There were cases where a do-while loop would be converted to a while
loop before finding out that it would be unsafe to expand the SCEV in
this situation and then bailing out of hardware loop conversion.
This patch checks if it would be unsafe to expand the SCEV and if so stops converting the do-while into a while, allowing conversion to a hardware loop.
Differential Revision: https://reviews.llvm.org/D83953
As explained in the comment:
// For a FLAT instruction the hardware decides whether to access
// global/scratch/shared memory based on the high bits of vaddr,
// ignoring the offset field, so we have to ensure that when we add
// remainder to vaddr it still points into the same underlying object.
// The easiest way to do that is to make sure that we split the offset
// into two pieces that are both >= 0 or both <= 0.
In particular FLAT (as opposed to SCRATCH and GLOBAL) instructions have
an unsigned immediate offset field, so we can't use it to help split a
negative offset.
Differential Revision: https://reviews.llvm.org/D83394
tryLatency compares two sched candidates. For the top zone it prefers
the one with lesser depth, but only if that depth is greater than the
total latency of the instructions we've already scheduled -- otherwise
its latency would be hidden and there would be no stall.
Unfortunately it only tests the depth of one of the candidates. This can
lead to situations where the TopDepthReduce heuristic does not kick in,
but a lower priority heuristic chooses the other candidate, whose depth
*is* greater than the already scheduled latency, which causes a stall.
The fix is to apply the heuristic if the depth of *either* candidate is
greater than the already scheduled latency.
All this also applies to the BotHeightReduce heuristic in the bottom
zone.
Differential Revision: https://reviews.llvm.org/D72392
MBBs are not allowed to have non-terminator instructions after the first
terminator. Currently in some cases (see the modified test),
EmitSchedule can add DBG_VALUEs after the last terminator, for example
when referring a debug value that gets folded into a TCRETURN
instruction on ARM.
This patch updates EmitSchedule to move inserted DBG_VALUEs just before
the first terminator. I am not sure if there are terminators produce
values that can in turn be used by a DBG_VALUE. In that case, moving the
DBG_VALUE might result in referencing an undefined register. But in any
case, it seems like currently there is no way to insert a proper DBG_VALUEs
for such registers anyways.
Alternatively it might make sense to just remove those extra DBG_VALUES.
I am not too familiar with the details of debug info in the backend and
would appreciate any suggestions on how to address the issue in the best
possible way.
Reviewers: vsk, aprantl, jpaquette, efriedma, paquette
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D83561
I meant to do this in D83913, but missed it while updating the
feature list.
Interestingly I think this is disabling the postRA scheduler. But
it does match our default 64-bit behavior.
Reviewed By: echristo
Differential Revision: https://reviews.llvm.org/D83996
When SCC is dead, but VCC is required then replace s_and / s_andn2
with s_mov into VCC when mask value is 0 or -1.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D83850
Accounting for the fact that Wasm function indices are 32-bit, but in wasm64 we want uniform 64-bit pointers.
Includes reloc types for 64-bit table indices.
Differential Revision: https://reviews.llvm.org/D83729
There was a lot of duplicate code here for checking the VT and
subtarget. Moving it into a helper avoids that.
It also fixes a bug that combineAdd reused Op0/Op1 after a call
to isHorizontalBinOp may have changed it. The new helper function
has its own local version of Op0/Op1 that aren't shared by other
code.
Fixes PR46455.
Reviewed By: spatel, bkramer
Differential Revision: https://reviews.llvm.org/D83971
Alternative to D83897. I believe the big change here is that I removed slow unaligned memory 16
Down side that it may adversely effect tuning if someone explicitly targets -march=pentium4 and expects pentium4 tuned code. Of course pentium4 is so old our default behavior with the previous settings may not have been the best either.
Reviewed By: echristo, RKSimon
Differential Revision: https://reviews.llvm.org/D83913
Although the SIMD spec proposal does not specifically include a
select instruction, the select instruction in MVP WebAssembly is
polymorphic over the selected types, so it is able to work on v128
values when they are enabled. This patch introduces a new variant of
the select instruction for each legal vector type. Additional ISel
patterns are adapted from the SELECT_I32 and SELECT_I64 patterns.
Depends on D83736.
Differential Revision: https://reviews.llvm.org/D83737
Updating the simd-select.ll tests manually with consistent named
regexps for the register numbers was taking more time than it was
worth, so this patch updates that test file to have autogenerated
output. This is not a significant readability regression because the
tests in that file are all very small.
Depends on D83734.
Differential Revision: https://reviews.llvm.org/D83736
We were previously expanding vselect and matching on the expansion to
generate bitselects, but in some cases the expansion would be further
combined and a bitselect would not get generated. This patch improves
codegen in those cases by legalizing vselect and lowering it to
v128.bitselect. The old pattern that matches the expansion is still
useful for lowering IR that already uses the expansion rather than a
select operation.
Differential Revision: https://reviews.llvm.org/D83734
In an upcoming AMDGPU patch, the scalar cases will be legal and vector
ops should be scalarized, rather than producing a long sequence of
vector ops which will also need to be scalarized.
Use a lazy heuristic that seems to work and improves the thumb2 MVE
test.
Basic support for variadic-def MIR Statepoint:
- Change TableGen STATEPOINT description to variadic out list
(For self-documentation purpose; by itself it does not affect
code generation in any way).
- Update StatepointOpers helper class to handle variadic defs.
- Update MachineVerifier to properly handle them, too.
With this change, new Statepoint instruction can be passed through
backend (excluding ISEL) without errors.
Full change set is available at D81603.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D81645
The hardware has created a real mess in the naming for add/sub, which
have been renamed basically every generation. Switch the carry out
pseudos to have the gfx9/gfx10 names. We were using the original SI/CI
v_add_i32/v_sub_i32 names. Later targets reintroduced these names as
carryless instructions with a saturating clamp bit, which we do not
define. Do this rename so we can unambiguously add these missing
instructions.
The carry-in versions should also be renamed, but at least those had a
consistent _u32 name to begin with. The 16-bit instructions were also
renamed, but aren't ambiguous.
This does regress assembler error message quality in some cases. In
mismatched wave32/wave64 situations, this will switch from
"unsupported instruction" to "invalid operand", with the error
pointing at the wrong position. I couldn't quite follow how the
assembler selects these, but the previous behavior seemed accidental
to me. It looked like there was a partial attempt to handle this which
was never completed (i.e. there is an AMDGPUOperand::isBoolReg but it
isn't used for anything).
Add widenScalar for TypeIdx == 0 for G_SITOFP/G_UITOFP.
Legailize, using widenScalar, as s64->s32 G_SITOFP/G_UITOFP
followed by s32->s16 G_FPTRUNC.
Differential Revision: https://reviews.llvm.org/D83880
This function has a bug which will incorrectly reschedule instructions
after an INLINEASM_BR (which can branch). (The bug may also allow
scheduling past a throwing-CALL, I'm not certain.)
I could fix that bug, but, as the removed FIXME notes, it's better to
attempt rescheduling before converting to 3-addr form, as that may
remove the need to convert in the first place. In fact, the code to do
such reordering was added to this pass only a few months later, in
2011, via the addition of the function rescheduleMIBelowKill. That
code does not contain the same bug.
The removal of the sink3AddrInstruction function is not a no-op: in
some cases it would move an instruction post-conversion, when
rescheduleMIBelowKill would not move the instruction pre-converison.
However, this does not appear to be important: the machine instruction
scheduler can reorder the after-conversion instructions, in any case.
