This reverts commit r222632 (and follow-up r222636), which caused a host
of LNT failures on an internal bot. I'll respond to the commit on the
list with a reproduction of one of the failures.
Conflicts:
lib/Target/X86/X86TargetTransformInfo.cpp
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222936 91177308-0d34-0410-b5e6-96231b3b80d8
The AAPCS treats small structs and homogeneous floating (or vector) aggregates
specially, and guarantees they either get passed as a contiguous block of
registers, or prevent any future use of those registers and get passed on the
stack.
This concept can fit quite neatly into LLVM's own type system, mapping an HFA
to [N x float] and so on, and small structs to [N x i64]. Doing so allows
front-ends to emit AAPCS compliant code without having to duplicate the
register counting logic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222903 91177308-0d34-0410-b5e6-96231b3b80d8
The string data for string-valued build attributes were being unconditionally
uppercased. There is no mention in the ARM ABI addenda about case conventions,
so it's technically implementation defined as to whether the data are
capitialised in some way or not. However, there are good reasons not to
captialise the data.
* It's less work.
* Some vendors may legitimately have case-sensitive checks for these
attributes which would fail on LLVM generated object files.
* There could be locale issues with uppercasing.
The original reasons for uppercasing appear to have stemmed from an
old codesourcery toolchain behaviour, see
http://comments.gmane.org/gmane.comp.compilers.llvm.cvs/87133
This patch makes the object file emitted no longer captialise string
data, it encodes as seen in the assembly source.
Change-Id: Ibe20dd6e60d2773d57ff72a78470839033aa5538
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222882 91177308-0d34-0410-b5e6-96231b3b80d8
This mostly entails adding relocations, however there are a couple of
changes to existing relocations:
1. R_AARCH64_NONE is defined to be zero rather than 256
R_AARCH64_NONE has been defined to be zero for a long time elsewhere
e.g. binutils and glibc since the submission of the AArch64 port in
2012 so this is required for compatibility.
2. R_AARCH64_TLSDESC_ADR_PAGE renamed to R_AARCH64_TLSDESC_ADR_PAGE21
I don't think there is any way for relocation names to leak out of LLVM
so this should not break anything.
Tested with check-all with no regressions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222821 91177308-0d34-0410-b5e6-96231b3b80d8
including SAE mode and memory operand.
Added AVX512_maskable_scalar template, that should cover all scalar instructions in the future.
The main difference between AVX512_maskable_scalar<> and AVX512_maskable<> is using X86select instead of vselect.
I need it, because I can't create vselect node for MVT::i1 mask for scalar instruction.
http://reviews.llvm.org/D6378
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222820 91177308-0d34-0410-b5e6-96231b3b80d8
Since (v)pslldq / (v)psrldq instructions resolve to a single input argument it is useful to match it much earlier than we currently do - this prevents more complicated shuffles (notably insertion into a zero vector) matching before it.
Differential Revision: http://reviews.llvm.org/D6409
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222796 91177308-0d34-0410-b5e6-96231b3b80d8
The pattern matching failed to recognize all instances of "-1", because when
comparing against "-1" we didn't use an APInt of the same bitwidth.
This commit fixes this and also adds inverse versions of the conditon to catch
more cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222722 91177308-0d34-0410-b5e6-96231b3b80d8
This does not matter on newer cores (where we can use reciprocal estimates in
fast-math mode anyway), but for older cores this allows us to generate better
fast-math code where we have multiple FDIVs with a common divisor.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222710 91177308-0d34-0410-b5e6-96231b3b80d8
This patch teaches function 'transformVSELECTtoBlendVECTOR_SHUFFLE' how to
convert VSELECT dag nodes to shuffles on targets that do not have SSE4.1.
On pre-SSE4.1 targets, we can still perform blend operations using movss/movsd.
Also, removed a target specific combine that performed a premature lowering of
VSELECT nodes to target specific MOVSS/MOVSD nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222647 91177308-0d34-0410-b5e6-96231b3b80d8
r222375 made some improvements to build_vector lowering of v4x32 and v4xf32 into an insertps, but it missed a case where:
1. A single extracted element is used twice.
2. The lower of the two non-zero indexes should be preserved, and the higher should be used for the dest mask.
This caused a crash, since the source value for the insertps ends-up uninitialized.
Differential Revision: http://reviews.llvm.org/D6377
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222635 91177308-0d34-0410-b5e6-96231b3b80d8
Introduced new target-independent intrinsics in order to support masked vector loads and stores. The loop vectorizer optimizes loops containing conditional memory accesses by generating these intrinsics for existing targets AVX2 and AVX-512. The vectorizer asks the target about availability of masked vector loads and stores.
Added SDNodes for masked operations and lowering patterns for X86 code generator.
Examples:
<16 x i32> @llvm.masked.load.v16i32(i8* %addr, <16 x i32> %passthru, i32 4 /* align */, <16 x i1> %mask)
declare void @llvm.masked.store.v8f64(i8* %addr, <8 x double> %value, i32 4, <8 x i1> %mask)
Scalarizer for other targets (not AVX2/AVX-512) will be done in a separate patch.
http://reviews.llvm.org/D6191
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222632 91177308-0d34-0410-b5e6-96231b3b80d8
has a remarkably unique and efficient lowering.
While we get this some of the time already, we miss a few cases and
there wasn't a principled reason we got it. We should at least test
this. v8 already has tests for this pattern.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222607 91177308-0d34-0410-b5e6-96231b3b80d8
This s_mov_b32 will write to a virtual register from the M0Reg
class and all the ds instructions now take an extra M0Reg explicit
argument.
This change is necessary to prevent issues with the scheduler
mixing together instructions that expect different values in the m0
registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222583 91177308-0d34-0410-b5e6-96231b3b80d8
filler such as if delay slot filler have to put NOP instruction into the
delay slot of microMIPS BEQ or BNE instruction which uses the register $0,
then instead of emitting NOP this instruction is replaced by the corresponding
microMIPS compact branch instruction, i.e. BEQZC or BNEZC.
Differential Revision: http://reviews.llvm.org/D3566
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222580 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds a feature flag to avoid unaligned 32-byte load/store AVX codegen
for Sandy Bridge and Ivy Bridge. There is no functionality change intended for
those chips. Previously, the absence of AVX2 was being used as a proxy to detect
this feature. But that hindered codegen for AVX-enabled AMD chips such as btver2
that do not have the 32-byte unaligned access slowdown.
Performance measurements are included in PR21541 ( http://llvm.org/bugs/show_bug.cgi?id=21541 ).
Differential Revision: http://reviews.llvm.org/D6355
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222544 91177308-0d34-0410-b5e6-96231b3b80d8
shuffle lowering to allow much better blend matching.
Specifically, with the new structure the code seems clearer to me and we
correctly can hit the cases where merging two 128-bit lanes is a clear
win and can be shuffled cheaply afterward.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222539 91177308-0d34-0410-b5e6-96231b3b80d8
a bunch more improvements.
Non-lane-crossing is fine, the key is that lane merging only makes sense
for single-input shuffles. Not sure why I got so turned around here. The
code all works, I was just using the wrong model for it.
This only updates v4 and v8 lowering. The v16 and v32 lowering requires
restructuring the entire check sequence.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222537 91177308-0d34-0410-b5e6-96231b3b80d8
Before this patch, the DAGCombiner only tried to convert build_vector dag nodes
into shuffles if all operands were either extract_vector_elt or undef.
This patch improves that logic and teaches the DAGCombiner how to deal with
build_vector dag nodes where one or more operands are zero. A build_vector
dag node with some zero operands is turned into a shuffle only if the resulting
shuffle mask is legal for the target.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222536 91177308-0d34-0410-b5e6-96231b3b80d8
lanes.
By special casing these we can often either reduce the total number of
shuffles significantly or reduce the number of (high latency on Haswell)
AVX2 shuffles that potentially cross 128-bit lanes. Even when these
don't actually cross lanes, they have much higher latency to support
that. Doing two of them and a blend is worse than doing a single insert
across the 128-bit lanes to blend and then doing a single interleaved
shuffle.
While this seems like a narrow case, it kept cropping up on me and the
difference is *huge* as you can see in many of the test cases. I first
hit this trying to perfectly fix the interleaving shuffle patterns used
by Halide for AVX2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222533 91177308-0d34-0410-b5e6-96231b3b80d8
merging 128-bit subvectors and also shuffling all the elements of those
subvectors. Currently we generate pretty bad code for many of these, but
I'm testing a patch that should dramatically improve this in addition to
making the shuffle lowering robust to other changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222525 91177308-0d34-0410-b5e6-96231b3b80d8
This mirrors r222331, which enabled SeparateConstOffsetFromGEP on AArch64, in
the PowerPC backend. Yields, on a POWER7 machine, a 30% speedup on
SingleSource/Benchmarks/Shootout/nestedloop (this might just be from LICM,
there is a store moved out of the inner loop) and a potential speedup on
MultiSource/Benchmarks/mediabench/mpeg2/mpeg2dec/mpeg2decode. Regardless, it
makes some code look cleaner, and synchronizing the backends in this regard
seems like a generally good thing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222504 91177308-0d34-0410-b5e6-96231b3b80d8
Windows itanium targets the MSVCRT, and the stack probe symbol is provided by
MSVCRT. This corrects the emission of stack probes on i686-windows-itanium.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222439 91177308-0d34-0410-b5e6-96231b3b80d8
This patch improves the lowering of v4f32 and v4i32 build_vector dag nodes
that are known to have at least two non-zero elements.
With this patch, a build_vector that performs a blend with zero is
converted into a shuffle. This is done to let the shuffle legalizer expand
the dag node in a optimal way. For example, if we know that a build_vector
performs a blend with zero, we can try to lower it as a movq/blend instead of
always selecting an insertps.
This patch also improves the logic that lowers a build_vector into a insertps
with zero masking. See for example the extra test cases added to test sse41.ll.
Differential Revision: http://reviews.llvm.org/D6311
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222375 91177308-0d34-0410-b5e6-96231b3b80d8
A register operand that has a common sub-class with its instruction's
defined register class is not always legal. For example,
SReg_32 and M0Reg both have a common sub-class, but we can't
use an SReg_32 in instructions that expect a M0Reg.
This prevents the llvm.SI.sendmsg.ll test from failing when the fold
operand pass is added.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222368 91177308-0d34-0410-b5e6-96231b3b80d8
This patch builds on http://reviews.llvm.org/D5598 to perform byte rotation shuffles (lowerVectorShuffleAsByteRotate) on pre-SSSE3 (palignr) targets - pre-SSSE3 is only enabled on i8 and i16 vector targets where it is a more definite performance gain.
I've also added a separate byte shift shuffle (lowerVectorShuffleAsByteShift) that makes use of the ability of the SLLDQ/SRLDQ instructions to implicitly shift in zero bytes to avoid the need to create a zero register if we had used palignr.
Differential Revision: http://reviews.llvm.org/D5699
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222340 91177308-0d34-0410-b5e6-96231b3b80d8
SeparateConstOffsetFromGEP can gives more optimizaiton opportunities related to GEPs, which benefits EarlyCSE
and LICM. By enabling these passes we can have better address calculations and generate a better addressing
mode. Some SPEC 2006 benchmarks (astar, gobmk, namd) have obvious improvements on Cortex-A57.
Reviewed in http://reviews.llvm.org/D5864.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222331 91177308-0d34-0410-b5e6-96231b3b80d8
This partially makes up for not having address spaces
used for alias analysis in some simple cases.
This is not yet enabled by default so shouldn't change anything yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222286 91177308-0d34-0410-b5e6-96231b3b80d8
Under many circumstances the stack is not 32-byte aligned, resulting in the use of the vmovups/vmovupd/vmovdqu instructions when inserting ymm reloads/spills.
This minor patch adds these instructions to the isFrameLoadOpcode/isFrameStoreOpcode helpers so that they can be correctly identified and not be treated as folded reloads/spills.
This has also been noticed by http://llvm.org/bugs/show_bug.cgi?id=18846 where it was causing redundant spills - I've added a reduced test case at test/CodeGen/X86/pr18846.ll
Differential Revision: http://reviews.llvm.org/D6252
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222281 91177308-0d34-0410-b5e6-96231b3b80d8
shift-right for booleans (i1).
Arithmetic shift-right immediate with sign-/zero-extensions also works for
boolean values. Update the assert and the test cases to reflect that fact.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222272 91177308-0d34-0410-b5e6-96231b3b80d8
shift-right for booleans (i1).
Logical shift-right immediate with sign-/zero-extensions also works for boolean
values. Update the assert and the test cases to reflect that fact.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222270 91177308-0d34-0410-b5e6-96231b3b80d8
Shifts also perform sign-/zero-extends to larger types, which requires us to emit
an integer extend instead of a simple COPY.
