As discussed on D47985, identity shuffle masks should probably be free.
I've limited this to the case where the input and output types all match - but we could probably accept all cases.
Differential Revision: https://reviews.llvm.org/D47986
llvm-svn: 334506
Implement default legalization of rotates: either in terms of the rotation
in the opposite direction (if legal), or in terms of shifts and ors.
Implement generating of rotate instructions for Hexagon. Hexagon only
supports rotates by an immediate value, so implement custom lowering of
ROTL/ROTR on Hexagon. If a rotate is not legal, use the default expansion.
Differential Revision: https://reviews.llvm.org/D47725
llvm-svn: 334497
Extend LONG_BRANCH_LUi and LONG_BRANCH_ADDiu pseudo instructions with
additional flag, so instead of always lowering to lui %hi(...),
addiu %lo(...) or addiu %hi(...), now they can lower to either %lo, %hi,
%higher or %highest depending on the added flag.
Differential Revision: https://reviews.llvm.org/D47941
llvm-svn: 334490
We were missing packed isel folding patterns for all of sse41, avx, and avx512.
For some reason avx512 had scalar load folding patterns under optsize(due to partial/undef reg update), but we didn't have the equivalent sse41 and avx patterns.
Sometimes we would get load folding due to peephole pass anyway, but we're also missing avx512 instructions from the load folding table. I'll try to fix that in another patch.
Some of this was spotted in the review for D47993.
This patch adds all the folds to isel, adds a few spot tests, and disables the peephole pass on a few tests to ensure we're testing some of these patterns.
llvm-svn: 334460
The use iterator, used within findMaskOperands(), can return anything which is
not a def. isUse() requires a register, so check isReg() before calling isUse().
Differential Revision: https://reviews.llvm.org/D48047
llvm-svn: 334459
Name table occupies a big chunk of size in current binary format sample profile.
In order to reduce its size, the patch changes the sample writer/reader to
save/restore MD5Hash of names in the name table. Sample annotation phase will
also use MD5Hash of name to query samples accordingly.
Experiment shows compact binary format can reduce the size of sample profile by
2/3 compared with binary format generally.
Differential Revision: https://reviews.llvm.org/D47955
llvm-svn: 334447
This would fail before because 1x vectors aren't legal,
so instead just use the scalar type.
Avoids regressions in a future AMDGPU commit to add
v4i16/v4f16 as legal types.
Test update is just the one test that this triggers
on in tree now. It wasn't checking anything before.
The result is completely changed since the selects
are eliminated. Not sure if it's considered better
or not.
llvm-svn: 334440
Summary:
Previously we would add them for adds, but not multiplies.
Reviewers: sanjoy
Subscribers: llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D48038
llvm-svn: 334428
Rational: if there is indirect access that is usually an issue
because load is not ready by the use. However, if use is inside a
loop and load is outside that is potentially an issue for a first
iteration only.
Differential Revision: https://reviews.llvm.org/D47740
llvm-svn: 334420
When program is compiled for mips3 with n64 abi, wrong register class
is used for creating an emergency spill slot. This patch fixes the
correct register class to be chosen.
This patch resolves PR35859.
Thanks to John Baldwin for reporting the issue!
Differential Revision: https://reviews.llvm.org/D47938
llvm-svn: 334419
This sets trackLivenessAfterRegAlloc on AVRRegisterInfo.
Most existing targets set this flag. Without it, specific IR inputs
cause LLVM to fail with:
Assertion failed: (getParent()->getProperties().hasProperty( MachineFunctionProperties::Property::TracksLiveness) &&
"Liveness information is accurate"), function livein_begin
file MachineBasicBlock.cpp, line 1354.
With this commit, this no longer happens.
Patch by Peter Nimmervoll.
llvm-svn: 334409
Summary:
This fixes most of the scheduling info for SKX vector operations.
I had to split a lot of the YMM/ZMM classes into separate classes for YMM and ZMM.
The before/after llvm-exegesis analysis are in the phabricator diff.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47721
llvm-svn: 334407
Summary:
The idiom recognition seems rather poor.
Only the `@bzhi32_d0` produces `v_bfe_u32`.
But they all should.
This needs to be fixed before D47980 can be re-landed.
Reviewers: mareko, bogner, rampitec, arsenm, tstellar, nhaehnle
Reviewed By: nhaehnle
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Tags: #amdgpu
Differential Revision: https://reviews.llvm.org/D48005
llvm-svn: 334398
Summary: When compiling with -fpic, in contrast to -fPIC, use only the
immediate field to index into the GOT. This saves space if the GOT is
known to be small. The linker will warn if the GOT is too large for
this method.