This patch fixes a kernel panic 4.4 LTS x86_64 Linux kernels, when
built with clang after 4b0aa5724f.
Link: https://github.com/ClangBuiltLinux/linux/issues/1085
Differential Revision: https://reviews.llvm.org/D83708
Lower the operations to predicated variants. This is prep work
required for fixed length code generation but also fixes a bug
whereby these operations fail selection when "unpacked" vector
types (e.g. MVT::nxv2f32) are used.
This patch also adds the missing "unpacked" patterns for FMA.
Differential Revision: https://reviews.llvm.org/D83765
Vector bitwise selects are matched by pseudo VBSP instruction
and expanded to VBSL/VBIT/VBIF after register allocation
depend on operands registers to minimize extra copies.
This adds a peephole optimisation to turn a t2MOVccr that could not be
folded into any other instruction into a CSEL on 8.1-m. The t2MOVccr
would usually be expanded into a conditional mov, that becomes an IT;
MOV pair. We can instead generate a CSEL instruction, which can
potentially be smaller and allows better register allocation freedom,
which can help reduce codesize. Performance is more variable and may
depend on the micrarchitecture details, but initial results look good.
If we need to control this per-cpu, we can add a subtarget feature as we
need it.
Original patch by David Penry.
Differential Revision: https://reviews.llvm.org/D83566
Summary:
This patch modifies IncrementMemoryAddress to use a vscale
when calculating the new address if the data type is scalable.
Also adds tablegen patterns which match an extract_subvector
of a legal predicate type with zip1/zip2 instructions
Reviewers: sdesmalen, efriedma, david-arm
Reviewed By: efriedma, david-arm
Subscribers: tschuett, hiraditya, psnobl, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83137
Previously, the vins* intrinsic was incorrectly defined to have its second and
third argument arguments as an i64. This patch fixes the second and third
argument of the vins* instruction and intrinsic to have i32s instead.
Differential Revision: https://reviews.llvm.org/D83497
Summary:
llc reports `fp32-denormals` is not recognized. I guess it was intended to be
`-denormal-fp-math-f32={preserve-sign|ieee} -mattr=+mad-mac-f32-insts`
Reviewers: rampitec
Reviewed By: rampitec
Subscribers: jvesely, nhaehnle, llvm-commits, kerbowa
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83883
This patch provides optimization of bit manipulation operations by
enabling the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific
bit-manip instructions from the ternary subset (zbt subextension) of the
experimental B extension of RISC-V.
It adds also the correspondent codegen tests.
This patch is based on Claire Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf
Differential Revision: https://reviews.llvm.org/D79875
This patch provides optimization of bit manipulation operations by
enabling the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific
bit-manip instructions from the single-bit subset (zbs subextension) of
the experimental B extension of RISC-V.
It adds also the correspondent codegen tests.
This patch is based on Claire Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf
Differential Revision: https://reviews.llvm.org/D79874
This patch provides optimization of bit manipulation operations by
enabling the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific
bit-manip instructions belonging to both the permutation and the base
subsets of the experimental B extension of RISC-V.
It adds also the correspondent codegen tests.
This patch is based on Claire Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf
Differential Revision: https://reviews.llvm.org/D79873
This patch provides optimization of bit manipulation operations by
enabling the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific
bit-manip instructions from the permutation subset (zbp subextension) of
the experimental B extension of RISC-V.
It adds also the correspondent codegen tests.
This patch is based on Claire Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf
Differential Revision: https://reviews.llvm.org/D79871
This patch provides optimization of bit manipulation operations by
enabling the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific
bit-manip instructions from the base subset (zbb subextension) of the
experimental B extension of RISC-V.
It adds also the correspondent codegen tests.
This patch is based on Claire Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf
Differential Revision: https://reviews.llvm.org/D79870
The existing code already considered this case. Unfortunately a typo in
the condition prevents it from triggering. Also the existing code, had
it run, forgot to do the folding.
This fixes PR42876.
Differential Revision: https://reviews.llvm.org/D65802
Add handling of s_andn2 and mask of 0.
This eliminates redundant instructions from uniform control flow.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D83641
This patch handles CFI with basic block sections, which unlike DebugInfo does
not support ranges. The DWARF standard explicitly requires emitting separate
CFI Frame Descriptor Entries for each contiguous fragment of a function. Thus,
the CFI information for all callee-saved registers (possibly including the
frame pointer, if necessary) have to be emitted along with redefining the
Call Frame Address (CFA), viz. where the current frame starts.
CFI directives are emitted in FDE’s in the object file with a low_pc, high_pc
specification. So, a single FDE must point to a contiguous code region unlike
debug info which has the support for ranges. This is what complicates CFI for
basic block sections.
Now, what happens when we start placing individual basic blocks in unique
sections:
* Basic block sections allow the linker to randomly reorder basic blocks in the
address space such that a given basic block can become non-contiguous with the
original function.
* The different basic block sections can no longer share the cfi_startproc and
cfi_endproc directives. So, each basic block section should emit this
independently.
* Each (cfi_startproc, cfi_endproc) directive will result in a new FDE that
caters to that basic block section.
* Now, this basic block section needs to duplicate the information from the
entry block to compute the CFA as it is an independent entity. It cannot refer
to the FDE of the original function and hence must duplicate all the stuff that
is needed to compute the CFA on its own.
* We are working on a de-duplication patch that can share common information in
FDEs in a CIE (Common Information Entry) and we will present this as a follow up
patch. This can significantly reduce the duplication overhead and is
particularly useful when several basic block sections are created.
* The CFI directives are emitted similarly for registers that are pushed onto
the stack, like callee saved registers in the prologue. There are cfi
directives that emit how to retrieve the value of the register at that point
when the push happened. This has to be duplicated too in a basic block that is
floated as a separate section.
Differential Revision: https://reviews.llvm.org/D79978
Because of the layout of stores (that don't have a destination operand)
this check is exactly the same as the one in
RISCVInstrInfo::isLoadFromStackSlot.
Differential Revision: https://reviews.llvm.org/D81805
ComputeNumSignBits and computeKnownBits both trigger "Scalable flag
may be dropped" warnings when a fixed length vector is extracted
from a scalable vector. This patch assumes nothing about the
demanded elements thus matching the behaviour when extracting a
scalable vector from a scalable vector.
Differential Revision: https://reviews.llvm.org/D83642
Fix two obvious errors in the code and also update the test check.
Also add one test to catch the failure.
Patch by Ruiling Song!
Differential Revision: https://reviews.llvm.org/D83280
The code already supports addressing a fixed-size stack object from
the frame-pointer, by first subtracting sizeof(SVE area) from FP.
Reviewers: efriedma, cameron.mcinally, david-arm, rengolin
Reviewed By: david-arm
Differential Revision: https://reviews.llvm.org/D83125
The hardware spec require src0 of s_cmpk should be a register. So, we
should not optimize s_cmp to s_cmpk if src0 is not register.
Patch by Ruiling Song!
I have added a new file:
llvm/test/CodeGen/AArch64/README
that describes what to do in the event one of the SVE codegen tests
fails the warnings check. In addition, I've added comments to all
the relevant SVE tests pointing users at the README file.
Differential Revision: https://reviews.llvm.org/D83467
In DAGCombiner::TransformFPLoadStorePair we were dropping the scalable
property of TypeSize when trying to create an integer type of equivalent
size. In fact, this optimisation makes no sense for scalable types
since we don't know the size at compile time. I have changed the code
to bail out when encountering scalable type sizes.