Related to PR21594.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222257 91177308-0d34-0410-b5e6-96231b3b80d8
This should expose more of the actually used VALU
instructions to the machine optimization passes.
This also should help getting i1 handling into a better state.
For not entirly understood reasons, this fixes the split-scalar-i64-add.ll
test where a 64-bit add would only partially be moved to the VALU
resulting in use of undefined VCC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222256 91177308-0d34-0410-b5e6-96231b3b80d8
This change emits a COPY for a shift-immediate with a "zero" shift value.
This fixes PR21594 where we emitted a shift instruction with an incorrect
immediate operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222247 91177308-0d34-0410-b5e6-96231b3b80d8
The triple parser should only accept existing architecture names
when the triple starts with armv, armebv, thumbv or thumbebv.
Patch by Gabor Ballabas.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222129 91177308-0d34-0410-b5e6-96231b3b80d8
This was motivated by a bug which caused code like this to be
miscompiled:
declare void @take_ptr(i8*)
define void @test() {
%addr1.32 = alloca i8
%addr2.32 = alloca i32, i32 1028
call void @take_ptr(i8* %addr1)
ret void
}
This was emitting the following assembly to get the value of %addr1:
add r0, sp, #1020
add r0, r0, #8
However, "add r0, r0, #8" is not a valid Thumb1 instruction, and this
could not be assembled. The generated object file contained this,
resulting in r0 holding SP+8 rather tha SP+1028:
add r0, sp, #1020
add r0, sp, #8
This function looked like it could have caused miscompilations for
other combinations of registers and offsets (though I don't think it is
currently called with these), and the heuristic it used did not match
the emitted code in all cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222125 91177308-0d34-0410-b5e6-96231b3b80d8
Some optimisations in DAGCombiner cause miscompilations for targets that use
TargetLowering::UndefinedBooleanContent, because they assume that the results
of a SELECT_CC node are boolean values, and can be safely ANDed, ORed and
XORed. These optimisations are only valid for targets that use
ZeroOrOneBooleanContent or ZeroOrNegativeOneBooleanContent.
This is a follow-up to D6210/r221693.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222123 91177308-0d34-0410-b5e6-96231b3b80d8
This patch teaches the DAGCombiner how to combine shuffles according to rules:
shuffle(shuffle(A, Undef, M0), B, M1) -> shuffle(B, A, M2)
shuffle(shuffle(A, B, M0), B, M1) -> shuffle(B, A, M2)
shuffle(shuffle(A, B, M0), A, M1) -> shuffle(B, A, M2)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222090 91177308-0d34-0410-b5e6-96231b3b80d8
Updated X86TargetLowering::isShuffleMaskLegal to match SHUFP masks with commuted inputs and PSHUFD masks that reference the second input.
As part of this I've refactored isPSHUFDMask to work in a more general manner and allow it to match against either the first or second input vector.
Differential Revision: http://reviews.llvm.org/D6287
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222087 91177308-0d34-0410-b5e6-96231b3b80d8
This gets the correct NaN behavior based on the compare type
the hardware uses. This now passes the new piglit test I have
for this on SI.
Add stricter tests for the operand order.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222079 91177308-0d34-0410-b5e6-96231b3b80d8
We use to track quite a few "adjusted" offsets through the FrameLowering code
to account for changes in the prologue instructions as we went and allow the
emission of correct CFA annotations. However, we were missing a couple of cases
and the code was almost impenetrable.
It's easier to just add any stack-adjusting instruction to a list and emit them
together.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222057 91177308-0d34-0410-b5e6-96231b3b80d8
When we folded the DPR alignment gap into a push, we weren't noting the extra
distance from the beginning of the push to the FP, and so FP ended up pointing
at an incorrect offset.
The .cfi_def_cfa_offset directives are still wrong in this case, but I think
that can be improved by refactoring.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222056 91177308-0d34-0410-b5e6-96231b3b80d8
The test's DWARF stubs were there just to trigger the emission of .cfi
directives. Fortunately, the NetBSD ABI already demands proper DWARF unwind
info, so it's easier to just use that triple.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222055 91177308-0d34-0410-b5e6-96231b3b80d8
If we have spilled the value of the m0 register, then we need to restore
it with v_readlane_b32 to a regular sgpr, because v_readlane_b32 can't
write to m0.
v_readlane_b32 can't write to m0, so
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222036 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds builtin support for xvdivdp and xvdivsp, along with a
test case. Straightforward stuff.
There's a companion patch for Clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221983 91177308-0d34-0410-b5e6-96231b3b80d8
getTargetConstant should only be used when you can guarantee the instruction
selected will be able to cope with the raw value. BUILD_VECTOR is rather too
generic for this so we should use getConstant instead. In that case, an
instruction can still consume the constant, but if it doesn't it'll be
materialised through its own round of ISel.
Should fix PR21352.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221961 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
This has most of what is needed for mips fast-isel call lowering for O32.
What is missing I will add on the next patch because this patch is already too large.
It should not be doing anything wrong but it will punt on some cases that it is basically
capable of doing.
The mechanism is there for parameters to be passed on the stack but I have not enabled it because it serves as a way for now to prevent some of the strange cases of O32 register passing that I have not fully checked yet and have some issues.
The Mips O32 abi rules are very complicated as far how data is passed in floating and integer registers.
However there is a way to think about this all very simply and this implementation reflects that.
Basically, the ABI rules are written as if everything is passed on the stack and aligned as such.
Once that is conceptually done, it is nearly trivial to reassign those locations to registers and
then all the complexity disappears.
So I have told tablegen that all the data is passed on the stack and during the lowering I fix
this by assigning to registers as per the ABI doc.
This has been my approach and you can line up what I did with the ABI document and see 1 to 1 what
is going on.
Test Plan: callabi.ll
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: jholewinski, echristo, ahatanak, llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D5714
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221948 91177308-0d34-0410-b5e6-96231b3b80d8
in-lane shuffles that aren't always handled well by the current vector
shuffle lowering.
No functionality change yet, that will follow in a subsequent commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221938 91177308-0d34-0410-b5e6-96231b3b80d8
The generic FastISel code would bail, because it can't emit a sign-extend for
AArch64. This copies the code over and uses AArch64 specific emit functions.
This is not ideal and 'computeAddress' should handles this, so it can fold the
address computation into the memory operation.
I plan to clean up 'computeAddress' anyways, so I will add that in a future
commit.
Related to rdar://problem/18962471.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221923 91177308-0d34-0410-b5e6-96231b3b80d8
If a function is just an unreachable, this would hit a
"this is not a MachO target" assertion because of setting
HasSubsectionViaSymbols.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221920 91177308-0d34-0410-b5e6-96231b3b80d8
e.g. v_mad_f32 a, b, c -> v_mad_f32 b, a, c
This simplifies matching v_madmk_f32.
This looks somewhat surprising, but it appears to be
OK to do this. We can commute src0 and src1 in all
of these instructions, and that's all that appears
to matter.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221910 91177308-0d34-0410-b5e6-96231b3b80d8
Normally entries can only move to a lower address, but when that wasn't viable,
the user's block was considered anyway. Unfortunately, it went via
createNewWater which wasn't designed to handle the case where there's already
an island after the block.
Unfortunately, the test we have is slow and fragile, and I couldn't reduce it
to anything sane even with the @llvm.arm.space intrinsic. The test change here
is recreating the previous one after the change.
rdar://problem/18545506
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221905 91177308-0d34-0410-b5e6-96231b3b80d8
We were using a naive heuristic to determine whether a basic block already had
an unconditional branch at the end. This mostly corresponded to reality
(assuming branches got optimised) because there's not much point in a branch to
the next block, but could go wrong.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221904 91177308-0d34-0410-b5e6-96231b3b80d8
Creating tests for the ConstantIslands pass is very difficult, since it depends
on precise layout details. Having the ability to precisely inject a number of
bytes into the stream helps greatly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221903 91177308-0d34-0410-b5e6-96231b3b80d8
between splitting a vector into 128-bit lanes and recombining them vs.
decomposing things into single-input shuffles and a final blend.
This handles a large number of cases in AVX1 where the cross-lane
shuffles would be much more expensive to represent even though we end up
with a fast blend at the root. Instead, we can do a better job of
shuffling in a single lane and then inserting it into the other lanes.
This fixes the remaining bits of Halide's regression captured in PR21281
for AVX1. However, the bug persists in AVX2 because I've made this
change reasonably conservative. The cases where it makes sense in AVX2
to split into 128-bit lanes are much more rare because we can often do
full permutations across all elements of the 256-bit vector. However,
the particular test case in PR21281 is an example of one of the rare
cases where it is *always* better to work in a single 128-bit lane. I'm
going to try to teach the logic to detect and form the good code even in
AVX2 next, but it will need to use a separate heuristic.
Finally, there is one pesky regression here where we previously would
craftily use vpermilps in AVX1 to shuffle both high and low halves at
the same time. We no longer pull that off, and not for any really good
reason. Ultimately, I think this is just another missing nuance to the
selection heuristic that I'll try to add in afterward, but this change
already seems strictly worth doing considering the magnitude of the
improvements in common matrix math shuffle patterns.
As always, please let me know if this causes a surprising regression for
you.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221861 91177308-0d34-0410-b5e6-96231b3b80d8
re-combining shuffles because nothing was available in the wider vector
type.
The key observation (which I've put in the comments for future
maintainers) is that at this point, no further combining is really
possible. And so even though these shuffles trivially could be combined,
we need to actually do that as we produce them when producing them this
late in the lowering.
This fixes another (huge) part of the Halide vector shuffle regressions.
As it happens, this was already well covered by the tests, but I hadn't
noticed how bad some of these got. The specific patterns that turn
directly into unpckl/h patterns were occurring *many* times in common
vector processing code.
There are still more problems here sadly, but trying to incrementally
tease them apart and it looks like this is the core of the problem in
the splitting logic.
There is some chance of regression here, you can see it in the test
changes. Specifically, where we stop forming pshufb in some cases, it is
possible that pshufb was in fact faster. Intel "says" that pshufb is
slower than the instruction sequences replacing it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221852 91177308-0d34-0410-b5e6-96231b3b80d8
Prior to this patch the TypePromotionHelper was promoting only sign extensions.
Supporting zero extensions changes:
- How constants are extended.
- How sign extensions, zero extensions, and truncate are composed together.
- How the type of the extended operation is recorded. Now we need to know the
kind of the extension as well as its type.
Each change is fairly small, unlike the diff.
Most of the diff are comments/variable renaming to say "extension" instead of
"sign extension".
The performance improvements on the test suite are within the noise.
Related to <rdar://problem/18310086>.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221851 91177308-0d34-0410-b5e6-96231b3b80d8
Optimize selects of i1 in the presence of 'true' and 'false' operands to simple
logic operations.
This fixes rdar://problem/18960150.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221848 91177308-0d34-0410-b5e6-96231b3b80d8
This folds the compare emission into the select emission when possible, so we
can directly use the flags and don't have to emit a separate compare.
Related to rdar://problem/18960150.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221847 91177308-0d34-0410-b5e6-96231b3b80d8
This is a follow-on to r221706 and r221731 and discussed in more detail in PR21385.
This patch also loosens the testcase checking for btver2. We know that the "1.0" will be loaded, but
we can't tell exactly when, so replace the CHECK-NEXT specifiers with plain CHECKs. The CHECK-NEXT
sequence relied on a quirk of post-RA-scheduling that may change independently of anything in these tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221819 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Large-model was added first. With the addition of support for multiple PIC
models in LLVM, now add small-model PIC for 32-bit PowerPC, SysV4 ABI. This
generates more optimal code, for shared libraries with less than about 16380
data objects.
Test Plan: Test cases added or updated
Reviewers: joerg, hfinkel
Reviewed By: hfinkel
Subscribers: jholewinski, mcrosier, emaste, llvm-commits
Differential Revision: http://reviews.llvm.org/D5399
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221791 91177308-0d34-0410-b5e6-96231b3b80d8
cases from Halide folks. This initial step was extracted from
a prototype change by Clay Wood to try and address regressions found
with Halide and the new vector shuffle lowering.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221779 91177308-0d34-0410-b5e6-96231b3b80d8
removes windows line endings and other noise. This is in prelude to
making substantive changes to these tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221776 91177308-0d34-0410-b5e6-96231b3b80d8
This patch enables the vec_vsx_ld and vec_vsx_st intrinsics for
PowerPC, which provide programmer access to the lxvd2x, lxvw4x,
stxvd2x, and stxvw4x instructions.
New LLVM intrinsics are provided to represent these four instructions
in IntrinsicsPowerPC.td. These are patterned after the similar
intrinsics for lvx and stvx (Altivec). In PPCInstrVSX.td, these
intrinsics are tied to the code gen patterns, with additional patterns
to allow plain vanilla loads and stores to still generate these
instructions.
