If analyzeBranch fails, on some targets, the out parameters point to
some blocks in the function. But we can't use that information, so make
sure to clear it out. (In some places in IfConversion, we assume that
any block with a TrueBB is analyzable.)
The change to the testcase makes it trigger a bug on builds without this
fix: IfConvertDiamond tries to perform a followup "merge" operation,
which isn't legal, and we somehow end up with a branch to a deleted MBB.
I'm not sure how this doesn't crash the compiler.
Differential Revision: https://reviews.llvm.org/D67306
llvm-svn: 371434
This enables GlobalISel to handle various intrinsics. The custom node
pattern will be ignored, and the intrinsic will work. This will also
allow SelectionDAG to directly select the intrinsics, but as they are
all custom lowered to the nodes, this ends up leaving dead code in the
table.
Eventually either GlobalISel should add the equivalent of custom nodes
equivalent, or intrinsics should be directly used. These each have
different tradeoffs.
There are a few more to handle, but these are easy to handle
ones. Some others fail for other reasons.
llvm-svn: 371432
Current for SAE instructions we only allow _MM_FROUND_CUR_DIRECTION(bit 2) or _MM_FROUND_NO_EXC(bit 3) to be used as the immediate passed to the inrinsics. But these instructions don't perform rounding so _MM_FROUND_CUR_DIRECTION is just sort of a default placeholder when you don't want to suppress exceptions. Using _MM_FROUND_NO_EXC by itself is really bit equivalent to (_MM_FROUND_NO_EXC | _MM_FROUND_TO_NEAREST_INT) since _MM_FROUND_TO_NEAREST_INT is 0. Since we aren't rounding on these instructions we should also accept (_MM_FROUND_CUR_DIRECTION | _MM_FROUND_NO_EXC) as equivalent to (_MM_FROUND_NO_EXC). icc allows this, but gcc does not.
Differential Revision: https://reviews.llvm.org/D67289
llvm-svn: 371430
Unfortunately MnemonicAlias defines a "Predicates" field just like an
instruction or pattern, with a somewhat different interpretation.
This ends up overriding the intended Predicates set by
PredicateControl on the pseudoinstruction defintions with an empty
list. This allowed incorrectly selecting instructions that should have
been rejected due to the SubtargetPredicate from patterns on the
instruction definition.
This does remove the divergent predicate from the 64-bit shift
patterns, which were already not used for the 32-bit shift, so I'm not
sure what the point was. This also removes a second, redundant copy of
the 64-bit divergent patterns.
llvm-svn: 371427
Just return once you emit the call, which is exactly what SelectionDAG does in
this situation.
Update call-translator-tail-call.ll.
Also update dllimport.ll to show that we tail call here in GISel again. Add
-verify-machineinstrs to the GISel line too, to defend against verifier
failures.
Differential revision: https://reviews.llvm.org/D67282
llvm-svn: 371425
Treat this as legal on gfx9 since it can use S_PACK_* instructions for
this.
This isn't used by anything yet. The same will probably apply to
16-bit G_BUILD_VECTOR without the trunc.
llvm-svn: 371423
These predicate vectors can usually be loaded and stored with a single
instruction, a VSTR_P0. However this instruction will store the entire P0
predicate, 16 bits, zeroextended to 32bits. Each lane of the the
v4i1/v8i1/v16i1 representing 4/2/1 bits.
As far as I understand, when llvm says "store this v4i1", it really does need
to store 4 bits (or 8, that being the size of a byte, with this bottom 4 as the
interesting bits). For example a bitcast from a v8i1 to a i8 is defined as a
store followed by a load, which is how the code is expanded.
So this instead lowers the v4i1/v8i1 load/store through some shuffles to get
the bits into the correct positions. This, as you might imagine, is not as
efficient as a single instruction. But I believe it is needed for correctness.
v16i1 equally should not load/store 32bits, only storing the 16bits of data.
Stack loads/stores are still using the VSTR_P0 (as can be seen by the test not
changing). This is fine as they are self-consistent, it is only "externally
observable loads/stores" (from our point of view) that need to be corrected.
Differential revision: https://reviews.llvm.org/D67085
llvm-svn: 371419
Reapply with fix to reduce resources required by the compiler - use
unsigned[2] instead of std::pair. This causes clang and gcc to compile
the generated file multiple times faster, and hopefully will reduce
the resource requirements on Visual Studio also. This fix is a little
ugly but it's clearly the same issue the previous author of
DFAPacketizer faced (the previous tables use unsigned[2] rather uglily
too).