I've added a test to
llvm/test/CodeGen/AArch64/sve-fp.ll
that exercises this code path. The test already emits an error if it
encounters warnings due to implicit TypeSize->uint64_t conversions.
Differential Revision: https://reviews.llvm.org/D83572
Preserve SCC dead flags in SIOptimizeExecMaskingPreRA.
This helps with removing redundant s_andn2 instructions later.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D83637
https://reviews.llvm.org/D78411 introduced test changes which relied on
the ability to strip debugify metadata even if module-level metadata is
missing. This introduces a more targeted test for that ability.
We have this generic transform in IR (instcombine),
but as shown in PR41098:
http://bugs.llvm.org/PR41098
...the pattern may emerge in codegen too.
x86 has a potential refinement/reversal opportunity here,
but that should come later or needs a target hook to
avoid the transform. Converting to bswap is the more
specific form, so we should use it if it is available.
This carves out an exception for a pair of consecutive loads that are
reversed from the consecutive order of a pair of stores. All of the
existing profitability/legality checks for the memops remain between
the 2 altered hunks of code.
This should give us the same x86 base-case asm that gcc gets in
PR41098 and PR44895:
http://bugs.llvm.org/PR41098http://bugs.llvm.org/PR44895
I think we are missing a potential subsequent conversion to use "movbe"
if the target supports that. That might be similar to what AArch64
would use to get "rev16".
Differential Revision: https://reviews.llvm.org/D83567
This carves out an exception for a pair of consecutive loads that are
reversed from the consecutive order of a pair of stores. All of the
existing profitability/legality checks for the memops remain between
the 2 altered hunks of code.
This should give us the same x86 base-case asm that gcc gets in
PR41098 and PR44895:i
http://bugs.llvm.org/PR41098http://bugs.llvm.org/PR44895
I think we are missing a potential subsequent conversion to use "movbe"
if the target supports that. That might be similar to what AArch64
would use to get "rev16".
Differential Revision:
This refactors option -disable-mve-tail-predication to take different arguments
so that we have 1 option to control tail-predication rather than several
different ones.
This is also a prep step for D82953, in which we want to reject reductions
unless that is requested with this option.
Differential Revision: https://reviews.llvm.org/D83133
Check that input size matches size of destination reg class.
Attempt to extend input size when needed.
Differential Revision: https://reviews.llvm.org/D83384
This patch adds support for constrained int/fp conversion between
signed/unsigned i32 and f32/f64.
Reviewed By: jhibbits
Differential Revision: https://reviews.llvm.org/D82747
Bit 7 of the index controls zeroing, the other bits are ignored when bit 7 is set. Shuffle lowering was using 128 and shuffle combining was using 255. Seems like we should be consistent.
This patch changes shuffle combining to use 128 to match lowering.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D83587
Currently, llvm when see a global variable in .maps section,
it ensures its type must be a struct type. Then pointee
will be further evaluated for the structure members.
In normal cases, the pointee type will be skipped.
Although this is what current all bpf programs are doing,
but it is a little bit restrictive. For example, it is legitimate
for users to have:
typedef struct { int key_size; int value_size; } __map_t;
__map_t map __attribute__((section(".maps")));
This patch lifts this restriction and typedef of
a struct type is also allowed for .maps section variables.
To avoid create unnecessary fixup entries when traversal
started with typedef/struct type, the new implementation
first traverse all map struct members and then traverse
the typedef/struct type. This way, in internal BTFDebug
implementation, no fixup entries are generated.
Two new unit tests are added for typedef and const
struct in .maps section. Also tested with kernel bpf selftests.
Differential Revision: https://reviews.llvm.org/D83638
fadd (fma A, B, (fmul C, D)), E --> fma A, B, (fma C, D, E)
This is only allowed when "reassoc" is present on the fadd.
As discussed in D80801, this transform goes beyond
what is allowed by "contract" FMF (-ffp-contract=fast).
That is because we are fusing the trailing add of 'E' with a
multiply, but without "reassoc", the code mandates that the
products A*B and C*D are added together before adding in 'E'.
I've added this example to the LangRef to try to clarify the
meaning of "contract". If that seems reasonable, we should
probably do something similar for the clang docs because
there does not appear to be any formal spec for the behavior
of -ffp-contract=fast.
Differential Revision: https://reviews.llvm.org/D82499
These test cases fail to use vpternlog because the AND was converted
to a blend shuffle and then converted back to AND during shuffle lowering.
This results in the AND having a different type than it started with.
This prevents our custom matching logic from seeing the two logic ops.
It is possible that LowerSwitch pass leaves certain blocks
unreachable from the entry. If not removed, these dead blocks
can cause undefined behavior in the subsequent passes.
It caused a crash in the AMDGPU backend after the instruction
selection when a PHI node has its incoming values coming from
these unreachable blocks.
In the AMDGPU pass flow, the last invocation of UnreachableBlockElim
precedes where LowerSwitch is currently placed and eventually
missed out on the opportunity to get these blocks eliminated.
This patch ensures that LowerSwitch pass get inserted earlier
to make use of the existing unreachable block elimination pass.
Reviewed By: sameerds, arsenm
Differential Revision: https://reviews.llvm.org/D83584
P9 is the only one with InstrSchedModel, but we may have more in the
future, we should not hardcoded it to P9, check hasInstrSchedModel
instead.
Reviewed By: hfinkel
Differential Revision: https://reviews.llvm.org/D83590
In BUILD_VECTOR lowering, we used to generally prefer using splats
over v128.const instructions because v128.const has a very large
encoding. However, in d5b7a4e2e8 we switched to preferring consts
because they are expected to be more efficient in engines. This patch
updates the ISel patterns to match this current preference.
Differential Revision: https://reviews.llvm.org/D83581
This doesn't appear used for anything, and is emitted incorrectly
based on the description. This also depends on the IR type, and
pointee element type.
This patch upstreams support for the Arm-v8 Cortex-A78 and Cortex-X1
processors for AArch64 and ARM.
In detail:
- Adding cortex-a78 and cortex-x1 as cpu options for aarch64 and arm targets in clang
- Adding Cortex-A78 and Cortex-X1 CPU names and ProcessorModels in llvm
details of the CPU can be found here:
https://www.arm.com/products/cortex-xhttps://www.arm.com/products/silicon-ip-cpu/cortex-a/cortex-a78
The following people contributed to this patch:
- Luke Geeson
- Mikhail Maltsev
Reviewers: t.p.northover, dmgreen
Reviewed By: dmgreen
Subscribers: dmgreen, kristof.beyls, hiraditya, danielkiss, cfe-commits,
llvm-commits, miyuki
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D83206
Truncations lowered as shuffles of multiple (concatenated) vectors often leave us with lane-crossing shuffles that feed a PACKSS/PACKUS, if both shuffles are fed from the same 2 vector sources, then we can PACK the sources directly and shuffle the result instead.
This is currently limited to whole i128 lanes in a 256-bit vector, but we can extend this if the need arises (but I'm not seeing many examples in real world code).
This patch builds on 0d7286a652 by simplifying the code for detecting
splat values and adding new tests demonstrating the lowering of
splatted absolute value shift amounts, which are common in code
generated by Halide. The lowering is very bad right now, but
subsequent patches will improve it considerably. The tests will be
useful for evaluating the improvements in those patches.
Reviewed By: aheejin
Differential Revision: https://reviews.llvm.org/D83493
As requested, these tests check for specific debug locs on the output of the
legalizer. The only one that I couldn't write was for moreElementsVector, which
AFAICT we don't trigger on AArch64.
This is currently bare-bones; we aren't taking advantage of any of the
FMA variant instructions. But it's enough to at least generate
code.
