Commit Graph

16078 Commits

Author SHA1 Message Date
Simon Pilgrim 1cd61b82bd [X86][SSE] Added non-temporal store tests for all 512-bit vector types
llvm-svn: 271393
2016-06-01 13:58:00 +00:00
Simon Pilgrim 288be8bab6 [X86][SSE] Added non-temporal store tests for all 256-bit vector types
Also added KNL AVX-512 checks

llvm-svn: 271391
2016-06-01 13:20:25 +00:00
Simon Pilgrim 80f5335969 [X86][SSE] Added non-temporal store tests for all 128-bit integer vector types
llvm-svn: 271389
2016-06-01 13:05:00 +00:00
Michael Zuckerman 6a894956fc Adding back-end support to two bit scanning intrinsics
Adding LLVM back-end support to two intrinsics dealing with bit scan: _bit_scan_forward and _bit_scan_reverse.
Their functionality is as described in Intel intrinsics guide:
https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=_bit_scan_forward&expand=371,370
https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=_bit_scan_reverse&expand=371,370

Commit on behalf of Omer Paparo Bivas


Differential Revision: http://reviews.llvm.org/D19915

llvm-svn: 271386
2016-06-01 12:02:37 +00:00
Oliver Stannard 92ca83cccd [ARM] Add additional matching for UBFX instructions
This adds an additional matcher to select UBFX(..) from SRL(AND(..)) in
ARMISelDAGToDAG to help with code size.

Patch by David Green.

Differential Revision: http://reviews.llvm.org/D20667

llvm-svn: 271384
2016-06-01 12:01:01 +00:00
Chris Dewhurst 53bde954db [Sparc] Allow passing of empty structs.
Passing an empty struct as a function call argument is now supported.

unit tests for various scenarios added.

llvm-svn: 271374
2016-06-01 08:48:56 +00:00
Craig Topper 4f2d5a68d3 Revert r271362 "[AVX512] Remove masked load intrinsics. Clang now emits generic masked load intrinsics instead."
Looks like something isn't quite right still. Also forgot to move the test cases to an autoupgrade test.

llvm-svn: 271363
2016-06-01 05:57:55 +00:00
Craig Topper dacd9d2bac [AVX512] Remove masked load intrinsics. Clang now emits generic masked load intrinsics instead.
The intrinsics will be autoupgraded to the same generic masked loads.

llvm-svn: 271362
2016-06-01 05:35:16 +00:00
Matthias Braun f9acacaa92 CodeGen: Refactor renameDisconnectedComponents() as a pass
Refactor LiveIntervals::renameDisconnectedComponents() to be a pass.
Also change the name to "RenameIndependentSubregs":

- renameDisconnectedComponents() worked on a MachineFunction at a time
  so it is a natural candidate for a machine function pass.

- The algorithm is testable with a .mir test now.

- This also fixes a problem where the lazy renaming as part of the
  MachineScheduler introduced IMPLICIT_DEF instructions after the number
  of a nodes in a region were counted leading to a mismatch.

Differential Revision: http://reviews.llvm.org/D20507

llvm-svn: 271345
2016-05-31 22:38:06 +00:00
Kevin B. Smith ed0b620a65 [X86]: Add a pattern that uses GR16_ABCD rather than GR32_ABCD to avoid falsely marking whole 32 bit register as live.
Differential Revision: http://reviews.llvm.org/D20649

llvm-svn: 271341
2016-05-31 22:00:12 +00:00
Matthias Braun ce0bcb78e6 ARM: Improve/fix comment in recently added test.
llvm-svn: 271340
2016-05-31 21:59:59 +00:00
Matthias Braun fe725c9241 ARM: Do not attempt to modify register class of physregs.
Physregs have no associated register class, do not attempt to modify it
in Thumb2InstrInfo::storeRegToStackSlot()/loadFromStackSlot().

