaron Watry
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Aaron Watry <awatry@gmail.com>
NOTE: This is a candidate for the 3.3 branch.
llvm-svn: 181576
2013-05-10 02:09:24 +00:00
f787ef1d96
R600/SI: Add intrinsic for MIMG IMAGE_GET_RESINFO opcode
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181269
2013-05-06 23:02:19 +00:00
e363dbf7eb
R600/SI: Handle arbitrary destination type in SITargetLowering::adjustWritemask
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181268
2013-05-06 23:02:15 +00:00
353b336e8c
R600/SI: Add intrinsic for texture image loading
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181267
2013-05-06 23:02:12 +00:00
c932d7329c
R600/SI: Add pattern for uint_to_fp
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181266
2013-05-06 23:02:07 +00:00
cf6452c7d4
R600/SI: Add patterns for integer maxima / minima
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181265
2013-05-06 23:02:04 +00:00
9b3d2535bf
R600/SI: Add pattern for AMDGPU.trunc intrinsic
...
Patch by: Michel Dänzer
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 181263
2013-05-06 23:02:00 +00:00
d93cede8e4
R600: Remove dead code from the CodeEmitter v2
...
v2:
- Replace switch statement with TSFlags query
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 181229
2013-05-06 17:50:57 +00:00
043de4c5af
R600: Emit config values in register / value pairs
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 181228
2013-05-06 17:50:51 +00:00
cfe2ef8fea
R600: Stop emitting the instruction type byte before each instruction
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 181225
2013-05-06 17:50:44 +00:00
dbbcaf31b6
R600: Emit ISA for CALL_FS_* instructions
...
Reviewed-by: Vincent Lejeune <vljn@ovi.com>
Tested-By: Aaron Watry <awatry@gmail.com>
llvm-svn: 181223
2013-05-06 17:50:26 +00:00
4489b85f2b
R600: Expand vector or, shl, srl, and xor nodes
...
llvm-svn: 181035
2013-05-03 17:21:31 +00:00
6a6ecedcb7
R600: BFI_INT is a vector-only instruction
...
llvm-svn: 181034
2013-05-03 17:21:24 +00:00
eac65dde30
R600: Add pattern for SHA-256 Ma function
...
This can be optimized using the BFI_INT instruction.
llvm-svn: 181033
2013-05-03 17:21:20 +00:00
c2516c6e40
R600: Clean up comments in Processors.td
...
llvm-svn: 181032
2013-05-03 17:21:14 +00:00
ddd43383ef
R600: Signed literals are 64bits wide
...
llvm-svn: 180960
2013-05-02 21:53:03 +00:00
2a44ae0053
R600: If previous bundle is dot4, PV valid chan is always X
...
llvm-svn: 180959
2013-05-02 21:52:55 +00:00
b0422e24a9
R600: Improve asmPrint of ALU clause
...
llvm-svn: 180957
2013-05-02 21:52:40 +00:00
f97af796a9
R600: Prettier asmPrint of Alu
...
llvm-svn: 180956
2013-05-02 21:52:30 +00:00
40b7f1f6c3
R600: Use new tablegen syntax for patterns
...
All but two patterns have been converted to the new syntax. The
remaining two patterns will require COPY_TO_REGCLASS instructions, which
the VLIW DAG Scheduler cannot handle.
llvm-svn: 180922
2013-05-02 15:30:12 +00:00
5447ae20ff
R600/SI: remove nonsense select pattern
...
Fortunately this pattern never matched, otherwise
we would have generated incorrect code.
Signed-off-by: Christian K??nig <christian.koenig@amd.com>
llvm-svn: 180921
2013-05-02 15:30:07 +00:00
3a8d78a2c3
R600: Always use texture cache for compute shaders
...
This will improve the performance of memory reads.
llvm-svn: 180762
2013-04-30 00:14:44 +00:00
3abdbf1cad
R600: use native for alu
...
llvm-svn: 180761
2013-04-30 00:14:38 +00:00
147700b8b4
R600: Packetize instructions
...
llvm-svn: 180760
2013-04-30 00:14:27 +00:00
076c0b28e3
R600: Rework Scheduling to handle difference between VLIW4 and VLIW5 chips
...
llvm-svn: 180759
2013-04-30 00:14:17 +00:00
22c4248213
R600: Add a Bank Swizzle operand
...
llvm-svn: 180758
2013-04-30 00:14:08 +00:00
7c395f77de
R600: Take inner dependency into tex/vtx clauses
...
llvm-svn: 180757
2013-04-30 00:14:00 +00:00
3f1d136b02
R600: Turn TEX/VTX into native instructions
...
llvm-svn: 180756
2013-04-30 00:13:53 +00:00
c299164284
R600: Add FetchInst bit to instruction defs to denote vertex/tex instructions
...
v2[Vincent Lejeune]: Split FetchInst into usesTextureCache/usesVertexCache
llvm-svn: 180755
2013-04-30 00:13:39 +00:00
7d820c0bef
R600: Add some new processor variants
...
llvm-svn: 180753
2013-04-30 00:13:27 +00:00
f501ea298b
R600: Clean up instruction class definitions
...
llvm-svn: 180752
2013-04-30 00:13:20 +00:00
4a0beb5207
R600: config section now reports use of killgt
...
llvm-svn: 180751
2013-04-30 00:13:13 +00:00
119ad03c67
R600: Use correct CF_END instruction on Northern Island GPUs
...
llvm-svn: 180735
2013-04-29 22:23:58 +00:00
8367067e02
R600: Fix encoding of CF_END_{EG, R600} instructions
...