At -O1 and higher the intrinsics are immediately converted to loads
and stores in InstCombineCalls.cpp. This will open up more
optimization opportunities while still allowing the correct
instructions to be generated. (Similar code exists for aligned
Altivec loads and stores.)
The new intrinsics are added to the code that checks for consecutive
loads and stores in PPCISelLowering.cpp, as well as to
PPCTargetLowering::getTgtMemIntrinsic().
There's a new test to verify the correct instructions are generated.
The loads and stores tend to be reordered, so the test just counts
their number. It runs at -O2, as it's not very effective to test this
at -O0, when many unnecessary loads and stores are generated.
I ended up having to modify vsx-fma-m.ll. It turns out this test case
is slightly unreliable, but I don't know a good way to prevent
problems with it. The xvmaddmdp instructions read and write the same
register, which is one of the multiplicands. Commutativity allows
either to be chosen. If the FMAs are reordered differently than
expected by the test, the register assignment can be different as a
result. Hopefully this doesn't change often.
There is a companion patch for Clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221767 91177308-0d34-0410-b5e6-96231b3b80d8
This commit adds a new pass that can inject checks before indirect calls to
make sure that these calls target known locations. It supports three types of
checks and, at compile time, it can take the name of a custom function to call
when an indirect call check fails. The default failure function ignores the
error and continues.
This pass incidentally moves the function JumpInstrTables::transformType from
private to public and makes it static (with a new argument that specifies the
table type to use); this is so that the CFI code can transform function types
at call sites to determine which jump-instruction table to use for the check at
that site.
Also, this removes support for jumptables in ARM, pending further performance
analysis and discussion.
Review: http://reviews.llvm.org/D4167
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221708 91177308-0d34-0410-b5e6-96231b3b80d8
This is a first step for generating SSE rcp instructions for reciprocal
calcs when fast-math allows it. This is very similar to the rsqrt optimization
enabled in D5658 ( http://reviews.llvm.org/rL220570 ).
For now, be conservative and only enable this for AMD btver2 where performance
improves significantly both in terms of latency and throughput.
We may never enable this codegen for Intel Core* chips because the divider circuits
are just too fast. On SandyBridge, divss can be as fast as 10 cycles versus the 21
cycle critical path for the rcp + mul + sub + mul + add estimate.
Follow-on patches may allow configuration of the number of Newton-Raphson refinement
steps, add AVX512 support, and enable the optimization for more chips.
More background here: http://llvm.org/bugs/show_bug.cgi?id=21385
Differential Revision: http://reviews.llvm.org/D6175
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221706 91177308-0d34-0410-b5e6-96231b3b80d8
My original support for the general dynamic and local dynamic TLS
models contained some fairly obtuse hacks to generate calls to
__tls_get_addr when lowering a TargetGlobalAddress. Rather than
generating real calls, special GET_TLS_ADDR nodes were used to wrap
the calls and only reveal them at assembly time. I attempted to
provide correct parameter and return values by chaining CopyToReg and
CopyFromReg nodes onto the GET_TLS_ADDR nodes, but this was also not
fully correct. Problems were seen with two back-to-back stores to TLS
variables, where the call sequences ended up overlapping with unhappy
results. Additionally, since these weren't real calls, the proper
register side effects of a call were not recorded, so clobbered values
were kept live across the calls.
The proper thing to do is to lower these into calls in the first
place. This is relatively straightforward; see the changes to
PPCTargetLowering::LowerGlobalTLSAddress() in PPCISelLowering.cpp.
The changes here are standard call lowering, except that we need to
track the fact that these calls will require a relocation. This is
done by adding a machine operand flag of MO_TLSLD or MO_TLSGD to the
TargetGlobalAddress operand that appears earlier in the sequence.
The calls to LowerCallTo() eventually find their way to
LowerCall_64SVR4() or LowerCall_32SVR4(), which call FinishCall(),
which calls PrepareCall(). In PrepareCall(), we detect the calls to
__tls_get_addr and immediately snag the TargetGlobalTLSAddress with
the annotated relocation information. This becomes an extra operand
on the call following the callee, which is expected for nodes of type
tlscall. We change the call opcode to CALL_TLS for this case. Back
in FinishCall(), we change it again to CALL_NOP_TLS for 64-bit only,
since we require a TOC-restore nop following the call for the 64-bit
ABIs.
During selection, patterns in PPCInstrInfo.td and PPCInstr64Bit.td
convert the CALL_TLS nodes into BL_TLS nodes, and convert the
CALL_NOP_TLS nodes into BL8_NOP_TLS nodes. This replaces the code
removed from PPCAsmPrinter.cpp, as the BL_TLS or BL8_NOP_TLS
nodes can now be emitted normally using their patterns and the
associated printTLSCall print method.
Finally, as a result of these changes, all references to get-tls-addr
in its various guises are no longer used, so they have been removed.
There are existing TLS tests to verify the changes haven't messed
anything up). I've added one new test that verifies that the problem
with the original code has been fixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221703 91177308-0d34-0410-b5e6-96231b3b80d8
The ISel lowering for global TLS access in PIC mode was creating a pseudo
instruction that is later expanded to a call, but the code was not
setting the hasCalls flag in the MachineFrameInfo alongside the adjustsStack
flag. This caused some functions to be mistakenly recognized as leaf functions,
and this in turn affected the decision to eliminate the frame pointer.
With the fix, hasCalls is properly set and the leaf frame pointer is correctly
preserved.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221695 91177308-0d34-0410-b5e6-96231b3b80d8
LLVM replaces the SelectionDAG pattern (xor (set_cc cc x y) 1) with
(set_cc !cc x y), which is only correct when the xor has type i1.
Instead, we should check that the constant operand to the xor is all
ones.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221693 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
This patch enables code generation for the MIPS II target. Pre-Mips32
targets don't have the MUL instruction, so we add the correspondent
pattern that uses the MULT/MFLO combination in order to retrieve the
product.
This is WIP as we don't support code generation for select nodes due to
the lack of conditional-move instructions.
Reviewers: dsanders
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D6150
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221686 91177308-0d34-0410-b5e6-96231b3b80d8
This fixes an issue with matching trunc -> assertsext -> zext on x86-64, which would not zero the high 32-bits. See PR20494 for details.
Recommitting - This time, with a hopefully working test.
Differential Revision: http://reviews.llvm.org/D6128
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221672 91177308-0d34-0410-b5e6-96231b3b80d8
AVX2 is available.
According to IACA, the new lowering has a throughput of 8 cycles instead of 13
with the previous one.
Althought this lowering kicks in some SPECs benchmarks, the performance
improvement was within the noise.
Correctness testing has been done for the whole range of uint32_t with the
following program:
uint4 v = (uint4) {0,1,2,3};
uint32_t i;
//Check correctness over entire range for uint4 -> float4 conversion
for( i = 0; i < 1U << (32-2); i++ )
{
float4 t = test(v);
float4 c = correct(v);
if( 0xf != _mm_movemask_ps( t == c ))
{
printf( "Error @ %vx: %vf vs. %vf\n", v, c, t);
return -1;
}
v += 4;
}
Where "correct" is the old lowering and "test" the new one.
The patch adds a test case for the two custom lowering instruction.
It also modifies the vector cost model, which is why cast.ll and uitofp.ll are
modified.
2009-02-26-MachineLICMBug.ll is also modified because we now hoist 7
instructions instead of 4 (3 more constant loads).
rdar://problem/18153096>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221657 91177308-0d34-0410-b5e6-96231b3b80d8
In the case we optimize an integer extend away and replace it directly with the
source register, we also have to clear all kill flags at all its uses.
This is necessary, because the orignal IR instruction might be trivially dead,
but we replaced it with a nop at MI level.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221628 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
... and after all that refactoring, it's possible to distinguish softfloat
floating point values from integers so this patch no longer breaks softfloat to
do it.
Remove direct handling of i32's in the N32/N64 ABI by promoting them to
i64. This more closely reflects the ABI documentation and also fixes
problems with stack arguments on big-endian targets.
We now rely on signext/zeroext annotations (already generated by clang) and
the Assert[SZ]ext nodes to avoid the introduction of unnecessary sign/zero
extends.
It was not possible to convert three tests to use signext/zeroext. These tests
are bswap.ll, ctlz-v.ll, ctlz-v.ll. It's not possible to put signext on a
vector type so we just accept the sign extends here for now. These tests don't
pass the vectors the same way clang does (clang puts multiple elements in the
same argument, these map 1 element to 1 argument) so we don't need to worry too
much about it.
With this patch, all known N32/N64 bugs should be fixed and we now pass the
first 10,000 tests generated by ABITest.py.
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D6117
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221534 91177308-0d34-0410-b5e6-96231b3b80d8
Reversing a CB* instruction used to drop the flags on the condition. On the
included testcase, this lead to a read from an undefined vreg.
Using addOperand keeps the flags, here <undef>.
Differential Revision: http://reviews.llvm.org/D6159
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221507 91177308-0d34-0410-b5e6-96231b3b80d8
Fixed an issue with the (v)cvttps2dq and (v)cvttpd2dq instructions being incorrectly put in the 2 source operand folding tables instead of the 1 source operand and added the missing SSE/AVX versions.
Also added missing (v)cvtps2dq and (v)cvtpd2dq instructions to the folding tables.
Differential Revision: http://reviews.llvm.org/D6001
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221489 91177308-0d34-0410-b5e6-96231b3b80d8
This test case was never actually testing the trivial spiller: the -spiller
option has not been hooked up for a while now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221475 91177308-0d34-0410-b5e6-96231b3b80d8
On 32 bit windows we use label differences and .set does not suppress
rolocations, a combination that was not used before r220256.
This fixes PR21497.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221456 91177308-0d34-0410-b5e6-96231b3b80d8
Example:
define <4 x i32> @test(<4 x i32> %a, <4 x i32> %b) {
%shuffle = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 5, i32 6, i32 3>
ret <4 x i32> %shuffle
}
Before llc (-mattr=+sse4.1), produced the following assembly instruction:
pblendw $4294967103, %xmm1, %xmm0
After
pblendw $63, %xmm1, %xmm0
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221455 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Currently, we give an error if %z is used with non-immediates, instead of continuing as if the %z isn't there.
For example, you use the %z operand modifier along with the "Jr" constraints ("r" makes the operand a register, and "J" makes it an immediate, but only if its value is 0).
In this case, you want the compiler to print "$0" if the inline asm input operand turns out to be an immediate zero and you want it to print the register containing the operand, if it's not.
We give an error in the latter case, and we shouldn't (GCC also doesn't).
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D6023
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221453 91177308-0d34-0410-b5e6-96231b3b80d8
If a section cannot be dead stripped, it is safe to use L symbols, since
the linker will keep all of it in the end.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221431 91177308-0d34-0410-b5e6-96231b3b80d8
condition to match a blend.
This prevents optimizations that work on VSELECT to perform invalid
transformations. Indeed, the optimized condition does not match the vector
boolean content that is expected and bad things may happen.
This patch yields the exact same code on the whole test-suite + specs (-O3 and
-O3 -march=core-avx2), it improves one test case (vector-blend.ll) and fixes a
bug reduced in vselect-avx.ll.
<rdar://problem/18819506>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221429 91177308-0d34-0410-b5e6-96231b3b80d8
Added missing memory folding for the (V)CVTDQ2PS instructions - we can safely fold these (but not the (V)CVTDQ2PD versions which have a register/memory size discrepancy in the source operand). I've added a test case demonstrating that stack folding now works.
Differential Revision: http://reviews.llvm.org/D5981
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221407 91177308-0d34-0410-b5e6-96231b3b80d8
This works around the limitation that PTX does not allow .param space
loads/stores with arbitrary pointers.
If a function has a by-val struct ptr arg, say foo(%struct.x *byval %d), then
add the following instructions to the first basic block :
%temp = alloca %struct.x, align 8
%tt1 = bitcast %struct.x * %d to i8 *
%tt2 = llvm.nvvm.cvt.gen.to.param %tt2
%tempd = bitcast i8 addrspace(101) * to %struct.x addrspace(101) *
%tv = load %struct.x addrspace(101) * %tempd
store %struct.x %tv, %struct.x * %temp, align 8
The above code allocates some space in the stack and copies the incoming
struct from param space to local space. Then replace all occurences of %d
by %temp.
Fixes PR21465.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221377 91177308-0d34-0410-b5e6-96231b3b80d8
This matches the format produced by the AMD proprietary driver.
//==================================================================//
// Shell script for converting .ll test cases: (Pass the .ll files
you want to convert to this script as arguments).
