RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 03:33:20 +01:00
Code Issues Projects Releases Wiki Activity
416f0c2389
llvm-mirror/test/CodeGen/AArch64
History
Kevin Qin 379441a4e6 [AArch64 NEON] Lower SELECT_CC with vector operand. 
When the scalar compare is between floating point and operands are
vector, we custom lower SELECT_CC to use NEON SIMD compare for
generating less instructions.

llvm-svn: 200365
2014-01-29 01:57:30 +00:00
..
128bit_load_store.ll [AArch64 NEON] Support poly128_t and implement relevant intrinsic. 2013-12-10 06:48:35 +00:00
adc.ll
addsub_ext.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
addsub-shifted.ll
addsub.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
alloca.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
analyze-branch.ll
assertion-rc-mismatch.ll Fix an over-constrained assertion in MachineFunction::addLiveIn. 2013-12-12 00:15:47 +00:00
atomic-ops-not-barriers.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
atomic-ops.ll
basic-pic.ll Convert a llc -filetype=obj test into a llvm-mc test. 2013-10-28 20:40:20 +00:00
bitfield-insert-0.ll
bitfield-insert.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
bitfield.ll
blockaddress.ll
bool-loads.ll
breg.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
callee-save.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
code-model-large-abs.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
compare-branch.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
complex-copy-noneon.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
concatvector-v8i8-bug.ll [AArch64] Removed unused i8 type from FPR8 register class. 2014-01-24 22:36:53 +00:00
cond-sel.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
directcond.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
dp1.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
dp2.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
dp-3source.ll
extern-weak.ll
extract.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
fastcc-reserved.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
fastcc.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
fcmp.ll
fcvt-fixed.ll
fcvt-int.ll AArch64: enable MISched by default. 2013-10-09 07:53:57 +00:00
flags-multiuse.ll
floatdp_1source.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
floatdp_2source.ll
fp128-folding.ll
fp128.ll AArch64: enable MISched by default. 2013-10-09 07:53:57 +00:00
fp-cond-sel.ll
fp-dp3.ll [AArch64] Check fmul node single use in fused multiply patterns 2013-12-24 00:47:29 +00:00
fpimm.ll AArch64: enable MISched by default. 2013-10-09 07:53:57 +00:00
frameaddr.ll add test cases for frameaddr and returnaddr for aarch64 2013-10-29 17:01:29 +00:00
func-argpassing.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
func-calls.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
global-alignment.ll
got-abuse.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
i128-align.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
illegal-float-ops.ll
init-array.ll Make sure -use-init-array has intended effect on all AArch64 ELF targets, not just linux. 2014-01-10 13:41:49 +00:00
inline-asm-constraints-badI.ll
inline-asm-constraints-badK2.ll
inline-asm-constraints-badK.ll
inline-asm-constraints-badL.ll
inline-asm-constraints.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
inline-asm-modifiers.ll AArch64: use default asm operand printing when modifier inapplicable 2013-11-04 23:04:07 +00:00
jump-table.ll Convert another llc -filetype=obj test. 2013-10-28 20:59:41 +00:00
large-consts.ll AArch64: add llc-based tests for previous commit. 2013-07-25 16:23:55 +00:00
large-frame.ll
ldst-regoffset.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
ldst-unscaledimm.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
ldst-unsignedimm.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
lit.local.cfg [tests] Cleanup initialization of test suffixes. 2013-08-16 00:37:11 +00:00
literal_pools.ll [AArch64] Make the use of FP instructions optional, but enabled by default. 2013-10-31 09:32:11 +00:00
local_vars.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
logical_shifted_reg.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
logical-imm.ll
movw-consts.ll
movw-shift-encoding.ll
mul-lohi.ll AArch64: don't try to handle [SU]MUL_LOHI nodes 2014-01-14 22:53:22 +00:00
neon-2velem-high.ll Add some missing pattern matches for AArch64 Neon intrinsics like vmull_high_n_s16 and friends. 2013-12-03 01:29:32 +00:00
