This allows vector-sized store merging of constants in DAGCombiner using the existing code in MergeConsecutiveStores().
All of the twisted logic that decides exactly what vector operations are legal and fast for each particular CPU are
handled separately in there using the appropriate hooks.
For the motivating tests in merge-store-constants.ll, we already produce the same vector code in IR via the SLP vectorizer.
So this is just providing a backend backstop for code that doesn't go through that pass (-O1). More details in PR24449:
https://bugs.llvm.org/show_bug.cgi?id=24449 (this change should be the last step to resolve that bug)
Differential Revision: https://reviews.llvm.org/D37451
llvm-svn: 313458
We just need to toggle bits 1 and 5 of the immediate and swap the sources. The peephole pass could trigger commuting/folding for this later, but its easy enough to fix in isel.
Disable the peephole pass on the main vperm2x128 test so we know we're doing this through isel.
llvm-svn: 313455
I've moved the test cases from the InstCombine optimizations to the backend to keep the coverage we had there. It covered every possible immediate so I've preserved the resulting shuffle mask for each of those immediates.
llvm-svn: 313450
I'm going to autoupgrade these intrinsics in a future commit. This bit will never be set in the resulting output so pre-removing the bit.
llvm-svn: 313434
This caused PR34629: asserts firing when building Chromium. It also broke some
buildbots building test-suite as reported on the commit thread.
> Summary:
> 1/ Operand folding during complex pattern matching for LEAs has been
> extended, such that it promotes Scale to accommodate similar operand
> appearing in the DAG.
> e.g.
> T1 = A + B
> T2 = T1 + 10
> T3 = T2 + A
> For above DAG rooted at T3, X86AddressMode will no look like
> Base = B , Index = A , Scale = 2 , Disp = 10
>
> 2/ During OptimizeLEAPass down the pipeline factorization is now performed over LEAs
> so that if there is an opportunity then complex LEAs (having 3 operands)
> could be factored out.
> e.g.
> leal 1(%rax,%rcx,1), %rdx
> leal 1(%rax,%rcx,2), %rcx
> will be factored as following
> leal 1(%rax,%rcx,1), %rdx
> leal (%rdx,%rcx) , %edx
>
> 3/ Aggressive operand folding for AM based selection for LEAs is sensitive to loops,
> thus avoiding creation of any complex LEAs within a loop.
>
> Reviewers: lsaba, RKSimon, craig.topper, qcolombet
>
> Reviewed By: lsaba
>
> Subscribers: spatel, igorb, llvm-commits
>
> Differential Revision: https://reviews.llvm.org/D35014
llvm-svn: 313376
The early out for AVX2 in lowerV2X128VectorShuffle is positioned in a weird spot below some shuffle mask equivalency checks.
But I think we want to allow VPERMQ for any unary shuffle.
Differential Revision: https://reviews.llvm.org/D37893
llvm-svn: 313373
When handling a v64i1 build vector of constants on 32-bit targets we were creating an illegal i64 constant that we then bitcasted back to v64i1. We need to instead create two 32-bit constants, bitcast them to v32i1 and concat the result. We should also take care to handle the halves being all zeros/ones after the split.
This patch splits the build vector and then recursively lowers the two pieces. This allows us to handle the all ones and all zeros cases with minimal effort. Ideally we'd just do the split and concat, and let lowering get called again on the new nodes, but getNode has special handling for CONCAT_VECTORS that reassembles the pieces back into a single BUILD_VECTOR. Hopefully the two temporary BUILD_VECTORS we had to create to do this that don't get returned don't cause any issues.
Fixes PR34605.
Differential Revision: https://reviews.llvm.org/D37858
llvm-svn: 313366
Currently if we're inserting 0s into the upper elements of a vector register we insert an explicit move of the smaller register to implicitly zero the upper bits. But if we can prove that they are already zero we can skip that. This is based on a similar idea of what we do to avoid emitting explicit zero extends for GR32->GR64.
Unfortunately, this is harder for vector registers because there are several opcodes that don't have VEX equivalent instructions, but can write to XMM registers. Among these are SHA instructions and a MMX->XMM move. Bitcasts can also get in the way.
So for now I'm starting with explicitly allowing only VPMADDWD because we emit zeros in combineLoopMAddPattern. So that is placing extra instruction into the reduction loop.