Differential Revision: https://reviews.llvm.org/D83444
The following: https://reviews.llvm.org/D82552
fixed an assert in the SelectionDag ISel legalizer for some CCs on armv7.
I noticed that this fix also fixes the assert when using fastcc, so I am
adding a fastcc regression test here.
Differential Revision: https://reviews.llvm.org/D82443
If we don't immediately lower the vector shift, the splat
constant vector we created may get turned into a constant pool
load before we get around to lowering the shift. This makes it
a lot more difficult to create a shift by constant. Sometimes we
fail to see through the constant pool at all and end up trying
to lower as if it was a variable shift. This requires custom
handling and may create an unsupported vselect on pre-sse-4.1
targets. Since we're after LegalizeVectorOps we are unable to
legalize the unsupported vselect as that code is in LegalizeVectorOps
rather than LegalizeDAG.
So calling LowerShift immediately ensures that we get see the
splat constant.
Fixes PR46527.
Differential Revision: https://reviews.llvm.org/D83455
The unfortunate split between finalizeLowering and the selector pass
means there's a point where the verifier fails. The DAG selector pass
skips the verifier, but this seems to not work when using the
GlobalISel fallback.
If we're extracting a subvector from a shuffle that is shuffling entire subvectors we can peek through and extract the subvector from the shuffle source instead.
This helps remove some cases where concat_vectors(extract_subvector(),extract_subvector()) legalizations has resulted in BLEND/VPERM2F128 shuffles of the subvectors.
Since the `RISCVExpandPseudo` pass has been split from
`RISCVExpandAtomicPseudo` pass, it would be nice to run the former as
early as possible (The latter has to be run as late as possible to
ensure correctness). Running earlier means we can reschedule these pairs
as we see fit.
Running earlier in the machine pass pipeline is good, but would mean
teaching many more passes about `hasLabelMustBeEmitted`. Splitting the
basic blocks also pessimises possible optimisations because some
optimisations are MBB-local, and others are disabled if the block has
its address taken (which is notionally what `hasLabelMustBeEmitted`
means).
This patch uses a new approach of setting the pre-instruction symbol on
the AUIPC instruction to a temporary symbol and referencing that. This
avoids splitting the basic block, but allows us to reference exactly the
instruction that we need to. Notionally, this approach seems more
correct because we do actually want to address a specific instruction.
This then allows the pass to be moved much earlier in the pass pipeline,
before both scheduling and register allocation. However, to do so we
must leave the MIR in SSA form (by not redefining registers), and so use
a virtual register for the intermediate value. By using this virtual
register, this pass now has to come before register allocation.
Reviewed By: luismarques, asb
Differential Revision: https://reviews.llvm.org/D82988
Fixed length vector code generation for SVE does not yet custom
lower BUILD_VECTOR and instead relies on expansion. At the same
time custom lowering for VECTOR_SHUFFLE is also not available so
this patch updates isShuffleMaskLegal to reject vector types that
require SVE.
Related to this it also prevents the merging of stores after
legalisation because this only works when BUILD_VECTOR is either
legal or can be elminated. When this is not the case the code
generator enters an infinite legalisation loop.
Differential Revision: https://reviews.llvm.org/D83408
Summary:
When legalizing a biscast operation from an fp16 operand to an i16 on a
target that requires both input and output types to be promoted to
32-bits, an assertion can fail when building the new node due to a
mismatch between the the operation's result size and the type specified to
the node.
This patches fix the issue by making sure the bit width of the types
match for the FP_TO_FP16 node, covering the difference with an extra
ANYEXTEND operation.
Reviewers: ostannard, efriedma, pirama, jmolloy, plotfi
Reviewed By: efriedma
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82552
On PPC64, for a variadic function, if va_start is not called, it won't
access any variadic argument on stack, thus we can save stores of
registers used to pass arguments.
Differential Revision: https://reviews.llvm.org/D82361
Updated the AArch64 tests the best I could with my vague, inferred
understanding of AArch64 register banks. As far as I can tell, there
is only one 32-bit/64-bit type which will use the gpr register bank,
so we have to use the fpr bank for the other operand.
GFX10 image instructions use one or more address operands starting at
vaddr0, instead of a single vaddr operand, to allow for NSA forms.
Differential Revision: https://reviews.llvm.org/D81675
Fix the division/remainder algorithm by adding a second quotient
refinement step, which is required in some cases like
0xFFFFFFFFu / 0x11111111u (https://bugs.llvm.org/show_bug.cgi?id=46212).
Also document, rewrite and simplify it by ensuring that we always have a
lower bound on inv(y), which simplifies the UNR step and the quotient
refinement steps.
Differential Revision: https://reviews.llvm.org/D83381
ExpandVectorBuildThroughStack is also used for CONCAT_VECTORS.
However, when calculating the offsets for each of the operands we
incorrectly use the element size rather than actual size and thus
the stores overlap.
Differential Revision: https://reviews.llvm.org/D83303
Summary:
The following combine currently breaks in the DAGCombiner:
```
extract_vector_elt (concat_vectors v4i16:a, v4i16:b), x
-> extract_vector_elt a, x
```
This happens because after we have combined these nodes we have inserted nodes
that use individual instances of the vector element type. In the above example
i16. However this isn't a legal type on all backends, and when the combining pass calls
the legalizer it breaks as it expects types to already be legal. The type legalizer has
already been run, and running it again would make a mess of the nodes.
In the example code at least, the generated code is still efficient after the change.
Reviewers: miyuki, arsenm, dmgreen, lebedev.ri
Reviewed By: miyuki, lebedev.ri
Subscribers: lebedev.ri, wdng, hiraditya, steven.zhang, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83231
vselect ((X & Pow2C) == 0), LHS, RHS --> vselect ((shl X, C') < 0), RHS, LHS
Follow-up to D83073 - the non-splat mask cases where we actually see an
improvement are quite limited from what I can tell. AVX1 needs multiply
and blend capabilities and AVX2 needs vector shift and blend capabilities.
The intersection of those 2 constraints is only vectors with 32-bit or
64-bit elements.
XOP is/was better.
Differential Revision: https://reviews.llvm.org/D83181
On SKX targets we end up loading a v16i8 PSHUFB mask from a v32i8 constant and scaling incorrectly indexes the demanded elts mask - we're missing a check that the constant pool is the same size as the loaded mask.
Test case from D81791 post-commit review.
We use extact_subvector and insert_subvector to "cast" between
fixed length and scalable vectors. This patch adds custom c++
based ISel for the following cases:
fixed_vector = ISD::EXTRACT_SUBVECTOR scalable_vector, 0
scalable_vector = ISD::INSERT_SUBVECTOR undef(scalable_vector), fixed_vector, 0
Which result in either EXTRACT_SUBREG/INSERT_SUBREG for NEON sized
vectors or COPY_TO_REGCLASS otherwise.
Differential Revision: https://reviews.llvm.org/D82871
For the GetElementPtr case in function
AddressingModeMatcher::matchOperationAddr
I've changed the code to use the TypeSize class instead of relying
upon the implicit conversion to a uint64_t. As part of this we now
check for scalable types and if we encounter one just bail out for
now as the subsequent optimisations doesn't currently support them.
