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
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
This should be a typo introduced in D69275, which may cause an unknown
segment fault in getNode.
Reviewed By: uweigand
Differential Revision: https://reviews.llvm.org/D83376
9cac4e6d1403554b06ec2fc9d834087b1234b695/D32628 intended to eliminate
this, and move all isel pseudo expansion to FinalizeISel. This was a
bad rebase or something, and failed to actually delete this call.
GlobalISel also has a redundant call of finalizeLowering. However, it
requires more work to remove it since it currently triggers a lot of
verifier errors in tests.
It looks like 9cac4e6d14 accidentally
added a second copy of this from a bad rebase or something. This
second copy was added, and the finalizeLowering call was not deleted
as intended.
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.
This removes existing code duplication and allows us to
assert that we are handling the expected cases.
We have a list of outstanding bugs that could benefit by
handling truncated source values, so that's a possible
addition going forward.
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
Occasionally we see absolutely massive basic blocks, typically in global
constructors that are vulnerable to heavy inlining. When these blocks are
dense with DBG_VALUE instructions, we can hit near quadratic complexity in
DwarfDebug's validThroughout function. The problem is caused by:
* validThroughout having to step through all instructions in the block to
examine their lexical scope,
* and a high proportion of instructions in that block being DBG_VALUEs
for a unique variable fragment,
Leading to us stepping through every instruction in the block, for (nearly)
each instruction in the block.
By adding this guard, we force variables in large blocks to use a location
list rather than a single-location expression, as shown in the added test.
This shouldn't change the meaning of the output DWARF at all: instead we
use a less efficient DWARF encoding to avoid a poor-performance code path.
Differential Revision: https://reviews.llvm.org/D83236
In DAGTypeLegalizer::SplitVecRes_ExtendOp I have replaced an invalid
call to getVectorNumElements() with a call to getVectorMinNumElements(),
since the code path works for both fixed and scalable vectors.
This fixes up a warning in the following test:
sve-sext-zext.ll
Differential Revision: https://reviews.llvm.org/D83197
Calling getVectorNumElements() is not safe for scalable vectors and we
should normally use getVectorElementCount() instead. However, for the
code changed in this patch I decided to simply move the instantiation of
the variable 'OutNumElems' lower down to the place where only fixed-width
vectors are used, and hence it is safe to call getVectorNumElements().
Fixes up one warning in this test:
sve-sext-zext.ll
Differential Revision: https://reviews.llvm.org/D83195
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
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.
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.
handleMoveDown or handleMoveUp cannot properly repair a main
range of a LiveInterval since they only get LiveRange. There
is a problem if certain use has moved few segments away and
there is a hole in the main range in between of these two
locations. We may get a SubRange with a very extended Segment
spanning several Segments of the main range and also spanning
that hole. If that happens then we end up with the main range
not covering its SubRange which is an error.
It might be possible to attempt fixing the main range in place
just between of the old and new index by extending all of its
Segments in between, but it is unclear this logic will be
faster than just straight constructMainRangeFromSubranges,
which itself is pretty cheap since it only contains interval
logic. That will also require shrinkToUses() call after which
is probably even more expensive.
In the test second move is from 64B to 92B for the sub1.
Subrange is correctly fixed:
L000000000000000C [16r,32B:0)[32B,92r:1) 0@16r 1@32B-phi
But the main range has a hole in between 80d and 88r after
updateRange():
%1 [16r,32B:0)[32B,80r:4)[80r,80d:3)[88r,96r:1)[96r,160B:2)
Since source position is 64B this segment is not even considered
by the updateRange().
Differential Revision: https://reviews.llvm.org/D82916
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
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
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
Summary:
Avoid exposing details about how children are stored. This will enable
subsequent type-erasure changes.
New methods are introduced to cover common access patterns.
Change-Id: Idb5f4b1b9c84e4cc71ddb39bb52a388682f5674f
Reviewers: arsenm, RKSimon, mehdi_amini, courbet
Subscribers: qcolombet, sdardis, wdng, hiraditya, jrtc27, zzheng, atanasyan, asbirlea, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83083
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
This matches the DAG behavior where this is called after the loop
checking for calls. The AMDGPU implementation depends on knowing if
there are calls in the function or not, so move this later.
Another problem is finalizeLowering is actually called twice; I was
seeing weird inconsistencies since the first call would produce
unexpected results and the second run would correct them in some
contexts. Since this requires disabling the verifier, and it's useful
to serialize the MIR immediately after selection, FinalizeISel should
probably not be a real pass.
Use a simpler code sequence when the shift amount is known not to be
zero modulo the bit width.
Nothing much uses this until D77152 changes the translation of fshl and
fshr intrinsics.
Differential Revision: https://reviews.llvm.org/D82540
Using a negation instead of a subtraction from a constant can save an
instruction on some targets.
Nothing much uses this until D77152 changes the translation of fshl and
fshr intrinsics.
Differential Revision: https://reviews.llvm.org/D82539
We need to ensure that the sign bits of the result all match
so we can't fold to undef.
Similar to PR46585.
Reviewed By: lebedev.ri
Differential Revision: https://reviews.llvm.org/D83163
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.
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
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
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
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
Whilst trying to assemble the following test:
clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_set2.c
I discovered we were hitting some warnings about possible invalid
calls to getVectorNumElements() in getCopyToPartsVector(). I've
tried to fix these by using ElementCount types where possible and
I've made the assumption that we don't support using a fixed width
vector to copy parts of a scalable vector, and vice versa. Looking
at how the copy is implemented I think that's the right thing for
now.
Differential Revision: https://reviews.llvm.org/D82744
This patch uses ranges for debug information when a function contains basic block sections rather than using [lowpc, highpc]. This is also the first in a series of patches for debug info and does not contain the support for linker relaxation. That will be done as a follow up patch.
Differential Revision: https://reviews.llvm.org/D78851
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.
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
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
As per documentation of `hasPairLoad`:
"`RequiredAlignment` gives the minimal alignment constraints that must be met to be able to select this paired load."
In this sense, `0` is strictly equivalent to `1`. We make this obvious by using `Align` instead of unsigned.
There is only one implementor of this interface.
Differential Revision: https://reviews.llvm.org/D82958
It's perfectly valid to do certain DAG combines where we extract
subvectors from a concat vector when we have scalable vector types.
However, we can do this in a way that avoids generating compiler
warnings by replacing calls to getVectorNumElements() with
getVectorMinNumElements(). Due to the way subvector extracts are
designed to work with scalable vector types this is ok.
This eliminates some warnings from existing tests in this file:
llvm/test/CodeGen/AArch64/sve-intrinsics-loads.ll
Differential Revision: https://reviews.llvm.org/D82655
Summary:
This is a fix for PR45009.
When working on D67492 I made DwarfExpression emit a single
DW_OP_entry_value operation covering the whole composite location
description that is produced if a register does not have a valid DWARF
number, and is instead composed of multiple register pieces. Looking
closer at the standard, this appears to not be valid DWARF. A
DW_OP_entry_value operation's block can only be a DWARF expression or a
register location description, so it appears to not be valid for it to
hold a composite location description like that.
See DWARFv5 sec. 2.5.1.7:
"The DW_OP_entry_value operation pushes the value that the described
location held upon entering the current subprogram. It has two
operands: an unsigned LEB128 length, followed by a block containing a
DWARF expression or a register location description (see Section
2.6.1.1.3 on page 39)."
Here is a dwarf-discuss mail thread regarding this:
http://lists.dwarfstd.org/pipermail/dwarf-discuss-dwarfstd.org/2020-March/004610.html
There was not a strong consensus reached there, but people seem to lean
towards that operations specified under 2.6 (e.g. DW_OP_piece) may not
be part of a DWARF expression, and thus the DW_OP_entry_value operation
can't contain those.
Perhaps we instead want to emit a entry value operation per each
DW_OP_reg* operation, e.g.:
- DW_OP_entry_value(DW_OP_regx sub_reg0),
DW_OP_stack_value,
DW_OP_piece 8,
- DW_OP_entry_value(DW_OP_regx sub_reg1),
DW_OP_stack_value,
DW_OP_piece 8,
[...]
The question then becomes how the call site should look; should a
composite location description be emitted there, and we then leave it up
to the debugger to match those two composite location descriptions?
Another alternative could be to emit a call site parameter entry for
each sub-register, but firstly I'm unsure if that is even valid DWARF,
and secondly it seems like that would complicate the collection of call
site values quite a bit. As far as I can tell GCC does not emit any
entry values / call sites in these cases, so we do not have something to
compare with, but the former seems like the more reasonable approach.
Currently when trying to emit a call site entry for a parameter composed
of multiple DWARF registers a (DwarfRegs.size() == 1) assert is
triggered in addMachineRegExpression(). Until the call site
representation is figured out, and until there is use for these entry
values in practice, this commit simply stops the invalid DWARF from
being emitted.
Reviewers: djtodoro, vsk, aprantl
Reviewed By: djtodoro, vsk
Subscribers: jyknight, hiraditya, fedor.sergeev, jrtc27, llvm-commits
Tags: #debug-info, #llvm
Differential Revision: https://reviews.llvm.org/D75270
While validating live-out values, record instructions that look like
a reduction. This will comprise of a vector op (for now only vadd),
a vorr (vmov) which store the previous value of vadd and then a vpsel
in the exit block which is predicated upon a vctp. This vctp will
combine the last two iterations using the vmov and vadd into a vector
which can then be consumed by a vaddv.
Once we have determined that it's safe to perform tail-predication,
we need to change this sequence of instructions so that the
predication doesn't produce incorrect code. This involves changing
the register allocation of the vadd so it updates itself and the
predication on the final iteration will not update the falsely
predicated lanes. This mimics what the vmov, vctp and vpsel do and
so we then don't need any of those instructions.
Differential Revision: https://reviews.llvm.org/D75533
It's pretty silly to diagnose on a scalar copy but the build does that:
loop variable 'SibReg' of type 'const llvm::Register' creates a copy from type 'const llvm::Register' [-Wrange-loop-analysis]
With an undef operand, it's possible for getVRegDef to fail and return
null. This is an edge case very little code bothered to
consider. Proper gMIR should use G_IMPLICIT_DEF instead.
I initially tried to apply this restriction to all SSA MIR, so then
getVRegDef would never fail anywhere. However, ProcessImplicitDefs
does technically run while the function is in SSA. ProcessImplicitDefs
and DetectDeadLanes would need to either move, or a new pseudo-SSA
type of function property would need to be introduced.
Basically a NFC, but allows subclasses access to the entire PeelingModuloScheduleExpander
class. We are doing this to allow backends, particularly one that are not necessarily
upstreamed, to inherit from PeelingModuloScheduleExpander and access its basic structures.
Renames Info into LoopInfo for consistency in PeelingModuloScheduleExpander.
Differential Revision: https://reviews.llvm.org/D82673
In RISC-V vector extension, users could group multiple vector registers
as one pseudo register. In mixed width operations, users could use
partial vector registers to reduce the register pressure. The parameter
to control register grouping and partial use is called LMUL. LMUL is a
part of the type. So, we have a bunch of vector types. In order to
support all these types, we need new MVT types in LLVM. In this patch, I
added several MVT types that are used in RISC-V vector implementation.
This is a standalone patch for MVT types without RISC-V related implementation.
Differential revision: https://reviews.llvm.org/D81724
This is a followup on D78403.
I'm unsure about `getAtomicOpAlign` overloads that take `AtomicRMWInst` and `AtomicCmpXchgInst`, shouldn't `getAlign` provide the correct answer already?
Differential Revision: https://reviews.llvm.org/D81369
Fix a warning in getNode() when extracting a subvector from a
concat vector. We can simply replace the call to getVectorNumElements
with getVectorMinNumElements as this follows the defined behaviour
for EXTRACT_SUBVECTOR.
Differential Revision: https://reviews.llvm.org/D82746
When trying to reduce a BUILD_VECTOR to a SHUFFLE_VECTOR it's
important that we carefully check the vector types that led to
that BUILD_VECTOR. In the test I have attached to this commit
there is a case where the results of two SVE faddv instructions
are being stored to consecutive memory locations. With my fix,
as part of merging those stores we discover that each BUILD_VECTOR
element came from an extract of a SVE vector element and
therefore bail out.
Differential Revision: https://reviews.llvm.org/D82564
If a constant is only allsignbits in the demanded/active bits, then sign extend it to an allsignbits bool pattern for OR/XOR ops.
This also requires SimplifyDemandedBits XOR handling to be modified to call ShrinkDemandedConstant on any (non-NOT) XOR pattern to account for non-splat cases.
Next step towards fixing PR45808 - with this patch we now get a <-1,-1,0,0> v4i64 constant instead of <1,1,0,0>.
Differential Revision: https://reviews.llvm.org/D82257
Pre-commit for D82257, this adds a DemandedElts arg to ShrinkDemandedConstant/targetShrinkDemandedConstant which will allow future patches to (optionally) add vector support.
reduceBuildVecExtToExtBuildVec was breaking a splat(zext(x)) pattern into buildvector(x, 0, x, 0, ..) resulting in much more complex insert+shuffle codegen.
We already go to some lengths to avoid this in SimplifyDemandedVectorElts etc. when we encounter splat buildvectors.
It should be OK to fold all splat(aext(x)) patterns - we might need to tighten this if we find a case where we mustn't introduce a buildvector(x, undef, x, undef, ..) but I can't find one.
Fixes PR46461.
