Justin Holewinski 0a7b783785 [NVPTX] Use approximate FP ops when unsafe-fp-math is used, and append ...

.ftz to instructions if the nvptx-f32ftz attribute is set to "true"

llvm-svn: 186820

2013-07-22 12:18:04 +00:00

..

add-128bit.ll

[NVPTX] 64-bit ADDC/ADDE are not legal

2013-07-01 12:59:04 +00:00

annotations.ll

[NVPTX] Remove support for SM < 2.0. This was never fully supported anyway.

2013-03-30 14:29:30 +00:00

arithmetic-fp-sm20.ll

…

arithmetic-int.ll

[NVPTX] Remove support for SM < 2.0. This was never fully supported anyway.

2013-03-30 14:29:30 +00:00

calling-conv.ll

[NVPTX] Remove support for SM < 2.0. This was never fully supported anyway.

2013-03-30 14:29:30 +00:00

compare-int.ll

[NVPTX] Calling conventions fix

2013-06-28 17:58:10 +00:00

convert-fp.ll

[NVPTX] Remove support for SM < 2.0. This was never fully supported anyway.

2013-03-30 14:29:30 +00:00

convert-int-sm20.ll

[NVPTX] Calling conventions fix

2013-06-28 17:58:10 +00:00

ctlz.ll

[NVPTX] Add support for cttz/ctlz/ctpop

2013-06-28 17:58:07 +00:00

ctpop.ll

Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion.

2013-07-13 20:38:47 +00:00

cttz.ll

[NVPTX] Add support for cttz/ctlz/ctpop

2013-06-28 17:58:07 +00:00

fast-math.ll

[NVPTX] Use approximate FP ops when unsafe-fp-math is used, and append

2013-07-22 12:18:04 +00:00

fma-disable.ll

…

fma.ll

…

generic-to-nvvm.ll

[NVPTX] Add support for selecting CUDA vs OCL mode based on triple

2013-06-21 18:51:49 +00:00

global-ordering.ll

[NVPTX] Order global variables in def-use order before emiting them in the final assembly

2012-11-16 21:03:51 +00:00

i1-global.ll

[NVPTX] Add support for selecting CUDA vs OCL mode based on triple

2013-06-21 18:51:49 +00:00

i1-param.ll

[NVPTX] Add support for selecting CUDA vs OCL mode based on triple

2013-06-21 18:51:49 +00:00

i8-param.ll

Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion.

2013-07-13 20:38:47 +00:00

intrin-nocapture.ll

[NVPTX] Remove NoCapture from address space conversion intrinsics. NoCapture is not valid in this case, and was causing incorrect optimizations.

2013-02-11 18:56:35 +00:00

intrinsic-old.ll

[NVPTX] Re-enable support for virtual registers in the final output

2013-05-31 12:14:49 +00:00

intrinsics.ll

[NVPTX] Re-enable support for virtual registers in the final output

2013-05-31 12:14:49 +00:00

ld-addrspace.ll

[NVPTX] Calling conventions fix

2013-06-28 17:58:10 +00:00

ld-generic.ll

[NVPTX] Calling conventions fix

2013-06-28 17:58:10 +00:00

ldu-i8.ll

[NVPTX] Make sure we zero out high-order 24 bits for 8-bit load into 32-bit value

2013-07-01 12:58:48 +00:00

ldu-reg-plus-offset.ll

Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion.

2013-07-13 20:38:47 +00:00

lit.local.cfg

…

load-sext-i1.ll

[NVPTX] Add support for selecting CUDA vs OCL mode based on triple

2013-06-21 18:51:49 +00:00

module-inline-asm.ll

[NVPTX] Add support for module-scope inline asm

2013-07-01 13:00:14 +00:00

nvvm-reflect.ll

[NVPTX] Add NVVMReflect pass to allow compile-time selection of

2013-03-30 14:29:25 +00:00

param-align.ll

[NVPTX] Use ABI alignment for parameters when alignment is not specified.

2012-11-09 23:50:24 +00:00

pr13291-i1-store.ll

[NVPTX] Clean up comparison/select/convert patterns and factor out PTX instructions from their patterns

2013-06-28 17:58:04 +00:00

pr16278.ll

[NVPTX] Remove old CONST_NOT_GEN address space that is not being used anymore and causes constants to be emitted in the global address space

2013-06-10 13:29:47 +00:00

ptx-version-30.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

ptx-version-31.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

refl1.ll

[NVPTX] Add support for selecting CUDA vs OCL mode based on triple

2013-06-21 18:51:49 +00:00

rsqrt.ll

Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion.

2013-07-13 20:38:47 +00:00

sched1.ll

Propagate DAG node ordering during type legalization and instruction selection

2013-03-20 00:10:32 +00:00

sched2.ll

Propagate DAG node ordering during type legalization and instruction selection

2013-03-20 00:10:32 +00:00

sext-in-reg.ll

Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion.

2013-07-13 20:38:47 +00:00

sext-params.ll

[NVPTX] Handle signext/zeroext attributes properly

2013-07-01 12:58:58 +00:00

simple-call.ll

…

sm-version-20.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

sm-version-21.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

sm-version-30.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

sm-version-35.ll

[NVPTX] Add more precise PTX/SM target attributes

2012-11-12 03:16:43 +00:00

st-addrspace.ll

[NVPTX] Remove i8 register class. PTX support for i8 (.b8, .u8, .s8) is rather poor and we're better off just ignoring it and letting LLVM expand all i8 ops out to i16.

2013-06-28 17:57:59 +00:00

st-generic.ll

[NVPTX] Remove i8 register class. PTX support for i8 (.b8, .u8, .s8) is rather poor and we're better off just ignoring it and letting LLVM expand all i8 ops out to i16.

2013-06-28 17:57:59 +00:00

tuple-literal.ll

[NVPTX] Remove support for SM < 2.0. This was never fully supported anyway.

2013-03-30 14:29:30 +00:00

vec-param-load.ll

[NVPTX] Fix vector loads from parameters that span multiple loads, and fix some typos

2013-07-01 12:59:01 +00:00

vector-args.ll

[NVPTX] Add support for vectorized function return values

2013-06-28 17:57:55 +00:00

vector-compare.ll

Allow targets to prefer TypeSplitVector over TypePromoteInteger when computing the legalization method for vectors

2012-11-29 14:26:24 +00:00

vector-loads.ll

Propagate DAG node ordering during type legalization and instruction selection

2013-03-20 00:10:32 +00:00

vector-select.ll

Teach the legalizer how to handle operands for VSELECT nodes

2012-11-29 14:26:28 +00:00