This changes fixes up all warnings in the following tests:
llvm/test/CodeGen/AArch64/sve-ld1-addressing-mode-reg-imm.ll
llvm/test/CodeGen/AArch64/sve-st1-addressing-mode-reg-imm.ll
Differential Revision: https://reviews.llvm.org/D83124
`__stack_chk_fail` does not return, but `unreachable` was not generated
following `call __stack_chk_fail`. This had a possibility to generate an
invalid binary for functions with a return type, because
`__stack_chk_fail`'s return type is void and `call __stack_chk_fail` can
be the last instruction in the function whose return type is non-void.
Generating `unreachable` after it makes sure CFGStackify's
`fixEndsAtEndOfFunction` handles it correctly.
Reviewed By: tlively
Differential Revision: https://reviews.llvm.org/D83277
For an addition with an immediate in specific ranges, a pair of
addi-addi can be generated instead of the ordinary lui-addi-add serial.
Reviewed By: MaskRay, luismarques
Differential Revision: https://reviews.llvm.org/D82262
... to shift/add or shift/sub.
Do not enable it on riscv32 with the M extension where decomposeMulByConstant
may not be an optimization.
Reviewed By: luismarques, MaskRay
Differential Revision: https://reviews.llvm.org/D82660
handleAssignments was assuming every argument type is an MVT, and
assignArg would always fail. This fixes one of the hacks in the
current AMDGPU calling convention code that pre-processes the
arguments.
The tests in a5b9ad7e9a actually failed
the verifier, which for some reason is not the default. Also add tests
for 0-sized function arguments, which do not add entries to the
expected register lists.
This is inspired by D81648. The basic idea is to have the set of SDValues which are lowered as either constants or direct frame references explicit in one place, and to separate them clearly from the spilling logic.
This is not NFC in that the handling of constants larger than > 64 bit has changed. The old lowering would crash on values which could not be encoded as a sign extended 64 bit value. The new lowering just spills all constants > 64 bits. We could be consistent about doing the sext(Con64) optimization, but I happen to know that this code path is utterly unexercised in practice, so simple is better for now.
This patch creates a clang flag to enable SESES. This flag also ensures that
lvi-cfi is on when using seses via clang.
SESES should use lvi-cfi to mitigate returns and indirect branches.
The flag to enable the SESES functionality only without lvi-cfi is now
-x86-seses-enable-without-lvi-cfi to warn users part of the mitigation is not
enabled if they use this flag. This is useful in case folks want to see the
cost of SESES separate from the LVI-CFI.
Reviewed By: sconstab
Differential Revision: https://reviews.llvm.org/D79910
We were checking the VBROADCAST_LOAD element size against the extraction destination size instead of the extracted vector element size - PEXTRW/PEXTB have implicit zext'ing so have i32 destination sizes for v8i16/v16i8 vectors, resulting in us extracting from the wrong part of a load.
This patch bails from the fold if the vector element sizes don't match, and we now use the target constant extraction code later on like the pre-AVX2 targets, fixing the test case.
Found by internal fuzzing tests.
Use SESES as the fallback at O0 where the optimized LVI pass isn't desired due
to its effect on build times at O0.
I updated the LVI tests since this changes the code gen for the tests touched in the parent revision.
This is a follow up to the comments I made here: https://reviews.llvm.org/D80964
Hopefully we can continue the discussion here.
Also updated SESES to handle LFENCE instructions properly instead of adding
redundant LFENCEs. In particular, 1) no longer add LFENCE if the current
instruction being processed is an LFENCE and 2) no longer add LFENCE if the
instruction right before the instruction being processed is an LFENCE
Reviewed By: sconstab
Differential Revision: https://reviews.llvm.org/D82037
Since WebAssembly's vector shift instructions take a scalar shift
amount rather than a vector shift amount, we have to check in ISel
that the vector shift amount is a splat. Previously, we were checking
explicitly for splat BUILD_VECTOR nodes, but this change uses the
standard utilities for detecting splat values that can handle more
complex splat patterns. Since the C++ ISel lowering is now more
general than the ISel patterns, this change also simplifies shift
lowering by using the C++ lowering for all SIMD shifts rather than
mixing C++ and normal pattern-based lowering.
This change improves ISel for shifts to the point that the
simd-shift-unroll.ll regression test no longer tests the code path it
was originally meant to test. The bug corresponding to that regression
test is no longer reproducible with its original reported reproducer,
so rather than try to fix the regression test, this change just
removes it.
Differential Revision: https://reviews.llvm.org/D83278
On AVX2 we tend to lower BUILD_VECTOR of constants as broadcasts if we can, in this case a <2 x i16> non-uniform constant has been lowered as a <4 x i32> broadcast.
The test case shows that the extraction folding code has incorrectly extracted the wrong part (lower WORD) of the resulting i32 memory source.
Found by internal fuzzing tests.
Provide the LLVM intrinsics needed to implement vector replace element
builtins in altivec.h which will be added in a subsequent patch.
Differential Revision: https://reviews.llvm.org/D83308
In the test based on PR46586:
https://bugs.llvm.org/show_bug.cgi?id=46586
...we are inserting 16-bits into the high element of the vector, shuffling it
to element 0, and extracting 32-bits. But xmm1 was never initialized, so the
top 16-bits of the extract are undef without this patch.
(It seems like we could do better than this by recognizing that we only demand
a subsection of the build vector, but I want to make sure we fix the
miscompile 1st.)
This path is only used for pre-SSE4.1, and simpler patterns get squashed
somewhere along the way, so the test still includes a 'urem' as it did in the
original test from the bug report.
Differential Revision: https://reviews.llvm.org/D83319
When an argument has 'byval' attribute and should be
passed on the stack according calling convention,
a stack copy would be emitted twice. This will cause
the real value will be put into stack where the pointer
should be passed.
Differential Revision: https://reviews.llvm.org/D83175
Summary:
When splitting a store of a scalable type, the new address is
calculated in SplitVecOp_STORE using a vscale and an add instruction.
Reviewers: sdesmalen, efriedma, david-arm
Reviewed By: david-arm
Subscribers: tschuett, hiraditya, psnobl, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83041
Summary:
When splitting a load of a scalable type, the new address is
calculated in SplitVecRes_LOAD using a vscale and an add instruction.
This patch also adds a DAG combiner fold to visitADD for vscale:
- Fold (add (vscale(C0)), (vscale(C1))) to (add (vscale(C0 + C1)))
Reviewers: sdesmalen, efriedma, david-arm
Reviewed By: david-arm
Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82792
There are now more SVE tests in LLVM and Clang that do not
emit warnings related to invalid use of EVT::getVectorNumElements()
and VectorType::getNumElements(). For these tests I have added
additional checks that there are no warnings in order to prevent
any future regressions.
Differential Revision: https://reviews.llvm.org/D82943
In an earlier commit 584d0d5c17 I
added functionality to allow AArch64 CodeGen support for falling
back to DAG ISel when Global ISel encounters scalable vector
types. However, it seems that we were not falling back early
enough as llvm::getLLTForType was still being invoked for scalable
vector types.
I've added a new fallback function to the call lowering class in
order to catch this problem early enough, rather than wait for
lowerFormalArguments to reject scalable vector types.
Differential Revision: https://reviews.llvm.org/D82524
This patch fixes all remaining warnings in:
llvm/test/CodeGen/AArch64/sve-trunc.ll
llvm/test/CodeGen/AArch64/sve-vector-splat.ll
I hit some warnings related to getCopyPartsToVector. I fixed two
issues:
1. In widenVectorToPartType() we assumed that we'd always be
using BUILD_VECTOR nodes to expand from one vector type to another,
which is incorrect for scalable vector types. I've fixed this for now
by simply bailing out immediately for scalable vectors.
2. In getCopyToPartsVector() I've changed the code to compare
the element counts of different types.
