Although this combine checks that there's no load folding barriers between
the loads that it's trying to merge, it was inserting the load at the
MIRBuilder's default insertion point, which is the G_OR use inst.
This was causing a miscompile in the test suite's
SingleSource/Regression/C/gcc-c-torture/execute/GCC-C-execute-bswap-2
Differential Revision: https://reviews.llvm.org/D106251
llvm::KnownBits::byteSwap() and reverse() don't modify in-place, so
we weren't actually computing anything. This was causing a miscompile on an
arm64 stage2 bootstrap clang build.
s56 stores are broken down into s32 + s24 stores. During this step
both of those new stores use an anyextended s64 value, resulting in
truncating stores. With s56, the s24 requires another lower step to
make it legal, and we were crashing because we didn't expect non-pow-2
stores to also be truncating as well.
Differential Revision: https://reviews.llvm.org/D106183
Since we're still building on top of the MVT based infrastructure, we
need to track the pointer type/address space on the side so we can end
up with the correct pointer LLTs when interpreting CCValAssigns.
This adds some level of type safety, allows helper functions to be added for
specific opcodes for free, and also allows us to succinctly check for class
membership with the usual dyn_cast/isa/cast functions.
To start off with, add variants for the different load/store operations with some
places using it.
Differential Revision: https://reviews.llvm.org/D105751
This fixes not respecting signext/zeroext in these cases. In the
anyext case, this avoids a larger merge with undef and should be a
better canonical form.
This should also handle this if a merge is needed, but I'm not aware
of a case where that can happen. In a future change this will also
allow AMDGPU to drop some custom code without introducing regressions.
Generalize the existing eq/ne case using `extractParts`. The original code only
handled narrowings for types of width 2n->n. This generalization allows for any
type that can be broken down by `extractParts`.
General overview is:
- Loop over each narrow-sized part and do exactly what the 2-register case did.
- Loop over the leftover-sized parts and do the same thing
- Widen the leftover-sized XOR results to the desired narrow size
- OR that all together and then do the comparison against 0 (just like the old
code)
This shows up a lot when building clang for AArch64 using GlobalISel, so it's
worth fixing. For the sake of simplicity, this doesn't handle the non-eq/ne
case yet.
Also remove the code in this case that notifies the observer; we're just going
to delete MI anyway so talking to the observer shouldn't be necessary.
Differential Revision: https://reviews.llvm.org/D105161
The original motivation for this was to implement moreElementsVector of shuffles
on AArch64, which resulted in complex sequences of artifacts like unmerge(unmerge(concat...))
which the combiner couldn't handle. It seemed here that the better option,
instead of writing ever-more-complex combines, was to have a way to find
the original "non-artifact" source registers for a given definition, walking
through arbitrary expressions of unmerge/concat/insert. As long as the bits
aren't extended or truncated, this is a pretty simple algorithm that avoids
the need for lots of combines and instead jumps straight to the final result
we want.
I've only used this new technique in 2 places within tryCombineUnmerge, using it
in more general situations resulted in infinite loops in AMDGPU. So for now
it's used when we would otherwise fail to combine and that seems to work.
In order to support looking through G_INSERTs, I also had to add it as an
artifact in isArtifact(), which caused a whole lot of issues in tests. AMDGPU
started infinite looping since full legalization of G_INSERT doensn't seem to
be there. To work around this, I've temporarily added a CLI option to use the
old behaviour so that the MIR tests will still run and terminate.
Other minor changes include no longer making >128b G_MERGE/UNMERGE legal.
We never had isel support for that anyway and it was a remnant of the legacy
legalizer rules. However being legal prevented the combiner from checking if it
was dead and deleting them.
Differential Revision: https://reviews.llvm.org/D104355
`LegalizerHelper::insertParts` uses `extractGCDType` on registers split into
a desired type and a smaller leftover type. This is used to populate a list
of registers. Each register in the list will have the same type as returned by
`extractGCDType`.
If we have
- `ResultTy` = s792
- `PartTy` = s64
- `LeftoverTy` = s24
When we call `extractGCDType`, we'll end up with two different types appended
to the list:
Part: gcd(792, 64, 24) => s8
Leftover: gcd(792, 24, 24) => s24
When this happens, we'll hit an assert while trying to build a G_MERGE_VALUES.
This patch changes the code for the leftover type so that we reuse the GCD from
the desired type.
e.g.
Leftover: gcd(792, 8, 24) => s8
https://llvm.godbolt.org/z/137Kqxj6j
Differential Revision: https://reviews.llvm.org/D105674
SelectionDAG's equivalents in ISD::InputArg/OutputArg track the
original argument index. Mips relies on this, and its currently
reinventing its own parallel CallLowering infrastructure which tracks
these indexes on the side. Add this to help move towards deleting the
custom mips handling.