Reviewers: jyknight, venkatra
Reviewed By: jyknight
Subscribers: brad, fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D47136
llvm-svn: 334383
Codeview references to unnamed structs and unions are expected to refer to the
complete type definition instead of a forward reference so Visual Studio can
resolve the type properly.
Differential Revision: https://reviews.llvm.org/D32498
llvm-svn: 334382
This patch started off much more general and ambitious, but it's been a nightmare
seeing all the ways x86 vector codegen can go wrong.
So the code is still structured to allow extending easily, but it's currently
limited in several ways:
1. Only handle cases with an extending load.
2. Only handle cases with a zero constant compare.
3. Ignore setcc with vector bitmask (SetCCWidth != 1) - so AVX512 should be unaffected.
The motivating case from PR37427:
https://bugs.llvm.org/show_bug.cgi?id=37427
...is the 1st test, and that shows the expected win - we eliminated the unnecessary
intermediate cast.
There's a clear regression in the last test (sgt_zero_fp_select) because we longer
recognize a 'SHRUNKBLEND' opportunity. I think that general problem is also present
in sgt_zero, so I'll try to fix that in a follow-up. We need to match a sign-bit
setcc from a sign-extended operand and remove it.
Differential Revision: https://reviews.llvm.org/D47330
llvm-svn: 334378
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47447
llvm-svn: 334361
It looks like this got left in by accident in r289794; I can't think of
any reason this check would be necessary. (Maybe it was meant to be a
check that the AND has one use? But we check that a few lines earlier.)
Differential Revision: https://reviews.llvm.org/D47921
llvm-svn: 334322
An expression like
(zext i2 {(trunc i32 (1 + %B) to i2),+,1}<%while.body> to i32)
will become zero exactly when the nested value becomes zero in its type.
Strip injective operations from the input value in howFarToZero to make
the value simpler.
Differential Revision: https://reviews.llvm.org/D47951
llvm-svn: 334318
Summary:
If we can use comdats, then we can make it so that the global metadata
is thrown away if the prevailing definition of the global was
uninstrumented. I have only tested this on COFF targets, but in theory,
there is no reason that we cannot also do this for ELF.
This will allow us to re-enable string merging with ASan on Windows,
reducing the binary size cost of ASan on Windows.
Reviewers: eugenis, vitalybuka
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D47841
llvm-svn: 334313
Extension to D46954 (PR37426), this patch adds support for v8i16/v16i16 rotations in a similar manner - the conversion of the shift/rotate amount to a multiplication factor and the use of PMULLW to shift left and PMULHUW (ISD::MULHU) to shift the wrapped bits back around to be ORd together.
Differential Revision: https://reviews.llvm.org/D47822
llvm-svn: 334309
Summary: This patch originated from D46562 and is a proper subset, with some issues addressed for fsub.
Reviewers: spatel, hfinkel, wristow, arsenm
Reviewed By: spatel
Subscribers: wdng
Differential Revision: https://reviews.llvm.org/D47910
llvm-svn: 334306
As detailed on Agner's Microarchitecture doc (21.8 AMD Bobcat and Jaguar pipeline - Dependency-breaking instructions), these instructions are dependency breaking and fast-path zero the destination register (and appropriate EFLAGS bits).
llvm-svn: 334303
Same fix as rL334110: I noticed while working on zero-idiom + dependency-breaking support (PR36671) that most of our binary instruction schedule tests were reusing the same src registers, which would cause the tests to fail once we enable scalar zero-idiom support on btver2.
llvm-svn: 334302
AMDGPU inline assembler support i16, half and i128 typed variables in constraints, but they were reported as error.
Needed to fix https://github.com/RadeonOpenCompute/ROCm/issues/341,
e.g. to be able to load with global_load_dwordx4 to a 128bit integer variable
Differential Revision: https://reviews.llvm.org/D44920
llvm-svn: 334301
Summary:
`%ret = add nuw i8 %x, C`
From [[ https://llvm.org/docs/LangRef.html#add-instruction | langref ]]:
nuw and nsw stand for “No Unsigned Wrap” and “No Signed Wrap”,
respectively. If the nuw and/or nsw keywords are present,
the result value of the add is a poison value if unsigned
and/or signed overflow, respectively, occurs.
So if `C` is `-1`, `%x` can only be `0`, and the result is always `-1`.