//==================================================================//
; This was necessary on my system so that A-Z in sed would match only
; upper case. I'm not sure why.
export LC_ALL='C'
TEST_FILES="$*"
MATCHES=`grep -v Patterns SIInstructions.td | grep -o '"[A-Z0-9_]\+["e]' | grep -o '[A-Z0-9_]\+' | sort -r`
for f in $TEST_FILES; do
# Check that there are SI tests:
grep -q -e 'verde' -e 'bonaire' -e 'SI' -e 'tahiti' $f
if [ $? -eq 0 ]; then
for match in $MATCHES; do
sed -i -e "s/\([ :]$match\)/\L\1/" $f
done
# Try to get check lines with partial instruction names
sed -i 's/\(;[ ]*SI[A-Z\\-]*: \)\([A-Z_0-9]\+\)/\1\L\2/' $f
fi
done
sed -i -e 's/bb0_1/BB0_1/g' ../../../test/CodeGen/R600/infinite-loop.ll
sed -i -e 's/SI-NOT: bfe/SI-NOT: {{[^@]}}bfe/g'../../../test/CodeGen/R600/llvm.AMDGPU.bfe.*32.ll ../../../test/CodeGen/R600/sext-in-reg.ll
sed -i -e 's/exp_IEEE/EXP_IEEE/g' ../../../test/CodeGen/R600/llvm.exp2.ll
sed -i -e 's/numVgprs/NumVgprs/g' ../../../test/CodeGen/R600/register-count-comments.ll
sed -i 's/\(; CHECK[-NOT]*: \)\([A-Z_0-9]\+\)/\1\L\2/' ../../../test/CodeGen/R600/select64.ll ../../../test/CodeGen/R600/sgpr-copy.ll
//==================================================================//
// Shell script for converting .td files (run this last)
//==================================================================//
export LC_ALL='C'
sed -i -e '/Patterns/!s/\("[A-Z0-9_]\+[ "e]\)/\L\1/g' SIInstructions.td
sed -i -e 's/"EXP/"exp/g' SIInstrInfo.td
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221350 91177308-0d34-0410-b5e6-96231b3b80d8
This patch improves the folding of vector AND nodes into blend operations for
targets that feature SSE4.1. A vector AND node where one of the operands is
a constant build_vector with elements that are either zero or all-ones can be
converted into a blend.
This allows for example to simplify the following code:
define <4 x i32> @test(<4 x i32> %A, <4 x i32> %B) {
%1 = and <4 x i32> %A, <i32 0, i32 0, i32 0, i32 -1>
%2 = and <4 x i32> %B, <i32 -1, i32 -1, i32 -1, i32 0>
%3 = or <4 x i32> %1, %2
ret <4 x i32> %3
}
Before this patch llc (-mcpu=corei7) generated:
andps LCPI1_0(%rip), %xmm0, %xmm0
andps LCPI1_1(%rip), %xmm1, %xmm1
orps %xmm1, %xmm0, %xmm0
retq
With this patch we generate a single 'vpblendw'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221343 91177308-0d34-0410-b5e6-96231b3b80d8
We currently try to push an even number of registers to preserve 8-byte
alignment during a function's prologue, but only when the stack alignment is
prcisely 8. Many of the reasons for doing it apply also when that alignment > 8
(the extra store is often free, and can save another stack adjustment, though
less frequently for 16-byte stack alignment).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221321 91177308-0d34-0410-b5e6-96231b3b80d8
We were making an attempt to do this by adding an extra callee-saved GPR (so
that there was an even number in the list), but when that failed we went ahead
and pushed anyway.
This had a couple of potential issues:
+ The .cfi directives we emit misplaced dN because they were based on
PrologEpilogInserter's calculation.
+ Unaligned stores can be less efficient.
+ Unaligned stores can actually fault (likely only an issue in niche cases,
but possible).
This adds a final explicit stack adjustment if all other options fail, so that
the actual locations of the registers match up with where they should be.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221320 91177308-0d34-0410-b5e6-96231b3b80d8
Patch to allow (v)blendps, (v)blendpd, (v)pblendw and vpblendd instructions to be commuted - swaps the src registers and inverts the blend mask.
This is primarily to improve memory folding (see new tests), but it also improves the quality of shuffles (see modified tests).
Differential Revision: http://reviews.llvm.org/D6015
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221313 91177308-0d34-0410-b5e6-96231b3b80d8
While fixing up the register classes in the machine combiner in a previous
commit I missed one.
This fixes the last one and adds a test case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221308 91177308-0d34-0410-b5e6-96231b3b80d8
Registers are not all equal. Some are not allocatable (infinite cost),
some have to be preserved but can be used, and some others are just free
to use.
Ensure there is a cost hierarchy reflecting this fact, so that the
allocator will favor scratch registers over callee-saved registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221293 91177308-0d34-0410-b5e6-96231b3b80d8
the tombstone or empty keys of a DenseMap<int64_t, T>. This patch
fixes the issue (and adds a tests case).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221214 91177308-0d34-0410-b5e6-96231b3b80d8
register class tGPRRegClass if the target is thumb1.
This commit fixes a crash that occurs during register allocation which was
triggered when a virtual register defined by an inline-asm instruction had to
be spilled.
rdar://problem/18740489
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221178 91177308-0d34-0410-b5e6-96231b3b80d8
For 8-bit divrems where the remainder is used, we used to generate:
divb %sil
shrw $8, %ax
movzbl %al, %eax
That was to avoid an H-reg access, which is problematic mainly because
it isn't possible in REX-prefixed instructions.
This patch optimizes that to:
divb %sil
movzbl %ah, %eax
To do that, we explicitly extend AH, and extract the L-subreg in the
resulting register. The extension is done using the NOREX variants of
MOVZX. To support signed operations, MOVSX_NOREX is also added.
Further, this introduces a new SDNode type, [us]divrem_ext_hreg, which is
then lowered to a sequence containing a single zext (rather than 2).
Differential Revision: http://reviews.llvm.org/D6064
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221176 91177308-0d34-0410-b5e6-96231b3b80d8
Function calls aren't supported yet.
This was reverted due to build breakages, which should be fixed now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221173 91177308-0d34-0410-b5e6-96231b3b80d8
call DAGCombiner. But we ran into a case (on Windows) where the
calling convention causes argument lowering to bail out of fast-isel,
and we end up in CodeGenAndEmitDAG() which does run DAGCombiner.
So, we need to make DAGCombiner check for 'optnone' after all.
Commit includes the test that found this, plus another one that got
missed in the original optnone work.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221168 91177308-0d34-0410-b5e6-96231b3b80d8
This CPU definition is redundant. The Cortex-A9 is defined as
supporting multiprocessing extensions. Remove its definition and
update appropriate tests.
LLVM defines both a cortex-a9 CPU and a cortex-a9-mp CPU. The only
difference between the two CPU definitions in ARM.td is that
cortex-a9-mp contains the feature FeatureMP for multiprocessing
extensions.
This is redundant since the Cortex-A9 is defined as having
multiprocessing extensions in the TRMs. armcc also defines the
Cortex-A9 as having multiprocessing extensions by default.
Change-Id: Ifcadaa6c322be0a33d9d2a39cfdd7da1d75981a7
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221166 91177308-0d34-0410-b5e6-96231b3b80d8
Some literals in the AArch64 backend had 15 'f's rather than 16, causing
comparisons with a constant 0xffffffffffffffff to be miscompiled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221157 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r221028. Later commits depend on this and
reverting just this one causes even more bots to fail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221041 91177308-0d34-0410-b5e6-96231b3b80d8
"[x86] Simplify vector selection if condition value type matches vselect value type and true value is all ones or false value is all zeros."
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221028 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r220996.
It introduced layering violations causing link errors in many
configurations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221020 91177308-0d34-0410-b5e6-96231b3b80d8
We need to figure out how to track ptrtoint values all the
way until result is converted back to a pointer in order
to correctly rewrite the pointer type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220997 91177308-0d34-0410-b5e6-96231b3b80d8
Now that we have initial support for VSX, we can begin adding
intrinsics for programmer access to VSX instructions. This patch adds
basic support for VSX intrinsics in general, and tests it by
implementing intrinsics for minimum and maximum for the vector double
data type.
The LLVM portion of this is quite straightforward. There is a
companion patch for Clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220988 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds an optimization in CodeGenPrepare to move an extractelement
right before a store when the target can combine them.
The optimization may promote any scalar operations to vector operations in the
way to make that possible.
** Context **
Some targets use different register files for both vector and scalar operations.
This means that transitioning from one domain to another may incur copy from one
register file to another. These copies are not coalescable and may be expensive.
For example, according to the scheduling model, on cortex-A8 a vector to GPR
move is 20 cycles.
** Motivating Example **
Let us consider an example:
define void @foo(<2 x i32>* %addr1, i32* %dest) {
%in1 = load <2 x i32>* %addr1, align 8
%extract = extractelement <2 x i32> %in1, i32 1
%out = or i32 %extract, 1
store i32 %out, i32* %dest, align 4
ret void
}
As it is, this IR generates the following assembly on armv7:
vldr d16, [r0] @vector load
vmov.32 r0, d16[1] @ cross-register-file copy: 20 cycles
orr r0, r0, #1 @ scalar bitwise or
str r0, [r1] @ scalar store
bx lr
Whereas we could generate much faster code:
vldr d16, [r0] @ vector load
vorr.i32 d16, #0x1 @ vector bitwise or
vst1.32 {d16[1]}, [r1:32] @ vector extract + store
bx lr
Half of the computation made in the vector is useless, but this allows to get
rid of the expensive cross-register-file copy.
** Proposed Solution **
To avoid this cross-register-copy penalty, we promote the scalar operations to
vector operations. The penalty will be removed if we manage to promote the whole
chain of computation in the vector domain.
Currently, we do that only when the chain of computation ends by a store and the
target is able to combine an extract with a store.
Stores are the most likely candidates, because other instructions produce values
that would need to be promoted and so, extracted as some point[1]. Moreover,
this is customary that targets feature stores that perform a vector extract (see
AArch64 and X86 for instance).
The proposed implementation relies on the TargetTransformInfo to decide whether
or not it is beneficial to promote a chain of computation in the vector domain.
Unfortunately, this interface is rather inaccurate for this level of details and
although this optimization may be beneficial for X86 and AArch64, the inaccuracy
will lead to the optimization being too aggressive.
Basically in TargetTransformInfo, everything that is legal has a cost of 1,
whereas, even if a vector type is legal, usually a vector operation is slightly
more expensive than its scalar counterpart. That will lead to too many
promotions that may not be counter balanced by the saving of the
cross-register-file copy. For instance, on AArch64 this penalty is just 4
cycles.
For now, the optimization is just enabled for ARM prior than v8, since those
processors have a larger penalty on cross-register-file copies, and the scope is
limited to basic blocks. Because of these two factors, we limit the effects of
the inaccuracy. Indeed, I did not want to build up a fancy cost model with block
frequency and everything on top of that.
[1] We can imagine targets that can combine an extractelement with other
instructions than just stores. If we want to go into that direction, the current
interfaces must be augmented and, moreover, I think this becomes a global isel
problem.
Differential Revision: http://reviews.llvm.org/D5921
<rdar://problem/14170854>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220978 91177308-0d34-0410-b5e6-96231b3b80d8
Since block address values can be larger than 2GB in 64-bit code, they
cannot be loaded simply using an @l / @ha pair, but instead must be
loaded from the TOC, just like GlobalAddress, ConstantPool, and
JumpTable values are.
The commit also fixes a bug in PPCLinuxAsmPrinter::doFinalization where
temporary labels could not be used as TOC values, since code would
attempt (and fail) to use GetOrCreateSymbol to create a symbol of the
same name as the temporary label.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220959 91177308-0d34-0410-b5e6-96231b3b80d8
r212242 introduced a legalizer hook, originally to let AArch64 widen
v1i{32,16,8} rather than scalarize, because the legalizer expected, when
scalarizing the result of a conversion operation, to already have
scalarized the operands. On AArch64, v1i64 is legal, so that commit
ensured operations such as v1i32 = trunc v1i64 wouldn't assert.
It did that by choosing to widen v1 types whenever possible. However,
v1i1 types, for which there's no legal widened type, would still trigger
the assert.
This commit fixes that, by only scalarizing a trunc's result when the
operand has already been scalarized, and introducing an extract_elt
otherwise.
This is similar to r205625.
Fixes PR20777.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220937 91177308-0d34-0410-b5e6-96231b3b80d8
Earlier this summer I fixed an issue where we were incorrectly combining
multiple loads that had different constraints such alignment, invariance,
temporality, etc. Apparently in one case I made copt paste error and swapped
alignment and invariance.
Tests included.
rdar://18816719
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220933 91177308-0d34-0410-b5e6-96231b3b80d8
It should be on for every target that supports unaligned accesses (e.g. not
v6m).