Differential Revision: https://reviews.llvm.org/D83028
When legalizing shuffles, we make an attempt to combine it into
a PPC specific canonical form that avoids a need for a swap. If the
combine is successful, we RAUW the node and the custom legalization
replaces the now dead node instead of the one it should replace.
Remove that erroneous call to RAUW.
This patch aims to exploit the xxsplti32dx XT, IX, IMM32 instruction when lowering VECTOR_SHUFFLEs.
We implement lowerToXXSPLTI32DX when lowering vector shuffles to check if:
- Element size is 4 bytes
- The RHS is a constant vector (and constant splat of 4-bytes)
- The shuffle mask is a suitable mask for the XXSPLTI32DX instruction where it is one of the 32 masks:
<0, 4-7, 2, 4-7>
<4-7, 1, 4-7, 3>
Differential Revision: https://reviews.llvm.org/D83245
X / (fabs(A) * sqrt(Z)) --> X / sqrt(A*A*Z) --> X * rsqrt(A*A*Z)
In the motivating case from PR46406:
https://bugs.llvm.org/show_bug.cgi?id=46406
...this is restoring the sequence that was originally in the source code.
We extracted a term from within the sqrt because we do not know in
instcombine whether a target will expand a sqrt call.
Note: we could say that the transform in IR should be restricted, but
that would not solve the problem if the source was originally in the
pattern shown here.
This is a gray area for fast-math-flag requirements. I think we should at
least check fast-math-flags on the fdiv and fmul because I view this
transform as 2 pieces: reassociate the fmul operands and form reciprocal
from the fdiv (as with the existing transform). We could argue that the
sqrt also needs FMF, but that was not required before, so we should change
that in a follow-up patch if that seems better.
We don't currently have a way to check that the target will produce a sqrt
or recip estimate without actually creating nodes (the APIs are SDValue
getSqrtEstimate() and SDValue getRecipEstimate()), so we clean up
speculatively created nodes if we are not able to create an estimate.
The x86 test with doubles verifies that we are not changing a test with
no estimate sequence.
Differential Revision: https://reviews.llvm.org/D82716
This covers both the existing memory functions as well as the new bulk memory proposal.
Added new test files since changes where also required in the inputs.
Also removes unused init/drop intrinsics rather than trying to make them work for 64-bit.
Differential Revision: https://reviews.llvm.org/D82821
This was resulting in a missing vreg def in the use select
instruction.
The output of the pseudo doesn't make sense, since it really shouldn't
have the vreg output in the first place, and instead an implicit scc
def to match the real scalar behavior.
We could have easier to understand tests if we selected scalar
versions of the [us]{add|sub}.with.overflow intrinsics.
This does still end up producing vector code in the end, since it gets
moved later.
We can often fold an ADDI into the offset of load/store instructions:
(load (addi base, off1), off2) -> (load base, off1+off2)
(store val, (addi base, off1), off2) -> (store val, base, off1+off2)
This is possible when the off1+off2 continues to fit the 12-bit immediate.
We remove the previous restriction where we would never fold the ADDIs if
the load/stores had nonzero offsets. We now do the fold the the resulting
constant still fits a 12-bit immediate, or if off1 is a variable's address
and we know based on that variable's alignment that off1+offs2 won't overflow.
Differential Revision: https://reviews.llvm.org/D79690
Summary:
When a desired symbol name contains invalid character that the
system assembler could not process, we need to emit .rename
directive in assembly path in order for that desired symbol name
to appear in the symbol table.
Reviewed By: hubert.reinterpretcast, DiggerLin, daltenty, Xiangling_L
Differential Revision: https://reviews.llvm.org/D82481
The default constructor wasn't setting isSet o the ArgDescriptor, so
while these had the value set, they were treated as missing. This only
ended up mattering in the indirect call case (and for regular calls in
GlobalISel, which current doesn't have a way to support the variable
ABI).
Summary: As Bugzilla-35090 reported, the rationale for using custom lowering SREM/UREM should no longer be true. At the IR level, the div-rem-pairs pass performs the transformation where the remainder is computed from the result of the division when both a required. We should now be able to lower these directly on P9. And the pass also fixed the problem that divide is in a different block than the remainder. This is a patch to remove redundant code and make SREM/UREM legal directly on P9.
Reviewed By: lkail
Differential Revision: https://reviews.llvm.org/D82145
Whether an instruction is deemed to have side effects in determined by
whether it has a tblgen pattern that emits a single instruction.
Because of the way a lot of the the vcvt instructions are specified
either in dagtodag code or with patterns that emit multiple
instructions, they don't get marked as not having side effects.
This just marks them as not having side effects manually. It can help
especially with instruction scheduling, to not create artificial
barriers, but one of these tests also managed to produce fewer
instructions.
Differential Revision: https://reviews.llvm.org/D81639
zext_vector_inreg needs to produces 0s in the extended bits and
sext_vector_inreg needs to produce upper bits that are all the
same. So we should fold them to a 0 vector instead of undef.
Fixes PR46585.
Probably not super important since there are no real CPUs with
avx512vl and not avx512bw. But vpternlog should be better than
vblendvb.
I do wonder if we should use vpternlog even with BWI. We
currently use vblendmb or vpblendmw by putting the mask into a GPR
and moving it to a k-register. But I don't think we hoist the
GPR to k-register copy in machine LICM. Using VPTERNLOG would use
a constant pool load, but has the advantage that we're pretty good
at hoisting and rematerializing those.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D83156
VPBLENDVB is multiple uops while VPTERNLOG is a single uop. So
we should use that instead.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D83155
Currently matchBinOpReduction only handles shufflevector reduction patterns, but in many cases these only occur in the final stages of a reduction, once we're down to legal vector widths.
Before this its likely that we are performing reductions using subvector extractions to repeatedly split the source vector in half and perform the binop on the halves.
Assuming we've found a non-partial reduction, this patch continues looking for subvector reductions as far as it can beyond the last shufflevector.
Fixes PR37890
Tests based on the PR37890 test cases - the vector combine pass should leave us with a reduction chain ending in extract(add(x,shuffle(x,1,-1,...))), but the higher reduction stages will be subvector extractions not shuffles.
Using PACK for truncations leaves us with intermediate shuffles that can be tricky to remove while the truncation tree is being formed.
This fold helps pull out the PERMQ case which is one of the most common, avoiding some costly lane-crossing shuffles.
A future patch will begin adding more general shuffle folding, which we should be able to use for HADD/HSUB as well.
AArch64ExpandPseudo::expand_DestructiveOp contains an assert to
ensure the destructive operand's register is unique. However,
this is only required when psuedo expansion emits a movprfx.
A simple example when a movprfx is not required is
Z0 = FADD_ZPZZ_UNDEF_S P0, Z0, Z0
which expands to an unprefixed FADD_ZPmZ_S instruction.
This patch moves the assert to the places where a movprfx is emitted.
Differential Revision: https://reviews.llvm.org/D83029
This patch is part of supporting `-fstack-clash-protection`. Implemented
probing when emitting prologue.
Differential Revision: https://reviews.llvm.org/D81460
Summary:
Since the br_table instruction takes an i32, switches over i64s (and
larger integers) must use the i32.wrap_i64 instruction to truncate the
table index. This truncation makes numbers just over 2^32
indistinguishable from small numbers, so it was a miscompilation to
omit the range check preceding these br_tables. This change fixes the
problem by skipping the "fixing" of the br_table when the range check
is an i64 instruction.
Fixes PR46447.