I'm not sure we want to use `KnownBits`/`LVI` here, because there is
exactly one possible value (all bits set, `-1`), so some other pass
should take care of replacing the known-all-ones with constant `-1`.
The `test/Transforms/InstCombine/set-lowbits-mask-canonicalize.ll` change *is* confusing.
What happening is, before this: (omitting `nuw` for simplicity)
1. First, InstCombine D47428/rL334127 folds `shl i32 1, %NBits`) to `shl nuw i32 -1, %NBits`
2. Then, InstSimplify D47883/rL334222 folds `shl nuw i32 -1, %NBits` to `-1`,
3. `-1` is inverted to `0`.
But now:
1. *This* InstSimplify fold `%ret = add nuw i32 %setbit, -1` -> `-1` happens first,
before InstCombine D47428/rL334127 fold could happen.
Thus we now end up with the opposite constant,
and it is all good: https://rise4fun.com/Alive/OA9https://rise4fun.com/Alive/sldC
Was mentioned in D47428 review.
Follow-up for D47883.
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47908
llvm-svn: 334298
Simplify combineVectorTruncationWithPACKUS to mask the upper bits followed by calling truncateVectorWithPACK instead of duplicating with similar code.
This results in the codegen using (V)PACKUSDW on SSE41+ targets for vXi64/vXi32 inputs where before it always used PACKUSWB (along with a lot more bitcasting).
I've raised PR37749 as until we avoid unnecessary concats back to 256-bit for bitwise ops, we can't avoid splitting the input value into 128-bit subvectors for masking.
llvm-svn: 334289
Currently the loop branch heuristic is applied before the invoke heuristic which makes us overestimate the probability of the unwind destination of invokes inside loops. This in turn makes us grossly underestimate the frequencies of loops with invokes.
Reviewed By: skatkov, vsk
Differential Revision: https://reviews.llvm.org/D47371
llvm-svn: 334285
The instruction makes use of a previously ignored field in the fence
instruction. It is introduced in the version 2.3 draft of the RISC-V
specification after much work by the Memory Model Task Group.
As clarified here <https://github.com/riscv/riscv-isa-manual/issues/186>,
the fence.tso assembler mnemonic does not have operands.
llvm-svn: 334278
We have some combines/lowerings that attempt to use PACKSS-then-PACKUS and others that use PACKUS-then-PACKSS.
PACKUS is much easier to combine with if we know the upper bits are zero as ComputeKnownBits can easily see through BITCASTs etc. especially now that rL333995 and rL334007 have landed. It also effectively works at byte level which further simplifies shuffle combines.
The only (minor) annoyances are that ComputeKnownBits can sometimes take longer as it doesn't fail as quickly as ComputeNumSignBits (but I'm not seeing any actual regressions in tests) and PACKUSDW only became available after SSE41 so we have more codegen diffs between targets.
llvm-svn: 334276
There could be more than one PHIs in exit block using same loop recurrence.
Don't assume there is only one and fix each user.
Differential Revision: https://reviews.llvm.org/D47788
llvm-svn: 334271
While trying to propagate AND masks back to loads, we currently allow
one non-load node to be included as a leaf in chain. This fix now
limits that node to produce only a single data value.
Differential Revision: https://reviews.llvm.org/D47878
llvm-svn: 334268
Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode.
Reviewers: spatel, arsenm, hfinkel, javed.absar
Reviewed By: spatel
Subscribers: nemanjai, wdng
Differential Revision: https://reviews.llvm.org/D47388
llvm-svn: 334242
- Make code easier to maintain.
- Avoid generating waitcnts for VMEM if the address sppace does not involve VMEM.
- Add support to generate waitcnts for LDS and GDS memory.
Differential Revision: https://reviews.llvm.org/D47504
llvm-svn: 334241
%ret = add nuw i8 %x, C
From langref:
nuw and nsw stand for “No Unsigned Wrap” and “No Signed Wrap”,
respectively. If the nuw and/or nsw keywords are present,
the result value of the add is a poison value if unsigned
and/or signed overflow, respectively, occurs.
So if C is -1, %x can only be 0, and the result is always -1.
https://rise4fun.com/Alive/sldC
Was mentioned in D47428 review.
llvm-svn: 334236
Summary:
`%r = shl nuw i8 C, %x`
As per langref:
```
If the nuw keyword is present, then the shift produces
a poison value if it shifts out any non-zero bits.