Patch by Charlie Turner.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220912 91177308-0d34-0410-b5e6-96231b3b80d8
This transformation worked if selector is produced by SETCC, however SETCC is needed only if we consider to swap operands. So I replaced SETCC check for this case.
Added tests for vselect of <X x i1> values.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220777 91177308-0d34-0410-b5e6-96231b3b80d8
Ffter commit at rev219046 512-bit broadcasts lowering become non-optimal. Most of tests on broadcasting and embedded broadcasting were changed and they doesn’t produce efficient code.
Example below is from commit changes (it’s the first test from test/CodeGen/X86/avx512-vbroadcast.ll):
define <16 x i32> @_inreg16xi32(i32 %a) {
; CHECK-LABEL: _inreg16xi32:
; CHECK: ## BB#0:
-; CHECK-NEXT: vpbroadcastd %edi, %zmm0
+; CHECK-NEXT: vmovd %edi, %xmm0
+; CHECK-NEXT: vpbroadcastd %xmm0, %ymm0
+; CHECK-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm0
; CHECK-NEXT: retq
%b = insertelement <16 x i32> undef, i32 %a, i32 0
%c = shufflevector <16 x i32> %b, <16 x i32> undef, <16 x i32> zeroinitializer
ret <16 x i32> %c
}
Here, 256-bit broadcast was generated instead of 512-bit one.
In this patch
1) I added vector-shuffle lowering through broadcasts
2) Removed asserts and branches likes because this is incorrect
- assert(Subtarget->hasDQI() && "We can only lower v8i64 with AVX-512-DQI");
3) Fixed lowering tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220774 91177308-0d34-0410-b5e6-96231b3b80d8
This is a Microsoft calling convention that supports both x86 and x86_64
subtargets. It passes vector and floating point arguments in XMM0-XMM5,
and passes them indirectly once they are consumed.
Homogenous vector aggregates of up to four elements can be passed in
sequential vector registers, but this part is not implemented in LLVM
and will be handled in Clang.
On 32-bit x86, it is similar to fastcall in that it uses ecx:edx as
integer register parameters and is callee cleanup. On x86_64, it
delegates to the normal win64 calling convention.
Reviewers: majnemer
Differential Revision: http://reviews.llvm.org/D5943
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220745 91177308-0d34-0410-b5e6-96231b3b80d8
Benchmarks have shown that it's harmless to the performance there, and having a
unified set of passes between the two cores where possible helps big.LITTLE
deployment.
Patch by Z. Zheng.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220744 91177308-0d34-0410-b5e6-96231b3b80d8
For a call to not return in to the stackmap shadow, the shadow must end with the call.
To do this, we must insert any required nops *before* the call, and not after it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220728 91177308-0d34-0410-b5e6-96231b3b80d8
This is a minor change to use the immediate version when the operand is a null
value. This should get rid of an unnecessary 'mov' instruction in debug
builds and align the code more with the one generated by SelectionDAG.
This fixes rdar://problem/18785125.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220713 91177308-0d34-0410-b5e6-96231b3b80d8
Minor enhancement to use 'tbz' for i1 compare-and-branch to get rid of an 'and'
instruction.
This fixes rdar://problem/18784953.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220712 91177308-0d34-0410-b5e6-96231b3b80d8
To avoid emitting too many nops, a stackmap shadow can include emitted instructions in the shadow, but these must not include branch targets.
A return from a call should count as a branch target as patching over the instructions after the call would lead to incorrect behaviour for threads currently making that call, when they return.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220710 91177308-0d34-0410-b5e6-96231b3b80d8
The pattern matching for a 'ConstantInt' value was too restrictive. Checking for
a 'Constant' with a bull value is sufficient for using an 'cbz/cbnz' instruction.
This fixes rdar://problem/18784732.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220709 91177308-0d34-0410-b5e6-96231b3b80d8
This fixes a bug where the input register was not defined for the 'tbz/tbnz'
instruction. This happened, because we folded the 'and' instruction from a
different basic block.
This fixes rdar://problem/18784013.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220704 91177308-0d34-0410-b5e6-96231b3b80d8
At higher optimization levels the LLVM IR may contain more complex patterns for
loads/stores from/to frame indices. The 'computeAddress' function wasn't able to
handle this and triggered an assertion.
This fix extends the possible addressing modes for frame indices.
This fixes rdar://problem/18783298.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220700 91177308-0d34-0410-b5e6-96231b3b80d8
Currently, the ARM backend will select the VMAXNM and VMINNM for these C
expressions:
(a < b) ? a : b
(a > b) ? a : b
but not these expressions:
(a > b) ? b : a
(a < b) ? b : a
This patch allows all of these expressions to be matched.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220671 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Fixes PR21100 which is caused by inconsistency between the declared return type
and the expected return type at the call site. The new behavior is consistent
with nvcc and the NVPTXTargetLowering::getPrototype function.
Test Plan: test/Codegen/NVPTX/vector-return.ll
Reviewers: jholewinski
Reviewed By: jholewinski
Subscribers: llvm-commits, meheff, eliben, jholewinski
Differential Revision: http://reviews.llvm.org/D5612
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220607 91177308-0d34-0410-b5e6-96231b3b80d8
Minor patch to fix an issue in XFormVExtractWithShuffleIntoLoad where a load is unary shuffled, then bitcast (to a type with the same number of elements) before extracting an element.
An undef was created for the second shuffle operand using the original (post-bitcasted) vector type instead of the pre-bitcasted type like the rest of the shuffle node - this was then causing an assertion on the different types later on inside SelectionDAG::getVectorShuffle.
Differential Revision: http://reviews.llvm.org/D5917
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220592 91177308-0d34-0410-b5e6-96231b3b80d8
This is a first step for generating SSE rsqrt instructions for
reciprocal square root calcs when fast-math is allowed.
For now, be conservative and only enable this for AMD btver2
where performance improves significantly - for example, 29%
on llvm/projects/test-suite/SingleSource/Benchmarks/BenchmarkGame/n-body.c
(if we convert the data type to single-precision float).
This patch adds a two constant version of the Newton-Raphson
refinement algorithm to DAGCombiner that can be selected by any target
via a parameter returned by getRsqrtEstimate()..
See PR20900 for more details:
http://llvm.org/bugs/show_bug.cgi?id=20900
Differential Revision: http://reviews.llvm.org/D5658
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220570 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Most structs were fixed by r218451 but those of between >32-bits and
<64-bits remained broken since they were not marked with [ASZ]ExtUpper.
This patch fixes the remaining cases by using
CCPromoteToUpperBitsInType<i64> on i64's in addition to i32 and smaller.
Reviewers: vmedic
Reviewed By: vmedic
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5963
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220556 91177308-0d34-0410-b5e6-96231b3b80d8
This fixes a miscompilation in the AArch64 fast-isel which was
triggered when a branch is based on an icmp with condition eq or ne,
and type i1, i8 or i16. The cbz instruction compares the whole 32-bit
register, so values with the bottom 1, 8 or 16 bits clear would cause
the wrong branch to be taken.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220553 91177308-0d34-0410-b5e6-96231b3b80d8
This adds support for legalization of instructions of the form:
[fp_conv] <1 x i1> %op to <1 x double>
where fp_conv is one of fpto[us]i, [us]itofp. This used to assert
because they were simply missing from the vector operand scalarizer.
A similar problem arose in r190830, with trunc instead.
Fixes PR20778.
Differential Revision: http://reviews.llvm.org/D5810
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220533 91177308-0d34-0410-b5e6-96231b3b80d8
x86's CMPXCHG -> EFLAGS consumer wasn't being recorded as a real EFLAGS
dependency because it was represented by a pair of CopyFromReg(EFLAGS) ->
CopyToReg(EFLAGS) nodes. ScheduleDAG was expecting the source to be an
implicit-def on the instruction, where the result numbers in the DAG and the
Uses list in TableGen matched up precisely.
The Copy notation seems much more robust, so this patch extends ScheduleDAG
rather than refactoring x86.
Should fix PR20376.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220529 91177308-0d34-0410-b5e6-96231b3b80d8
Currently, @llvm.smul.with.overflow.i8 expands to 9 instructions, where
3 are really needed.
This adds X86ISD::UMUL8/SMUL8 SD nodes, and custom lowers them to
MUL8/IMUL8 + SETO.
i8 is a special case because there is no two/three operand variants of
(I)MUL8, so the first operand and return value need to go in AL/AX.
Also, we can't write patterns for these instructions: TableGen refuses
patterns where output operands don't match SDNode results. In this case,
instructions where the output operand is an implicitly defined register.
A related special case (and FIXME) exists for MUL8 (X86InstrArith.td):
// FIXME: Used for 8-bit mul, ignore result upper 8 bits.
// This probably ought to be moved to a def : Pat<> if the
// syntax can be accepted.
[(set AL, (mul AL, GR8:$src)), (implicit EFLAGS)]
Ideally, these go away with UMUL8, but we still need to improve TableGen
support of implicit operands in patterns.
Before this change:
movsbl %sil, %eax
movsbl %dil, %ecx
imull %eax, %ecx
movb %cl, %al
sarb $7, %al
movzbl %al, %eax
movzbl %ch, %esi
cmpl %eax, %esi
setne %al
After:
movb %dil, %al
imulb %sil
seto %al
Also, remove a made-redundant testcase for PR19858, and enable more FastISel
ALU-overflow tests for SelectionDAG too.
Differential Revision: http://reviews.llvm.org/D5809
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220516 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r214981.
I'm not sure what I was thinking when I wrote this. Testing with MSVC
shows that this function is mangled to '@f@8':
int __fastcall f(int a, int b);
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220492 91177308-0d34-0410-b5e6-96231b3b80d8
This updates check for double precision zero floating point constant to allow
use of instruction with immediate value rather than temporary register.
Currently "a == 0.0", where "a" is of "double" type generates:
vmov.i32 d16, #0x0
vcmpe.f64 d0, d16
With this change it becomes:
vcmpe.f64 d0, #0
Patch by Sergey Dmitrouk.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220486 91177308-0d34-0410-b5e6-96231b3b80d8
A previous patch enabled SELECT_VSRC and SELECT_CC_VSRC for VSX to
handle <2 x double> cases. This patch adds SELECT_VSFRC and
SELECT_CC_VSFRC to allow use of all 64 vector-scalar registers for the
f64 type when VSX is enabled. The changes are analogous to those in
the previous patch. I've added a new variant to vsx.ll to test the
code generation.
(I also cleaned up a little formatting in PPCInstrVSX.td from the
previous patch.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220395 91177308-0d34-0410-b5e6-96231b3b80d8
This enables targets to adapt their pass pipeline to the register
allocator in use. For example, with the AArch64 backend, using PBQP
with the cortex-a57, the FPLoadBalancing pass is no longer necessary.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220321 91177308-0d34-0410-b5e6-96231b3b80d8
With VSX enabled, test/CodeGen/PowerPC/recipest.ll exposes a bug in
the FMA mutation pass. If we have a situation where a killed product
register is the same register as the FMA target, such as:
%vreg5<def,tied1> = XSNMSUBADP %vreg5<tied0>, %vreg11, %vreg5,
%RM<imp-use>; VSFRC:%vreg5 F8RC:%vreg11
then the substitution makes no sense. We end up getting a crash when
we try to extend the interval associated with the killed product
register, as there is already a live range for %vreg5 there. This
patch just disables the mutation under those circumstances.
Since recipest.ll generates different code with VMX enabled, I've
modified that test to use -mattr=-vsx. I've borrowed the code from
that test that exposed the bug and placed it in fma-mutate.ll, where
it tests several mutation opportunities including the "bad" one.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220290 91177308-0d34-0410-b5e6-96231b3b80d8
Every target we support has support for assembly that looks like
a = b - c
.long a
What is special about MachO is that the above combination suppresses the
production of a relocation.
With this change we avoid producing the intermediary labels when they don't
add any value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220256 91177308-0d34-0410-b5e6-96231b3b80d8
X86 code to lower VSELECT messed a bit with the bits set in the mask of VSELECT
when it knows it can be lowered into BLEND. Indeed, only the high bits need to be
set for those and it optimizes those accordingly.
However, when the mask is a compile time constant, the lowering will be handled
by the generic optimizer and those modifications will generate bad code in the
generic optimizer.
This patch fixes that by preventing the optimization if the VSELECT will be
handled by the generic optimizer.
<rdar://problem/18675020>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220242 91177308-0d34-0410-b5e6-96231b3b80d8
This patch improves support for commutative instructions in the x86 memory folding implementation by attempting to fold a commuted version of the instruction if the original folding fails - if that folding fails as well the instruction is 're-commuted' back to its original order before returning.