The EOP bit was not being encoded.
llvm-svn: 180734
2013-04-29 22:23:54 +00:00
456adc6c4e
R600: Initialize AMDGPUMachineFunction::ShaderType to ShaderType::COMPUTE
...
We need to intialize this to something and since clang does not set
the shader type attribute and clang is used only for compute shaders,
initializing it to COMPUTE seems like the best choice.
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 180620
2013-04-26 18:32:24 +00:00
87047f69ad
R600: Initialize BooleanVectorContents
...
Fixes test/CodeGen/R600/setcc.ll
llvm-svn: 180231
2013-04-24 23:56:18 +00:00
34e4068d05
R600: Use SHT_PROGBITS for the .AMDGPU.config section
...
The libelf implementation that is distributed here:
http://www.mr511.de/software/english.html
will not parse sections that are marked SHT_NULL.
llvm-svn: 180230
2013-04-24 23:56:14 +00:00
117f075f6e
R600: Use .AMDGPU.config section to emit stacksize
...
llvm-svn: 180124
2013-04-23 17:34:12 +00:00
b6bfe85a07
R600: Add CF_END
...
llvm-svn: 180123
2013-04-23 17:34:00 +00:00
034ca0fe41
Remove unused DwarfSectionOffsetDirective string
...
The value isn't actually used, and setting it emits a COFF specific
directive.
llvm-svn: 180064
2013-04-22 22:49:11 +00:00
b53d8963ce
ArrayRefize getMachineNode(). No functionality change.
...
llvm-svn: 179901
2013-04-19 22:22:57 +00:00
9d10c4ce86
R600: Add pattern for the BFI_INT instruction
...
llvm-svn: 179830
2013-04-19 02:11:06 +00:00
ea977bc0e3
R600/SI: Use InstFlag for VOP3 modifier operands
...
InstFlag has a default value of 0 and will simplify the VOP3 patterns.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 179829
2013-04-19 02:11:00 +00:00
2d5c341cee
R600: Make Export Instruction not duplicable
...
llvm-svn: 179686
2013-04-17 15:17:39 +00:00
218093e834
R600: Export is emitted as a CF_NATIVE inst
...
llvm-svn: 179685
2013-04-17 15:17:32 +00:00
98a7380859
R600: Emit used GPRs count
...
llvm-svn: 179684
2013-04-17 15:17:25 +00:00
cb97e3acfa
R600/SI: Emit config values in register value pairs.
...
Instead of emitting config values in a predefined order, the code
emitter will now emit a 32-bit register index followed by the 32-bit
config value.
llvm-svn: 179546
2013-04-15 17:51:35 +00:00
3a7beafb32
R600/SI: Emit configuration value in the .AMDGPU.config ELF section
...
llvm-svn: 179545
2013-04-15 17:51:30 +00:00
9991659fab
R600: Emit ELF formatted code rather than raw ISA.
...
llvm-svn: 179544
2013-04-15 17:51:21 +00:00
3ee2b1e26f
R600ControlFlowFinalizer.cpp: Fix a warning. [-Wunused-variable]
...
llvm-svn: 179263
2013-04-11 04:16:27 +00:00
3b0853be56
Whitespace.
...
llvm-svn: 179262
2013-04-11 04:16:22 +00:00
8caa904bde
R600/SI: Add pattern for AMDGPUurecip
...
21 more little piglits with radeonsi.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 179186
2013-04-10 17:17:56 +00:00
04d9aa4822
R600: Add VTX_READ_* and RAT_WRITE_CACHELESS_* when computing cf addr
...
llvm-svn: 179174
2013-04-10 13:29:20 +00:00
8b1ed28ef1
R600/SI: dynamical figure out the reg class of MIMG
...
Depending on the number of bits set in the writemask.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 179166
2013-04-10 08:39:16 +00:00
8e06e2a8c4
R600/SI: adjust writemask to only the used components
...
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 179165
2013-04-10 08:39:08 +00:00
4ace663255
R600/SI: remove image sample writemask
...
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 179164
2013-04-10 08:39:01 +00:00
5f11dd390a
R600: Control Flow support for pre EG gen
...
llvm-svn: 179020
2013-04-08 13:05:49 +00:00
754f80ff3a
R600/SI: Add support for buffer stores v2
...
v2:
- Use the ADDR64 bit
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178931
2013-04-05 23:31:51 +00:00
6db08eb42f
R600/SI: Use same names for corresponding MUBUF operands and encoding fields
...
The code emitter knows how to encode operands whose name matches one of
the encoding fields. If there is no match, the code emitter relies on
the order of the operand and field definitions to determine how operands
should be encoding. Matching by order makes it easy to accidentally break
the instruction encodings, so we prefer to match by name.