Reviewers: aheejin, dschuff, kripken
Reviewed By: kripken
Subscribers: sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83017
We canonicalize patterns like:
%s = lshr i32 %a0, 1
%t = trunc i32 %s to i1
to:
%a = and i32 %a0, 2
%c = icmp ne i32 %a, 0
...in IR, but the bit-shifting original sequence may be better for x86 vector codegen.
I tried several variants of the transform, and it's tricky to not induce regressions.
In particular, I did not find a way to cleanly handle non-splat constants, so I've left
that as a TODO item here (currently negative tests for those are included). AVX512
resulted in some diffs, but didn't look meaningful, so I left that out too. Some of
the 256-bit AVX1 diffs are questionable, but close enough that they are probably
insignificant.
Differential Revision: https://reviews.llvm.org/D83073.
Summary:
D80831 changed part of the prefix usage for AIX.
But there are other places getting prefix from DataLayout.
This patch intends to make prefix usage consistent on AIX.
Reviewed by: hubert.reinterpretcast, daltenty
Differential Revision: https://reviews.llvm.org/D81270
Given a loop with two subloops, it should be possible for both to be
converted to hardware loops. That's what this patch does, simply enough.
It slightly alters the loop iterating order to try and convert all
subloops. If one (or more) succeeds, it stops as before.
Differential Revision: https://reviews.llvm.org/D78502
Summary:
Teach LLVM to recognize the above pattern, which is usually a
transformation of (a + b + 1) >> 1, where the operands are either
signed or unsigned types.
Subscribers: kristof.beyls, hiraditya, danielkiss, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82669
This patch upstreams support for the Arm-v8 Cortex-A77
processor for AArch64 and ARM.
In detail:
- Adding cortex-a77 as a cpu option for aarch64 and arm targets in clang
- Cortex-A77 CPU name and ProcessorModel in llvm
details of the CPU can be found here:
https://www.arm.com/products/silicon-ip-cpu/cortex-a/cortex-a77
and a similar submission to GCC can be found here:
e0664b7a63
The following people contributed to this patch:
- Luke Geeson
- Mikhail Maltsev
Reviewers: t.p.northover, dmgreen, ostannard, SjoerdMeijer
Reviewed By: dmgreen
Subscribers: dmgreen, kristof.beyls, hiraditya, danielkiss, cfe-commits,
llvm-commits, miyuki
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D82887
This patch is part of supporting `-fstack-clash-protection`. Mainly do
such things compared to existing `lowerDynamicAlloc`
- Added a new pseudo instruction PPC::PREPARE_PROBED_ALLOC to get
actual frame pointer and final stack pointer.
- Synthesize a loop to probe by blocks.
- Use DYNAREAOFFSET to get MaxCallFrameSize which is calculated in
prologepilog.
Differential Revision: https://reviews.llvm.org/D81358
I think this mostly looks ok. The only weird thing I noticed was
a couple rotate vXi8 tests picked up an extra logic op where we have
(and (or (and), (andn)), X). Previously we matched the (or (and), (andn))
to vpternlog, but now we match the (and (or), X) and leave the and/andn
unmatched.
Exit early if the exec mask is zero at the end of control flow.
Mark the ends of control flow during control flow lowering and
convert these to exits during the insert skips pass.
Reviewed By: nhaehnle
Differential Revision: https://reviews.llvm.org/D82737
Exit early if the exec mask is zero at the end of control flow.
Mark the ends of control flow during control flow lowering and
convert these to exits during the insert skips pass.
Reviewed By: nhaehnle
Differential Revision: https://reviews.llvm.org/D82737
Generate a single early exit block out-of-line and branch to this
if all lanes are killed. This avoids branching if lanes are active.
Reviewed By: nhaehnle
Differential Revision: https://reviews.llvm.org/D82641
We consider v32i16/v64i8 to be legal types on avx512f, but we
don't have most operations until avx512bw. But we can use
and/or/xor operations. So try those before splitting.
This is especially helpful since we turn some ands with constant
masks into shuffles in early DAG combines. So we should make sure
we recover those back to AND.
SelectionDAGBuilder converts logic-of-compares into multiple branches based
on a boolean TLI setting in isJumpExpensive(). But that probably never
considered the pattern of extracted bools from a vector compare - it seems
unlikely that we would want to turn vector logic into control-flow.
The motivating x86 reduction case is shown in PR44565:
https://bugs.llvm.org/show_bug.cgi?id=44565
...and that test shows the expected improvement from using pmovmsk codegen.
For AArch64, I modified the test to include an extra op because the simpler
test gets transformed by a codegen invocation of SimplifyCFG.
Differential Revision: https://reviews.llvm.org/D82602
The comments here indicate that we prefer to promote the shifts
instead of allowing rotate to be pattern matched. But we weren't
taking into account whether 512-bit registers are enabled or
whethever we have vpsllvw/vpsrlvw instructions.
splatvar_rotate_v32i8 is a slight regrssion, but the other cases
are neutral or improved.
We currently don't mark ROTL as custom when avx512bw is enabled
under the assumption we'll be able to promote the shifts in the
rotate idiom. But if we don't have 512-bit registers enabled we
can't promote.
As of 1fed131660, we have code that
changes shuffle masks so that we can put the shuffle in a canonical
form that can be matched to a single instruction. However, it
does not properly account for undef elements in the BUILD_VECTOR
that is the RHS splat so we can end up with undefs where they
shouldn't be. This patch converts the splat input with undefs to
one without.
This patch puts the _ZERO pseudos and corresponding patterns
under the predicate 'UseExperimentalZeroingPseudos', so that they
can be enabled/disabled through compile flags.
This is done because the zeroing pseudos use MOVPRFX to do merging of
the inactive lanes, but it depends on the uarch whether this operation
is actually merged with the destructive operation. If not, it may be
more profitable to use a SELECT and to give the compiler the freedom to
schedule these instructions as normal, rather than keeping them bundled
together. Additionally, this feature is not yet fully implemented and
there are still known bugs (see D80410) that need to be resolved before
the 'experimental' can be dropped from the name.
Reviewers: paulwalker-arm, cameron.mcinally, efriedma
Reviewed By: paulwalker-arm
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82780
I have added CHECK lines to the following tests:
llvm/test/CodeGen/AArch64/sve-breakdown-scalable-vectortype.ll
llvm/test/CodeGen/AArch64/sve-calling-convention-tuple-types.ll
llvm/test/CodeGen/AArch64/sve-intrinsics-create-tuple.ll
llvm/test/CodeGen/AArch64/sve-intrinsics-loads.ll
since they are now free of warnings related to invalid use of
EVT::getVectorNumElements() and VectorType::getNumElements().
Differential Revision: https://reviews.llvm.org/D82957
There was a rogue 'assert' in AArch64ISelLowering for the tuple.get intrinsics,
that shouldn't really have been there (I suspect this was a remnant from when
we expected the wider vector always to have come from a vector CONCAT).
When I tried to create a more minimal reproducer, I found a bug in
DAGCombiner where it drops the scalable flag when trying to fold:
extract_subv (bitcast X), Index --> bitcast (extract_subv X, Index')
This patch fixes both issues.
Reviewers: david-arm, efriedma, spatel
Reviewed By: efriedma
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82910
Move the Thumb2SizeReduce pass to before IfConversion when optimising
for minimal code size.
Running the Thumb2SizeReduction pass before IfConversionallows T1
instructions to propagate to the final output, rather than the
ifConverter modifying T2 instructions and preventing them from being
reduced later.
This change does introduce a regression regarding execution time, so
it's only applied when optimising for size.