Updated version of r219584 (reverted in r219595) - the commutation attempt now explicitly ensures that neither of the commuted source operands are tied to the destination operand / register, which was the source of all the regressions that occurred with the original patch attempt.
Added additional regression test case provided by Joerg Sonnenberger.
Differential Revision: http://reviews.llvm.org/D5818
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220239 91177308-0d34-0410-b5e6-96231b3b80d8
The previous code had a few problems, motivating the choices here.
1. It could create instructions clobbering CPSR, but the incoming MachineInstr
didn't reflect this. A potential source of corruption. This is why the patch
has a new PseudoInst for before lowering.
2. Similarly, there was some code to handle the incoming instruction not being
ARMCC::AL, but this would have caused massive problems if it was actually
invoked when a complex offset needing more than one instruction was requested.
3. It wasn't designed to handle unaligned pointers (or offsets). These should
probably be minimised anyway, but the code needs to deal with them properly
regardless.
4. It had some rather dubious ad-hoc code to avoid calling
emitThumbRegPlusImmediate, a function which should be designed to do precisely
this job.
We seem to cover the common cases correctly now, and hopefully can enhance
emitThumbRegPlusImmediate to handle any extra optimisations we need to add in
future.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220236 91177308-0d34-0410-b5e6-96231b3b80d8
The current instruction selection patterns for SMULW[BT] and SMLAW[BT]
are incorrect. These instructions multiply a 32-bit and a 16-bit value
(both signed) and return the top 32 bits of the 48-bit result. This
preserves the 16 bits of overflow, whereas the patterns they currently
match truncate the result to 16 bits then sign extend.
To select these instructions, we would need to match an ISD::SMUL_LOHI,
a sign extend, two shifts and an or. There is no way to match SMUL_LOHI
in an instruction pattern as it defines multiple values, so this would
have to be done in C++. I have raised
http://llvm.org/bugs/show_bug.cgi?id=21297 to cover allowing correct
selection of these instructions.
This fixes http://llvm.org/bugs/show_bug.cgi?id=19396
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220196 91177308-0d34-0410-b5e6-96231b3b80d8
This function can, for some offsets from the SP, split one instruction
into two. Since it re-uses the original instruction as the first
instruction of the result, we need ensure its result register is not
marked as dead before we use it in the second instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220194 91177308-0d34-0410-b5e6-96231b3b80d8
We recently discovered an issue that reinforces what a good idea it is
to always specify -mcpu in our code generation tests, particularly for
-mattr=+vsx. This patch ensures that all tests that specify
-mattr=+vsx also specify -mcpu=pwr7 or -mcpu=pwr8, as appropriate.
Some of the uses of -mattr=+vsx added recently don't make much sense
(when specified for -mtriple=powerpc-apple-darwin8 or -march=ppc32,
for example). For cases like this I've just removed the extra VSX
test commands; there's enough coverage without them.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220173 91177308-0d34-0410-b5e6-96231b3b80d8
Patch by Bill Seurer; some comment formatting changes by me.
There are a few PowerPC test cases for FastISel support that currently
fail with VSX support enabled. The temporary workaround under
discussion in http://reviews.llvm.org/D5362 helps, but the tests still
fail because they specify -fast-isel-abort, and the VSX workaround
punts back to SelectionDAG. We have plans to fix FastISel permanently
for VSX, but until that's in place these tests are preventing us from
enabling VSX by default. Therefore we are adding -mattr=-vsx to these
tests until the full support is ready.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220172 91177308-0d34-0410-b5e6-96231b3b80d8
The VSX testing variant in test/CodeGen/PowerPC/fma.ll had to be
disabled because of unexpected behavior on many of the builders. I
tracked this down to a situation that occurs when the VSX attribute is
enabled for a target that disables the MI early scheduling pass. This
patch adds -mcpu=pwr7 to make this predictable. The other issue will
be addressed separately.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220171 91177308-0d34-0410-b5e6-96231b3b80d8
The previous tests claimed to test constant offsets in the function name,
but the tests weren't actually testing them.
Clone the tests, and do testing of all combinations of the following:
1) with/without constant pointer offset
2) 32/64-bit addressing modes
3) Usage and non-usage of the return value from the atomicrmw
Reviewed-by: Matt Arsenault <matthew.arsenault@amd.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220103 91177308-0d34-0410-b5e6-96231b3b80d8
The function name now matches what it's actually testing.
Signed-off-by: Aaron Watry <awatry@gmail.com>
Reviewed-by: Matt Arsenault <matthew.arsenault@amd.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220102 91177308-0d34-0410-b5e6-96231b3b80d8
TL;DR: Indexing maps with [] creates missing entries.
The long version:
When selecting lifetime intrinsics, we index the *static* alloca map with the AllocaInst we find for that lifetime. Trouble is, we don't first check to see if this is a dynamic alloca.
On the attached example, this causes a dynamic alloca to create an entry in the static map, and returns 0 (the default) as the frame index for that lifetime. 0 was used for the frame index of the stack protector, which given that it now has a lifetime, is coloured, and merged with other stack slots.
PEI would later trigger an assert because it expects the stack protector to not be dead.
This fix ensures that we only get frame indices for static allocas, ie, those in the map. Dynamic ones are effectively dropped, which is suboptimal, but at least isn't completely broken.
rdar://problem/18672951
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220099 91177308-0d34-0410-b5e6-96231b3b80d8
With VSX enabled, LLVM crashes when compiling
test/CodeGen/PowerPC/fma.ll. I traced this to the liveness test
that's revised in this patch. The interval test is designed to only
work for virtual registers, but in this case the AddendSrcReg is
physical. Since there is already a walk of the MIs between the
AddendMI and the FMA, I added a check for def/kill of the AddendSrcReg
in that loop. At Hal Finkel's request, I converted the liveness test
to an assert restricted to virtual registers.
I've changed the fma.ll test to have VSX and non-VSX variants so we
can test both kinds of multiply-adds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220090 91177308-0d34-0410-b5e6-96231b3b80d8
When the input to a store instruction was a zero vector, the backend
always selected a normal vector store regardless of the non-temporal
hint. This is fixed by this patch.
This fixes PR19370.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220054 91177308-0d34-0410-b5e6-96231b3b80d8
We should be talking about the number of source elements, not the number of destination elements, given we know at this point that the source and dest element numbers are not the same.
While we're at it, avoid writing to std::vector::end()...
Bug found with random testing and a lot of coffee.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220051 91177308-0d34-0410-b5e6-96231b3b80d8
Currently the VSX support enables use of lxvd2x and stxvd2x for 2x64
types, but does not yet use lxvw4x and stxvw4x for 4x32 types. This
patch adds that support.
As with lxvd2x/stxvd2x, this involves straightforward overriding of
the patterns normally recognized for lvx/stvx, with preference given
to the VSX patterns when VSX is enabled.
In addition, the logic for permitting misaligned memory accesses is
modified so that v4r32 and v4i32 are treated the same as v2f64 and
v2i64 when VSX is enabled. Finally, the DAG generation for unaligned
loads is changed to just use a normal LOAD (which will become lxvw4x)
on P8 and later hardware, where unaligned loads are preferred over
lvsl/lvx/lvx/vperm.
A number of tests now generate the VSX loads/stores instead of
lvx/stvx, so this patch adds VSX variants to those tests. I've also
added <4 x float> tests to the vsx.ll test case, and created a
vsx-p8.ll test case to be used for testing code generation for the
P8Vector feature. For now, that simply tests the unaligned load/store
behavior.
This has been tested along with a temporary patch to enable the VSX
and P8Vector features, with no new regressions encountered with or
without the temporary patch applied.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220047 91177308-0d34-0410-b5e6-96231b3b80d8
v2: use dyn_cast
fixup comments
v3: use cast
Reviewed-by: Matt Arsenault <arsenm2@gmail.com>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220044 91177308-0d34-0410-b5e6-96231b3b80d8
Patch by Bill Seurer; committed on his behalf.
These test cases generate slightly different code sequences when VSX
is activated and thus fail. The update turns off VSX explicitly for
the existing checks and then adds a second set of checks for most of
them that test the VSX instruction output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220019 91177308-0d34-0410-b5e6-96231b3b80d8
The bug is in ARMConstantIslands::createNewWater where the upper bound of the
new water split point is computed:
// This could point off the end of the block if we've already got constant
// pool entries following this block; only the last one is in the water list.
// Back past any possible branches (allow for a conditional and a maximally
// long unconditional).
if (BaseInsertOffset + 8 >= UserBBI.postOffset()) {
BaseInsertOffset = UserBBI.postOffset() - UPad - 8;
DEBUG(dbgs() << format("Move inside block: %#x\n", BaseInsertOffset));
}
The split point is supposed to be somewhere between the machine instruction that
loads from the constant pool entry and the end of the basic block, before branch
instructions. The code above is fine if the basic block is large enough and
there are a sufficient number of instructions following the machine instruction.
However, if the machine instruction is near the end of the basic block,
BaseInsertOffset can point to the machine instruction or another instruction
that precedes it, and this can lead to convergence failure.
This commit fixes this bug by ensuring BaseInsertOffset is larger than the
offset of the instruction following the constant-loading instruction.
rdar://problem/18581150
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220015 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Backends can use setInsertFencesForAtomic to signal to the middle-end that
montonic is the only memory ordering they can accept for
stores/loads/rmws/cmpxchg. The code lowering those accesses with a stronger
ordering to fences + monotonic accesses is currently living in
SelectionDAGBuilder.cpp. In this patch I propose moving this logic out of it
for several reasons:
- There is lots of redundancy to avoid: extremely similar logic already
exists in AtomicExpand.
- The current code in SelectionDAGBuilder does not use any target-hooks, it
does the same transformation for every backend that requires it
- As a result it is plain *unsound*, as it was apparently designed for ARM.
It happens to mostly work for the other targets because they are extremely
conservative, but Power for example had to switch to AtomicExpand to be
able to use lwsync safely (see r218331).
- Because it produces IR-level fences, it cannot be made sound ! This is noted
in the C++11 standard (section 29.3, page 1140):
```
Fences cannot, in general, be used to restore sequential consistency for atomic
operations with weaker ordering semantics.
```
It can also be seen by the following example (called IRIW in the litterature):
```
atomic<int> x = y = 0;
int r1, r2, r3, r4;
Thread 0:
x.store(1);
Thread 1:
y.store(1);
Thread 2:
r1 = x.load();
r2 = y.load();
Thread 3:
r3 = y.load();
r4 = x.load();
```
r1 = r3 = 1 and r2 = r4 = 0 is impossible as long as the accesses are all seq_cst.
But if they are lowered to monotonic accesses, no amount of fences can prevent it..
This patch does three things (I could cut it into parts, but then some of them
would not be tested/testable, please tell me if you would prefer that):
- it provides a default implementation for emitLeadingFence/emitTrailingFence in
terms of IR-level fences, that mimic the original logic of SelectionDAGBuilder.
As we saw above, this is unsound, but the best that can be done without knowing
the targets well (and there is a comment warning about this risk).
- it then switches Mips/Sparc/XCore to use AtomicExpand, relying on this default
implementation (that exactly replicates the logic of SelectionDAGBuilder, so no
functional change)
- it finally erase this logic from SelectionDAGBuilder as it is dead-code.
Ideally, each target would define its own override for emitLeading/TrailingFence
using target-specific fences, but I do not know the Sparc/Mips/XCore memory model
well enough to do this, and they appear to be dealing fine with the ARM-inspired
default expansion for now (probably because they are overly conservative, as
Power was). If anyone wants to compile fences more agressively on these
platforms, the long comment should make it clear why he should first override
emitLeading/TrailingFence.
Test Plan: make check-all, no functional change
Reviewers: jfb, t.p.northover
Subscribers: aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D5474
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219957 91177308-0d34-0410-b5e6-96231b3b80d8
When the constant divisor was larger than 32bits, then the optimized code
generated for the AArch64 backend would emit the wrong code, because the shift
was defined as a shift of a 32bit constant '(1<<Lg2(divisor))' and we would
loose the upper 32bits.
This fixes rdar://problem/18678801.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219934 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
In order to support big endian targets for the BuildPairF64 nodes we
just need to swap the low/high pair registers. Additionally, for the
ExtractElementF64 nodes we have to calculate the correct stack offset
with respect to the node's register/operand that we want to extract.
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5753
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219931 91177308-0d34-0410-b5e6-96231b3b80d8
In AVX512f we support 64x2 and 32x8 inserts via matching them to 32x4 and 64x4
respectively. These are matched by "Alt" Pat<>'s (Alt stands for alternative
VTs).
Since DQ has native support for these intructions, I peeled off the non-"Alt"
part of the baseclass into vinsert_for_size_no_alt. The DQ instructions are
derived from this multiclass. The "Alt" Pat<>'s are disabled with DQ.