```
Thus, if the sign bit is set on `C`, then `%x` can only be `0`,
which means that `%r` can only be `C`.
Or in other words, set sign bit means that the signed value
is negative, so the constant is `<= 0`.
https://rise4fun.com/Alive/WMkhttps://rise4fun.com/Alive/udv
Was mentioned in D47428 review.
We already handle the `0` constant, https://godbolt.org/g/UZq1sJ, so this only handles negative constants.
Could use computeKnownBits() / LazyValueInfo,
but the cost-benefit analysis (https://reviews.llvm.org/D47891)
suggests it isn't worth it.
Reviewers: spatel, craig.topper
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47883
llvm-svn: 334222
These weren't included in D19544 - probably just an oversight.
D40044 made it more likely that we'll have LLVM math intrinsics rather
than libcalls, so this bug was more easily exposed.
As the tests/code show, we already have the complete mappings for pow/exp/log.
I don't have any experience with SVML, so I don't know if anything else is
missing. It's also not clear to me that we should be doing this transform in
IR rather than DAG/isel, but that's a separate issue.
Differential Revision: https://reviews.llvm.org/D47610
llvm-svn: 334211
The implementation follows the MIPS backend and expands the pseudo instruction
directly during asm parsing. As the result, only real MC instructions are
emitted to the MCStreamer. The actual expansion to real instructions is
similar to the expansion performed by the GNU Assembler.
This patch supersedes D41949.
Differential Revision: https://reviews.llvm.org/D46118
Patch by Mario Werner.
llvm-svn: 334203
%r = shl nuw i8 C, %x
As per langref: If the nuw keyword is present, then the shift produces
a poison value if it shifts out any non-zero bits.
Thus, if the sign bit is set on C, then %x can only be 0,
which means that %r can only be C.
https://rise4fun.com/Alive/WMk
Was mentioned in D47428 review.
llvm-svn: 334200
BitPermutationSelector sets Repl32 flag for bit groups which can be (potentially) benefit from 32-bit rotate-and-mask instructions with bit replication, i.e. rlwinm/rlwimi copies lower 32 bits into upper 32 bits on 64-bit PowerPC before rotation.
However, enforcing 32-bit instruction sometimes results in redundant generated code.
For example, the following simple code is compiled into rotldi + rlwimi while it can be compiled into only rldimi instruction if Repl32 flag is not set on the bit group for (a & 0xFFFFFFFF).
uint64_t func(uint64_t a, uint64_t b) {
return (a & 0xFFFFFFFF) | (b << 32) ;
}
To avoid such problem, this patch checks the potential benefit of Repl32 flag before setting it. If a bit group does not require rotation (i.e. RLAmt == 0) and won't be merged into another group, we do not benefit from Repl32 flag on this group.
Differential Revision: https://reviews.llvm.org/D47867
llvm-svn: 334195
Simplify combineVectorTruncationWithPACKSS to just a SIGN_EXTEND_INREG followed by using the existing truncateVectorWithPACK instead of duplicating code.
llvm-svn: 334193
The existing trunc_shl_17_v8i16_v8i32 test case should (but doesn't) fold to zero, I've added 2 new test cases:
- trunc_shl_16_v8i16_v8i32 which folds to zero (this is actually testing the target faux shuffle combine)
- trunc_shl_15_v8i16_v8i32 which should perform the full shl + truncate
llvm-svn: 334188
We do the same thing in rewriteSingleStoreAlloca.
Fixes PR37632.
Reviewers: chandlerc, davide, efriedma
Reviewed By: davide
Differential Revision: https://reviews.llvm.org/D47825
llvm-svn: 334187
This has two main components. First, widen
widen short constant loads in DAG when they have
the correct alignment. This is already done a bit in
AMDGPUCodeGenPrepare, since that has access to
DivergenceAnalysis. This can't help kernarg loads
created in the DAG. Start to use DAG divergence analysis
to help this case.
The second part is to avoid kernel argument lowering
breaking the alignment of short vector elements because
calling convention lowering wants to split everything
into legal register types.
When loading a split type, load the nearest 4-byte aligned
segment and shift to get the desired bits. This extra
load of the earlier argument piece ends up merging,
and the bit extract hopefully folds out.
There are a number of improvements and regressions with
this, but I think as-is this is a better compromise between
several of the worst parts of SelectionDAG.
Particularly when i16 is legal, this produces worse code
for i8 and i16 element vector kernel arguments. This is
partially due to the very weak load merging the DAG does.