This patch allows the DFAPacketizer to be queried after a packet is formed to work out which
resources were allocated to the packetized instructions.
This is particularly important for targets that do their own bundle packing - it's not
sufficient to know simply that instructions can share a packet; which slots are used is
also required for encoding.
This extends the emitter to emit a side-table containing resource usage diffs for each
state transition. The packetizer maintains a set of all possible resource states in its
current state. After packetization is complete, all remaining resource states are
possible packetization strategies.
The sidetable is only ~500K for Hexagon, but the extra tracking is disabled by default
(most uses of the packetizer like MachinePipeliner don't care and don't need the extra
maintained state).
Differential Revision: https://reviews.llvm.org/D66936
llvm-svn: 371399
This patch allows the DFAPacketizer to be queried after a packet is formed to work out which
resources were allocated to the packetized instructions.
This is particularly important for targets that do their own bundle packing - it's not
sufficient to know simply that instructions can share a packet; which slots are used is
also required for encoding.
This extends the emitter to emit a side-table containing resource usage diffs for each
state transition. The packetizer maintains a set of all possible resource states in its
current state. After packetization is complete, all remaining resource states are
possible packetization strategies.
The sidetable is only ~500K for Hexagon, but the extra tracking is disabled by default
(most uses of the packetizer like MachinePipeliner don't care and don't need the extra
maintained state).
Differential Revision: https://reviews.llvm.org/D66936
........
Reverted as this is causing "compiler out of heap space" errors on MSVC 2017/19 NDEBUG builds
llvm-svn: 371393
Summary:
This patch implements two arithmetic intrinsics:
* int_aarch64_sve_abs
* int_aarch64_sve_neg
testing the support for scalable vector types in intrinsics added in D65930.
Reviewed By: greened
Differential Revision: https://reviews.llvm.org/D65931
llvm-svn: 371388
Loosely based on DAGCombiner version, but this part is slightly simpler in
GlobalIsel because all address calculation is performed by G_GEP. That makes
the inc/dec distinction moot so there's just pre/post to think about.
No targets can handle it yet so testing is via a special flag that overrides
target hooks.
llvm-svn: 371384
The incoming accumulator value can be discovered through a sext, in
which case there will be a mismatch between the input and the result.
So sign extend the accumulator input if we're performing a 64-bit mac.
Differential Revision: https://reviews.llvm.org/D67220
llvm-svn: 371370
Summary:
After tailduplication, we have redundant copies. We can remove these
copies in machine-cp if it's safe to, i.e.
```
$reg0 = OP ...
... <<< No read or clobber of $reg0 and $reg1
$reg1 = COPY $reg0 <<< $reg0 is killed
...
<RET>
```
will be transformed to
```
$reg1 = OP ...
...
<RET>
```
Differential Revision: https://reviews.llvm.org/D65267
llvm-svn: 371359
This patch decodes target and faux shuffles with getTargetShuffleInputs - a reduced version of resolveTargetShuffleInputs that doesn't resolve SM_SentinelZero cases, so we can correctly remove zero vectors if they aren't demanded.
llvm-svn: 371353
If the two zero vectors have undefs in different places they
won't get combined by simplifySelect.
This fixes a regression from an earlier commit.
llvm-svn: 371351
The change to avx512-vec-cmp.ll is a regression, but should be
easy to fix. It occurs because the getZeroVector call was
canonicalizing both sides to the same node, then SimplifySelect
was able to simplify it. But since only called getZeroVector
on some VTs this isn't a robust way to combine this.
The change to vector-shuffle-combining-ssse3.ll is more
instructions, but removes a constant pool load so its unclear
if its a regression or not.
llvm-svn: 371350
As reported in post-commit review of r370327,
there is some case where the code crashes.
As discussed with Craig Topper, the problem is that getConstant()
internally calls getSplatBuildVector(), so we don't insert
the constant itself.
If we do that manually we're good.
llvm-svn: 371346
This generalizes the existing <32 x i1> pre-AVX2 split code to support reductions from <64 x i1> as well, we can probably generalize to any larger pow2 case in the future if the (unlikely) need ever arises.