The translation of cmpxchg added by
9481399c0f specifically skipped weak
cmpxchg due to not understanding the meaning. Weak cmpxchg was added
in 420a216817. As explained in the
commit message, the weak mode is implicit in how
ATOMIC_CMP_SWAP_WITH_SUCCESS is lowered. If it's expanded to a regular
ATOMIC_CMP_SWAP, it's replaced with a strong cmpxchg.
This handling seems weird to me, but this was already following the
DAG behavior. I would expect the strong IR instruction to not have the
boolean output. Failing that, I might expect the IRTranslator to emit
ATOMIC_CMP_SWAP and a constant for the boolean.
This lowers intrinsic @llvm.get.active.lane.mask to a setcc node, i.e. an icmp
ule, and creates vectors for its 2 arguments on which the comparison is
performed.
Differential Revision: https://reviews.llvm.org/D82292
Summary:
The printer seems to intend to not print the trailing comma but has a
copy-paste error for the last value in the escape, and the parser
enforces having no trailing comma, but somehow a test was never included
to actually confirm it.
Reviewers: thegameg, arsenm
Reviewed By: thegameg, arsenm
Subscribers: wdng, arsenm, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82478
This function is deceptive at best: it doesn't return what you'd expect.
If you have an arbitrary GlobalValue and you want to determine the
alignment of that pointer, Value::getPointerAlignment() returns the
correct value. If you want the actual declared alignment of a function
or variable, GlobalObject::getAlignment() returns that.
This patch switches all the users of GlobalValue::getAlignment to an
appropriate alternative.
Differential Revision: https://reviews.llvm.org/D80368
Implement them on top of sdiv/udiv, similar to what we do for integer
types.
Potential future work: implementing i8/i16 srem/urem, optimizations for
constant divisors, optimizing the mul+sub to mls.
Differential Revision: https://reviews.llvm.org/D81511
Summary:
This patch adds base support for code generating fixed length
vector operations targeting a known SVE vector length. To achieve
this we lower fixed length vector operations to equivalent scalable
vector operations, whereby SVE predication is used to limit the
elements processed to those present within the fixed length vector.
Specifically this patch implements load and store operations, which
get lowered to their masked counterparts thusly:
V = load(Addr) =>
V = extract_fixed_vector(masked_load(make_pred(V.NumElts), Addr))
store(V, (Addr)) =>
masked_store(insert_fixed_vector(V), make_pred(V.NumElts), Addr))
Reviewers: rengolin, efriedma
Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80385
Summary:
- AssertAlign node records the guaranteed alignment on its source node,
where these alignments are retrieved from alignment attributes in LLVM
IR. These tracked alignments could help DAG combining and lowering
generating efficient code.
- In this patch, the basic support of AssertAlign node is added. So far,
we only generate AssertAlign nodes on return values from intrinsic
calls.
- Addressing selection in AMDGPU is revised accordingly to capture the
new (base + offset) patterns.
Reviewers: arsenm, bogner
Subscribers: jvesely, wdng, nhaehnle, tpr, hiraditya, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81711
Following on from this RFC[0] from a while back, this is the first patch towards
implementing variadic debug values.
This patch specifically adds a set of functions to MachineInstr for performing
operations specific to debug values, and replacing uses of the more general
functions where appropriate. The most prevalent of these is replacing
getOperand(0) with getDebugOperand(0) for debug-value-specific code, as the
operands corresponding to values will no longer be at index 0, but index 2 and
upwards: getDebugOperand(x) == getOperand(x+2). Similar replacements have been
added for the other operands, along with some helper functions to replace
oft-repeated code and operate on a variable number of value operands.
[0] http://lists.llvm.org/pipermail/llvm-dev/2020-February/139376.html<Paste>
Differential Revision: https://reviews.llvm.org/D81852
We have many cases where we call SimplifyMultipleUseDemandedBits and demand specific vector elements, but all the bits from them - this adds a helper wrapper to handle this.
For little endian targets, if we only need the lowest element and none of the extended bits then we can just use the (bitcasted) source vector directly.
We already do this in SimplifyDemandedBits, this adds the SimplifyMultipleUseDemandedBits equivalent.
If a collection of interconnected phi nodes is only ever loaded, stored
or bitcast then we can convert the whole set to the bitcast type,
potentially helping to reduce the number of register moves needed as the
phi's are passed across basic block boundaries. This has to be done in
CodegenPrepare as it naturally straddles basic blocks.
The alorithm just looks from phi nodes, looking at uses and operands for
a collection of nodes that all together are bitcast between float and
integer types. We record visited phi nodes to not have to process them
more than once. The whole subgraph is then replaced with a new type.
Loads and Stores are bitcast to the correct type, which should then be
folded into the load/store, changing it's type.
This comes up in the biquad testcase due to the way MVE needs to keep
values in integer registers. I have also seen it come up from aarch64
partner example code, where a complicated set of sroa/inlining produced
integer phis, where float would have been a better choice.
I also added undef and extract element handling which increased the
potency in some cases.
This adds it with an option that defaults to off, and disabled for 32bit
X86 due to potential issues around canonicalizing NaNs.
Differential Revision: https://reviews.llvm.org/D81827
At the moment we use Global ISel by default at -O0, however it is
currently not capable of dealing with scalable vectors for two
reasons:
1. The register banks know nothing about SVE registers.
2. The LLT (Low Level Type) class knows nothing about scalable
vectors.
For now, the easiest way to avoid users hitting issues when using
the SVE ACLE is to fall back on normal DAG ISel when encountering
instructions that operate on scalable vector types.
I've added a couple of RUN lines to existing SVE tests to ensure
we can compile at -O0. I've also added some new tests to
CodeGen/AArch64/GlobalISel/arm64-fallback.ll
that demonstrate we correctly fallback to DAG ISel at -O0 when
lowering formal arguments or translating instructions that involve
scalable vector types.
Differential Revision: https://reviews.llvm.org/D81557
Without this fix, handleMoveUp can create an invalid live range like
this:
[98904e,98908r:0)[98908e,227504r:1)
where the two segments overlap, but only because we have lost the "e"
(early-clobber) on the end point of the first segment.
Differential Revision: https://reviews.llvm.org/D82110
For now I have changed SimplifyDemandedBits and it's various callers
to assume we know nothing for scalable vectors and to ignore the
demanded bits completely. I have also done something similar for
SimplifyDemandedVectorElts. These changes fix up lots of warnings
due to calls to EVT::getVectorNumElements() for types with scalable
vectors. These functions are all used for optimisations, rather than
functional requirements. In future we can revisit this code if
there is a need to improve code quality for SVE.
Differential Revision: https://reviews.llvm.org/D80537
When trying to calculate the number of sign bits for scalable vectors
we should just bail out for now and pretend we know nothing.
Differential Revision: https://reviews.llvm.org/D81093
Summary:
Extend StackLifetime with option to calculate liveliness
where alloca is only considered alive on basic block entry
if all non-dead predecessors had it alive at terminators.
Depends on D82043.
Reviewers: eugenis
Reviewed By: eugenis
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D82124
This was passing in all the parameters needed to construct a
LegalizerHelper in the custom legalization, when it's simpler to just
pass in the existing helper.
This is slightly more annoying to use in the common case where you
don't need the legalizer helper, but we could add back the common
parameters back in addition to the helper.
I didn't propagate this to all the internal target changes that this
logically implies, but did update a sample one for
legalizeMinNumMaxNum.
This is in preparation for moving AMDGPU load/store legalization
entirely into custom lowering. The current set of legalization actions
is really constraining and not really capable of expressing all the
actions needed to legalize loads/stores. In particular there's no way
to express when the memory access itself needs to change size vs. the
result type. There's also a lot of redundancy since the same
split/widen actions need to be applied in both vector and scalar
cases. All of the sub-cases logically belong as steps in the legalizer
helper, but it will be easier to consider everything at once in custom
lowering.
This patch adds some missing information to the LF_BUILDINFO which allows for rebuilding an .OBJ without any external dependency but the .OBJ itself (other than the compiler executable).
Some tools need this information to reproduce a build without any knowledge of the build system. The LF_BUILDINFO therefore stores a full path to the compiler, the PWD (which is the CWD at program startup), a relative or absolute path to the TU, and the full CC1 command line. The command line needs to be freestanding (not depend on any environment variable). In the same way, MSVC doesn't store the provided command-line, but an expanded version (somehow their equivalent of CC1) which is also freestanding.
For more information see PR36198 and D43002.
Differential Revision: https://reviews.llvm.org/D80833
Summary:
Half-precision floating point arguments and returns are currently
promoted to either float or int32 in clang's CodeGen and there's
no existing support for the lowering of `half` arguments and returns
from IR in AArch32's backend.
Such frontend coercions, implemented as coercion through memory
in clang, can cause a series of issues in argument lowering, as causing
arguments to be stored on the wrong bits on big-endian architectures
and incurring in missing overflow detections in the return of certain
functions.
This patch introduces the handling of half-precision arguments and returns in
the backend using the actual "half" type on the IR. Using the "half"
type the backend is able to properly enforce the AAPCS' directions for
those arguments, making sure they are stored on the proper bits of the
registers and performing the necessary floating point convertions.
Reviewers: rjmccall, olista01, asl, efriedma, ostannard, SjoerdMeijer
Reviewed By: ostannard
Subscribers: stuij, hiraditya, dmgreen, llvm-commits, chill, dnsampaio, danielkiss, kristof.beyls, cfe-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D75169
We're missing a plain English explanation of how this pass is supposed
to operate -- add one to the file comment.
Differential Revision: https://reviews.llvm.org/D80929
Added NextPowerOf2() routine to TypeSize and rewritten the code
in getVectorTypeBreakdown to avoid warnings being generated.
Differential Revision: https://reviews.llvm.org/D81578
Instead of asserting the number of elements is the same, we should be
comparing the element counts instead. In addition, when looking at
concats of extract_subvectors it's fine to use getVectorMinNumElements()
for scalable vectors.
I discovered these warnings when compiling the structured loads tests in
this file:
test/CodeGen/AArch64/sve-intrinsics-loads.ll
Differential Revision: https://reviews.llvm.org/D81936
Summary:
This invariant is being violated in the test case
https://reviews.llvm.org/D77849, related to the use of the relatively
new ability for callbr to have return values, and MachineBasicBlocks
with INLINEASM_BR terminators to emit live out register defs.
As noted in the comment, this triggers invariant violations in
MachineVerifier via `llc -verify-machineinstrs` or
`llc -verify-regalloc`, since only MachineInstrs that are terminators
are allowed to follow the first terminator.
https://reviews.llvm.org/D75098 may rework this very assertion if we're
spilling via a (proposed) TCOPY MachineInstr.
Reviewers: void, efriedma, arsenm
Reviewed By: efriedma
Subscribers: qcolombet, wdng, hiraditya, llvm-commits, srhines
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D78166
When the zext gets promoted, it used to retain the original location,
which pessimizes the debugging experience causing an unexpected
jump in stepping at -Og.
Fixes https://bugs.llvm.org/show_bug.cgi?id=46120 (which also
contains a full C repro).
Differential Revision: https://reviews.llvm.org/D81437
Summary:
Add a flag to omit the xray_fn_idx to cut size overhead and relocations
roughly in half at the cost of reduced performance for single function
patching. Minor additions to compiler-rt support per-function patching
without the index.
Reviewers: dberris, MaskRay, johnislarry
Subscribers: hiraditya, arphaman, cfe-commits, #sanitizers, llvm-commits
Tags: #clang, #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D81995
Summary:
This code is going to be used in StackSafety.
This patch is file move with minimal changes. Identifiers
will be fixed in the followup patch.
Reviewers: eugenis, pcc
Reviewed By: eugenis
Subscribers: mgorny, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81831
It's possible to end up with a zext or something in the way of a G_CONSTANT,
even pre-legalization. This can happen with memsets.
e.g.
https://godbolt.org/z/Bjc8cw
To make sure we can catch these cases, use `getConstantVRegValWithLookThrough`
instead of `mi_match`.
Differential Revision: https://reviews.llvm.org/D81875
The promotion machinery in CGP moves instructions retaining
debug locations. When the transformation is local, this is mostly
correct, but when instructions are moved cross-BBs, this is not
always true and causes jumpiness in line tables. This is the first
of a series of commits. sext(s) and zext(s) need to be treated
similarly.
Differential Revision: https://reviews.llvm.org/D81879
This implements the following combines:
((0-A) + B) -> B-A
(A + (0-B)) -> A-B
Porting over the basic algebraic combines from the DAGCombiner. There are
several combines which fold adds away into subtracts. This is just the simplest
one.
I noticed that add combines are some of the most commonly hit across CTMark,
(via print statements when they fire), so I'm porting over some of the obvious
ones.
This gives some minor code size improvements on CTMark at -O3 on AArch64.
Differential Revision: https://reviews.llvm.org/D77453
Summary:
Teach MachineVerifier to check branches for MBB operands if they are not declared indirect.
Add `isBarrier`, `isIndirectBranch` to `G_BRINDIRECT` and `G_BRJT`.
Without these, `MachineInstr.isConditionalBranch()` was giving a
false-positive for those instructions.
Reviewers: aemerson, qcolombet, dsanders, arsenm
Reviewed By: dsanders
Subscribers: hiraditya, wdng, simoncook, s.egerton, arsenm, rovka, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81587
This patch tries to reassociate two patterns related to FMA to expose
more ILP on PowerPC.