neon-2velem.ll [AArch64]Fix a problem that the register order of fmls/fmla by element is incorrect. 2013-12-25 07:12:34 +00:00
neon-3vdiff.ll [AArch64 NEON] Support poly128_t and implement relevant intrinsic. 2013-12-10 06:48:35 +00:00
neon-aba-abd.ll Fix pattern match for movi with 0D result 2013-12-09 19:29:14 +00:00
neon-across.ll [AArch64] Refactor NEON floating-point Max/Min/Maxnm/Minnm across vector AArch64 2013-12-11 23:21:25 +00:00
neon-add-pairwise.ll [AArch64]Add missing pair intrinsics such as: 2013-12-09 03:51:42 +00:00
neon-add-sub.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-bitcast.ll [AArch64] Removed unnecessary copy patterns with v1fx types. 2013-12-12 15:46:29 +00:00
neon-bitwise-instructions.ll Teach the DAGCombiner how to fold 'vselect' dag nodes according 2014-01-08 18:33:04 +00:00
neon-bsl.ll [AArch64] Added vselect patterns with float and double types 2014-01-23 19:18:57 +00:00
neon-compare-instructions.ll Implement aarch64 neon instruction class SIMD misc. 2013-11-14 02:44:13 +00:00
neon-copy.ll [AArch64 NEON] Try to generate CONCAT_VECTOR when lowering BUILD_VECTOR or SHUFFLE_VECTOR. 2014-01-27 02:53:54 +00:00
neon-copyPhysReg-tuple.ll [AArch64]Add support to copy D tuples such as DPair/DTriple/DQuad and Q tuples such as QPair/QTriple/QQuad. There is no test case for D tuple as the original test cases are too large. As the copy of the D tuple is similar to the Q tuple, the correctness can be guaranteed. 2014-01-07 10:00:03 +00:00
neon-crypto.ll Add predicate for AArch64 crypto instructions. 2013-11-19 01:38:31 +00:00
neon-diagnostics.ll AArch64: The pattern match should check the range of the immediate value. 2013-11-29 02:11:22 +00:00
neon-extract.ll [AArch64 NEON] Fix a bug caused by undef lane when generating VEXT. 2014-01-21 01:48:52 +00:00
neon-facge-facgt.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
neon-fma.ll [AArch64] Check fmul node single use in fused multiply patterns 2013-12-24 00:47:29 +00:00
neon-fpround_f128.ll [AArch64 NEON] Add test case for vector FP_ROUND. 2014-01-26 02:23:33 +00:00
neon-frsqrt-frecp.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
neon-halving-add-sub.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
neon-load-store-v1i32.ll [AArch64]Fix the problem can't select f16_to_f32 and f32_to_f16. 2014-01-17 06:23:30 +00:00
neon-max-min-pairwise.ll [AArch64]Add missing pair intrinsics such as: 2013-12-09 03:51:42 +00:00
neon-max-min.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
neon-misc-scalar.ll Add some missing pattern matches for AArch64 Neon intrinsics like vuqadd_s64 and friends. 2013-12-03 01:33:52 +00:00
neon-misc.ll [AArch64 NEON] Custom lower conversion between vector integer and vector floating point if element bit-width doesn't match. 2014-01-17 05:52:35 +00:00
neon-mla-mls.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-mov.ll [AArch64 NEON]Implment loading vector constant form constant pool. 2013-12-18 06:26:04 +00:00
neon-mul-div.ll [AArch64 NEON] Expand vector for UDIV/SDIV/UREM/SREM/FREM as neon doesn't support these operations. 2014-01-17 09:54:30 +00:00
neon-or-combine.ll [AArch64] Fix assertion failure caused by an invalid comparison between APInt values. 2014-01-13 16:51:00 +00:00
neon-perm.ll [AArch64 NEON] Add more scenarios to use perm instructions when lowering shuffle_vector 2014-01-13 01:56:29 +00:00
neon-rounding-halving-add.ll AArch64: add initial NEON support 2013-08-01 09:20:35 +00:00
neon-rounding-shift.ll Initial support for Neon scalar instructions. 2013-09-24 02:47:27 +00:00
neon-saturating-add-sub.ll Initial support for Neon scalar instructions. 2013-09-24 02:47:27 +00:00
neon-saturating-rounding-shift.ll Initial support for Neon scalar instructions. 2013-09-24 02:47:27 +00:00
neon-saturating-shift.ll Initial support for Neon scalar instructions. 2013-09-24 02:47:27 +00:00
neon-scalar-abs.ll [AArch64] Add support for NEON scalar absolute value instruction. 2013-10-16 21:04:34 +00:00
neon-scalar-add-sub.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-by-elem-fma.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-by-elem-mul.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-compare.ll [AArch64]The compare to zero intrinsics should be implemented by 'icmp/fcmp' and 'sext' not 'zext'. Modify the test cases. 2013-12-23 02:42:10 +00:00
neon-scalar-copy.ll For AArch64 Neon, simplify scalar dup by lane0 for fp. 2013-12-30 02:44:35 +00:00
neon-scalar-cvt.ll [AArch64] Refactor the NEON signed/unsigned floating-point convert to fixed-point 2013-12-10 21:33:56 +00:00