llvm-svn: 271339
2016-05-31 21:39:12 +00:00
Ahmed Bougacha 96ef87e910 [CodeGen] Promote FMINNAN/FMAXNAN like other binops.
We think it's OK to generate half fminnan because it's legal for the
transform-to type (f32; r245196). However, PromoteFloatRes was missing
the case; simply promote like the other binops, including minnum.

llvm-svn: 271317
2016-05-31 18:50:25 +00:00
Rafael Espindola 4d29099f7f Delete AArch64II::MO_CONSTPOOL.
A constant pool holding the address of a variable in equivalent to
a got entry. It produces exactly the same instruction sequence as a
got use and unlike a got use this is not uniqued by the linker.

llvm-svn: 271311
2016-05-31 18:31:14 +00:00
Rafael Espindola 7ad97b2fe4 Add a use of shouldAssumeDSOLocal to ARM.
Now this code path knows about position independent executables.

llvm-svn: 271290
2016-05-31 15:31:55 +00:00
Ranjeet Singh 16c24f4d6e [ARM] Add backend support for load/store intrinsics.
Added support to map intrinsics
__builtin_arm_{ldc,ldcl,ldc2,ldc2l,stc,stcl,stc2,stc2l}
to their ARM instructions.

Differential Revision: http://reviews.llvm.org/D20564

llvm-svn: 271271
2016-05-31 12:39:30 +00:00
Simon Pilgrim e05dc45897 [X86][SSE] Add load-folding patterns for (V)CVTDQ2PD (PR27291)
Added patterns for (V)CVTDQ2PD -> 2f64 loading from a 64-bit source.

llvm-svn: 271269
2016-05-31 12:04:35 +00:00
Simon Dardis 03676dc969 [mips] bnec/beqc register constraint fix
beqc and bnec cannot have $rs == $rt. Inhibit compact branch creation
if that would occur.

Reviewers: vkalintiris, dsanders

Differential Revision: http://reviews.llvm.org/D20624

llvm-svn: 271260
2016-05-31 09:54:55 +00:00
Igor Breger 73ee8ba9b0 [AVX512] Fix intrinsic vcvtps2ph lowering.
Differential Revision: http://reviews.llvm.org/D20788

llvm-svn: 271255
2016-05-31 08:04:21 +00:00
Igor Breger 52bd1d5fcc Fix intrinsic vbroadcast{i32|f32}x2 lowering.
Differential Revision: http://reviews.llvm.org/D20780

llvm-svn: 271254
2016-05-31 07:43:39 +00:00
Craig Topper 50f85c22c5 [AVX512] Remove masked store intrinsics. Clang now emits generic masked store intrinsics instead.
The intrinsics will be autoupgraded to the same generic masked stores.

llvm-svn: 271245
2016-05-31 01:50:02 +00:00
Saleem Abdulrasool d2f705ddf9 X86: permit using SjLj EH on x86 targets as an option
This adds support to the backed to actually support SjLj EH as an exception
model.  This is *NOT* the default model, and requires explicitly opting into it
from the frontend.  GCC supports this model and for MinGW can still be enabled
via the `--using-sjlj-exceptions` options.

Addresses PR27749!

llvm-svn: 271244
2016-05-31 01:48:07 +00:00
Craig Topper 8287fd8abd [X86] Remove SSE/AVX unaligned store intrinsics as clang no longer uses them. Auto upgrade to native unaligned store instructions.
llvm-svn: 271236
2016-05-30 23:15:56 +00:00
Craig Topper 39716f8358 [X86] Use update_llc_test_checks.py to re-generate a test in preparation for an upcoming commit. NFC
llvm-svn: 271234
2016-05-30 22:54:14 +00:00
Simon Pilgrim d788c9d83d [X86][XOP] Split off auto-upgraded xop intrinsics
llvm-svn: 271228
2016-05-30 19:50:56 +00:00
Simon Pilgrim 582d75b0eb [X86][SSE] Renamed pmovxrm tests
These aren't intrinsics anymore - as discussed on D20686