// Pattern 1:
// A = FADD X, Y (Leaf)
// B = FMA A, M21, M22 (Prev)
// C = FMA B, M31, M32 (Root)
// -->
// A = FMA X, M21, M22
// B = FMA Y, M31, M32
// C = FADD A, B
// Pattern 2:
// A = FMA X, M11, M12 (Leaf)
// B = FMA A, M21, M22 (Prev)
// C = FMA B, M31, M32 (Root)
// -->
// A = FMUL M11, M12
// B = FMA X, M21, M22
// D = FMA A, M31, M32
// C = FADD B, D
Reviewed By: jsji
Differential Revision: https://reviews.llvm.org/D80175
Current implementation of division estimation isn't correct for some
cases like 1.0/0.0 (result is nan, not expected inf).
And this change exposes a potential infinite loop: we use
isConstOrConstSplatFP in combineRepeatedFPDivisors to look up if the
divisor is some constant. But it doesn't work after legalized on some
platforms. This patch restricts the method to act before LegalDAG.
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D80542
This decreases the time consumed by the pass [during RawSpeed unity build]
by 25% (0.0586 s -> 0.04388 s).
While that isn't really impressive overall, that wasn't the goal here.
The memory results here are noticeable.
The baseline results are:
```
total runtime: 55.65s.
calls to allocation functions: 19754254 (354960/s)
temporary memory allocations: 4951609 (88974/s)
peak heap memory consumption: 239.13MB
peak RSS (including heaptrack overhead): 463.79MB
total memory leaked: 198.01MB
```
While with this patch the results are:
```
total runtime: 55.37s.
calls to allocation functions: 19068237 (344403/s) # -3.47 %
temporary memory allocations: 4261772 (76974/s) # -13.93 % (!!!)
peak heap memory consumption: 239.13MB
peak RSS (including heaptrack overhead): 463.73MB
total memory leaked: 198.01MB
```
So we get rid of *a lot* of temporary allocations.
Using `SmallSet<8>` makes sense to me because at least here
for x86 BdVer2, the size of that set is *never* more than 3,
over all of llvm test-suite + RawSpeed.
The story might be different on other targets,
not sure if it will ever justify whole DenseSet,
but if it does SmallDenseSet might be a compromise.
SUMMARY:
Since we deal with aix emitLinkage in the PPCAIXAsmPrinter::emitLinkage() in the patch https://reviews.llvm.org/D75866. It do not go to AsmPrinter::emitLinkage() any more, we clean up some aix related code in the AsmPrinter::emitLinkage()
Reviewers: Jason liu
Differential Revision: https://reviews.llvm.org/D81613
Put AND before ADD in LegalizerHelper::lowerFPTRUNC_F64_TO_F16
in order to match algorithm from AMDGPUTargetLowering::LowerFP_TO_FP16.
Differential Revision: https://reviews.llvm.org/D81666
Summary:
Fix crash when using -debug caused by the GlobalISel observer trying to print
an incomplete DBG_VALUE instruction. This was caused by the MachineIRBuilder
using buildInstr, which immediately inserts the instruction causing print,
instead of using BuildMI to first build up the instruction and using
insertInstr when finished.
Add RUN-line to existing debug-insts.ll test with -debug flag set to make sure
no crash is happening.
Also fixed a missing %s in the 2nd RUN-line of the same test.
Reviewers: t.p.northover, aditya_nandakumar, aemerson, dsanders, arsenm
Reviewed By: arsenm
Subscribers: wdng, arsenm, rovka, hiraditya, volkan, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D76934
Until we have a real need for computing known bits for scalable
vectors I have simply changed the code to bail out for now and
pretend we know nothing. I've also fixed up some simple callers of
computeKnownBits too.
Differential Revision: https://reviews.llvm.org/D80437
If the target explicitly requested custom legalization, it should be
required to implement this. Also move default legalizeIntrinsic
implementation into the header so it's next to the related
legalizeCustom.
The memory folding raplaced the old instruction without copying the symbols assigned. Which will resulted in built fail due to the lost symbols.
Reviewed by craig.topper
Differential Revision: https://reviews.llvm.org/D78471
SUMMARY:
in the aix assembly , it do not have .hidden and .protected directive.
in current llvm. if a function or a variable which has visibility attribute, it will generate something like the .hidden or .protected , it can not recognize by aix as.
in aix assembly, the visibility attribute are support in the pseudo-op like
.extern Name [ , Visibility ]
.globl Name [, Visibility ]
.weak Name [, Visibility ]
in this patch, we implement the visibility attribute for the global variable, function or extern function .
for example.
extern __attribute__ ((visibility ("hidden"))) int
bar(int* ip);
__attribute__ ((visibility ("hidden"))) int b = 0;
__attribute__ ((visibility ("hidden"))) int
foo(int* ip){
return (*ip)++;
}
the visibility of .comm linkage do not support , we will have a separate patch for it.
we have the unsupported cases ("default" and "internal") , we will implement them in a a separate patch for it.
Reviewers: Jason Liu ,hubert.reinterpretcast,James Henderson
Differential Revision: https://reviews.llvm.org/D75866
It was annoying enough that every custom lowering needed to set the
insert point, but this was made worse since now these all needed to be
updated to setInstrAndDebugLoc. Consolidate these so every
legalization action has the right insert position by default.
This should fix dropping debug info in every custom AMDGPU
legalization.
The current relationship between LegalizerHelper and MachineIRBuilder
confuses me, because the LegalizerHelper modifies the MachineIRBuilder
which it does not own. Constructing a LegalizerHelper destroys the
insert point, since the constructor calls setMF, which clears all the
fields. Try to separate these functions, so it's possible to construct
a LegalizerHelper from an existing MachineIRBuilder without losing the
insert point/debug loc.
The construction APIs for MachineIRBuilder don't make much sense, and
it's been annoying to sort through it with these trivial functions
separate from the declaration.
New instructions were getting printed both in createdInstr, and in the
final printNewInstrs, so it made it look like the same instructions
were created twice. This overall made reading the debug output
harder. Stop printing the initial construction and only print new
instructions in the summary at the end. This avoids printing the less
useful case where instructions are sometimes initially created with no
operands.
I'm not sure this is the correct instance to remove; now the visible
ordering is different. Now you will typically see the one erased
instruction message before all the new instructions in order. I think
this is the more logical view of typical legalization changes,
although it's mechanically backwards from the normal
insert-new-erase-old pattern.
If a resource can be held for multiple cycles in the schedule model
then an instruction can be placed into the available queue, another
instruction can be scheduled, but the first will not be taken back out if
the two instructions hazard. To fix this make sure that we update the
available queue even on the first MOp of a cycle, pushing available
instructions back into the pending queue if they now conflict.
This happens with some downstream schedules we have around MVE
instruction scheduling where we use ResourceCycles=[2] to show the
instruction executing over two beats. Apparently the test changes here
are OK too.
Differential Revision: https://reviews.llvm.org/D76909
If fmul and fadd are separated by an fma, we can fold them together
to save an instruction:
fadd (fma A, B, (fmul C, D)), N1 --> fma(A, B, fma(C, D, N1))
The fold implemented here is actually a specialization - we should
be able to peek through >1 fma to find this pattern. That's another
patch if we want to try that enhancement though.
This transform was guarded by the TLI hook enableAggressiveFMAFusion(),
so it was done for some in-tree targets like PowerPC, but not AArch64
or x86. The hook is protecting against forming a potentially more
expensive computation when fma takes longer to execute than a single
fadd. That hook may be needed for other transforms, but in this case,
we are replacing fmul+fadd with fma, and the fma should never take
longer than the 2 individual instructions.
'contract' FMF is all we need to allow this transform. That flag
corresponds to -ffp-contract=fast in Clang, so we are allowed to form
fma ops freely across expressions.
Differential Revision: https://reviews.llvm.org/D80801
Summary:
The naked function attribute is meant to suppress all function
prologue/epilogue instructions.
On ARM, some are still emitted if an argument greater than 64 bytes in size
(the threshold for using the byval attribute in IR) is passed partially
in registers.
Perform the check for Attribute::Naked and early exit in
SelectionDAGISel::LowerArguments().
Checking in ARMFrameLowering::determineCalleeSaves() is too late.
A test case is included.
Reviewers: llvm-commits, olista01, danielkiss
Reviewed By: danielkiss
Subscribers: kristof.beyls, hiraditya, danielkiss
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80715
Change-Id: Icedecf2a4ad31bc3c35ab0df7489a9d346e1f7cc
Summary:
Note to downstream target maintainers: this might silently change the semantics of your code if you override `TargetLowering::allowsMisalignedMemoryAccesses` without marking it override.
This patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790
Reviewers: courbet
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81374
Summary:
Note to downstream target maintainers: this might silently change the semantics of your code if you override `TargetLowering::allowsMemoryAccess` without marking it override.
This patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790
Reviewers: courbet
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81379
Summary:
Currently, MachineVerifier will attempt to verify that tied operands
satisfy register constraints as soon as the function is no longer in
SSA form. However, PHIElimination will take the function out of SSA
form while TwoAddressInstructionPass will actually rewrite tied operands
to match the constraints. PHIElimination runs first in the pipeline.
Therefore, whenever the MachineVerifier is run after PHIElimination,
it will encounter verification errors on any tied operands.
This patch adds a function property called TiedOpsRewritten that will be
set by TwoAddressInstructionPass and will control when the verifier checks
tied operands.
Reviewed By: nemanjai
Differential Revision: https://reviews.llvm.org/D80538
In two instances of CreateStackTemporary we are sometimes promoting
alignments beyond the stack alignment. I have introduced a new function
called getReducedAlign that will return the alignment for the broken
down parts of illegal vector types. For example, on NEON a <32 x i8>
type is made up of two <16 x i8> types - in this case the sensible
alignment is 16 bytes, not 32.
In the legalization code wherever we create stack temporaries I have
started using the reduced alignments instead for illegal vector types.
I added a test to
CodeGen/AArch64/build-one-lane.ll
that tries to insert an element into an illegal fixed vector type
that involves creating a temporary stack object.
Differential Revision: https://reviews.llvm.org/D80370
Commit d77ae1552f ("[DebugInfo] Support to emit debugInfo
for extern variables") added support to emit debuginfo
for extern variables. Currently, only BPF target enables to
emit debuginfo for extern variables.
But if the extern variable has "void" type, the compilation will
fail.
-bash-4.4$ cat t.c
extern void bla;
void *test() {
void *x = &bla;
return x;
}
-bash-4.4$ clang -target bpf -g -O2 -S t.c
missing global variable type
!1 = distinct !DIGlobalVariable(name: "bla", scope: !2, file: !3, line: 1,
isLocal: false, isDefinition: false)
...
fatal error: error in backend: Broken module found, compilation aborted!
PLEASE submit a bug report to https://bugs.llvm.org/ and include the crash backtrace,
preprocessed source, and associated run script.
Stack dump:
...
The IR requires a DIGlobalVariable must have a valid type and the
"void" type does not generate any type, hence the above fatal error.
Note that if the extern variable is defined as "const void", the
compilation will succeed.
-bash-4.4$ cat t.c
extern const void bla;
const void *test() {
const void *x = &bla;
return x;
}
-bash-4.4$ clang -target bpf -g -O2 -S t.c
-bash-4.4$ cat t.ll
...
!1 = distinct !DIGlobalVariable(name: "bla", scope: !2, file: !3, line: 1,
type: !6, isLocal: false, isDefinition: false)
!6 = !DIDerivedType(tag: DW_TAG_const_type, baseType: null)
...
Since currently, "const void extern_var" is supported by the
debug info, it is natural that "void extern_var" should also
be supported. This patch disabled assertion of "void extern_var"
in IR verifier and add proper guarding when emiting potential
null debug info type to dwarf types.
Differential Revision: https://reviews.llvm.org/D81131
This moves the SuffixTree test used in the Machine Outliner and moves it into Support for use in other outliners elsewhere in the compilation pipeline.
Differential Revision: https://reviews.llvm.org/D80586
We sometimes have functions with large numbers of sibling basic
blocks (usually with an error path exit from each one). This was
triggering the qudratic behavior in this function - after visiting
each child llvm would re-scan the parent from the beginning again. We
modify the work stack to record the next index to be worked on
alongside the pointer. This avoids the need to linearly search for
the next unfinished child.
Differential Revision: https://reviews.llvm.org/D80029
Summary: This is a followup on D81196.
Reviewers: courbet
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81362
Summary: Note to downstream target maintainers: this might silently change the semantics of your code if you override `TargetLowering::HandleByVal` without marking it `override`.
This patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790
Reviewers: courbet
Subscribers: sdardis, hiraditya, jrtc27, atanasyan, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D81365
Scalable vectors cannot use 'BUILD_VECTOR', so it is necessary to
properly split and widen scalable vectors when passing them
to CopyToReg/CopyFromReg.
This functionality is added to TargetLoweringBase::getVectorTypeBreakdown().
This patch only adds support for 'splitting' scalable vectors that
are a multiple of some legal type, e.g.
<vscale x 6 x i64> -> 3 x <vscale x 2 x i64>
Reviewers: efriedma, c-rhodes
Reviewed By: efriedma
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80139
There's two properties we want to verify:
1. That the successors returned by analyzeBranch are in the CFG
successor list, and
2. That there are no extraneous successors are in the CFG successor
list.
The previous implementation mostly accomplished this, but in a very
convoluted manner.
Differential Revision: https://reviews.llvm.org/D79793
Previously, it tried to infer the correct destination block from the
successor list, but this is a rather tricky propspect, given the
existence of successors that occur mid-block, such as invoke, and
potentially in the future, callbr/INLINEASM_BR. (INLINEASM_BR, in
particular would be problematic, because its successor blocks are not
distinct from "normal" successors, as EHPads are.)