neon-scalar-ext.ll Improve pattern match from v1i8 to v1i32 for AArch64 Neon. 2014-01-26 04:55:53 +00:00
neon-scalar-extract-narrow.ll [AArch64] Add support for NEON scalar extract narrow instructions. 2013-10-18 14:03:24 +00:00
neon-scalar-fabd.ll [AArch64] Refactor the NEON floating-point absolute difference LLVM AArch64 2013-12-10 21:33:59 +00:00
neon-scalar-fcvt.ll [AArch64] Refactor the Neon vector/scalar floating-point convert intrinsics so 2013-12-10 16:11:39 +00:00
neon-scalar-fp-compare.ll [AArch64]The compare to zero intrinsics should be implemented by 'icmp/fcmp' and 'sext' not 'zext'. Modify the test cases. 2013-12-23 02:42:10 +00:00
neon-scalar-mul.ll Implemented Neon scalar by element intrinsics. 2013-11-21 07:37:04 +00:00
neon-scalar-neg.ll [AArch64] Add support for NEON scalar negate instruction. 2013-10-16 21:04:39 +00:00
neon-scalar-recip.ll [AArch64] Refactor the NEON scalar floating-point reciprocal step and 2013-12-11 21:03:43 +00:00
neon-scalar-reduce-pairwise.ll [AArch64] Refactor the NEON scalar reduce pairwise intrinsics, so that they use 2013-12-09 22:47:38 +00:00
neon-scalar-rounding-shift.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-saturating-add-sub.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-saturating-rounding-shift.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-saturating-shift.ll Fix Incorrect CHECK message [0-31]+ in test case. 2013-12-12 02:19:13 +00:00
neon-scalar-shift-imm.ll AArch64: The pattern match should check the range of the immediate value. 2013-11-29 02:11:22 +00:00
neon-scalar-shift.ll Add missing pattern matches to support ACLE intrinsics of AArch64 NEON. 2013-12-25 01:22:51 +00:00
neon-select_cc.ll [AArch64 NEON] Lower SELECT_CC with vector operand. 2014-01-29 01:57:30 +00:00
neon-shift-left-long.ll Fix a bug in DAGcombiner about zero-extend after setcc. 2013-12-30 02:05:13 +00:00
neon-shift.ll Initial support for Neon scalar instructions. 2013-09-24 02:47:27 +00:00
neon-shl-ashr-lshr.ll [AArch64]Can't select shift left 0 of type v1i64 2013-12-30 02:12:46 +00:00
neon-simd-ldst-multi-elem.ll Fix the bugs about AArch64 Load/Store vector types and bitcast between i64 and vector types. 2013-11-22 08:47:22 +00:00
neon-simd-ldst-one.ll Add missing pattern matches to support ACLE intrinsics of AArch64 NEON. 2013-12-25 01:22:51 +00:00
neon-simd-ldst.ll For AArch64, add missing register cost calculation for big value types like v4i64 and v8i64. 2013-12-05 02:12:01 +00:00
neon-simd-post-ldst-multi-elem.ll Implement the newly added ACLE functions for ld1/st1 with 2/3/4 vectors. 2013-11-18 06:31:53 +00:00
neon-simd-post-ldst-one.ll Implement AArch64 neon instructions class SIMD lsone and SIMD lone-post. 2013-11-19 02:17:05 +00:00
neon-simd-shift.ll [AArch64]Add missing floating point convert, round and misc intrinsics. 2013-12-03 06:06:55 +00:00
neon-simd-tbl.ll Implement AArch64 NEON instruction set AdvSIMD (table). 2013-11-14 01:57:32 +00:00
neon-simd-vget.ll For AArch64 back-end instruction selection, lower Neon_Lowxxx with EXTRCT_SUBREG. 2013-11-22 02:45:13 +00:00
neon-truncStore-extLoad.ll [AArch64]Pattern match failures for truncate store and extend load 2013-12-09 03:34:08 +00:00
neon-vector-list-spill.ll [AArch64]Add support to spill/fill D tuples such as DPair/DTriple/DQuad. There is no test cases for D tuple as the original test cases are too large. As the spill/fill of the D tuple is similar to the Q tuple, the correctness can be guaranteed. 2014-01-07 10:50:43 +00:00
pic-eh-stubs.ll
regress-bitcast-formals.ll
regress-f128csel-flags.ll
regress-fp128-livein.ll [AArch64] Fix NZCV reg live-in bug in F128CSEL codegen. 2013-10-24 08:28:24 +00:00
regress-tail-livereg.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
regress-tblgen-chains.ll
regress-w29-reserved-with-fp.ll
regress-wzr-allocatable.ll
returnaddr.ll add test cases for frameaddr and returnaddr for aarch64 2013-10-29 17:01:29 +00:00
setcc-takes-i32.ll
sext_inreg.ll For AArch64, lowering sext_inreg and generate optimized code by using SXTL. 2014-01-15 05:08:01 +00:00
sibling-call.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
sincos-expansion.ll
tail-call.ll Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. 2013-07-14 06:24:09 +00:00
tls-dynamic-together.ll
tls-dynamics.ll AArch64: enable MISched by default. 2013-10-09 07:53:57 +00:00
tls-execs.ll AArch64: add llc-based tests for previous commit. 2013-07-25 16:23:55 +00:00
tst-br.ll
variadic.ll Fix non-deterministic SDNodeOrder-dependent codegen 2014-01-12 14:09:17 +00:00
zero-reg.ll