It only looks for fairly specific combines between pairs
of loads which no longer appear. In particular this
causes v4i16 loads to be split into 2 components when
previously the two halves were merged.
Worse, because of the newly introduced shifts, there
is a lot more unnecessary vector packing and unpacking code
emitted. At least some of this is due to reporting
false for isTypeDesirableForOp for i16 as a workaround for
the lack of divergence information in the DAG. The cases
where this happens it doesn't actually matter, but the
relevant code in SimplifyDemandedBits doens't have the context
to know to ignore this.
The use of the scalar cache is probably more important
than the mess of mostly scalar instructions doing this packing
and unpacking. Future work can fix this, possibly by making better
use of the new DAG divergence information for controlling promotion
decisions, or adding another version of shift + trunc + shift
combines that doesn't only know about the used types.
llvm-svn: 334180
Summary: Prevent folding of operations with memory loads when one of the sources has undefined register update.
Reviewers: craig.topper
Subscribers: llvm-commits, mike.dvoretsky, ashlykov
Differential Revision: https://reviews.llvm.org/D47621
llvm-svn: 334175
Summary:
When the branch folder hoist code into a predecessor it adjust live-in's
in the blocks it hoist code from. However it fail to handle hoisted code
that contain a defed register that originally is live-in in the block
through a super register.
This is fixed by replacing the live-in handling code with calls to
utility functions in LivePhysRegs.
Reviewers: kparzysz, gberry, MatzeB, uweigand, aprantl
Reviewed By: kparzysz
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47529
llvm-svn: 334163
With '-elf-output-style=GNU -relocations', a header containing the number
of entries is printed before all the relocation entries in the section.
For Android packed format, we need to perform the unpacking first before
we can get the actual number of relocations in the section.
Patch by Rahul Chaudhry!
Differential Revision: https://reviews.llvm.org/D47800
llvm-svn: 334147
When denormals are supported we are producing a full division for
1.0f / x. That still can be replaced by the faster version:
bool c = fabs(x) > 0x1.0p+96f;
float s = c ? 0x1.0p-32f : 1.0f;
x *= s;
return s * v_rcp_f32(x)
in case if requested accuracy is 2.5ulp or less. The same version
is used if denormals are not supported for non 1.0 numerators, where
just v_rcp_f32 is then used for 1.0 numerator.
The optimization of 1/x is extended to the case -1/x, which is the
same except for the resulting sign bit.
OpenCL conformance passed with both enabled and disabled denorms.
Differential Revision: https://reviews.llvm.org/D47805
llvm-svn: 334142
The bug report:
https://bugs.llvm.org/show_bug.cgi?id=36036
...requests a DAG change for this, but an IR canonicalization
probably handles most cases. If we still want to match this
pattern in the backend, there's a proposal for that too:
D47831
Alive proofs including nsw/nuw cases that were first noted in:
D46988
https://rise4fun.com/Alive/Kmp
This patch is largely copied from the existing code that was
initially added with:
D40984
...but I didn't see much gain from trying to share code.
llvm-svn: 334137
Fixes terrible code on targets without f16 support. The
legalization creates a mess that is difficult to recover
from. Also should avoid randomly breaking these tests
multiple times in sequence in future commits.
Some regressions in cases where it happens to be better
to pull the source modifier after the conversion.
llvm-svn: 334132
Summary:
This is [[ https://bugs.llvm.org/show_bug.cgi?id=37603 | PR37603 ]].
https://godbolt.org/g/VCMNpShttps://rise4fun.com/Alive/idM
When doing bit manipulations, it is quite common to calculate some bit mask,
and apply it to some value via `and`.
The typical C code looks like:
```
int mask_signed_add(int nbits) {
return (1 << nbits) - 1;
}
```
which is translated into (with `-O3`)
```
define dso_local i32 @mask_signed_add(int)(i32) local_unnamed_addr #0 {
%2 = shl i32 1, %0
%3 = add nsw i32 %2, -1
ret i32 %3
}
```
But there is a second, less readable variant:
```
int mask_signed_xor(int nbits) {
return ~(-(1 << nbits));
}
```
which is translated into (with `-O3`)
```
define dso_local i32 @mask_signed_xor(int)(i32) local_unnamed_addr #0 {
%2 = shl i32 -1, %0
%3 = xor i32 %2, -1
ret i32 %3
}
```
Since we created such a mask, it is quite likely that we will use it in `and` next.
And then we may get rid of `not` op by folding into `andn`.