llvm-svn: 271226
2016-05-30 19:14:37 +00:00
Simon Pilgrim 24da61058a [X86][AVX2] Regenerated AVX2 extension tests
llvm-svn: 271224
2016-05-30 18:49:57 +00:00
Simon Pilgrim d64af65f6d [X86][SSE] Updated storeu fast-isel tests to match clang builtin tests
Since rL271214 the headers have no longer used the storeu intrinsic

llvm-svn: 271222
2016-05-30 18:42:51 +00:00
Simon Pilgrim 4ed0e07b23 [X86][SSE2] Updated _mm_store_pd1/_mm_store1_pd fast-isel tests to match D20617
llvm-svn: 271220
2016-05-30 18:18:44 +00:00
Diana Picus f353a5e06d [BPF] Remove exit-on-error from tests (PR27768, PR27769)
The exit-on-error flag is necessary to avoid some assertions/unreachables. We
can get past them by creating a few dummy nodes.

Fixes PR27768, PR27769.

Differential Revision: http://reviews.llvm.org/D20726

llvm-svn: 271200
2016-05-30 08:28:34 +00:00
Simon Pilgrim 9602d678cb [X86][SSE] (Reapplied) Replace (V)PMOVSX and (V)PMOVZX integer extension intrinsics with generic IR (llvm)
This patch removes the llvm intrinsics VPMOVSX and (V)PMOVZX sign/zero extension intrinsics and auto-upgrades to SEXT/ZEXT calls instead. We already did this for SSE41 PMOVSX sometime ago so much of that implementation can be reused.

Reapplied now that the the companion patch (D20684) removes/auto-upgrade the clang intrinsics has been committed.

Differential Revision: http://reviews.llvm.org/D20686

llvm-svn: 271131
2016-05-28 18:03:41 +00:00
Sanjay Patel 97c2c108fd [x86] avoid printing unnecessary sign bits of hex immediates in asm comments (PR20347)
It would be better to check the valid/expected size of the immediate operand, but this is
generally better than what we print right now.

Differential Revision: http://reviews.llvm.org/D20385

llvm-svn: 271114
2016-05-28 14:58:37 +00:00
Ahmed Bougacha a3dc1ba142 [X86] Try to zero elts when lowering 256-bit shuffle with PSHUFB.
Otherwise we fallback to a blend of PSHUFBs later on.

Differential Revision: http://reviews.llvm.org/D19661

llvm-svn: 271113
2016-05-28 14:38:04 +00:00
Rafael Espindola fe796dca90 Fix default reloc model on ARM.
llvm-svn: 271111
2016-05-28 10:41:15 +00:00
Renato Golin 9be88629d5 Revert "Revert "Map DynamicNoPIC to Static on non-darwin.""
This reverts commit r271096, as reverting it broke even more buildbots!

But that also means I'll break on ARM again... :(

llvm-svn: 271099
2016-05-28 04:47:13 +00:00
Renato Golin 4f22c51b09 Revert "Map DynamicNoPIC to Static on non-darwin."
This reverts commit r271052, as it broke some ARM buildbots.

llvm-svn: 271096
2016-05-28 04:24:26 +00:00
Matt Arsenault 1ff389a7bf AMDGPU: Cleanup vector insert/extract tests
This mostly makes sure that 3-vector dynamic inserts
and extracts are covered.

llvm-svn: 271082
2016-05-28 00:51:06 +00:00
Matt Arsenault 7401516985 AMDGPU: Add fract intrinsic
Remove broken patterns matching it. This was matching the
unsafe math pattern and expanding the fix for the buggy instruction
from the pattern. The problems are also on CI. Remove the workarounds
and only use fract with unsafe math or from the intrinsic.

llvm-svn: 271078
2016-05-28 00:19:52 +00:00
Rafael Espindola f9bda6805b Map DynamicNoPIC to Static on non-darwin.
DynamicNoPIC was only every used on darwin. This maps it to static on
ELF. It matches what is done on X86.

llvm-svn: 271052
2016-05-27 21:44:18 +00:00
Michael Kuperstein a75c77b127 [X86] Detect SAD patterns and emit psadbw instructions.
This recommits r267649 with a fix for PR27539.