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178930
2013-04-05 23:31:44 +00:00
60174bb9ca
R600: Add RV670 processor
...
This is an R600 GPU with double support.
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178929
2013-04-05 23:31:40 +00:00
2f21c7e551
R600/SI: Add processor types for each SI variant
...
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178928
2013-04-05 23:31:35 +00:00
edbf1eb42b
R600/SI: Avoid generating S_MOVs with 64-bit immediates v2
...
SITargetLowering::analyzeImmediate() was converting the 64-bit values
to 32-bit and then checking if they were an inline immediate. Some
of these conversions caused this check to succeed and produced
S_MOV instructions with 64-bit immediates, which are illegal.
v2:
- Clean up logic
Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178927
2013-04-05 23:31:20 +00:00
bcbb13d691
R600: Use a mask for offsets when encoding instructions
...
llvm-svn: 178763
2013-04-04 14:00:09 +00:00
8e377fdba6
R600: Fix wrong address when substituting ENDIF
...
llvm-svn: 178762
2013-04-04 14:00:03 +00:00
c44fa99719
R600: Take export into account when computing cf address
...
llvm-svn: 178761
2013-04-04 13:59:59 +00:00
c3d3f9b66e
R600: Fix last ALU of a clause being emitted in a separate clause
...
llvm-svn: 178675
2013-04-03 18:24:47 +00:00
80031d9fc4
R600: Factorize maximum alu per clause in a single location
...
llvm-svn: 178667
2013-04-03 16:49:34 +00:00
b6d6c0d458
R600: Simplify data structure and add DEBUG to R600ControlFlowFinalizer
...
llvm-svn: 178665
2013-04-03 16:24:09 +00:00
9931298b30
R600: Consider KILLGT as an ALU instruction
...
Mesa does not override llvm behavior wrt KILLGT anymore so llvm
has to handle KILLGT on its own.
llvm-svn: 178664
2013-04-03 16:24:04 +00:00
fd98f7f2b6
Target/R600: Fix CMake build to add missing files.
...
llvm-svn: 178508
2013-04-01 22:05:58 +00:00
bfaa63a6db
R600: Add support for native control flow
...
llvm-svn: 178505
2013-04-01 21:48:05 +00:00
ace6f7351e
R600/SI: Share code recording ShaderTypeAttribute between generations
...
llvm-svn: 178504
2013-04-01 21:47:53 +00:00
f43bc57b66
R600: Emit CF_ALU and use true kcache register.
...
llvm-svn: 178503
2013-04-01 21:47:42 +00:00
53f3525d35
R600: Emit native instructions for tex
...
llvm-svn: 178452
2013-03-31 19:33:04 +00:00
6c75232cf0
These two are default in the constructor for MCAsmInfo.
...
llvm-svn: 178293
2013-03-28 21:37:18 +00:00
08f5929942
R600/SI: add SETO/SETUO patterns
...
6 more piglit tests.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 178145
2013-03-27 15:27:31 +00:00
3c14580acb
R600/SI: add cummuting of rev instructions
...
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 178127
2013-03-27 09:12:59 +00:00
70a5032c1b
R600/SI: add mulhu/mulhs patterns
...
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 178126
2013-03-27 09:12:51 +00:00
20a7e6b764
R600/SI: add srl/sha patterns for SI
...
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 178125
2013-03-27 09:12:44 +00:00
3234178bf9
R600/SIMCCodeEmitter.cpp: Prune a couple of unused members, STI and Ctx. [-Wunused-private-field]
...
llvm-svn: 178065
2013-03-26 19:42:48 +00:00
8370dbbffd
R600/SI: improve post ISel folding
...
Not only fold immediates, but avoid unnecessary copies as well.
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178024
2013-03-26 14:04:17 +00:00
082c661f94
R600/SI: improve vector interpolation
...
Prevent loading M0 multiple times.
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178023
2013-03-26 14:04:12 +00:00
25ce3e9f4c
R600/SI: avoid unecessary subreg extraction in IMAGE_SAMPLE
...
Just define the address as unknown instead of VReg_32.
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178022
2013-03-26 14:04:07 +00:00
eecebd0bab
R600/SI: switch back to RegPressure scheduling
...
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178021
2013-03-26 14:04:02 +00:00
727d06de1d
R600/SI: mark most intrinsics as readnone v2
...
They read from constant register space anyway.
v2: fix lit tests
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178020
2013-03-26 14:03:57 +00:00
737d4a1665
R600/SI: replace WQM intrinsic
...
Just enable WQM when we see an LDS interpolation instruction.
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178019
2013-03-26 14:03:50 +00:00
6a9d390b6b
R600/SI: fix ELSE pseudo op handling
...
Restore the EXEC mask early, otherwise a copy might end up not beeing executed.
Candidate for the mesa stable branch.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 178018
2013-03-26 14:03:44 +00:00
90b45124cd
R600: fix DenseMap with pointer key iteration in the structurizer
...
Use a MapVector on types where the iteration order matters.
Otherwise we doesn't always produce a deterministic output.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
llvm-svn: 177999
2013-03-26 10:24:20 +00:00
Aaron Watry