But now that i have actually looked:
https://godbolt.org/g/VTUDmU
_some_ backend changes will be needed too.
We clearly loose `bzhi` recognition.
Reviewers: spatel, craig.topper, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D47428
llvm-svn: 334127
Summary:
In D47428, i propose to choose the `~(-(1 << nbits))` as the canonical form of low-bit-mask formation.
As it is seen from these tests, there is a reason for that.
AArch64 currently better handles `~(-(1 << nbits))`, but not the more traditional `(1 << nbits) - 1` (sic!).
The other way around for X86.
It would be much better to canonicalize.
This patch is completely monkey-typing.
I don't really understand how this works :)
I have based it on `// x & (-1 >> (32 - y))` pattern.
Also, when we only have `BMI`, i wonder if we could use `BEXTR` with `start=0` ?
Related links:
https://bugs.llvm.org/show_bug.cgi?id=36419https://bugs.llvm.org/show_bug.cgi?id=37603https://bugs.llvm.org/show_bug.cgi?id=37610https://rise4fun.com/Alive/idM
Reviewers: craig.topper, spatel, RKSimon, javed.absar
Reviewed By: craig.topper
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D47453
llvm-svn: 334125
Summary:
In D47428, i propose to choose the `~(-(1 << nbits))` as the canonical form of low-bit-mask formation.
As it is seen from these tests, there is a reason for that.
AArch64 currently better handles `~(-(1 << nbits))`, but not the more traditional `(1 << nbits) - 1` (sic!).
The other way around for X86.
It would be much better to canonicalize.
It would seem that there is too much tests, but this is most of all the auto-generated possible variants
of C code that one would expect for BZHI to be formed, and then manually cleaned up a bit.
So this should be pretty representable, which somewhat good coverage...
Related links:
https://bugs.llvm.org/show_bug.cgi?id=36419https://bugs.llvm.org/show_bug.cgi?id=37603https://bugs.llvm.org/show_bug.cgi?id=37610https://rise4fun.com/Alive/idM
Reviewers: javed.absar, craig.topper, RKSimon, spatel
Reviewed By: RKSimon
Subscribers: kristof.beyls, llvm-commits, RKSimon, craig.topper, spatel
Differential Revision: https://reviews.llvm.org/D47452
llvm-svn: 334124
These encodings correspond to the cases in the normal encoding scheme where there is no index and our modrm reading code initially decodes it as such. The VSIB handling code tried to compensate for this, but failed to add the base needed to make later code do the right thing.
Fixes PR37712.
llvm-svn: 334121
As detailed on Agner's Microarchitecture doc (21.8 AMD Bobcat and Jaguar pipeline - Dependency-breaking instructions), all these instructions are dependency breaking and zero the destination register.
llvm-svn: 334119
Before this patch, debugify would insert debug value intrinsics before the
terminating instruction in a block. This had the advantage of being simple,
but was a bit too simple/unrealistic.
This patch teaches debugify to insert debug values immediately after their
operand defs. This enables better testing of the compiler.
For example, with this patch, `opt -debugify-each` is able to identify a
vectorizer DI-invariance bug fixed in llvm.org/PR32761. In this bug, the
vectorizer produced different output with/without debug info present.
Reverting Davide's bugfix locally, I see:
$ ~/scripts/opt-check-dbg-invar.sh ./bin/opt \
.../SLPVectorizer/AArch64/spillcost-di.ll -slp-vectorizer
Comparing: -slp-vectorizer .../SLPVectorizer/AArch64/spillcost-di.ll
Baseline: /var/folders/j8/t4w0bp8j6x1g6fpghkcb4sjm0000gp/T/tmp.iYYeL1kf
With DI : /var/folders/j8/t4w0bp8j6x1g6fpghkcb4sjm0000gp/T/tmp.sQtQSeet
9,11c9,11
< %5 = getelementptr inbounds %0, %0* %2, i64 %0, i32 1
< %6 = bitcast i64* %4 to <2 x i64>*
< %7 = load <2 x i64>, <2 x i64>* %6, align 8, !tbaa !0
---
> %5 = load i64, i64* %4, align 8, !tbaa !0
> %6 = getelementptr inbounds %0, %0* %2, i64 %0, i32 1
> %7 = load i64, i64* %6, align 8, !tbaa !5
12a13
> store i64 %5, i64* %8, align 8, !tbaa !0
14,15c15
< %10 = bitcast i64* %8 to <2 x i64>*
< store <2 x i64> %7, <2 x i64>* %10, align 8, !tbaa !0
---
> store i64 %7, i64* %9, align 8, !tbaa !5
:: Found a test case ^
Running this over the *.ll files in tree, I found four additional examples
which compile differently with/without DI present. I plan on filing bugs for
these.
llvm-svn: 334118
This causes "permission denied" error in some controlled test environment where source tree is read-only.