Instead, require the caller to pass in the expected fallthrough
successor explicitly. In most callers, the correct block is
immediately clear. But, in MachineBlockPlacement, we do need to record
the original ordering, before starting to reorder blocks.
Unfortunately, the goal of decoupling the behavior of end-of-block
jumps from the successor list has not been fully accomplished in this
patch, as there is currently no other way to determine whether a block
is intended to fall-through, or end as unreachable. Further work is
needed there.
Differential Revision: https://reviews.llvm.org/D79605
Just computing the alignment makes sense without caring about the
general known bits, such as for non-integral pointers. Separate the
two and start calling into the TargetLowering hooks for frame indexes.
Start calling the TargetLowering implementation for FrameIndexes,
which improves the AMDGPU matching for stack addressing modes. Also
introduce a new hook for returning known alignment of target
instructions. For AMDGPU, it would be useful to report the known
alignment implied by certain intrinsic calls.
Also stop using MaybeAlign.
PendingInLocs ends up having the same value as InLocs, just computed
a bit more indirectly. It is a leftover of a previous implementation
approach.
This patch drops PendingInLocs, as well as the Diff and Removed
calulations, which are no longer needed.
Differential Revision: https://reviews.llvm.org/D80868
This patch updates TargetLoweringBase::computeRegisterProperties and
TargetLoweringBase::getTypeConversion to support scalable vectors,
and make the right calls on how to legalise them. These changes are required
to legalise both MVTs and EVTs.
Reviewers: efriedma, david-arm, ctetreau
Reviewed By: efriedma
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80640
Current implementation of emitPatchpoint() is very inefficient:
for every FrameIndex operand if creates new MachineInstr with
that operand expanded and all other copied as is.
Since PATCHPOINT/STATEPOINT instructions may have *a lot* of
FrameIndex operands, we end up creating and erasing many
machine instructions. But we can do it in single pass, with only
one new machine instruction generated.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D81181
Summary:
This patch adds legalisation of extensions where the operand
of the extend is a legal scalable type but the result is not.
EXTRACT_SUBVECTOR is used to split the result, before
being replaced by target-specific [S|U]UNPK[HI|LO] operations.
For example:
```
zext <vscale x 16 x i8> %a to <vscale x 16 x i16>
```
should emit:
```
uunpklo z2.h, z0.b
uunpkhi z1.h, z0.b
```
Reviewers: sdesmalen, efriedma, david-arm
Reviewed By: efriedma
Subscribers: tschuett, hiraditya, rkruppe, psnobl, huihuiz, cfe-commits, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79587
Summary:
Cache the results from getMachineBasicBlocks in LexicalScopes to speed
up UserValueScopes::dominates queries. This replaces the caching done
in UserValueScopes. Compared to the old caching method, this reduces
memory traffic when a VarLoc is copied (e.g. when a VarLocMap grows),
and enables caching across basic blocks.
When compiling sqlite 3.5.7 (CTMark version), this patch reduces the
number of calls to getMachineBasicBlocks from 10,207 to 1,093. I also
measured a small compile-time reduction (~ 0.1% of total wall time, on
average, on my machine).
As a drive-by, I made the DebugLoc in UserValueScopes a const reference
to cut down on MetadataTracking traffic.
Reviewers: jmorse, Orlando, aprantl, nikic
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80957
This wasn't getting much value from the DAG or depth arguments, since
it's only called on the frame index root nodes. FrameIndexes can also
only return a scalar value, so it also didn't need DemandedElts.
D79003/rG9fa58d1bf2f8 exposed an issue with scalarizeBinOpOfSplats that we were extracting from the splatted vector result instead of the source, the splat index is only valid for the source vector not the result, which may contain undefs, including at the splat index.
This reverts commit 21dadd774f.
In at least PromoteIntBinOps, they wanted to know about users of *all* values
produced by the node not just the integer being promoted. For example not
replacing chain users if the operation was a load breaks the ordering of the
DAG.
Summary:
This patch adds support for dumping .dot
representation of SelectionDAG. It is inspired from the fact that,
a developer may want to just dump the graph at
a predictable path with a simple name to compare.
The exisitng utility (i.e. viewGraph) are overkill
for this motive hence this patch adds the requires support
while using the core routines from GraphWriter.
Example usage: DAG.dumpDotGraph("/tmp/graph.dot", "MyGraph")
will create /tmp/graph.dot file when DAG is an
object of SelectionDAG class.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D80711
To do so, I had to sink the old school inline operand handling into GCStatepointInst which is non ideal. This code should be removed shortly and I was able to at least clean it up a bunch.
The AMDGPU lowering for unconstrained G_FDIV sometimes needs to
introduce a mode switch in the middle, so it's helpful to have
constrained instructions available to legalize this. Right now nothing
is preventing reordering of the mode switch with the other
instructions in the expansion.
When we rematerialize a value as part of the coalescing, we may
widen the register class of the destination register.
When this happens, updateRegDefUses may create additional subranges
to account for the wider register class.
The created subranges are empty and if they are not defined by
the rematerialized instruction we clean them up.
However, if they are defined by the rematerialized instruction but
unused, we failed to flag them as dead definition and would leave
them as empty live-range.
This is wrong because empty live-ranges don't interfere with anything,
thus if we don't fix them, we would fail to account that the
rematerialized instruction clobbers some lanes.
E.g., let us consider the following pseudo code:
def.lane_low64:reg128 = ldimm
newdef:reg32 = COPY def.lane_low64_low32
When rematerialization happens for newdef, we end up with:
newdef.lane_low64:reg128 = ldimm
= use newdef.lane_low64_low32
Let's look at the live interval of newdef.
Before rematerialization, we would get:
newdef [defIdx, useIdx:0) 0@defIdx
Right after updateRegDefUses, newdef register class is widen to reg128
and the subrange definitions will be augmented to fill the subreg that
is used at the definition point, here lane_low64.
The resulting live interval would be:
newdef [newDefIdx, useIdx:0) 0@newDefIdx
* lane_low64_high32 EMPTY
* lane_low64_low32 [newDefIdx, useIdx:0)
Before this patch this would be the final status of the live interval.
Therefore we miss that lane_low64_high32 is actually live on the
definition point of newdef.
With this patch, after rematerializing, we check all the added subranges
and for the ones that are defined but empty, we flag them as dead def.
Thus, in that case, newdef would look like this:
newdef [newDefIdx, useIdx:0) 0@newDefIdx
* lane_low64_high32 [newDefIdx, newDefIdxDead) ; <-- instead of EMPTY
* lane_low64_low32 [newDefIdx, useIdx:0)
This fixes https://www.llvm.org/PR46154
Record internal state based on register units. This is often more
efficient as there are typically fewer register units to update
compared to iterating over all the aliases of a register.
Original patch by Matthias Braun, but I've been rebasing and fixing it
for almost 2 years and fixed a few bugs causing intermediate failures
to make this patch independent of the changes in
https://reviews.llvm.org/D52010.
In the function "Analysis.cpp:isInTailCallPosition", it only checks whether
a call is in a tail call position if the call has side effects, access memory
or it is not safe to speculative execute. Therefore, a speculatable function
will not go through tail call position check and improperly tail called when
it is not in a tail-call position. This patch enables tail call position check
for speculatable functions.
Differential Revision: https://reviews.llvm.org/D80661
Summary:
In the patch D73152, it adds a new function LiveVariables::addNewBlock.
This new function will add the reg which PHI used to the MBB which reg
is from.
But the new function may cause LiveVariable Verification failed when the
Src reg in PHI is undef.
Reviewed By: bjope
Differential Revision: https://reviews.llvm.org/D80077
If we're only demanding the (shifted) sign bits of the shift source value, then we can use the value directly.
This handles SimplifyDemandedBits/SimplifyMultipleUseDemandedBits for both ISD::SHL and X86ISD::VSHLI.
Differential Revision: https://reviews.llvm.org/D80869
Move TargetFrameLowering.h include to the top of the TargetFrameLoweringImpl.cpp includes (clang-format doesn't do this by default as the filenames don't match).
This adds call site info support for call instructions with delay slot.
Search for instructions inside call delay slot, which load value
into parameter forwarding registers.
Return address of the call points to instruction after call delay slot,
which is not the one, immediately after the call instruction.
Patch by Nikola Tesic
Differential revision: https://reviews.llvm.org/D78107
This patch implements a target independent DAG combine to produce multiply-high
instructions from shifts. This DAG combine will combine shifts for any type as
long as the MULH on the narrow type is legal.
For now, it is enabled on PowerPC as PowerPC is the only target that has an
implementation of the isMulhCheaperThanMulShift TLI hook introduced in
D78271.
Moreover, this DAG combine focuses on catching the pattern:
(shift (mul (ext <narrow_type>:$a to <wide_type>), (ext <narrow_type>:$b to <wide_type>)), <narrow_width>)
to produce mulhs when we have a sign-extend, and mulhu when we have
a zero-extend.
The patch performs the following checks:
- Operation is a right shift arithmetic (sra) or logical (srl)
- Input to the shift is a multiply
- Both operands to the shift are sext/zext nodes
- The extends into the multiply are both the same
- The narrow type is half the width of the wide type
- The shift amount is the width of the narrow type
- The respective mulh operation is legal
Differential Revision: https://reviews.llvm.org/D78272
The collectCallSiteParameters() method searches for instructions
which load values into registers used for parameters passing.
Previously, interpretation of those values, loaded by one such
instruction, was implemented inside collectCallSiteParameters() method.
This patch moves the interpretation code from collectCallSiteParameters()
method into a separate static method named interpretValue. New method is
called from collectCallSiteParameters() to process each instruction from
targeted instruction scope.
The collectCallSiteParameters() searches for loaded parameter value
among instructions which precede the call instruction, inside the same
basic block. When needed, new method (interpretValue) could be used for
searching any instruction scope.
This is preparation for search of parameter value, loaded inside call
delay slot.
Patch by Nikola Tesic
Differential revision: https://reviews.llvm.org/D78106
This patch adds clang options:
-fbasic-block-sections={all,<filename>,labels,none} and
-funique-basic-block-section-names.
LLVM Support for basic block sections is already enabled.
+ -fbasic-block-sections={all, <file>, labels, none} : Enables/Disables basic
block sections for all or a subset of basic blocks. "labels" only enables
basic block symbols.
+ -funique-basic-block-section-names: Enables unique section names for
basic block sections, disabled by default.
Differential Revision: https://reviews.llvm.org/D68049
Do not spill UNDEF GC values. Instead, replace corresponding
gc.relocate intrinsic with an (arbitrary, but recognizable) constant.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D80714
These cases all follow the same pattern:
struct A {
friend class X;
//...
class X {};
};
But 'friend class X;' injects 'X' into the surrounding namespace scope,
rather than introducing a class member. So the second 'class X {}' is a
completely different type, which changes the meaning of the earlier name
'X' from '::X' to 'A::X'.
Additionally, the friend declaration is pointless -- members of a class
don't need to be befriended to be able to access private members.
Summary:
Instead of iterating over all VarLoc IDs in removeEntryValue(), just
iterate over the interval reserved for entry value VarLocs. This changes
the iteration order, hence the test update -- otherwise this is NFC.
This appears to give an ~8.5x wall time speed-up for LiveDebugValues when
compiling sqlite3.c 3.30.1 with a Release clang (on my machine):
```
---User Time--- --System Time-- --User+System-- ---Wall Time--- --- Name ---
Before: 2.5402 ( 18.8%) 0.0050 ( 0.4%) 2.5452 ( 17.3%) 2.5452 ( 17.3%) Live DEBUG_VALUE analysis
After: 0.2364 ( 2.1%) 0.0034 ( 0.3%) 0.2399 ( 2.0%) 0.2398 ( 2.0%) Live DEBUG_VALUE analysis
```
The change in removeEntryValue() is the only one that appears to affect
wall time, but for consistency (and to resolve a pending TODO), I made
the analogous changes for iterating over SpillLocKind VarLocs.
Reviewers: nikic, aprantl, jmorse, djtodoro
Subscribers: hiraditya, dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80684
The AMDGPU non-strict fdiv lowering needs to introduce an FP mode
switch in some cases, and has custom nodes to provide chain/glue for
the intermediate FP operations. We need to propagate nofpexcept here,
but getNode was dropping the flags.
Adding nofpexcept in the AMDGPU custom lowering is left to a future
patch.
Also fix a second case where flags were dropped, but in this case it
seems it just didn't handle this number of operands.
Test will be included in future AMDGPU patch.
Summary:
While clustering mem ops, AMDGPU target needs to consider number of clustered bytes
to decide on max number of mem ops that can be clustered. This patch adds support to pass
number of clustered bytes to target mem ops clustering logic.
Reviewers: foad, rampitec, arsenm, vpykhtin, javedabsar
Reviewed By: foad
Subscribers: MatzeB, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, javed.absar, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80545
I inverted the mask when I ported to the new form of G_PTRMASK in
8bc03d2168.
I don't think this really broke anything, since G_VASTART isn't
handled for types with an alignment higher than the stack alignment.