I'd like to allow PSADBW as well after D37453, but that's currently blocked by a bitcast. We either need to peek through bitcasts or canonicalize insert_subvectors with zeros to remove bitcasts on the value being inserted.
Longer term we should probably have a cleanup pass that removes superfluous zeroing moves even when the producer is in another basic block which is something these isel tricks can't do. See PR32544.
Differential Revision: https://reviews.llvm.org/D37653
llvm-svn: 313365
This adds support for allowing v8f16 vector types, thus avoiding conversions
from/to single precision for these types. This is a follow up patch of
commits r311154 and r312104, which added support for scalars and v4f16
types, respectively.
Differential Revision: https://reviews.llvm.org/D37802
llvm-svn: 313351
Summary:
1/ Operand folding during complex pattern matching for LEAs has been
extended, such that it promotes Scale to accommodate similar operand
appearing in the DAG.
e.g.
T1 = A + B
T2 = T1 + 10
T3 = T2 + A
For above DAG rooted at T3, X86AddressMode will no look like
Base = B , Index = A , Scale = 2 , Disp = 10
2/ During OptimizeLEAPass down the pipeline factorization is now performed over LEAs
so that if there is an opportunity then complex LEAs (having 3 operands)
could be factored out.
e.g.
leal 1(%rax,%rcx,1), %rdx
leal 1(%rax,%rcx,2), %rcx
will be factored as following
leal 1(%rax,%rcx,1), %rdx
leal (%rdx,%rcx) , %edx
3/ Aggressive operand folding for AM based selection for LEAs is sensitive to loops,
thus avoiding creation of any complex LEAs within a loop.
Reviewers: lsaba, RKSimon, craig.topper, qcolombet
Reviewed By: lsaba
Subscribers: spatel, igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D35014
llvm-svn: 313343
Because the stack growth direction and addressing is done
in the same direction, modifying SP at the beginning of the
call sequence was incorrect. If we had a stack passed argument,
we would end up skipping that number of bytes before pushing
arguments, leaving unused/inconsistent space.
The callee creates fixed stack objects in its frame, so
the space necessary for these is already logically allocated
in the callee, so we just let the callee increment SP if
it really requires it.
llvm-svn: 313279
Using SplitCSR for the frame register was very broken. Often
the copies in the prolog and epilog were optimized out, in addition
to them being inserted after the true prolog where the FP
was clobbered.
I have a hacky solution which works that continues to use
split CSR, but for now this is simpler and will get to working
programs.
llvm-svn: 313274
This patch complements D16810 "[mips] Make isel select the correct DEXT variant
up front.". Now ISel picks the right variant of DINS, so now there is no need
to replace DINS with the appropriate variant during
MipsMCCodeEmitter::encodeInstruction().
This patch also enables target specific instruction verification for ins, dins,
dinsm, dinsu, ext, dext, dextm, dextu. These instructions have constraints that
are checked when generating MipsISD::Ins and MipsISD::Ext nodes, but these
constraints are not checked during instruction selection. Adding machine
verification should catch outstanding cases.
Finally, correct a bug that instruction verification uncovered, where the
position operand of a DINSU generated during lowering was being silently
and accidently corrected to the correct value.
Reviewers: slthakur
Differential Revision: https://reviews.llvm.org/D34809
llvm-svn: 313254
We already have a combine for this pattern when the input to shl is add, so we just need to enable the transformation when the input is or.
Original patch by @tstellar
Differential Revision: https://reviews.llvm.org/D19325
llvm-svn: 313251
Summary:
XRay had been assuming that the previous section is the "text" section
of the function when lowering the instrumentation map. Unfortunately
this is not a safe assumption, because we may be coming from lowering
debug type information for the function being lowered.
This fixes an issue with combining -gsplit-dwarf, -generate-type-units,
-debug-compile and -fxray-instrument for sole member functions. When the
split dwarf section is stripped, we're left with references from the
xray_instr_map to the debug section. The change now uses the function's
symbol instead of the previous section's start symbol.
We found the bug while attempting to strip the split debug sections off
an XRay-instrumented object file, which had a peculiar edge-case for
single-function classes where the single function is being lowered.
Because XRay had assocaited the instrumentation map for a function to
the debug types section instead of the function's section, the objcopy
call will fail due to the misplaced reference from the xray_instr_map
section.