This patch prevents GlobalISel from optimizing out redundant branch
instructions when compiling without optimizations.
The motivating example is code like the following common pattern in
Swift, where users expect to be able to set a breakpoint on the early
exit:
public func f(b: Bool) {
guard b else {
return // I would like to set a breakpoint here.
}
...
}
The patch modifies two places in GlobalISEL: The first one is in
IRTranslator.cpp where the removal of redundant branches is made
conditional on the optimization level. The second one is in
AArch64InstructionSelector.cpp where an -O0 *only* optimization is
being removed.
Disabling these optimizations increases code size at -O0 by
~8%. However, doing so improves debuggability, and debug builds are
the primary reason why developers compile without optimizations. We
thus concluded that this is the right trade-off.
rdar://79515454
Differential Revision: https://reviews.llvm.org/D105238
We're trying to match a few pointer computation patterns here for
re-association opportunities.
1) Isolating a constant operand to be on the RHS, e.g.:
G_PTR_ADD(BASE, G_ADD(X, C)) -> G_PTR_ADD(G_PTR_ADD(BASE, X), C)
2) Folding two constants in each sub-tree as long as such folding
doesn't break a legal addressing mode.
G_PTR_ADD(G_PTR_ADD(BASE, C1), C2) -> G_PTR_ADD(BASE, C1+C2)
AArch64 code size improvements on CTMark with -Os:
Program before after diff
pairlocalalign 251048 251044 -0.0%
consumer-typeset 421820 421812 -0.0%
kc 431348 431320 -0.0%
SPASS 413404 413300 -0.0%
clamscan 384396 384220 -0.0%
tramp3d-v4 370640 370412 -0.1%
lencod 432096 431772 -0.1%
bullet 479400 478796 -0.1%
sqlite3 288504 288072 -0.1%
7zip-benchmark 573796 570768 -0.5%
Geomean difference -0.1%
Differential Revision: https://reviews.llvm.org/D105069
In `IRTranslator::translateGetElementPtr`, when we run into a vector gep with
some scalar operands, we try to normalize those operands using
`buildSplatVector`.
This is fine except for when the getelementptr has a <1 x N> type. In that case
it is treated as a scalar. If we run into one of these then every call to
```
// With VectorWidth = 1
LLT::fixed_vector(VectorWidth, PtrTy)
```
will assert.
Here's an example (equivalent to the added testcase):
https://godbolt.org/z/hGsTnMYdW
To get around this, this patch adds a variable, `WantSplatVector`, which
is true when our vector type ought to actually be represented using a vector.
When it's false, we'll translate as a scalar. This checks if `VectorWidth > 1`.
This fixes this bug:
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=35496
Differential Revision: https://reviews.llvm.org/D105316
This enables proper lowering of non-byte sized loads. We still aren't
faithfully preserving memory types everywhere, so the legality checks
still only consider the size.
Previously we didn't preserve the memory type and had to blindly
interpret a number of bytes. Now that non-byte memory accesses are
representable, we can handle these correctly.
Ported from DAG version (minus some weird special case i1 legality
checking which I don't fully understand, and we don't have a way to
query for)
For now, this is NFC and the test changes are placeholders. Since the
legality queries are still relying on byte-flattened memory sizes, the
legalizer can't actually see these non-byte accesses. This keeps this
change self contained without merging it with the larger patch to
switch to LLT memory queries.
GlobalISel is relying on regular MachineMemOperands to track all of
the memory properties of accesses. Just the raw byte size is
insufficent to disambiguate all situations. For example, if we need to
split an unaligned extending load, we need to know the number of bits
in the original source value and can't infer it from the result
type. This is also a problem for extending vector loads.
This does decrease the maximum representable size from the full
uint64_t bytes to a maximum of 16-bits. No in tree testcases hit this,
other than places using UINT64_MAX for unknown sizes. This may be an
issue for G_MEMCPY and co., although they can just use unknown size
for large static sizes. This also has potential for backend abuse by
relying on the type when it really shouldn't be relevant after
selection.
This does not include the necessary MIR printer/parser changes to
represent this.
This patch relands https://reviews.llvm.org/D104454, but fixes some failing
builds on Mac OS which apparently has a different definition for size_t,
that caused 'ambiguous operator overload' for the implicit conversion
of TypeSize to a scalar value.
This reverts commit b732e6c9a8.
Adds legalizer, register bank select, and instruction
select support for G_SBFX and G_UBFX. These opcodes generate
scalar or vector ALU bitfield extract instructions for
AMDGPU. The instructions allow both constant or register
values for the offset and width operands.
The 32-bit scalar version is expanded to a sequence that
combines the offset and width into a single register.
There are no 64-bit vgpr bitfield extract instructions, so the
operations are expanded to a sequence of instructions that
implement the operation. If the width is a constant,
then the 32-bit bitfield extract instructions are used.