In some cases ScheduleDAGRRList has to add new nodes to resolve problems
with interfering physical registers. When new nodes are added, it
completely re-computes the topological order, which can take a long
time, but is unnecessary. We only add nodes one by one, and initially
they do not have any predecessors. So we can just insert them at the end
of the vector. Later we add predecessors, but the helper function
properly updates the topological order much more efficiently. With this
change, the compile time for the program below drops from 300s to 30s on
my machine.
define i11129 @test1() {
%L1 = load i11129, i11129* undef
%B30 = ashr i11129 %L1, %L1
store i11129 %B30, i11129* undef
ret i11129 %L1
}
This should be generally beneficial, as we can skip a large amount of
work. Theoretically there are some scenarios where we might not safe
much, e.g. when we add a dependency between the first and last node.
Then we would have to shift all nodes. But we still do not have to spend
the time re-computing the initial order.
Reviewers: MatzeB, atrick, efriedma, niravd, paquette
Reviewed By: paquette
Differential Revision: https://reviews.llvm.org/D59722
This code was repeated in two callers of CommitTargetLoweringOpt.
But CommitTargetLoweringOpt is also called from TargetLowering.
We should print a message for those calls to. So sink the
repeated code into CommitTargetLoweringOpt to catch those calls.
We are calling getValidShiftAmountConstant first followed by getValidMinimumShiftAmountConstant/getValidMaximumShiftAmountConstant if that failed. But both are used in the same way in ComputeNumSignBits and the Min/Max variants call getValidShiftAmountConstant internally anyhow.
This patch adds support for emission of following DWARFv5 macro
forms in .debug_macro.dwo section:
- DW_MACRO_start_file
- DW_MACRO_end_file
- DW_MACRO_define_strx
- DW_MACRO_undef_strx
Reviewed By: dblaikie
Differential Revision: https://reviews.llvm.org/D78866
Summary:
This caused incorrect debug information for parameters:
Previously, after a COPY of a parameter that changes the width,
we would emit a DBG_VALUE that continues to be associated to that
parameter, even though it now used a different width.
This made the LiveDebugValues pass assume the parameter value
got clobbered and it stopped tracking the parameter entry
value, leading to incorrect debug information.
Fixes https://bugs.llvm.org/show_bug.cgi?id=39715
Subscribers: aprantl, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80819
Let the codegen recognized the nomerge attribute and disable branch folding when the attribute is given
Differential Revision: https://reviews.llvm.org/D79537
DW_MACRO_define_strx forms are supported now in llvm-dwarfdump and these
forms can be used in both debug_macro[.dwo] sections. An added advantage
for using strx forms over strp forms is that it uses indices
approach instead of a relocation to debug_str section.
This patch unify the emission for debug_macro section.
Reviewed by: dblaikie, ikudrin
Differential Revision: https://reviews.llvm.org/D78865
Since on AIX, our strategy is to not use -u to suppress any undefined
symbols, we need to emit .extern for the symbols with AvailableExternally
linkage.
Differential Revision: https://reviews.llvm.org/D80642
optimizations
As discussed in the thread http://lists.llvm.org/pipermail/llvm-dev/2020-May/141838.html,
some bit field access width can be reduced by ReduceLoadOpStoreWidth, some
can't. If two accesses are very close, and the first access width is reduced,
the second is not. Then the wide load of second access will be stalled for long
time.
This patch add command line options to guard ReduceLoadOpStoreWidth and
ShrinkLoadReplaceStoreWithStore, so users can use them to disable these
store width reduction optimizations.
Differential Revision: https://reviews.llvm.org/D80745
Currently combineInsertEltToShuffle turns insert_vector_elt into a
vector_shuffle, even if the inserted element is a vector with a single
element. In this case, it should be unlikely that the additional shuffle
would be more efficient than a insert_vector_elt.
Additionally, this fixes a infinite cycle in DAGCombine, where
combineInsertEltToShuffle turns a insert_vector_elt into a shuffle,
which gets turned back into a insert_vector_elt/extract_vector_elt by
a custom AArch64 lowering (in visitVECTOR_SHUFFLE).
Such insert_vector_elt and extract_vector_elt combinations can be
lowered efficiently using mov on AArch64.
There are 2 test changes in arm64-neon-copy.ll: we now use one or two
mov instructions instead of a single zip1. The reason that we need a
second mov in ins1f2 is that we have to move the result to the result
register and is not really related to the DAGCombine fold I think.
But in any case, on most uarchs, mov should be cheaper than zip1. On a
Cortex-A75 for example, zip1 is twice as expensive as mov
(https://developer.arm.com/docs/101398/latest/arm-cortex-a75-software-optimization-guide-v20)
Reviewers: spatel, efriedma, dmgreen, RKSimon
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D80710
AddressingModeMatcher::matchScaledValue was calling getSExtValue for a constant before ensuring that we can actually represent the value as int64_t
Fixes OSSFuzz#22723 which is a followup to rGc479052a74b2 (PR46004 / OSSFuzz#22357)
Summary:
The description of EXTACT_SUBVECTOR and INSERT_SUBVECTOR has been
changed to accommodate scalable vectors (see ISDOpcodes.h). This
patch updates the asserts used to verify these requirements when
using SelectionDAG's getNode interface.
This patch introduces the MVT function getVectorMinNumElements
that can be used against fixed-length and scalable vectors when
only the known minimum vector length is required.
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80709
We should be using getVectorElementCount() to assert that two types
have the same numbers of elements. I encountered the warnings while
compiling this test:
CodeGen/AArch64/sve-intrinsics-ld1.ll
Differential Revision: https://reviews.llvm.org/D80616
I have tried to ensure that SelectionDAG and DAGCombiner do
sensible things for scalable vectors, and added support for a
limited number of simple folds. Codegen support for the vector
extract patterns have also been added to the AArch64 backend.
New vector extract tests have been added here:
CodeGen/AArch64/sve-extract-element.ll
and I have also added new folds using inserts and extracts here:
CodeGen/AArch64/sve-insert-element.ll
Differential Revision: https://reviews.llvm.org/D80208
During legalization we can end up with extends of loads, which in the case of
zexts causes us to not hit tablegen imported patterns.
The caveat here is that we don't want anyext load forming, since some variants
are illegal. This change also prevents the combine from creating any illegal
loads.
Differential Revision: https://reviews.llvm.org/D80458
I get confused by a lot of the predicate names here, since I would
assume they apply to vectors as well. Rename to reflect they only
apply to scalars.
Also add a few predicates AMDGPU uses that should be generally useful.
Also add any() to complement all. I've wanted to use this a few times
but then worked around it not being there.
Summary:
We received a report of LiveDebugValues consuming 25GB+ of RAM when
compiling code generated by Unity's IL2CPP scripting backend.
There's an initial 5GB spike due to repeatedly copying cached lists of
MachineBasicBlocks within the UserValueScopes members of VarLocs.
But the larger scaling issue arises due to the fact that prior to range
extension, there are 81K basic blocks and 156K DBG_VALUEs: given enough
memory, LiveDebugValues would insert 101 million MIs (I counted this by
incrementing a counter inside of VarLoc::BuildDbgValue).
It seems like LiveDebugValues would have to be rearchitected to support
this kind of input (we'd need some new represntation for DBG_VALUEs that
get inserted into ~every block via flushPendingLocs). OTOH, large globs
of auto-generated code are typically not debugged interactively.
So: add cutoffs to disable range extension when the input is too big. I
chose the cutoffs experimentally, erring on the conservative side. When
compiling a large collection of Apple software, range extension never
got disabled.
rdar://63418929
Reviewers: aprantl, friss, jmorse, Orlando
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80662
Summary:
Verify that each DBG_VALUE has a debug location. This is required by
LiveDebugValues, and perhaps by other late passes.
There's an exception for tests: lots of tests use a two-operand form of
DBG_VALUE for convenience. There's no reason to prevent that.
This is an extension of D80665, but there's no dependency.
Reviewers: aprantl, jmorse, davide, chrisjackson
Subscribers: hiraditya, asb, rbar, johnrusso, simoncook, sabuasal, niosHD, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, PkmX, jocewei, Jim, lenary, s.egerton, pzheng, sameer.abuasal, apazos, luismarques, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80670
Summary:
Assert that MachineLICM does not move a debug instruction and then drop
its debug location. Later passes require each debug instruction to have
a location.
Testing: check-llvm, clang stage2 RelWithDebInfo build (x86_64)
Reviewers: aprantl, davide, chrisjackson, jmorse
Subscribers: hiraditya, asbirlea, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80665
These are the two operand sets which are expected to survive more than another week or so. Instead of bothering to update the deopt and gc-transition operands, we'll just wait until those are removed and delete the code.
For those following along, this is likely to be the last (major) change in this sequence for about a week. I want to wait until all of this has been merged downstream to ensure I haven't introduced any bugs (and migrate some downstream code to the new interfaces). Once that's done, we should be able to delete Statepoint/ImmutableStatepoint without too much work.
I'd apparently only grepped in the lib directories and missed a few used in the Statepoint header itself. Beyond simple mechanical cleanup, changed the type of one routine to reflect the fact it also returns a statepoint.
Sinking logic around actual callee from Statepoint to GCStatepointInst. While doing so, adjust naming to be consistent about refering to "actual" callee and follow precedent on naming from CallBase otherwise.
Use the result to simplify one consumer. This is mostly just to ensure the new code is exercised, but is also a helpful cleanup on it's own.
While LazyBlockFrequencyInfo itself is lazy, the dominator tree
and loop info analyses it requires are not. Drop the dependency
on this pass in SelectionDAGIsel at O0.
This makes for a ~0.6% O0 compile-time improvement.
Differential Revision: https://reviews.llvm.org/D80387
This patch upgrades DISubrange to support fortran requirements.
Summary:
Below are the updates/addition of fields.
lowerBound - Now accepts signed integer or DIVariable or DIExpression,
earlier it accepted only signed integer.
upperBound - This field is now added and accepts signed interger or
DIVariable or DIExpression.
stride - This field is now added and accepts signed interger or
DIVariable or DIExpression.
This is required to describe bounds of array which are known at runtime.
Testing:
unit test cases added (hand-written)
check clang
check llvm
check debug-info
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D80197
Now that all of the statepoint related routines have classes with isa support, let's cleanup.
I'm leaving the (dead) utitilities in tree for a few days so that I can do the same cleanup downstream without breakage.
Can't test this since I can't directly use the default expansion for
AMDGPU. It needs to scale the amount by the wave size, rather than use
the raw byte size value.
If we have a memory instruction (e.g. a load), we shouldn't combine it away in
some trivial combine.
It's possible that, say, a call lives between the instructions. This could
modify the value loaded, making the load instructions not safe to fold.
Differential Revision: https://reviews.llvm.org/D80053
Use getFunctionEntryPointSymbol whenever possible to enclose the
implementation detail and reduce duplicate logic.
Differential Revision: https://reviews.llvm.org/D80402
In the current statepoint design, we have four distinct groups of operands to the call: call args, gc transition args, deopt args, and gc args. This format prexisted the support in IR for operand bundles and was in fact one of the inspirations for the extension. However, we never went back and rearchitected statepoints to fully leverage bundles.
This change is the first in a small sequence to do so. All this does is extend the SelectionDAG lowering code to allow deopt and gc transition operands to be specified in either inline argument bundles or operand bundles.
Differential Revision: https://reviews.llvm.org/D8059
Summary:
Previously, we only added early-clobber flags to the 'group' immediate flag operand
of an inline asm operand.
However, we also have to add the EarlyClobber flag to the MachineOperand itself.
This fixes PR46028
Reviewers: arsenm, leonardchan
Reviewed By: arsenm, leonardchan
Subscribers: phosek, wdng, rovka, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80467
Summary:
A struct argument can be passed-by-value to a callee via a pointer to a
temporary stack copy. Add support for emitting an entry value DBG_VALUE
when an indirect parameter DBG_VALUE becomes unavailable. This is done
by omitting DW_OP_stack_value from the entry value expression, to make
the expression describe the location of an object.
rdar://63373691
Reviewers: djtodoro, aprantl, dstenb
Subscribers: hiraditya, lldb-commits, llvm-commits
Tags: #lldb, #llvm
Differential Revision: https://reviews.llvm.org/D80345
Confusingly, these were unrelated and had different semantics. The
G_PTR_MASK instruction predates the llvm.ptrmask intrinsic, but has a
different format. G_PTR_MASK only allows clearing the low bits of a
pointer, and only a constant number of bits. The ptrmask intrinsic
allows an arbitrary mask. Replace G_PTR_MASK to match the intrinsic.
Only selects the cases that look like the old instruction. More work
is needed to select the general case. Also new legalization code is
still needed to deal with the case where the incoming mask size does
not match the pointer size, which has a specified behavior in the
langref.
This intrinsic implements IEEE-754 operation roundToIntegralTiesToEven,
and performs rounding to the nearest integer value, rounding halfway
cases to even. The intrinsic represents the missed case of IEEE-754
rounding operations and now llvm provides full support of the rounding
operations defined by the standard.
Differential Revision: https://reviews.llvm.org/D75670
binop (splat X), (splat C) --> splat (binop X, C)
binop (splat C), (splat X) --> splat (binop C, X)
We do this in IR, and there's a similar fold for the case with 2
non-constant operands just above the code diff in this patch.
This was discussed in D79718, and the extra shuffle in the test
(llvm/test/CodeGen/X86/vector-fshl-128.ll::sink_splatvar) where it
was noticed disappears because demanded elements analysis is no
longer blocked. The large majority of the test diffs seem to be
benign code scheduling changes, but I do see another type of win:
moving the splat later allows binop narrowing in some cases.
Regressions were avoided on x86 and ARM with the INSERT_VECTOR_ELT
restriction.