We still need to tweak combineBitcastvxi1 to improve AVX512F codegen as its assumes vXi1 types should be handled on the mask registers even when they aren't legal.
Differential Revision: https://reviews.llvm.org/D67070
llvm-svn: 371328
isel used to require zero vectors to be canonicalized to a single
type to minimize the number of patterns needed to match. This is
no longer required.
I plan to do this to integers too, but floating point was simpler
to start with. Integer has a complication where v32i16/v64i8 aren't
legal when the other 512-bit integer types are.
llvm-svn: 371325
This patch enables generation of fused multiply add/sub for instructions operating on fp16.
Tested on aarch64-linux.
Differential Revision: https://reviews.llvm.org/D67297
llvm-svn: 371321
Summary:
Normally TargetLowering::expandFixedPointMul would handle
SMULFIXSAT with scale zero by using an SMULO to compute the
product and determine if saturation is needed (if overflow
happened). But if SMULO isn't custom/legal it falls through
and uses the same technique, using MULHS/SMUL_LOHI, as used
for non-zero scales.
Problem was that when checking for overflow (handling saturation)
when not using MULO we did not expect to find a zero scale. So
we ended up in an assertion when doing
APInt::getLowBitsSet(VTSize, Scale - 1)
This patch fixes the problem by adding a new special case for
how saturation is computed when scale is zero.
Reviewers: RKSimon, bevinh, leonardchan, spatel
Reviewed By: RKSimon
Subscribers: wuzish, nemanjai, hiraditya, MaskRay, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D67071
llvm-svn: 371309
Summary:
Add an intrinsic that takes 2 unsigned integers with
the scale of them provided as the third argument and
performs fixed point multiplication on them. The
result is saturated and clamped between the largest and
smallest representable values of the first 2 operands.
This is a part of implementing fixed point arithmetic
in clang where some of the more complex operations
will be implemented as intrinsics.
Patch by: leonardchan, bjope
Reviewers: RKSimon, craig.topper, bevinh, leonardchan, lebedev.ri, spatel
Reviewed By: leonardchan
Subscribers: ychen, wuzish, nemanjai, MaskRay, jsji, jdoerfert, Ka-Ka, hiraditya, rjmccall, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57836
llvm-svn: 371308
Fix for https://bugs.llvm.org/show_bug.cgi?id=43230.
When creating PSHUFLW from a repeated shuffle mask, we have to apply
the checks to the repeated mask, not the original one. For the test
case from PR43230 the inspected part of the original mask is all undef.
Differential Revision: https://reviews.llvm.org/D67314
llvm-svn: 371307
Despite the fact that the localizer's original motivation was to fix horrendous
constant spilling at -O0, shortening live ranges still has net benefits even
with optimizations enabled.
On an -Os build of CTMark, doing this improves code size by 0.5% geomean.
There are a few regressions, bullet increasing in size by 0.5%. One example from
bullet where code size increased slightly was due to GlobalISel actually now
generating the same code as SelectionDAG. So we actually have an opportunity
in future to implement better heuristics for localization and therefore be
*better* than SDAG in some cases. In relation to other optimizations though that
one is relatively minor.
Differential Revision: https://reviews.llvm.org/D67303
llvm-svn: 371266
Trying to minimize the features we need to manipulate when this
is updated for D67259.
The VBMI is interesting because it enables some improved combining
for truncates.
I enabled fast-variable-shuffle because all the CPUs we're going
to add implicitly enable it. So they can share check lines.
llvm-svn: 371261
Test verified that we could compile an empty module and produce an XCOFF
object file. Newer tests superssed this coverage, its safe to remove.
llvm-svn: 371247
We can use a MOVSX16 here then rely on FixupBWInst to change to
MOVSX32 if the upper bits are dead. With a special case to
not promote if it could be turned into CBW.
Then we can rely on X86MCInstLower to turn the MOVSX into CBW
very late if register allocation worked out.
Using MOVSX gives an opportunity to use the MOVSX as a both a
copy and a sign extend since the input and output register aren't
tied together.
Differential Revision: https://reviews.llvm.org/D67192
llvm-svn: 371243
We can rely on X86FixupBWInsts to turn these into MOVZX32. This
simplifies a follow up commit to use MOVSX for i8 sdivrem with
a late optimization to use CBW when register allocation works out.
llvm-svn: 371242
Extend the common/local-common testing for object files to also verify the
symbol table now that the needed functionality has landed in llvm-readobj.