Moved the AArch64 specific code for creating G_SBFX to
CombinerHelper.cpp so that it can be used by other targets.
Only bitfield extracts with constant offset and width values
are handled currently.
Differential Revision: https://reviews.llvm.org/D100149
To reflect that the size may be scalable, a TypeSize is returned
instead of an unsigned. In places where the result is used,
it currently relies on an implicit cast of TypeSize -> uint64_t,
which asserts that the type is not scalable.
This patch is NFC for fixed-width vectors.
Reviewed By: aemerson
Differential Revision: https://reviews.llvm.org/D104454
This also adds new interfaces for the fixed- and scalable case:
* LLT::fixed_vector
* LLT::scalable_vector
The strategy for migrating to the new interfaces was as follows:
* If the new LLT is a (modified) clone of another LLT, taking the
same number of elements, then use LLT::vector(OtherTy.getElementCount())
or if the number of elements is halfed/doubled, it uses .divideCoefficientBy(2)
or operator*. That is because there is no reason to specifically restrict
the types to 'fixed_vector'.
* If the algorithm works on the number of elements (as unsigned), then
just use fixed_vector. This will need to be fixed up in the future when
modifying the algorithm to also work for scalable vectors, and will need
then need additional tests to confirm the behaviour works the same for
scalable vectors.
* If the test used the '/*Scalable=*/true` flag of LLT::vector, then
this is replaced by LLT::scalable_vector.
Reviewed By: aemerson
Differential Revision: https://reviews.llvm.org/D104451
Since this method can apply to cmpxchg operations, make sure it's clear
what value we're actually retrieving. This will help ensure we don't
accidentally ignore the failure ordering of cmpxchg in the future.
We could potentially introduce a getOrdering() method on AtomicSDNode
that asserts the operation isn't cmpxchg, but not sure that's
worthwhile.
Differential Revision: https://reviews.llvm.org/D103338
This only applies to FastIsel. GlobalIsel seems to sidestep
the issue.
This fixes https://bugs.llvm.org/show_bug.cgi?id=46996
One of the things we do in llvm is decide if a type needs
consecutive registers. Previously, we just checked if it
was an array or not.
(plus an SVE specific check that is not changing here)
This causes some confusion when you arbitrary IR like:
```
%T1 = type { double, i1 };
define [ 1 x %T1 ] @foo() {
entry:
ret [ 1 x %T1 ] zeroinitializer
}
```
We see it is an array so we call CC_AArch64_Custom_Block
which bails out when it sees the i1, a type we don't want
to put into a block.
This leaves the location of the double in some kind of
intermediate state and leads to odd codegen. Which then crashes
the backend because it doesn't know how to implement
what it's been asked for.
You get this:
```
renamable $d0 = FMOVD0
$w0 = COPY killed renamable $d0
```
Rather than this:
```
$d0 = FMOVD0
$w0 = COPY $wzr
```
The backend knows how to copy 64 bit to 64 bit registers,
but not 64 to 32. It can certainly be taught how but the real
issue seems to be us even trying to assign a register block
in the first place.
This change makes the logic of
AArch64TargetLowering::functionArgumentNeedsConsecutiveRegisters
a bit more in depth. If we find an array, also check that all the
nested aggregates in that array have a single member type.
Then CC_AArch64_Custom_Block's assumption of a type that looks
like [ N x type ] will be valid and we get the expected codegen.
New tests have been added to exercise these situations. Note that
some of the output is not ABI compliant. The aim of this change is
to simply handle these situations and not to make our processing
of arbitrary IR ABI compliant.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D104123
<string> is currently the highest impact header in a clang+llvm build:
https://commondatastorage.googleapis.com/chromium-browser-clang/llvm-include-analysis.html
One of the most common places this is being included is the APInt.h header, which needs it for an old toString() implementation that returns std::string - an inefficient method compared to the SmallString versions that it actually wraps.
This patch replaces these APInt/APSInt methods with a pair of llvm::toString() helpers inside StringExtras.h, adjusts users accordingly and removes the <string> from APInt.h - I was hoping that more of these users could be converted to use the SmallString methods, but it appears that most end up creating a std::string anyhow. I avoided trying to use the raw_ostream << operators as well as I didn't want to lose having the integer radix explicit in the code.
Differential Revision: https://reviews.llvm.org/D103888
G_INSERT legalization is incomplete and doesn't work very
well. Instead try to use sequences of G_MERGE_VALUES/G_UNMERGE_VALUES
padding with undef values (although this can get pretty large).
For the case of load/store narrowing, this is still performing the
load/stores in irregularly sized pieces. It might be cleaner to split
this down into equal sized pieces, and rely on load/store merging to
optimize it.