Differential Revision: https://reviews.llvm.org/D79886
Summary:
Clean-up code around mem ops clustering logic. This patch cleans up code within
the function clusterNeighboringMemOps(). It is WIP, and this patch is a first cut.
Reviewers: foad, rampitec, arsenm, vpykhtin, javedabsar
Reviewed By: foad
Subscribers: MatzeB, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, javed.absar, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80119
-fno-PIC and -fPIE code generally cannot be linked in -shared mode and there is no benefit accessing via local aliases.
Actually, a .Lfoo$local reference will be converted to a STT_SECTION (if no section relaxation) reference which will cause the section symbol (sizeof(Elf64_Sym)=24) to be generated.
-fno-semantic-interposition is currently the CC1 default. (The opposite
disables some interprocedural optimizations.) However, it does not infer
dso_local: on most targets accesses to ExternalLinkage functions/variables
defined in the current module still need PLT/GOT.
This patch makes explicit -fno-semantic-interposition infer dso_local,
so that PLT/GOT can be eliminated if targets implement local aliases
for AsmPrinter::getSymbolPreferLocal (currently only x86).
Currently we check whether the module flag "SemanticInterposition" is 0.
If yes, infer dso_local. In the future, we can infer dso_local unless
"SemanticInterposition" is 1: frontends other than clang will also
benefit from the optimization if they don't bother setting the flag.
(There will be risks if they do want ELF interposition: they need to set
"SemanticInterposition" to 1.)
For the supported binops (basic arithmetic, logicals + shifts), if we fail to simplify the demanded vector elts, then call SimplifyMultipleUseDemandedBits and try to peek through ops to remove unnecessary dependencies.
This helps with PR40502.
Differential Revision: https://reviews.llvm.org/D79003
Fixes a build issue with libc++ configured with _LIBCPP_RAW_ITERATORS (ADL not effective)
```
llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp:1602:3: error: no matching function for call to 'transform'
transform(HexString.begin(), HexString.end(), HexString.begin(), tolower);
^~~~~~~~~
```
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D80475
For the 'inverse shift', we currently always perform a subtraction of the original (masked) shift amount.
But for the case where we are handling power-of-2 type widths, we can replace:
(sub bw-1, (and amt, bw-1) ) -> (and (xor amt, bw-1), bw-1) -> (and ~amt, bw-1)
This allows x86 shifts to fold away the and-mask.
Followup to D77301 + D80466.
http://volta.cs.utah.edu:8080/z/Nod0Gr
Differential Revision: https://reviews.llvm.org/D80489
This patch introduces a TargetLowering query, isMulhCheaperThanMulShift.
Currently in DAG Combine, it will transform mulhs/mulhu into a
wider multiply and a shift if the wide multiply is legal.
This TLI function is implemented on 64-bit PowerPC, as it is more desirable to
have multiply-high over multiply + shift for words and doublewords. Having
multiply-high can also aid in further transformations that can be done.
Differential Revision: https://reviews.llvm.org/D78271
Disable pruning of unreachable resumes in the DwarfEHPrepare pass
at optnone. While I expect the pruning itself to be essentially free,
this does require a dominator tree calculation, that is not used for
anything else. Saving this DT construction makes for a 0.4% O0
compile-time improvement.
Differential Revision: https://reviews.llvm.org/D80400
Replace with forward declaration and move dependency down to source files that actually need it.
Both TargetLowering.h and TargetMachine.h are 2 of the most expensive headers (top 10) in the ClangBuildAnalyzer report when building llc.
When performing codegen at optnone, don't add alias analysis to
the pipeline. We don't need it, but it causes an unnecessary
dominator tree calculation.
I've also moved the module verifier call to the top so that a bunch
of disabled-at-optnone passes group more nicely.
Differential Revision: https://reviews.llvm.org/D80378
If the caller needs to reponsible for making sure the MaybeAlign
has a value, then we should just make the caller convert it to an Align
with operator*.
I explicitly deleted the relational comparison operators that
were being inherited from Optional. It's unclear what the meaning
of two MaybeAligns were one is defined and the other isn't
should be. So make the caller reponsible for defining the behavior.
I left the ==/!= operators from Optional. But now that exposed a
weird quirk that ==/!= between Align and MaybeAlign required the
MaybeAlign to be defined. But now we use the operator== from
Optional that takes an Optional and the Value.
Differential Revision: https://reviews.llvm.org/D80455
This temporarily reverts commit 7019cea26d.
It seems that, for some targets, there are instructions with a lot of memory operands (probably more than would be expected). This causes a lot of buildbots to timeout and notify failed builds. While investigations are ongoing to find out why this happens, revert the changes.
AddressingModeMatcher::matchAddr was calling getSExtValue for a constant before ensuring that we can actually represent the value as int64_t
Fixes PR46004 / OSSFuzz#22357
(This patch is by Jessica, I'm just committing it on her behalf because I need
a post-legalizer combiner for something else).
This supersedes D77250, which did equivalent work in the selector. This can be
done pre-legalization or post-legalization. Post-legalization is more likely to
hit, since G_IMPLICIT_DEFs tend to appear during legalization. There's no reason
to not do it pre-legalization though-- if it can be caught earlier, great.
(I also think that it might be worth reimplementing D78769 using a
target-specific post-legalization combine too after thinking about it for a
while.)
Differential Revision: https://reviews.llvm.org/D78852
Summary:
To support all targets, the mayAlias member function needs to support instructions with multiple operands.
This revision also changes the order of the emitted instructions in some test cases.
Reviewers: efriedma, hfinkel, craig.topper, dmgreen
Reviewed By: efriedma
Subscribers: MatzeB, dmgreen, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80161
Summary:
For some targets generic combines don't really do much and they
consume a disproportionate amount of time.
There's not really a mechanism in SDISel to tactically disable
combines, but we can have a switch to disable all of them and
let the targets just implement what they specifically need.
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79112
When moving an instruction into a block where it was referenced by a phi when peeling,
refer to the phi's register number and assert that the instruction has it in its destinations.
This way, it also covers instructions with more than one destination.
Patch by Hendrik Greving!
Differential Revision: https://reviews.llvm.org/D80027
This is split off from D80316, slightly tightening the definition of overloaded
hardwareloop intrinsic llvm.loop.decrement.reg specifying that both operands
its result have the same type.
We do not have any special handling for constant FP deopt arguments.
They are just spilled to stack or generated in register by MOVS
instruction. This is inefficient and, when we have too many such
constant arguments, may result in register allocation failure.
Instead, we can bitcast such constant FP operands to appropriately
sized integer and record as constant into statepoint and later, into
StackMap.
Reviewed By: skatkov
Differential Revision: https://reviews.llvm.org/D80318
Will make it easier to pass the pointer info and alignment
correctly to the loads/stores.
While there also make the i32 stores independent and use a token
factor to join before the load.
If we don't know anything about the alignment of a pointer, Align(1) is
still correct: all pointers are at least 1-byte aligned.
Included in this patch is a bugfix for an issue discovered during this
cleanup: pointers with "dereferenceable" attributes/metadata were
assumed to be aligned according to the type of the pointer. This
wasn't intentional, as far as I can tell, so Loads.cpp was fixed to
stop making this assumption. Frontends may need to be updated. I
updated clang's handling of C++ references, and added a release note for
this.
Differential Revision: https://reviews.llvm.org/D80072
Previously this code just used a default constructed
MachinePointerInfo. But we know the accesses are to a fixed stack
object or at least somewhere on the stack.
While there fix the alignment passed to the full vector load/stores.
I don't think this function is currently exercised in tree so I
don't know how to test it. I just noticed it when I removed
non-constant index support in this function.
Differential Revision: https://reviews.llvm.org/D80058
See https://reviews.llvm.org/D74651 for the preallocated IR constructs
and LangRef changes.
In X86TargetLowering::LowerCall(), if a call is preallocated, record
each argument's offset from the stack pointer and the total stack
adjustment. Associate the call Value with an integer index. Store the
info in X86MachineFunctionInfo with the integer index as the key.
This adds two new target independent ISDOpcodes and two new target
dependent Opcodes corresponding to @llvm.call.preallocated.{setup,arg}.
The setup ISelDAG node takes in a chain and outputs a chain and a
SrcValue of the preallocated call Value. It is lowered to a target
dependent node with the SrcValue replaced with the integer index key by
looking in X86MachineFunctionInfo. In
X86TargetLowering::EmitInstrWithCustomInserter() this is lowered to an
%esp adjustment, the exact amount determined by looking in
X86MachineFunctionInfo with the integer index key.
The arg ISelDAG node takes in a chain, a SrcValue of the preallocated
call Value, and the arg index int constant. It produces a chain and the
pointer fo the arg. It is lowered to a target dependent node with the
SrcValue replaced with the integer index key by looking in
X86MachineFunctionInfo. In
X86TargetLowering::EmitInstrWithCustomInserter() this is lowered to a
lea of the stack pointer plus an offset determined by looking in
X86MachineFunctionInfo with the integer index key.
Force any function containing a preallocated call to use the frame
pointer.
Does not yet handle a setup without a call, or a conditional call.
Does not yet handle musttail. That requires a LangRef change first.
Tried to look at all references to inalloca and see if they apply to
preallocated. I've made preallocated versions of tests testing inalloca
whenever possible and when they make sense (e.g. not alloca related,
inalloca edge cases).
Aside from the tests added here, I checked that this codegen produces
correct code for something like
```
struct A {
A();
A(A&&);
~A();
};
void bar() {
foo(foo(foo(foo(foo(A(), 4), 5), 6), 7), 8);
}
```
by replacing the inalloca version of the .ll file with the appropriate
preallocated code. Running the executable produces the same results as
using the current inalloca implementation.
Reverted due to unexpectedly passing tests, added REQUIRES: asserts for reland.
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D77689
See https://reviews.llvm.org/D74651 for the preallocated IR constructs
and LangRef changes.
In X86TargetLowering::LowerCall(), if a call is preallocated, record
each argument's offset from the stack pointer and the total stack
adjustment. Associate the call Value with an integer index. Store the
info in X86MachineFunctionInfo with the integer index as the key.
This adds two new target independent ISDOpcodes and two new target
dependent Opcodes corresponding to @llvm.call.preallocated.{setup,arg}.
The setup ISelDAG node takes in a chain and outputs a chain and a
SrcValue of the preallocated call Value. It is lowered to a target
dependent node with the SrcValue replaced with the integer index key by
looking in X86MachineFunctionInfo. In
X86TargetLowering::EmitInstrWithCustomInserter() this is lowered to an
%esp adjustment, the exact amount determined by looking in
X86MachineFunctionInfo with the integer index key.
The arg ISelDAG node takes in a chain, a SrcValue of the preallocated
call Value, and the arg index int constant. It produces a chain and the
pointer fo the arg. It is lowered to a target dependent node with the
SrcValue replaced with the integer index key by looking in
X86MachineFunctionInfo. In
X86TargetLowering::EmitInstrWithCustomInserter() this is lowered to a
lea of the stack pointer plus an offset determined by looking in
X86MachineFunctionInfo with the integer index key.
Force any function containing a preallocated call to use the frame
pointer.
Does not yet handle a setup without a call, or a conditional call.
Does not yet handle musttail. That requires a LangRef change first.
Tried to look at all references to inalloca and see if they apply to
preallocated. I've made preallocated versions of tests testing inalloca
whenever possible and when they make sense (e.g. not alloca related,
inalloca edge cases).
Aside from the tests added here, I checked that this codegen produces
correct code for something like
```
struct A {
A();
A(A&&);
~A();
};
void bar() {
foo(foo(foo(foo(foo(A(), 4), 5), 6), 7), 8);
}
```
by replacing the inalloca version of the .ll file with the appropriate
preallocated code. Running the executable produces the same results as
using the current inalloca implementation.
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D77689
SCEVExpander modifies the underlying function so it is more suitable in
Transforms/Utils, rather than Analysis. This allows using other
transform utils in SCEVExpander.
This patch was originally committed as b8a3c34eee, but broke the
modules build, as LoopAccessAnalysis was using the Expander.
The code-gen part of LAA was moved to lib/Transforms recently, so this
patch can be landed again.
Reviewers: sanjoy.google, efriedma, reames
Reviewed By: sanjoy.google
Differential Revision: https://reviews.llvm.org/D71537
Replace with forward declarations and move necessary includes down to source files.
Exposes an implicit dependency on TargetMachine.h in llvm-opt-fuzzer.cpp
We have the getNegatibleCost/getNegatedExpression to evaluate the cost and negate the expression.
However, during negating the expression, the cost might change as we are changing the DAG,
and then, hit the assertion if we negated the wrong expression as the cost is not trustful anymore.
This patch is target to remove the getNegatibleCost to avoid the out of sync with getNegatedExpression,
and check the cost during negating the expression. It also reduce the duplicated code between
getNegatibleCost and getNegatedExpression. And fix the crash for the test in D76638
Reviewed By: RKSimon, spatel
Differential Revision: https://reviews.llvm.org/D77319
This was looking for a compare condition, and copying the compare
flags. I don't think this was ever correct outside of certain min/max
patterns which aren't checked, but this probably predates select
instructions having fast math flags.
Replace with forward declarations and move includes down to source files where required.
I also needed to move the TargetLoweringObjectFile::SectionForGlobal wrapper implementation down into TargetLoweringObjectFile.cpp
This reverts commit 525a591f0f.