Differential Revision: https://reviews.llvm.org/D47839
llvm-svn: 334114
Summary:
This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483.
It contains only context for fsqrt.
Reviewers: spatel, hfinkel, arsenm
Reviewed By: spatel
Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai
Differential Revision: https://reviews.llvm.org/D47749
llvm-svn: 334113
I noticed while working on zero-idiom + dependency-breaking support (PR36671) that most of our binary instruction tests were reusing the same src registers, which would cause the tests to fail once we enable scalar zero-idiom support on btver2. Fixed in all targets to keep them in sync.
llvm-svn: 334110
These were added for D5603 / rL219542, and there's a proposal to
change one side in D47807.
These are tests of constant propagation, so they shouldn't have
ever been tested/housed under InstCombine.
llvm-svn: 334107
As detailed on Agner's Microarchitecture doc (21.8 AMD Bobcat and Jaguar pipeline - Dependency-breaking instructions), all these instructions are dependency breaking and zero the destination register.
TODO: Scalar instructions still need to be tested (need to check EFLAGS handling).
llvm-svn: 334104
If no alignment is set, the abi/preferred alignment of structs will be
used which may be higher than required. This can lead to extra padding
and in the end an increase in data size.
Differential Revision: https://reviews.llvm.org/D47633
llvm-svn: 334099
We should never get different CodeGen based on whether the code is being
compiled in debug mode so we must skip over @llvm.dbg.value (and similar)
calls.
Should fix at least the worst part of PR37690.
llvm-svn: 334090
The test changes in r334031 give unstable pass/fail results on the
llvm-clang-x86_64-expensive-checks-win buildbot. Revert the test changes to
turn the bot green.
llvm-svn: 334084
Only the bottom 16-bits of BEXTR's control op are required (0:8 INDEX, 15:8 LENGTH).
Differential Revision: https://reviews.llvm.org/D47690
llvm-svn: 334083
On targets like Arm some relaxations may only be performed when certain
architectural features are available. As functions can be compiled with
differing levels of architectural support we must make a judgement on
whether we can relax based on the MCSubtargetInfo for the function. This
change passes through the MCSubtargetInfo for the function to
fixupNeedsRelaxation so that the decision on whether to relax can be made
per function. In this patch, only the ARM backend makes use of this
information. We must also pass the MCSubtargetInfo to applyFixup because
some fixups skip error checking on the assumption that relaxation has
occurred, to prevent code-generation errors applyFixup must see the same
MCSubtargetInfo as fixupNeedsRelaxation.
Differential Revision: https://reviews.llvm.org/D44928
llvm-svn: 334078
Add minimal support to lower function calls.
Support only functions with arguments/return that go through registers
and have type i32.
Patch by Petar Avramovic.
Differential Revision: https://reviews.llvm.org/D45627
llvm-svn: 334071
This is a fix for the problem arising in D47374 (PR37678):
https://bugs.llvm.org/show_bug.cgi?id=37678
We may not have throughput info because it's not specified in the model
or it's not available with variant scheduling, so assume that those
instructions can execute/complete at max-issue-width.
Differential Revision: https://reviews.llvm.org/D47723
llvm-svn: 334055
CodeGenPrepare pass move extension instructions close to load instructions in different BB, so they can be combined later. But the extension instructions can't move through logical and shift instructions in current implementation. This patch enables this enhancement, so we can eliminate more extension instructions.
Differential Revision: https://reviews.llvm.org/D45537
This is re-commit of r331783, which was reverted by r333305. The performance regression was caused by some unlucky alignment, not a code generation problem.
llvm-svn: 334049
Preserves the low bound of the !range. I don't think
it's legal to do anything with the top half since it's
theoretically reading garbage.
llvm-svn: 334045
Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483.
Reviewers: spatel, hfinkel
Reviewed By: spatel
Subscribers: nemanjai
Differential Revision: https://reviews.llvm.org/D47389
llvm-svn: 334037
This patch fixe the logic in ReadState::cycleEvent(). That method was not
correctly updating field `TotalCycles`.