Differential Revision: https://reviews.llvm.org/D66944
llvm-svn: 371237
-tailcallopt requires that we perform different stack adjustments than with
sibling calls. For example, the `@caller_to0_from8` function in
test/CodeGen/AArch64/tail-call.ll requires that we adjust SP. Without
-tailcallopt, this adjustment does not happen. With it, however, it is expected.
So, to ensure that adding sibling call support doesn't break -tailcallopt,
make CallLowering always fall back on possible tail calls when -tailcallopt
is passed in.
Update test/CodeGen/AArch64/tail-call.ll with a GlobalISel line to make sure
that we don't differ from the SDAG implementation at any point.
Differential Revision: https://reviews.llvm.org/D67245
llvm-svn: 371227
This patch sinks add/mul(shufflevector(insertelement())) into the basic block in which they are used so that they can then be selected together.
This is useful for various MVE instructions, such as vmla and others that take R registers.
Loop tests have been added to the vmla test file to make sure vmlas are generated in loops.
Differential revision: https://reviews.llvm.org/D66295
llvm-svn: 371218
This patch allows the DFAPacketizer to be queried after a packet is formed to work out which
resources were allocated to the packetized instructions.
This is particularly important for targets that do their own bundle packing - it's not
sufficient to know simply that instructions can share a packet; which slots are used is
also required for encoding.
This extends the emitter to emit a side-table containing resource usage diffs for each
state transition. The packetizer maintains a set of all possible resource states in its
current state. After packetization is complete, all remaining resource states are
possible packetization strategies.
The sidetable is only ~500K for Hexagon, but the extra tracking is disabled by default
(most uses of the packetizer like MachinePipeliner don't care and don't need the extra
maintained state).
Differential Revision: https://reviews.llvm.org/D66936
llvm-svn: 371198
Summary:
This fixes poor scheduling in a function containing a barrier and a few
load instructions.
Without this fix, ScheduleDAGInstrs::buildSchedGraph adds an artificial
edge in the dependency graph from the barrier instruction to the exit
node representing live-out latency, with a latency of about 500 cycles.
Because of this it thinks the critical path through the graph also has
a latency of about 500 cycles. And because of that it does not think
that any of the load instructions are on the critical path, so it
schedules them with no regard for their (80 cycle) latency, which gives
poor results.
Reviewers: arsenm, dstuttard, tpr, nhaehnle
Subscribers: kzhuravl, jvesely, wdng, yaxunl, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D67218
llvm-svn: 371192
The MVE and LOB extensions of Armv8.1m can be combined to enable
'tail predication' which removes the need for a scalar remainder
loop after vectorization. Lane predication is performed implicitly
via a system register. The effects of predication is described in
Section B5.6.3 of the Armv8.1-m Arch Reference Manual, the key points
being:
- For vector operations that perform reduction across the vector and
produce a scalar result, whether the value is accumulated or not.
- For non-load instructions, the predicate flags determine if the
destination register byte is updated with the new value or if the
previous value is preserved.
- For vector store instructions, whether the store occurs or not.
- For vector load instructions, whether the value that is loaded or
whether zeros are written to that element of the destination
register.
This patch implements a pass that takes a hardware loop, containing
masked vector instructions, and converts it something that resembles
an MVE tail predicated loop. Currently, if we had code generation,
we'd generate a loop in which the VCTP would generate the predicate
and VPST would then setup the value of VPR.PO. The loads and stores
would be placed in VPT blocks so this is not tail predication, but
normal VPT predication with the predicate based upon a element
counting induction variable. Further work needs to be done to finally
produce a true tail predicated loop.
Because only the loads and stores are predicated, in both the LLVM IR
and MIR level, we will restrict support to only lane-wise operations
(no horizontal reductions). We will perform a final check on MIR
during loop finalisation too.
Another restriction, specific to MVE, is that all the vector
instructions need operate on the same number of elements. This is
because predication is performed at the byte level and this is set
on entry to the loop, or by the VCTP instead.
Differential Revision: https://reviews.llvm.org/D65884
llvm-svn: 371179
Summary:
Fix a bug of not update the jump table and recommit it again.
In `block-placement` pass, it will create some patterns for unconditional we can do the simple early retrun.