Differential Revision: http://reviews.llvm.org/D20598

llvm-svn: 271033
2016-05-27 18:53:22 +00:00
Simon Pilgrim 7e67a22298 [X86][AVX] Removed some remains of old (pre-regeneration) filechecks
llvm-svn: 271007
2016-05-27 15:56:19 +00:00
Than McIntosh 4daf7f13b6 Disable lifetime-start-on-first-use analysis.
Summary:
Turn off lifetime-start-on-first-use enhancement for the moment
pending a fix for bug 27903.

Bug: 27903

Reviewers: tejohnson, wmi, qcolombet, gbiv

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D20731

llvm-svn: 271003
2016-05-27 15:27:51 +00:00
Simon Pilgrim 4642a57fbf Revert: r270973 - [X86][SSE] Replace (V)PMOVSX and (V)PMOVZX integer extension intrinsics with generic IR (llvm)
llvm-svn: 270976
2016-05-27 09:02:25 +00:00
Simon Pilgrim c013e5737b [X86][SSE] Replace (V)PMOVSX and (V)PMOVZX integer extension intrinsics with generic IR (llvm)
This patch removes the llvm intrinsics VPMOVSX and (V)PMOVZX sign/zero extension intrinsics and auto-upgrades to SEXT/ZEXT calls instead. We already did this for SSE41 PMOVSX sometime ago so much of that implementation can be reused.

A companion patch (D20684) removes/auto-upgrade the clang intrinsics.

Differential Revision: http://reviews.llvm.org/D20686

llvm-svn: 270973
2016-05-27 08:49:15 +00:00
Mitch Bodart 05aeeb5cf1 [CodeGen] Fix problem with X86 byte registers in CriticalAntiDepBreaker
CriticalAntiDepBreaker was not correctly tracking defs of the high X86 byte
registers, leading to incorrect use of a busy register to break an
antidependence.

Fixes pr27681, and its duplicates pr27580, pr27804.

Differential Revision: http://reviews.llvm.org/D20456

llvm-svn: 270935
2016-05-26 23:08:52 +00:00
Krzysztof Parzyszek da0b9a959e [Hexagon] Enable the post-RA scheduler
The aggressive anti-dependency breaker can rename the restored callee-
saved registers. To prevent this, mark these registers are live on all
paths to the return/tail-call instructions, and add implicit use operands
for them to these instructions.

llvm-svn: 270898
2016-05-26 19:44:28 +00:00
Chad Rosier 14aa2ad1f4 [AArch64] Generate rev16/rev32 from bswap + srl when upper bits are known zero.
Canonicalize (srl (bswap i32 x), 16) to (rotr (bswap i32 x), 16), if the high
16-bits of x are zero. Similarly, canonicalize (srl (bswap i64 x), 32) to
(rotr (bswap i64 x), 32), if the high 32-bits of x are zero.

test_rev_w_srl16:            test_rev_w_srl16:
  and w8, w0, #0xffff          and     w8, w0, #0xffff
  rev w8, w8           --->    rev16   w0, w8
  lsr     w0, w8, #16

test_rev_x_srl32:            test_rev_x_srl32:
  rev x8, x8           --->    rev32   x0, x8
  lsr x0, x8, #32

llvm-svn: 270896
2016-05-26 19:41:33 +00:00
Changpeng Fang 71369b3a39 AMDGPU/SI: Enable load-store-opt by default.
Summary: Enable load-store-opt by default, and update LIT tests.

Reviewers: arsenm

Differential Revision: http://reviews.llvm.org/D20694

llvm-svn: 270894
2016-05-26 19:35:29 +00:00
Krzysztof Parzyszek 729e7ad31f Add test/CodeGen/MIR/Hexagon/lit.local.cfg
Require that Hexagon is a registered target.

llvm-svn: 270887
2016-05-26 18:35:45 +00:00
Krzysztof Parzyszek 143f684a79 Do not rename registers that do not start an independent live range
llvm-svn: 270885
2016-05-26 18:22:53 +00:00