Fixed an issue with pointers to members based on typedefs. In this case,
LLVM would emit a second UDT. I fixed it by not passing the class type
to getTypeIndex when the base type is not a function type. lowerType
only uses the class type for direct function types. This suggests if we
have a PMF with a function typedef, there may be an issue, but that can
be solved separately.
verifyFunction/verifyModule don't assert or error internally. They
also don't print anything if you don't pass a raw_ostream to them.
So the caller needs to check the result and ideally pass a stream
to get the messages. Otherwise they're just really expensive no-ops.
I've filed PR45965 for another instance in SLPVectorizer
that causes a lit test failure.
Differential Revision: https://reviews.llvm.org/D80106
I have changed the pass so that we ignore shuffle vectors with
scalable vector types, and replaced VectorType with FixedVectorType
in the rest of the pass. I couldn't think of an easy way to test
this change, since for scalable vectors we shouldn't be using
shufflevectors for interleaving. This change fixes up some
type size assert warnings I found in the following test:
CodeGen/AArch64/sve-intrinsics-int-arith-imm.ll
Differential Revision: https://reviews.llvm.org/D79700
> Before this patch, S_[L|G][THREAD32|DATA32] records were emitted with a simple name, not the fully qualified name (namespace + class scope).
>
> Differential Revision: https://reviews.llvm.org/D79447
This causes asserts in Chromium builds:
CodeViewDebug.cpp:2997: void llvm::CodeViewDebug::emitDebugInfoForUDTs(const std::vector<std::pair<std::string, const DIType *>> &):
Assertion `OriginalSize == UDTs.size()' failed.
I will follow up on the Phabricator issue.
for variables in nested scopes (including inlined functions) if there is a
single location which covers the entire scope and the scope is contained in a
single block.
Based on work by @jmorse.
Reviewed By: vsk, aprantl
Differential Revision: https://reviews.llvm.org/D79571
Now that load/store alignment is required, we no longer need most
of them. Also switch the getLoadStoreAlignment() helper to return
Align instead of MaybeAlign.
This is a no-op/NFC at the moment & generally makes the code /somewhat/
cleaner/less reliant on assumptions about what will produce a debug_addr
section.
It's still a bit "spooky action at a distance" - the add ranges code
pre-emptively inserts addresses into the address pool it knows will
eventually be used by the range emission code (or low/high pc).
The 'ideal' would be either to actually compute the addresses needed for
range (& loc) emission earlier - which would mean decanonicalizing the
range/loc representation earlier to account for whether it was going to
use addrx encodings or not (which would be unfortunate, but could be
refactored to be relatively unobtrusive).
Alternatively, emitting the range/loc sections earlier would cause them
to request the needed addresses sooner - but then you endup having to
split finalizeModuleInfo because some things need to be handled there
before the ranges/locs are emitted, I think...
We know the pointer somewhere on the stack, we just don't know
exactly where since the index may be variable.
Differential Revision: https://reviews.llvm.org/D80060
Along the lines of D77454 and D79968. Unlike loads and stores, the
default alignment is getPrefTypeAlign, to match the existing handling in
various places, including SelectionDAG and InstCombine.
Differential Revision: https://reviews.llvm.org/D80044
This is basically the same patch as D63233, but converted to
funnel shifts rather than regular shifts. I did not see a
way to effectively share code for these 2 cases though.
This follows D79718 and D79827 to re-fix PR37426 because
that gets canonicalized to funnel shift intrinsics in IR.
I did draft an alternative patch as an enhancement to
"shouldSinkOperands()", but that was awkward because
we have to key the transform from the select, but then
look at both its users and its operands.
The code was calculating an offset from a stack pointer SDValue.
This is exactly what getMemBasePlusOffset does. I also replaced
sizeof(int) with a hardcoded 4. We know the type we're operating
on is 4 bytes. But the size of int that the source code is being
compiled with isn't guaranteed to be 4 bytes.
While here replace another use of getMemBasePlusOffset that was
proceeded with a call to getConstant with the other signature
that call getConstant internally.
This bug is exposed by Test7 of ehthrow.cxx in MSVC EH suite where
a rethrow occurs in a try-catch inside a catch (i.e., a nested Catch
handlers). See the test code in
https://github.com/microsoft/compiler-tests/blob/master/eh/ehthrow.cxx#L346
When an object is rethrown in a Catch handler, the copy-ctor of this
object must be executed after the destructions of live objects, but
BEFORE the dtors of live objects in parent handlers.
Today Windows 64-bit runtime (__CxxFrameHandler3 & 4) expects nested Catch
handers
are stored in pre-order (outer first, inner next) in $tryMap$ table, so
that given a State, its Catch's beginning State can be properly
retrieved. The Catch beginning state (which is also the ending State) is
the State where rethrown object's copy-ctor must take place.
LLVM currently stores nested catch handlers in post-ordering because
it's the natural way to compute the highest State in Catch.
The fix is to simply store TryCatch handler in pre-order, but update
Catch's highest State after child Catches are all processed.
Differential Revision: https://reviews.llvm.org/D79474?id=263919
Summary:
In the the given example, a stack slot pointer is merged
between a setjmp and longjmp. This pointer is spilled,
so it does not get correctly restored, addinga undefined
behaviour where it shouldn't.
Change-Id: I60ec010844f2a24ce01ceccf12eb5eba5ab94abb
Reviewers: eli.friedman, thanm, efriedma
Reviewed By: efriedma
Subscribers: MatzeB, qcolombet, tpr, rnk, efriedma, hiraditya, llvm-commits, chill
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D77767
This is D77454, except for stores. All the infrastructure work was done
for loads, so the remaining changes necessary are relatively small.
Differential Revision: https://reviews.llvm.org/D79968
Before this patch, S_[L|G][THREAD32|DATA32] records were emitted with a simple name, not the fully qualified name (namespace + class scope).
Differential Revision: https://reviews.llvm.org/D79447
For now I have changed FoldConstantVectorArithmetic to return early
if we encounter a scalable vector, since the subsequent code assumes
you can perform lane-wise constant folds. However, in future work we
should be able to extend this to look at splats of a constant value
and fold those if possible. I have also added the same code to
FoldConstantArithmetic, since that deals with vectors too.
The warnings I fixed in this patch were being generated by this
existing test:
CodeGen/AArch64/sve-int-arith.ll
Differential Revision: https://reviews.llvm.org/D79421
Summary:
The BFloat IR type is introduced to provide support for, initially, the BFloat16
datatype introduced with the Armv8.6 architecture (optional from Armv8.2
onwards). It has an 8-bit exponent and a 7-bit mantissa and behaves like an IEEE
754 floating point IR type.
This is part of a patch series upstreaming Armv8.6 features. Subsequent patches
will upstream intrinsics support and C-lang support for BFloat.
Reviewers: SjoerdMeijer, rjmccall, rsmith, liutianle, RKSimon, craig.topper, jfb, LukeGeeson, sdesmalen, deadalnix, ctetreau
Subscribers: hiraditya, llvm-commits, danielkiss, arphaman, kristof.beyls, dexonsmith
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D78190
Summary:
D78319 introduced basic support for inline asm input operands in GlobalISel.
However, that patch did not handle the case where a memory input operand still needs to
be indirectified. Later code asserts that the memory operand is already indirect.
This patch adds an early return false to trigger the SelectionDAG fallback for now.
Reviewers: arsenm, paquette
Reviewed By: arsenm
Subscribers: thakis, wdng, rovka, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79955
I've created a new variant of CreateStackTemporary that takes
TypeSize and Align arguments, and made the older instances of
CreateStackTemporary call this new function. This refactoring is
in preparation for more patches in this area related to scalable
vectors and improving the alignment calculations.
Differential Revision: https://reviews.llvm.org/D79933
This patch adds support for DWARF attribute DW_AT_data_location.
Summary:
Dynamic arrays in fortran are described by array descriptor and
data allocation address. Former is mapped to DW_AT_location and
later is mapped to DW_AT_data_location.
Testing:
unit test cases added (hand-written)
check llvm
check debug-info
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D79592
llvm rejects DWARF operator DW_OP_push_object_address.This DWARF
operator is needed for Flang to support allocatable array.
Summary:
Currently llvm rejects DWARF operator DW_OP_push_object_address.
below error is produced when llvm finds this operator.
[..]
invalid expression
!DIExpression(151)
warning: ignoring invalid debug info in pushobj.ll
[..]
There are some parts missing in support of this operator, need to
be completed.
Testing
-added a unit testcase
-check-debuginfo
-check-llvm
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D79306
Summary:
In the patch D78849, it uses llvm::any_of to instead of for loop to
simplify the function addRequired().
It's obvious that above code is not a NFC conversion. Because any_of
will return if any addRequired(Reg) is true immediately, but we want
every element to call addRequired(Reg).
This patch uses for_range loop to fix above any_of bug.
Reviewed By: MaskRay, nickdesaulniers
Differential Revision: https://reviews.llvm.org/D79872
Summary:
D78319 introduced basic support for inline asm input operands in GlobalISel.
However, that patch did not handle the case where a memory input operand still needs to
be indirectified. Later code asserts that the memory operand is already indirect.
This patch adds an early return false to trigger the SelectionDAG fallback for now.
Reviewers: arsenm, paquette
Reviewed By: arsenm
Subscribers: wdng, rovka, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79955
We need to use it to handle <16 x double> indirect indexes
in the AMDGPU BE.
The only visible change from adding it is in ARM cost model.
To me it looks reasonable. With doubling a vector size it
quadruples the cost up to the size 8 and then it did only
double it. Now it also quadruples, which seems a logical
progression to me.
Actual AMDGPU code is to follow, this is a common part, plus
load/store legalization in the AMDGPU BE not to break what
works now.
Differential Revision: https://reviews.llvm.org/D79952
The fact that loads and stores can have the alignment missing is a
constant source of confusion: code that usually works can break down in
rare cases. So fix the LoadInst API so the alignment is never missing.
To reduce the number of changes required to make this work, IRBuilder
and certain LoadInst constructors will grab the module's datalayout and
compute the alignment automatically. This is the same alignment
instcombine would eventually apply anyway; we're just doing it earlier.
There's a minor risk that the way we're retrieving the datalayout
could break out-of-tree code, but I don't think that's likely.
This is the last in a series of patches, so most of the necessary
changes have already been merged.
Differential Revision: https://reviews.llvm.org/D77454
For IR generated by a compiler, this is really simple: you just take the
datalayout from the beginning of the file, and apply it to all the IR
later in the file. For optimization testcases that don't care about the
datalayout, this is also really simple: we just use the default
datalayout.
The complexity here comes from the fact that some LLVM tools allow
overriding the datalayout: some tools have an explicit flag for this,
some tools will infer a datalayout based on the code generation target.
Supporting this properly required plumbing through a bunch of new
machinery: we want to allow overriding the datalayout after the
datalayout is parsed from the file, but before we use any information
from it. Therefore, IR/bitcode parsing now has a callback to allow tools
to compute the datalayout at the appropriate time.
Not sure if I covered all the LLVM tools that want to use the callback.
(clang? lli? Misc IR manipulation tools like llvm-link?). But this is at
least enough for all the LLVM regression tests, and IR without a
datalayout is not something frontends should generate.
This change had some sort of weird effects for certain CodeGen
regression tests: if the datalayout is overridden with a datalayout with
a different program or stack address space, we now parse IR based on the
overridden datalayout, instead of the one written in the file (or the
default one, if none is specified). This broke a few AVR tests, and one
AMDGPU test.
Outside the CodeGen tests I mentioned, the test changes are all just
fixing CHECK lines and moving around datalayout lines in weird places.
Differential Revision: https://reviews.llvm.org/D78403
Use an extra shift-by-1 instead of a compare and select to handle the
shift-by-zero case. This sometimes saves one instruction (if the compare
couldn't be combined with a previous instruction). It also works better
on targets that don't have good select instructions.
Note that currently this change doesn't affect most targets because
expandFunnelShift is not used because funnel shift intrinsics are
lowered early in SelectionDAGBuilder. But there is work afoot to change
that; see D77152.
Differential Revision: https://reviews.llvm.org/D77301
Expands on the enablement of the shouldSinkOperands() TLI hook in:
D79718
The last codegen/IR test diff shows what I suspected could happen - we were
sinking all splat shift operands into a loop. But that's not what we want in
general; we only want to sink the *shift amount* operand if it is a splat.
Differential Revision: https://reviews.llvm.org/D79827
It sounds like an interesting idea in theory, but nothing is actually
taking advantage of it, and specifying/implementing the edge cases is
painful. So just forbid it.
Differential Revision: https://reviews.llvm.org/D79814
Under MVE a vdup will always take a gpr register, not a floating point
value. During DAG combine we convert the types to a bitcast to an
integer in an attempt to fold the bitcast into other instructions. This
is OK, but only works inside the same basic block. To do the same trick
across a basic block boundary we need to convert the type in
codegenprepare, before the splat is sunk into the loop.
This adds a convertSplatType function to codegenprepare to do that,
putting bitcasts around the splat to force the type to an integer. There
is then some adjustment to the code in shouldSinkOperands to handle the
extra bitcasts.
Differential Revision: https://reviews.llvm.org/D78728
This patch extends DIModule Debug metadata in LLVM to support
Fortran modules. DIModule is extended to contain File and Line
fields, these fields will be used by Flang FE to create debug
information necessary for representing Fortran modules at IR level.
Furthermore DW_TAG_module is also extended to contain these fields.
If these fields are missing, debuggers like GDB won't be able to
show Fortran modules information correctly.