But the `early-ret` pass is before `block-placement`, we don't want to run it again.
This patch is to do the simple early return to optimize the blocks at the last of `block-placement`.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D63972
llvm-svn: 371177
The
;CHECK: bb
;CHECK-NEXT: %namedVReg1353:_(p0) = COPY $d0
parts of the test case failed when the tests were placed in a directory
including "bb" in the path, since the full path of the file is then
output in the
; ModuleID = '/repo/bb/
line which the CHECK matched on and then the CHECK-NEXT failed.
llvm-svn: 371171
This was only using the correct register constraints if this was the
final result instruction. If the extract was a sub instruction of the
result, it would attempt to use GIR_ConstrainSelectedInstOperands on a
COPY, which won't work. Move the handling to
createAndImportSubInstructionRenderer so it works correctly.
I don't fully understand why runOnPattern and
createAndImportSubInstructionRenderer both need to handle these
special cases, and constrain them with slightly different methods. If
I remove the runOnPattern handling, it does break the constraint when
the final result instruction is EXTRACT_SUBREG.
llvm-svn: 371150
The same stack is loaded for each workitem ID, and each use. Nothing
prevents you from creating multiple fixed stack objects with the same
offsets, so this was creating a load for each unique frame index,
despite them being the same offset. Re-use the same frame index so the
loads are CSEable.
llvm-svn: 371148
This patch reuses the MIR vreg renamer from the MIRCanonicalizerPass to cleanup
names of vregs in a MIR file for MIR test authors. I found it useful when
writing a regression test for a globalisel failure I encountered recently and
thought it might be useful for other folks as well.
Differential Revision: https://reviews.llvm.org/D67209
llvm-svn: 371121
Recommit basic sibling call lowering (https://reviews.llvm.org/D67189)
The issue was that if you have a return type other than void, call lowering
will emit COPYs to get the return value after the call.
Disallow sibling calls other than ones that return void for now. Also
proactively disable swifterror tail calls for now, since there's a similar issue
with COPYs there.
Update call-translator-tail-call.ll to include test cases for each of these
things.
llvm-svn: 371114
The code was incorrectly counting the number of identical instructions,
and therefore tried to predicate an instruction which should not have
been predicated. This could have various effects: a compiler crash,
an assembler failure, a miscompile, or just generating an extra,
unnecessary instruction.
Instead of depending on TargetInstrInfo::removeBranch, which only
works on analyzable branches, just remove all branch instructions.
Fixes https://bugs.llvm.org/show_bug.cgi?id=43121 and
https://bugs.llvm.org/show_bug.cgi?id=41121 .
Differential Revision: https://reviews.llvm.org/D67203
llvm-svn: 371111
As noted in PR43197, we can use test+add+cmov+sra to implement
signed division by a power of 2.
This is based off the similar version in AArch64, but I've
adjusted it to use target independent nodes where AArch64 uses
target specific CMP and CSEL nodes. I've also blocked INT_MIN
as the transform isn't valid for that.
I've limited this to i32 and i64 on 64-bit targets for now and only
when CMOV is supported. i8 and i16 need further investigation to be
sure they get promoted to i32 well.
I adjusted a few tests to enable cmov to demonstrate the new
codegen. I also changed twoaddr-coalesce-3.ll to 32-bit mode
without cmov to avoid perturbing the scenario that is being
set up there.
Differential Revision: https://reviews.llvm.org/D67087
llvm-svn: 371104
As discussed on D64551 and PR43227, we don't correctly handle cases where the base load has a non-zero byte offset.
Until we can properly handle this, we must bail from EltsFromConsecutiveLoads.
llvm-svn: 371078
This attempts to just fix the creation of VPT blocks, fixing up the iterating,
which instructions are considered in the bundle, and making sure that we do not
overrun the end of the block.
Differential Revision: https://reviews.llvm.org/D67219
llvm-svn: 371064
G_FENCE comes form fence instruction. For MIPS fence is generated in
AtomicExpandPass when atomic instruction gets surrounded with fence
instruction when needed.
G_FENCE arguments don't have LLT, because of that there is no job for
legalizer and regbankselect. Instruction select G_FENCE for MIPS32.
Differential Revision: https://reviews.llvm.org/D67181
llvm-svn: 371056
Matt Arsenault