Running the LLVM Test Suite with this change produces a geomean
difference of -0.1% for the size..text metric.
Differential Revision: https://reviews.llvm.org/D82439
Summary:
If amdgpu-flat-work-group-size is not specified in LLVM IR, the backend
uses default value of 1024. For this, minimum waves per EU should be 4.
However, backend is still setting minimum value to 1 instead of calculated
value. This is not observed normally as frontend always provide
amdgpu-flat-work-group-size attribute.
Reviewers: rampitec, b-sumner, sameerds, msearles
Reviewed By: rampitec
Subscribers: qcolombet, arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81991
The caller can't handle the node having multiple results like a
masked load does. So we need to detect the case and do our own
result replacement.
Fixes PR46532.
If the original register operand had a subregister, it wasn't getting
cleared. This resulted in reinterpreted the subreg index as
unrecognized target flags, which produced unparseable MIR.
In visitSCALAR_TO_VECTOR we try to optimise cases such as:
scalar_to_vector (extract_vector_elt %x)
into vector shuffles of %x. However, it led to numerous warnings
when %x is a scalable vector type, so for now I've changed the
code to only perform the combination on fixed length vectors.
Although we probably could change the code to work with scalable
vectors in certain cases, without a proper profit analysis it
doesn't seem worth it at the moment.
This change fixes up one of the warnings in:
llvm/test/CodeGen/AArch64/sve-merging-stores.ll
I've also added a simplified version of the same test to:
llvm/test/CodeGen/AArch64/sve-fp.ll
which already has checks for no warnings.
Differential Revision: https://reviews.llvm.org/D82872
Currently, BTF generation stops at pointer struct members
if the pointee type is a struct. This is to avoid bloating
generated BTF size. The following is the process to
correctly record types for these pointee struct types.
- During type traversal stage, when a struct member, which
is a pointer to another struct, is encountered,
the pointee struct type, keyed with its name, is
remembered in a Fixup map.
- Later, when all type traversal is done, the Fixup map
is scanned, based on struct name matching, to either
resolve as pointing to a real already generated type
or as a forward declaration.
Andrii discovered a bug if the struct member pointee struct
is anonymous. In this case, a struct with empty name is
recorded in Fixup map, and later it happens another anonymous
struct with empty name is defined in BTF. So wrong type
resolution happens.
To fix the problem, if the struct member pointee struct
is anonymous, pointee struct type will be generated in
stead of being put in Fixup map.
Differential Revision: https://reviews.llvm.org/D82976
Before this instruction supported output values, it fit fairly
naturally as a terminator. However, being a terminator while also
supporting outputs causes some trouble, as the physreg->vreg COPY
operations cannot be in the same block.
Modeling it as a non-terminator allows it to be handled the same way
as invoke is handled already.
Most of the changes here were created by auditing all the existing
users of MachineBasicBlock::isEHPad() and
MachineBasicBlock::hasEHPadSuccessor(), and adding calls to
isInlineAsmBrIndirectTarget or mayHaveInlineAsmBr, as appropriate.
Reviewed By: nickdesaulniers, void
Differential Revision: https://reviews.llvm.org/D79794
This prevents the outlined functions from pulling in a lot of unnecessary code
in our downstream libraries/linker. Which stops outlining making codesize
worse in c++ code with no-exceptions.
Differential Revision: https://reviews.llvm.org/D57254
The pass to split atomic and non-atomic RISC-V pseudo-instructions was itself
split into two passes in D79635 / commit rG2cb0644f90b7, with the splitting of
non-atomic instructions being moved to the PreSched2 phase. A comment was
added to D79635 detailing a case where this caused problems, so this commit
moves the non-atomic split pass back to the PreEmitPass2 phase. This allows
the bulk of the changes from D79635 to remain committed, while addressing the
the reported problem (the pass split is now almost NFC). Once the root problem
is fixed we can move the (non-atomic) instruction splitting pass back to
earlier in the pipeline.
Summary: Support symbol with offset value as a VEMCExpr.
Reviewers: simoll, k-ishizaka
Reviewed By: simoll
Subscribers: hiraditya, llvm-commits
Tags: #llvm, #ve
Differential Revision: https://reviews.llvm.org/D82734
The situation where the caller uses a TOC and the callee does not
but is marked as clobbers the TOC (st_other=1) was not being compiled
correctly if both functions where in the same object file.
The call site where we had `callee` was missing a nop after the call.
This is because it was assumed that since the two functions where in
the same DSO they would be sharing a TOC. This is not the case if the
callee uses PC Relative because in that case it may clobber the TOC.
This patch makes sure that we add the cnop correctly so that the
linker has a place to restore the TOC.
Reviewers: sfertile, NeHuang, saghir
Differential Revision: https://reviews.llvm.org/D81126
D82257/rG3521ecf1f8a3 was incorrectly sign-extending a constant vector from the lsb, this is fine if all the constant elements are 'allsignbits' in the active bits, but if only some of the elements are, then we are corrupting the constant values for those elements.
This fix ensures we sign extend from the msb of the active/demanded bits instead.
Summary:
This implements two hooks that attempt to avoid control flow for RISC-V. RISC-V
will lower SELECTs into control flow, which is not a great idea.
The hook `hasMultipleConditionRegisters()` turns off the following
DAGCombiner folds:
select(C0|C1, x, y) <=> select(C0, x, select(C1, x, y))
select(C0&C1, x, y) <=> select(C0, select(C1, x, y), y)
The second hook `setJumpIsExpensive` controls a flag that has a similar purpose
and is used in CodeGenPrepare and the SelectionDAGBuilder.
Both of these have the effect of ensuring more logic is done before fewer jumps.
Note: with the `B` extension, we may be able to lower select into a conditional
move instruction, so at some point these hooks will need to be guarded based on
enabled extensions.
Reviewed By: luismarques
Differential Revision: https://reviews.llvm.org/D79268
Summary:
As shown, LLVM is keen to avoid logic and introduce selects (in DAGCombiner, and
other places). This leads to control flow on RISC-V which we should attempt to
avoid.
Reviewed By: luismarques
Differential Revision: https://reviews.llvm.org/D79267
Change imm with timm in pattern for SI_INIT_EXEC_LO and
remove regbank mappings for non register operands.
Differential Revision: https://reviews.llvm.org/D82885
We currently lower SDIV to SDIV_MERGE_OP1. This forces the value
for inactive lanes in a way that can hamper register allocation,
however, the lowering has no requirement for inactive lanes.
Instead this patch replaces SDIV_MERGE_OP1 with SDIV_PRED thus
freeing the register allocator. Once done the only user of
SDIV_MERGE_OP1 is intrinsic lowering so I've removed the node
and perform ISel on the intrinsic directly. This also allows
us to implement MOVPRFX based zeroing in the same manner as SUB.
This patch also renames UDIV_MERGE_OP1 and [F]ADD_MERGE_OP1 for
the same reason but in the ADD cases the ISel code is already
as required.
Differential Revision: https://reviews.llvm.org/D82783
In case of more than wavesize CSR SGPR spills, lanes of reserved VGPR were getting
overwritten due to wrap around.
Reserve a VGPR (when NumVGPRSpillLanes = 0, WaveSize, 2*WaveSize, ..) and when one
of the two conditions is true:
1. One reserved VGPR being tracked by VGPRReservedForSGPRSpill is not yet reserved.
2. All spill lanes of reserved VGPR(s) are full and another spill lane is required.
Reviewed By: arsenm, kerbowa
Differential Revision: https://reviews.llvm.org/D82463
Jay Foad