Fixes <rdar://problem/18426089>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219874 91177308-0d34-0410-b5e6-96231b3b80d8
The SelectDS1Addr1Offset complex pattern always tries to store constant
lds pointers in the offset operand and store a zero value in the addr operand.
Since the addr operand does not accept immediates, the zero value
needs to first be copied to a register.
This newly created zero value will not go through normal instruction
selection, so we need to manually insert a V_MOV_B32_e32 in the complex
pattern.
This bug was hidden by the fact that if there was another zero value
in the DAG that had not been selected yet, then the CSE done by the DAG
would use the unselected node for the addr operand rather than the one
that was just created. This would lead to the zero value being selected
and the DAG automatically inserting a V_MOV_B32_e32 instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219848 91177308-0d34-0410-b5e6-96231b3b80d8
This is mostly a copy of the existing FastISel GEP code, but we have to
duplicate it for AArch64, because otherwise we would bail out even for simple
cases. This is because the standard fastEmit functions don't cover MUL at all
and ADD is lowered very inefficientily.
The original commit had a bug in the add emit logic, which has been fixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219831 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Fixes a FIXME in MachineSinking. Instead of using the simple heuristics in
isPostDominatedBy, use the real MachinePostDominatorTree and MachineLoopInfo.
The old heuristics caused instructions to sink unnecessarily, and might create
register pressure.
This is the second try of the fix. The first one (D4814) caused a performance
regression due to failing to sink instructions out of loops (PR21115). This
patch fixes PR21115 by sinking an instruction from a deeper loop to a shallower
one regardless of whether the target block post-dominates the source.
Thanks Alexey Volkov for reporting PR21115!
Test Plan:
Added a NVPTX codegen test to verify that our change prevents the backend from
over-sinking. It also shows the unnecessary register pressure caused by
over-sinking.
Added an X86 test to verify we can sink instructions out of loops regardless of
the dominance relationship. This test is reduced from Alexey's test in PR21115.
Updated an affected test in X86.
Also ran SPEC CINT2006 and llvm-test-suite for compilation time and runtime
performance. Results are attached separately in the review thread.
Reviewers: Jiangning, resistor, hfinkel
Reviewed By: hfinkel
Subscribers: hfinkel, bruno, volkalexey, llvm-commits, meheff, eliben, jholewinski
Differential Revision: http://reviews.llvm.org/D5633
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219773 91177308-0d34-0410-b5e6-96231b3b80d8
Peephole optimization that generates a single conditional branch
for csinc-branch sequences like in the examples below. This is
possible when the csinc sets or clears a register based on a condition
code and the branch checks that register. Also the condition
code may not be modified between the csinc and the original branch.
Examples:
1. Convert csinc w9, wzr, wzr, <CC>;tbnz w9, #0, 0x44
to b.<invCC>
2. Convert csinc w9, wzr, wzr, <CC>; tbz w9, #0, 0x44
to b.<CC>
rdar://problem/18506500
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219742 91177308-0d34-0410-b5e6-96231b3b80d8
Thumb1 has legitimate reasons for preferring 32-bit alignment of types
i1/i8/i16, since the 16-bit encoding of "add rD, sp, #imm" requires #imm to be
a multiple of 4. However, this is a trade-off betweem code size and RAM usage;
the DataLayout string is not the best place to represent it even if desired.
So this patch removes the extra Thumb requirements, hopefully making ARM and
Thumb completely compatible in this respect.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219734 91177308-0d34-0410-b5e6-96231b3b80d8
There's no hard requirement on LLVM to align local variable to 32-bits, so the
Thumb1 frame handling needs to be able to deal with variables that are only
naturally aligned without falling over.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219733 91177308-0d34-0410-b5e6-96231b3b80d8
This is mostly a copy of the existing FastISel GEP code, but on AArch64 we bail
out even for simple cases, because the standard fastEmit functions don't cover
MUL and ADD is lowered inefficientily.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219726 91177308-0d34-0410-b5e6-96231b3b80d8
Before, ARM and Thumb mode code had different preferred alignments, which could
lead to some rather unexpected results. There's justification for reducing it
from the default 64-bits (wasted space), but I don't think there is for going
below 32-bits.
There's no actual ABI change here, just to reassure people.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219719 91177308-0d34-0410-b5e6-96231b3b80d8
Sign-/zero-extend folding depended on the load and the integer extend to be
both selected by FastISel. This cannot always be garantueed and SelectionDAG
might interfer. This commit adds additonal checks to load and integer extend
lowering to catch this.
Related to rdar://problem/18495928.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219716 91177308-0d34-0410-b5e6-96231b3b80d8
This effectively reverts revert 219707. After fixing the test to work with
new function name format and renamed intrinsic.
Reviewed-by: Tom Stellard <tom@stellard.net>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219710 91177308-0d34-0410-b5e6-96231b3b80d8
v2: Add SI lowering
Add test
v3: Place work dimensions after the kernel arguments.
v4: Calculate offset while lowering arguments
v5: rebase
v6: change prefix to AMDGPU
Reviewed-by: Tom Stellard <tom@stellard.net>
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219705 91177308-0d34-0410-b5e6-96231b3b80d8
Use 0 as the base address for a constant address, so if
we have a constant address we can save moves and form
read2/write2s.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219698 91177308-0d34-0410-b5e6-96231b3b80d8
e.g Currently we'll generate following instructions if the immediate is too wide:
MOV X0, WideImmediate
ADD X1, BaseReg, X0
LDR X2, [X1, 0]
Using [Base+XReg] addressing mode can save one ADD as following:
MOV X0, WideImmediate
LDR X2, [BaseReg, X0]
Differential Revision: http://reviews.llvm.org/D5477
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219665 91177308-0d34-0410-b5e6-96231b3b80d8
Some early revisions of the Cortex-A53 have an erratum (835769) whereby it is
possible for a 64-bit multiply-accumulate instruction in AArch64 state to
generate an incorrect result. The details are quite complex and hard to
determine statically, since branches in the code may exist in some
circumstances, but all cases end with a memory (load, store, or prefetch)
instruction followed immediately by the multiply-accumulate operation.
The safest work-around for this issue is to make the compiler avoid emitting
multiply-accumulate instructions immediately after memory instructions and the
simplest way to do this is to insert a NOP.
This patch implements such work-around in the backend, enabled via the option
-aarch64-fix-cortex-a53-835769.
The work-around code generation is not enabled by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219603 91177308-0d34-0410-b5e6-96231b3b80d8
This patch improves support for commutative instructions in the x86 memory folding implementation by attempting to fold a commuted version of the instruction if the original folding fails - if that folding fails as well the instruction is 're-commuted' back to its original order before returning.
This mainly helps the stack inliner better fold reloads of 3 (or more) operand instructions (VEX encoded SSE etc.) but by performing this in the lowest foldMemoryOperandImpl implementation it also replaces the X86InstrInfo::optimizeLoadInstr version and is now used by FastISel too.
Differential Revision: http://reviews.llvm.org/D5701
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219584 91177308-0d34-0410-b5e6-96231b3b80d8
Summary: Implement the most basic form of conditional branches in Mips fast-isel.
Test Plan:
br1.ll
run 4 flavors of test-suite. mips32 r1/r2 and at -O0/O2
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D5583
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219556 91177308-0d34-0410-b5e6-96231b3b80d8
Currently this only functions to match simple cases
where ds_read2_* / ds_write2_* instructions can be used.
In the future it might match some of the other weird
load patterns, such as direct to LDS loads.
Currently enabled only with a subtarget feature to enable
easier testing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219533 91177308-0d34-0410-b5e6-96231b3b80d8
Fixes a logic error in the MachineScheduler found by Steve Montgomery (and
confirmed by Andy). This has gone unfixed for months because the fix has been
found to introduce some small performance regressions. However, Andy has
recommended that, at this point, we fix this to avoid further dependence on the
incorrect behavior (and then follow-up separately on any regressions), and I
agree.
Fixes PR18883.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219512 91177308-0d34-0410-b5e6-96231b3b80d8
Summary: Add the ability to convert 64 or 32 bit floating point values to integer in mips fast-isel
Test Plan:
fpintconv.ll
ran 4 flavors of test-suite with no errors, misp32 r1/r2 O0/O2
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: llvm-commits, rfuhler, mcrosier
Differential Revision: http://reviews.llvm.org/D5562
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219511 91177308-0d34-0410-b5e6-96231b3b80d8
This patch changes the fast-math implementation for calculating sqrt(x) from:
y = 1 / (1 / sqrt(x))
to:
y = x * (1 / sqrt(x))
This has 2 benefits: less code / faster code and one less estimate instruction
that may lose precision.
The only target that will be affected (until http://reviews.llvm.org/D5658 is approved)
is PPC. The difference in codegen for PPC is 2 less flops for a single-precision sqrtf
or vector sqrtf and 4 less flops for a double-precision sqrt.
We also eliminate a constant load and extra register usage.
Differential Revision: http://reviews.llvm.org/D5682
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219445 91177308-0d34-0410-b5e6-96231b3b80d8
The current implementation of GPR->FPR register moves uses a stack slot. This mechanism writes a double word and reads a word. In big-endian the load address must be displaced by 4-bytes in order to get the right value. In little endian this is no longer required. This patch fixes the issue and adds LE regression tests to fast-isel-conversion which currently expose this problem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219441 91177308-0d34-0410-b5e6-96231b3b80d8
The instruction emitter will crash if it encounters a CopyToReg
node with a non-register operand like FrameIndex.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219428 91177308-0d34-0410-b5e6-96231b3b80d8
LLVM assumes INSERT_SUBREG will always have register operands, so
we need to legalize non-register operands, like FrameIndexes, to
avoid random assertion failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219420 91177308-0d34-0410-b5e6-96231b3b80d8
This adds the Pat<>'s for the intrinsics. These are necessary because we
don't lower these intrinsics to SDNodes but match them directly. See the
rational in the previous commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219362 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
I had forgotten to check for NotSlowIncDec in the patterns that can generate
inc/dec for the above pattern (added in D4796).
This currently applies to Atom Silvermont, KNL and SKX.
Test Plan: New checks on atomic_mi.ll
Reviewers: jfb, nadav
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5677
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219336 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Fix pr21099
The pseudocode of what we were doing (spread through two functions) was:
if (operand.doesNotFitIn32Bits())
Opc.initializeWithFoo();
if (operand < 0)
operand = -operand;
if (operand.doesFitIn8Bits())
Opc.initializeWithBar();
else if (operand.doesFitIn32Bits())
Opc.initializeWithBlah();
doStuff(Opc);
So for operand == INT32_MIN, Opc was never initialized because the operand changes
from fitting in 32 bits to not fitting, causing the various bugs/error messages
noted by pr21099.
This patch adds an extra test at the beginning for this case, and an
llvm_unreachable to have better error message if the operand ends up
not fitting in 32-bits at the end.
Test Plan: new test + make check
Reviewers: jfb
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5655
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219257 91177308-0d34-0410-b5e6-96231b3b80d8
Added a FIXME coment instead, we need to handle the case where the
two DS instructions being compared have different numbers of operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219236 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
According to the ABI documentation, f128 and {f128} should both be returned
in $f0 and $f2. However, this doesn't match GCC's behaviour which is to
return f128 in $f0 and $f2, but {f128} in $f0 and $f1.
Reviewers: vmedic
Reviewed By: vmedic
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5578
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219196 91177308-0d34-0410-b5e6-96231b3b80d8
The code already folds sign-/zero-extends, but only if they are arguments to
mul and shift instructions. This extends the code to also fold them when they
are direct inputs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219187 91177308-0d34-0410-b5e6-96231b3b80d8
Tiny enhancement to the address computation code to also fold sub instructions
if the rhs is constant and can be folded into the offset.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219186 91177308-0d34-0410-b5e6-96231b3b80d8
This commit fixes an issue with sign-/zero-extending loads that was discovered
by Richard Barton.
We use now the correct load instructions for sign-extending loads to 64bit. Also
updated and added more unit tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219185 91177308-0d34-0410-b5e6-96231b3b80d8
The patch's author points out that, despite the function's documentation,
getSetCCResultType is only used to get the SETCC result type (with one
here-removed problematic exception). In one case, getSetCCResultType was being
used to get the predicate type to use for a SELECT node, and then
SIGN_EXTENDing (or truncating) to get the input predicate to match that type.
Unfortunately, this was happening inside visitSIGN_EXTEND, and creating new
SIGN_EXTEND nodes was causing an infinite loop. In addition, this behavior was
wrong if a target was not using ZeroOrNegativeOneBooleanContent. Lastly, the
extension/truncation seems unnecessary here: SELECT is defined as:
Select(COND, TRUEVAL, FALSEVAL). If the type of the boolean COND is not i1
then the high bits must conform to getBooleanContents.