Added extra code comments in class ReadState to better describe each field.
llvm-svn: 334028
Similar to v4i32 SHL, convert v8i16 shift amounts to scale factors instead to improve performance and reduce instruction count. We were already doing this for constant shifts, this adds variable shift support.
Reduces the serial nature of the codegen, which relies on chains of plendvb/pand+pandn+por shifts.
This is a step towards adding support for vXi16 vector rotates.
Differential Revision: https://reviews.llvm.org/D47546
llvm-svn: 334023
Summary:
Allow extended parsing of variable assembler assignment syntax and modify X86 to permit
VAR = register assignment. As we emit these as .set directives when possible, we inline
such expressions in output assembly.
Fixes PR37425.
Reviewers: rnk, void, echristo
Reviewed By: rnk
Subscribers: nickdesaulniers, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D47545
llvm-svn: 334022
When legalizing illegal FP load results, this was
for some reason dropping the invariant and dereferencable
memory flags. There doesn't seem to be any reason for this,
and the equivalent isn't done for integer loads.
Fixes an issue in a future AMDGPU commit where some identical
loads fail to merge because one of the loads ends up
dropping the flags.
llvm-svn: 334020
When adjusting a cmp in order to canonicalize an abs/nabs select pattern we need
to use the type of the existing operand when creating a new operand not the
type of a select operand, as the two may be different.
This fixes PR37686.
llvm-svn: 334019
BitPermutationSelector builds the output value by repeating rotate-and-mask instructions with input registers.
Here, we may avoid one rotate instruction if we start building from an input register that does not require rotation.
For example of the test case bitfieldinsert.ll, it first rotates left r4 by 8 bits and then inserts some bits from r5 without rotation.
This can be executed by one rlwimi instruction, which rotates r4 by 8 bits and inserts its bits into r5.
This patch adds a check for rotation amounts in the comparator used in sorting to process the input without rotation first.
Differential Revision: https://reviews.llvm.org/D47765
llvm-svn: 334011
Ideally we'd use resolveTargetShuffleInputs to handle faux shuffles as well but:
(a) that code path doesn't handle general/pre-legalized ops/types very well.
(b) I'm concerned about the compute time as they recurse to calls to computeKnownBits/ComputeNumSignBits which would need depth limiting somehow.
llvm-svn: 334007
When the branch target of a Thumb2 unconditional or conditonal branch is
resolved at assembly time, no range checking is performed on the result
leading to incorrect immediates. This change adds a range check:
+- 16 Megabytes for unconditional branches, +- 1 Megabyte for the
conditional branch.
Differential Revision: https://reviews.llvm.org/D46306
llvm-svn: 333997
The Thumb BL range is + or - either 16 Megabytes or 4 Megabytes depending
on whether the CPU supports Thumb2 or the v8-m baseline ops. The existing
check for BL range is incorrectly set at +- 32 Megabytes. This change
corrects the higher range and uses the lower range if the featurebits
don't have the necessary support for it.
Differential Revision: https://reviews.llvm.org/D46305
llvm-svn: 333991
This is the new version of D46181, allowing setjmp/longjmp
to work correctly with the Intel CET shadow stack by storing
SSP on setjmp and fixing it on longjmp. The patch has been
updated to use the cf-protection-return module flag instead
of HasSHSTK, and the bug that caused D46181 to be reverted
has been fixed with the test expanded to track that fix.
patch by mike.dvoretsky
Differential Revision: https://reviews.llvm.org/D47311
llvm-svn: 333990
Passing -mattr=-mmx needs to disable these instructions since the MMX register class won't have been set up. But we don't want -mattr=-mmx to disable SSE so we have to do it separately.
llvm-svn: 333984
Start by emitting remarks for very basic unsupported cases such as
irreducible CFGs and EHFunclets. The end goal is to be able to cover all
the cases where we give up with an explanation.
llvm-svn: 333972
We already output true and false in the printer, but the parser isn't able to
read it.
Differential Revision: https://reviews.llvm.org/D47424
llvm-svn: 333970
As noted in rL333782, we can be both better for optimization and
safer with this transform:
BinOp (shuffle V1, Mask), C --> shuffle (BinOp V1, NewC), Mask
The only potentially unsafe-to-speculate binops are integer div/rem.
All other binops are always safe (although I don't see a way to
assert that in code here).
For opcodes like shifts that can produce poison, it can't matter
here because we know the lanes with undef are dropped by the
subsequent shuffle.
Differential Revision: https://reviews.llvm.org/D47686
llvm-svn: 333962
Krzysztof Parzyszek