Reviewers: pcc, dblaikie, echristo
Subscribers: llvm-commits, aprantl
Differential Revision: https://reviews.llvm.org/D37791
llvm-svn: 313233
This caused PR34596.
> [MachineCombiner] Update instruction depths incrementally for large BBs.
>
> Summary:
> For large basic blocks with lots of combinable instructions, the
> MachineTraceMetrics computations in MachineCombiner can dominate the compile
> time, as computing the trace information is quadratic in the number of
> instructions in a BB and it's relevant successors/predecessors.
>
> In most cases, knowing the instruction depth should be enough to make
> combination decisions. As we already iterate over all instructions in a basic
> block, the instruction depth can be computed incrementally. This reduces the
> cost of machine-combine drastically in cases where lots of instructions
> are combined. The major drawback is that AFAIK, computing the critical path
> length cannot be done incrementally. Therefore we only compute
> instruction depths incrementally, for basic blocks with more
> instructions than inc_threshold. The -machine-combiner-inc-threshold
> option can be used to set the threshold and allows for easier
> experimenting and checking if using incremental updates for all basic
> blocks has any impact on the performance.
>
> Reviewers: sanjoy, Gerolf, MatzeB, efriedma, fhahn
>
> Reviewed By: fhahn
>
> Subscribers: kiranchandramohan, javed.absar, efriedma, llvm-commits
>
> Differential Revision: https://reviews.llvm.org/D36619
llvm-svn: 313213
MachineScheduler when clustering loads or stores checks if base
pointers point to the same memory. This check is done through
comparison of base registers of two memory instructions. This
works fine when instructions have separate offset operand. If
they require a full calculated pointer such instructions can
never be clustered according to such logic.
Changed shouldClusterMemOps to accept base registers as well and
let it decide what to do about it.
Differential Revision: https://reviews.llvm.org/D37698
llvm-svn: 313208
This is to fix PR34502. After rL311401, the live range of spilled vreg will be
cleared. HoistSpill need to use the live range of the original vreg before splitting
to know the moving range of the spills. The patch saves a copy of live interval for
the spilled vreg inside of HoistSpillHelper.
Differential Revision: https://reviews.llvm.org/D37578
llvm-svn: 313197
NFC.
Replacing -mcpu=skx by -mattr in the run command of the codegen test: avx512-gather-scatter-intrin.ll.
Reviewers: delena
Revision: https://reviews.llvm.org/D37799
llvm-svn: 313144
NFC.
Updating the code gen regression test bmi2-schedule.ll to use the KNL and SKYLAKE prefixes for the run commands that use the knl and Skylake mcpu options.
The fix is in preparation for a large patch of adding all SKL scheduling information.
Reviewers: delena, zvi, RKSimon
Revision: https://reviews.llvm.org/D37796
llvm-svn: 313138
NFC.
Updating the code gen regression test bmi2-schedule.ll to use the KNL and SKYLAKE prefixes for the run commands that use the knl and Skylake mcpu options.
The fix is in preparation for a large patch of adding all SKL scheduling information.
Reviewers: delena, zvi
Revision: https://reviews.llvm.org/D37796
llvm-svn: 313137
Updating codegen test bmi2-schedule.ll to use the SKYLAKE and KNL prefix as preparatipn for an upcoming patch to add all SKL scheduling information.
llvm-svn: 313136
Add explicit mc-encoding checks showing that the AVX512VL ABS intrinsics are actually mapped to EVEX encoding.
This is a pre-commit for a soon to come patch which will lower x86 target specific ABS intrinsics to IR.
Differential Revision: https://reviews.llvm.org/D37688
llvm-svn: 313131
Load with zero-extend and sign-extend from v2i8 to v2i32 is "Legal" since SSE4.1 and may be performed using PMOVZXBD , PMOVSXBD instructions.
llvm-svn: 313121
The masked store instruction only cares about the sign-bit of each mask element,
so the compare s<0 isn't needed.
As noted in PR11210:
https://bugs.llvm.org/show_bug.cgi?id=11210
...fixing this should allow us to eliminate x86-specific masked store intrinsics in IR.
(Although more testing will be needed to confirm that.)
I filed a bug to track improvements for AVX512:
https://bugs.llvm.org/show_bug.cgi?id=34584
Differential Revision: https://reviews.llvm.org/D37446
llvm-svn: 313089