So here we remove this use of getSetCCResultType and update
getSetCCResultType's documentation to reflect its actual uses.
Patch by deadal nix!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219141 91177308-0d34-0410-b5e6-96231b3b80d8
This trades a (register-renamer-friendly) movaps for a floating point
/ integer domain cross. That is a very bad trade, even on architectures
where domain crossing is relatively fast. On any chip where there is
even a cycle stall, this is a Very Bad Idea. It doesn't even seem likely
to cause a spill to be introduced because the reason for the copy is to
destructively shuffle in place.
Thanks to Ben Kramer for fixing a bug in this code that my new shuffle
lowering exposed and highlighting that perhaps it should just go away.
=]
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219090 91177308-0d34-0410-b5e6-96231b3b80d8
that are unused.
This allows the combiner to delete math feeding shuffles where the math
isn't actually necessary. This improves some of the vperm2x128 tests
that regressed when the vector shuffle lowering started actually
generating vperm instructions rather than forcibly decomposing them.
Sadly, this isn't enough to get this *really* right because we still
form a completely unnecessary permutation. To fix that, we also need to
fold shuffles which just rearrange concatenated or inserted subvectors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219086 91177308-0d34-0410-b5e6-96231b3b80d8
It's debatable whether this transform is useful at all, but for now make sure
we don't generate invalid asm.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219084 91177308-0d34-0410-b5e6-96231b3b80d8
new vector shuffle lowering.
This is loosely based on a patch by Marius Wachtler to the PR (thanks!).
I refactored it a bi to use std::count_if and a mutable array ref but
the core idea was exactly right. I also added some direct testing of
this case.
I believe PR21137 is now the only remaining regression.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219081 91177308-0d34-0410-b5e6-96231b3b80d8
shuffles using AVX and AVX2 instructions. This fixes PR21138, one of the
few remaining regressions impacting benchmarks from the new vector
shuffle lowering.
You may note that it "regresses" many of the vperm2x128 test cases --
these were actually "improved" by the naive lowering that the new
shuffle lowering previously did. This regression gave me fits. I had
this patch ready-to-go about an hour after flipping the switch but
wasn't sure how to have the best of both worlds here and thought the
correct solution might be a completely different approach to lowering
these vector shuffles.
I'm now convinced this is the correct lowering and the missed
optimizations shown in vperm2x128 are actually due to missing
target-independent DAG combines. I've even written most of the needed
DAG combine and will submit it shortly, but this part is ready and
should help some real-world benchmarks out.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219079 91177308-0d34-0410-b5e6-96231b3b80d8
Update the entire regression test suite for the new shuffles. Remove
most of the old testing which was devoted to the old shuffle lowering
path and is no longer relevant really. Also remove a few other random
tests that only really exercised shuffles and only incidently or without
any interesting aspects to them.
Benchmarking that I have done shows a few small regressions with this on
LNT, zero measurable regressions on real, large applications, and for
several benchmarks where the loop vectorizer fires in the hot path it
shows 5% to 40% improvements for SSE2 and SSE3 code running on Sandy
Bridge machines. Running on AMD machines shows even more dramatic
improvements.
When using newer ISA vector extensions the gains are much more modest,
but the code is still better on the whole. There are a few regressions
being tracked (PR21137, PR21138, PR21139) but by and large this is
expected to be a win for x86 generated code performance.
It is also more correct than the code it replaces. I have fuzz tested
this extensively with ISA extensions up through AVX2 and found no
crashes or miscompiles (yet...). The old lowering had a few miscompiles
and crashers after a somewhat smaller amount of fuzz testing.
There is one significant area where the new code path lags behind and
that is in AVX-512 support. However, there was *extremely little*
support for that already and so this isn't a significant step backwards
and the new framework will probably make it easier to implement lowering
that uses the full power of AVX-512's table-based shuffle+blend (IMO).
Many thanks to Quentin, Andrea, Robert, and others for benchmarking
assistance. Thanks to Adam and others for help with AVX-512. Thanks to
Hal, Eric, and *many* others for answering my incessant questions about
how the backend actually works. =]
I will leave the old code path in the tree until the 3 PRs above are at
least resolved to folks' satisfaction. Then I will rip it (and 1000s of
lines of code) out. =] I don't expect this flag to stay around for very
long. It may not survive next week.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219046 91177308-0d34-0410-b5e6-96231b3b80d8
the various ways in which blends can be used to do vector element
insertion for lowering with the scalar math instruction forms that
effectively re-blend with the high elements after performing the
operation.
This then allows me to bail on the element insertion lowering path when
we have SSE4.1 and are going to be doing a normal blend, which in turn
restores the last of the blends lost from the new vector shuffle
lowering when I got it to prioritize insertion in other cases (for
example when we don't *have* a blend instruction).
Without the patterns, using blends here would have regressed
sse-scalar-fp-arith.ll *completely* with the new vector shuffle
lowering. For completeness, I've added RUN-lines with the new lowering
here. This is somewhat superfluous as I'm about to flip the default, but
hey, it shows that this actually significantly changed behavior.
The patterns I've added are just ridiculously repetative. Suggestions on
making them better very much welcome. In particular, handling the
commuted form of the v2f64 patterns is somewhat obnoxious.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219033 91177308-0d34-0410-b5e6-96231b3b80d8
perform a load to use blendps rather than movss when it is available.
For non-loads, blendps is *much* faster. It can execute on two ports in
Sandy Bridge and Ivy Bridge, and *three* ports on Haswell. This fixes
one of the "regressions" from aggressively taking the "insertion" path
in the new vector shuffle lowering.
This does highlight one problem with blendps -- it isn't commuted as
heavily as it should be. That's future work though.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219022 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r218918, effectively reapplying r218914 after fixing
an Ocaml bindings test and an Asan crash. The root cause of the latter
was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a
PR to investigate who requires the loose check (and why).
Original commit message follows.
--
This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString. Integers are stringified and
a `\0` character is used as a separator.
Part of PR17891.
Note: I've attached my testcases upgrade scripts to the PR. If I've
just broken your out-of-tree testcases, they might help.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219010 91177308-0d34-0410-b5e6-96231b3b80d8
In the X86 backend, matching an address is initiated by the 'addr' complex
pattern and its friends. During this process we may reassociate and-of-shift
into shift-of-and (FoldMaskedShiftToScaledMask) to allow folding of the
shift into the scale of the address.
However as demonstrated by the testcase, this can trigger CSE of not only the
shift and the AND which the code is prepared for but also the underlying load
node. In the testcase this node is sitting in the RecordedNode and MatchScope
data structures of the matcher and becomes a deleted node upon CSE. Returning
from the complex pattern function, we try to access it again hitting an assert
because the node is no longer a load even though this was checked before.
Now obviously changing the DAG this late is bending the rules but I think it
makes sense somewhat. Outside of addresses we prefer and-of-shift because it
may lead to smaller immediates (FoldMaskAndShiftToScale is an even better
example because it create a non-canonical node). We currently don't recognize
addresses during DAGCombiner where arguably this canonicalization should be
performed. On the other hand, having this in the matcher allows us to cover
all the cases where an address can be used in an instruction.
I've also talked a little bit to Dan Gohman on llvm-dev who added the RAUW for
the new shift node in FoldMaskedShiftToScaledMask. This RAUW is responsible
for initiating the recursive CSE on users
(http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-September/076903.html) but it
is not strictly necessary since the shift is hooked into the visited user. Of
course it's safer to keep the DAG consistent at all times (e.g. for accurate
number of uses, etc.).
So rather than changing the fundamentals, I've decided to continue along the
previous patches and detect the CSE. This patch installs a very targeted
DAGUpdateListener for the duration of a complex-pattern match and updates the
matching state accordingly. (Previous patches used HandleSDNode to detect the
CSE but that's not practical here). The listener is only installed on X86.
I tested that there is no measurable overhead due to this while running
through the spec2k BC files with llc. The only thing we pay for is the
creation of the listener. The callback never ever triggers in spec2k since
this is a corner case.
Fixes rdar://problem/18206171
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219009 91177308-0d34-0410-b5e6-96231b3b80d8
The test Atomics-32.ll was both redundant (all operations are also checked by
atomics.ll at least) and not actually checking correctness (it was not using
FileCheck, just verifying that the compiler does not crash).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218997 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
hwsync is only required for seq_cst fences, acquire and release one can use
the cheaper lwsync.
Test Plan: Added some cases to atomics.ll + make check-all
Reviewers: jfb, wschmidt
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5317
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218995 91177308-0d34-0410-b5e6-96231b3b80d8
and MOVSD nodes for single element vector inserts.
This is particularly important because a number of patterns in the
backend detect these patterns and leverage them to simplify things. It
also fixes quite a few of the insertion bad code examples. However, it
regresses a specific area: when available, blendps and blendpd are
*dramatically* faster than movss and movsd respectively. But it doesn't
really work to form the blend logic first because the blends *aren't* as
crazy efficient when the data is coming from memory anyways, and thus
will have a movss or movsd regardless. Also, doing that would block
a bunch of the patterns that this is designed to hit.
So my plan is to go into the patterns for lowering MOVSS and MOVSD and
lower them via blends when available. However that's a pretty invasive
restructuring so it will need to be a follow-up patch.
I have already gone into the patterns to lower MOVSS and MOVSD from
memory using MOVLPD, etc. Without that, several of the test cases
I already have regress.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218985 91177308-0d34-0410-b5e6-96231b3b80d8
That commit was introduced in order to help investigate a problem in ARM
codegen breaking from commit 202304 (Add a limit to the heuristic that register
allocates instructions in local order). Recent analisys indicated that the
problem no longer exists, so I'm reverting this change.
See PR18996.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218981 91177308-0d34-0410-b5e6-96231b3b80d8
I got them quite wrong when updating it and had the SSE4.1 run checked
for SSE2 and the SSE2 run checked for SSE4.1. I think everything was
actually generic SSE, but this still seems good to fix. While here,
hoist the triple into the IR and make the flag set a bit more direct in
what it is trying to test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218978 91177308-0d34-0410-b5e6-96231b3b80d8
lowering to match VZEXT_MOVL patterns.
I hadn't realized that these had sufficient pattern smarts in the
backend to lower zext-ing from the low element of a vector without it
being a scalar_to_vector node. They do, and this is how to match a bunch
of patterns for movq, movss, etc.
There is a weird propensity to end up using pshufd to place the element
afterward even though it means domain crossing (or rather, to use
xorps+movss to zext the element rather than movq) but that's an
orthogonal problem with VZEXT_MOVL that someone should probably look at.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218977 91177308-0d34-0410-b5e6-96231b3b80d8
vector to a zero vector for the v2 cases and fix the v4 integer cases to
actually blend from a vector.
There are already seprate tests for the case of inserting from a scalar.
These cases cover a lot of the regressions I've seen in the regression
test suite for the new vector shuffle lowering and specifically cover
the reported lack of using various zext-ing instruction patterns. My
next patch should fix a big chunk of this, but wanted to get a nice
baseline for these patterns in the test cases first.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218976 91177308-0d34-0410-b5e6-96231b3b80d8
element types to form illegal vector types.
I've added a special SSE1 test case here that makes sure we don't break
this going forward.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218974 91177308-0d34-0410-b5e6-96231b3b80d8
testing that we generated divps and divss but not in a very systematic
way. There are other tests for widening binary operations already that
make these unnecessary.
The second one seems mostly about testing Atom as well as normal X86,
but despite the comment claiming it is testing a different instruction
sequence, it then tests for exactly the same div instruction sequence!
(The sequence of instructions is actually quite different on Atom, but
not the sequence of div instructions....)
And then it has an "execution" test that simply isn't run? Very strange.
Anyways, none of this is really needed so clean this up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218972 91177308-0d34-0410-b5e6-96231b3b80d8
a test case that was just grepping the debug stats output rather than
actually checking the generated code for anything useful.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218951 91177308-0d34-0410-b5e6-96231b3b80d8
intergrated much more fully into some logical part of the backend to
really understand what it is trying to accomplish and how to update it.
I suspect it no longer holds enough value to be worth having.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218950 91177308-0d34-0410-b5e6-96231b3b80d8
shufle switch.
I nuked a win64 config from one test as it doesn't really make sense to
cover that ABI specially for generic v2f32 tests...
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218948 91177308-0d34-0410-b5e6-96231b3b80d8
two functions that really didn't have any interesting assertions, and
generated more precise tests for one of the others.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218946 91177308-0d34-0410-b5e6-96231b3b80d8
test cases that will change with the new vector shuffle lowering. This
gives us a nice baseline for deltas against. I've checked and removed
the cases where there were weird register usage being pinned down, and
all of these are extremely pin-pointed tests so fully checking them
seems very appropriate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218941 91177308-0d34-0410-b5e6-96231b3b80d8