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D79484
GNU ld's internal linker script uses (https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=add44f8d5c5c05e08b11e033127a744d61c26aee)
.text :
{
*(.text.unlikely .text.*_unlikely .text.unlikely.*)
*(.text.exit .text.exit.*)
*(.text.startup .text.startup.*)
*(.text.hot .text.hot.*)
*(SORT(.text.sorted.*))
*(.text .stub .text.* .gnu.linkonce.t.*)
/* .gnu.warning sections are handled specially by elf.em. */
*(.gnu.warning)
}
Because `*(.text.exit .text.exit.*)` is ordered before `*(.text .text.*)`, in a -ffunction-sections build, the C library function `exit` will be placed before other functions.
gold's `-z keep-text-section-prefix` has the same problem.
In lld, `-z keep-text-section-prefix` recognizes `.text.{exit,hot,startup,unlikely,unknown}.*`, but not `.text.{exit,hot,startup,unlikely,unknown}`, to avoid the strange placement problem.
In -fno-function-sections or -fno-unique-section-names mode, a function whose `function_section_prefix` is set to `.exit"`
will go to the output section `.text` instead of `.text.exit` when linked by lld.
To address the problem, append a dot to become `.text.exit.`
Reviewed By: grimar
Differential Revision: https://reviews.llvm.org/D79600
Summary:
ConstantExprs involving operations on <1 x Ty> could translate into MIR
that failed to verify with:
*** Bad machine code: Reading virtual register without a def ***
The problem was that translate(const Constant &C, Register Reg) had
recursive calls that passed the same Reg in for the translation of a
subexpression, but without updating VMap for the subexpression first as
translate(const Constant &C, Register Reg) expects.
Fix this by using the same translateCopy helper function that we use for
translating Instructions. In some cases this causes extra G_COPY
MIR instructions to be generated.
Fixes https://bugs.llvm.org/show_bug.cgi?id=45576
Reviewers: arsenm, volkan, t.p.northover, aditya_nandakumar
Subscribers: jvesely, wdng, nhaehnle, rovka, hiraditya, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D78378
I have fixed up some places in SelectionDAG::getNode() where we
used to assert that the number of vector elements for two types
are the same. I have changed such cases to assert that the
element counts are the same instead. I've added new tests that
exercise the code paths for all the truncations. All the extend
operations are covered by this existing test:
CodeGen/AArch64/sve-sext-zext.ll
For the ISD::SETCC case I fixed this code path is exercised by
these existing tests:
CodeGen/AArch64/sve-fcmp.ll
CodeGen/AArch64/sve-intrinsics-int-compares-with-imm.ll
Differential Revision: https://reviews.llvm.org/D79399
allocas in LLVM IR have a specified alignment. When that alignment is
specified, the alloca has at least that alignment at runtime.
If the specified type of the alloca has a higher preferred alignment,
SelectionDAG currently ignores that specified alignment, and increases
the alignment. It does this even if it would trigger stack realignment.
I don't think this makes sense, so this patch changes that.
I was looking into this for SVE in particular: for SVE, overaligning
vscale'ed types is extra expensive because it requires realigning the
stack multiple times, or using dynamic allocation. (This currently isn't
implemented.)
I updated the expected assembly for a couple tests; in particular, for
arg-copy-elide.ll, the optimization in question does not increase the
alignment the way SelectionDAG normally would. For the rest, I just
increased the specified alignment on the allocas to match what
SelectionDAG was inferring.
Differential Revision: https://reviews.llvm.org/D79532
It is bad practice to capture by default (via [&] in this case) when
using lambdas, so we should avoid that as much as possible.
This patch fixes that in the getForwardingRegsDefinedByMI
from DwarfDebug module.
Differential Revision: https://reviews.llvm.org/D79616
We should use explicit type instead of auto type deduction when
the type is so obvious. In addition, we remove ambiguity, since auto
type deduction sometimes is not that intuitive, so that could lead
us to some unwanted behavior.
This patch fixes that in the collectCallSiteParameters() from
DwarfDebug module.
Differential Revision: https://reviews.llvm.org/D79624
We have the getNegatibleCost/getNegatedExpression to evaluate the cost and negate the expression.
However, during negating the expression, the cost might change as we are changing the DAG,
and then, hit the assertion if we negated the wrong expression as the cost is not trustful anymore.
This patch is target to remove the getNegatibleCost to avoid the out of sync with getNegatedExpression,
and check the cost during negating the expression. It also reduce the duplicated code between
getNegatibleCost and getNegatedExpression. And fix the crash for the test in D76638
Reviewed By: RKSimon, spatel
Differential Revision: https://reviews.llvm.org/D77319
I don't have any test cases since X86 doesn't return any tied
operands from getUndefRegClearance today. But conceivably we could
want BreakFalseDeps to insert a dependency breaking XOR for
a tied operand in the future.
Currently this code exists in widenScalar for G_MERGE_VALUE
sources. I'm not sure if the existing expansion in widenScalar should
be removed or not. The widenScalar variant tries to extend to the
requested size, but this just uses the original bitwidth.
This patch stores the alignment for ConstantPoolSDNode as an
Align and updates the getConstantPool interface to take a MaybeAlign.
Removing getAlignment() will be done as a follow up.
Differential Revision: https://reviews.llvm.org/D79436
This fixes a verifier failure on a bot:
http://green.lab.llvm.org/green/job/test-suite-verify-machineinstrs-aarch64-O0-g/
```
*** Bad machine code: MBB has duplicate entries in its successor list. ***
- function: foo
- basic block: %bb.5 indirectgoto (0x7fe3d687ca08)
```
One of the GCC torture suite tests (pr70460.c) has an indirectbr instruction
which has duplicate blocks in its destination list.
According to the langref this is allowed:
> Blocks are allowed to occur multiple times in the destination list, though
> this isn’t particularly useful.
(https://www.llvm.org/docs/LangRef.html#indirectbr-instruction)
We don't allow this in MIR. So, when we translate such an instruction, the
verifier screams.
This patch makes `translateIndirectBr` check if a successor has already been
added to a block. If the successor is present, it is skipped rather than added
twice.
Differential Revision: https://reviews.llvm.org/D79609
them in a special text section.
For sampleFDO, because the optimized build uses profile generated from
previous release, previously we couldn't tell a function without profile
was truely cold or just newly created so we had to treat them conservatively
and put them in .text section instead of .text.unlikely. The result was when
we persuing the best performance by locking .text.hot and .text in memory,
we wasted a lot of memory to keep cold functions inside.
In https://reviews.llvm.org/D66374, we introduced profile symbol list to
discriminate functions being cold versus functions being newly added.
This mechanism works quite well for regular use cases in AutoFDO. However,
in some case, we can only have a partial profile when optimizing a target.
The partial profile may be an aggregated profile collected from many targets.
The profile symbol list method used for regular sampleFDO profile is not
applicable to partial profile use case because it may be too large and
introduce many false positives.
To solve the problem for partial profile use case, we provide an option called
--profile-unknown-in-special-section. For functions without profile, we will
still treat them conservatively in compiler optimizations -- for example,
treat them as warm instead of cold in inliner. When we use profile info to
add section prefix for functions, we will discriminate functions known to be
not cold versus functions without profile (being unknown), and we will put
functions being unknown in a special text section called .text.unknown.
Runtime system will have the flexibility to decide where to put the special
section in order to achieve a balance between performance and memory saving.
Differential Revision: https://reviews.llvm.org/D62540
If the SimplifyMultipleUseDemandedBits calls BITCASTs that peek through back to the original type then we can remove the BITCASTs entirely.
Differential Revision: https://reviews.llvm.org/D79572
With a fix to uninitialized EndOffset.
DW_OP_call_ref is the only operation that has an operand which depends
on the DWARF format. The patch fixes handling that operation in DWARF64
units.
Differential Revision: https://reviews.llvm.org/D79501
DW_OP_call_ref is the only operation that has an operand which depends
on the DWARF format. The patch fixes handling that operation in DWARF64
units.
Differential Revision: https://reviews.llvm.org/D79501
CorrectExtraCFGEdges function.
The latter was a workaround for "Various pieces of code" leaving bogus
extra CFG edges in place. Where by "various" it meant only
IfConverter::MergeBlocks, which failed to clear all of the successors
of dead blocks it emptied out. This wouldn't matter a whole lot,
except that the dead blocks remained listed as predecessors of
still-useful blocks, inhibiting optimizations.
This fix slightly changed two thumb tests, because the correct CFG
successors allowed for the "diamond" if-conversion pattern to be
detected, when it could only use "simple" before.
Additionally, the removal of a now-redundant call to analyzeBranch
(with AllowModify=true) in BranchFolder::OptimizeFunction caused a
later check for an empty block in BranchFolder::OptimizeBlock to
fail. Correct this by moving the call to analyzeBranch in
OptimizeBlock higher.
Differential Revision: https://reviews.llvm.org/D79527
Summary:
This helps detect some missed BFI updates during CodeGenPrepare.
This is debug build only and disabled behind a flag.
Fix a missed update in CodeGenPrepare::dupRetToEnableTailCallOpts().
Reviewers: davidxl
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D77417
When peeling out the epilogue we need to ignore illegal phis coming from stages
greater than the producer stage. Otherwise we end up with circular phi
dependencies.
Differential Revision: https://reviews.llvm.org/D79581
Summary:
This patch handles illegal scalable types when lowering IR operations,
addressing several places where the value of isScalableVector() is
ignored.
For types such as <vscale x 8 x i32>, this means splitting the
operations. In this example, we would split it into two
operations of type <vscale x 4 x i32> for the low and high halves.
In cases such as <vscale x 2 x i32>, the elements in the vector
will be promoted. In this case they will be promoted to
i64 (with a vector of type <vscale x 2 x i64>)
Reviewers: sdesmalen, efriedma, huntergr
Reviewed By: efriedma
Subscribers: david-arm, tschuett, hiraditya, rkruppe, psnobl, cfe-commits, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D78812
Calling getShiftAmountTy with LegalTypes set may return a type that's too narrow to hold the shift amount for integer type it's applied to.
Fixes the regression introduced by D79096
Differential Revision: https://reviews.llvm.org/D79405
Try to combine N short vector cast ops into 1 wide vector cast op:
concat (cast X), (cast Y)... -> cast (concat X, Y...)
This is part of solving PR45794:
https://bugs.llvm.org/show_bug.cgi?id=45794
As noted in the code comment, this is uglier than I was hoping because
the opcode determines whether we pass the source or destination type
to isOperationLegalOrCustom(). Also IIUC, there's no way to validate
what the other (dest or src) type is. Without the extra legality check
on that, there's an ARM regression test in:
test/CodeGen/ARM/isel-v8i32-crash.ll
...that will crash trying to lower an unsupported v8f32 to v8i16.
Differential Revision: https://reviews.llvm.org/D79360
This patch adds ORE for MachinePipeliner, so that people can anaylyze
their code using opt-viewer or other tools, then optimize the code to
catch more piplining opportunities.
Reviewed By: bcahoon
Differential Revision: https://reviews.llvm.org/D79368
Make the kind of cost explicit throughout the cost model which,
apart from making the cost clear, will allow the generic parts to
calculate better costs. It will also allow some backends to
approximate and correlate the different costs if they wish. Another
benefit is that it will also help simplify the cost model around
immediate and intrinsic costs, where we currently have multiple APIs.
RFC thread:
http://lists.llvm.org/pipermail/llvm-dev/2020-April/141263.html
Differential Revision: https://reviews.llvm.org/D79002
Summary:
I have fixed several places in getSplatSourceVector and isSplatValue
to work correctly with scalable vectors. I added new support for
the ISD::SPLAT_VECTOR DAG node as one of the obvious cases we can
support with scalable vectors. In other places I have tried to do
the sensible thing, such as bail out for vector types we don't yet
support or don't intend to support.
It's not possible to add IR test cases to cover these changes, since
they are currently only ever exercised on certain targets, e.g.
only X86 targets use the result of getSplatSourceVector. I've
assumed that X86 tests already exist to test these code paths for
fixed vectors. However, I have added some AArch64 unit tests that
test the specific functions I have changed.
Differential revision: https://reviews.llvm.org/D79083
Register live ranges may have had gaps that after coalescing should be
removed. This is done by adding a new segment to the range, and merging
it with neighboring segments. When doing so, do not assume that each
subrange of the register ended at the same index. If a subrange ended
earlier, adding this segment could make the live range invalid.
Instead, if the subrange is not live at the start of the segment,
extend it first.
Today symbol names generated for machine basic block sections use a
unary encoding to reduce bloat. This is essential when every basic block
in the binary is assigned a symbol however with basic block clusters
(rG05192e585ce175b55f2a26b83b4ed7882785c8e6) when we only need to
generate a few non-temporary symbols we can assign more descriptive
names making them more user friendly. With this change -
Cold cluster section for function foo is named "foo.cold"
Exception cluster section for function foo is named "foo.eh"
Other cluster sections identified by their ids are named "foo.ID"
Using this format works well with existing tools. It will demangle as
expected and works with existing symbolizers, profilers and debuggers
out of the box.
$ c++filt _Z3foov.cold
foo() [clone .cold]
$ c++filt _Z3foov.eh
foo() [clone .eh]
$c++filt _Z3foov.1234
foo() [clone 1234]
Tests for basicblock-sections are updated with some cleanup where
appropriate.
Differential Revision: https://reviews.llvm.org/D79221
Before this patch, global variables didn't have their namespace prepended in the Codeview debug symbol stream. This prevented Visual Studio from displaying them in the debugger (they appeared as 'unspecified error')
Differential Revision: https://reviews.llvm.org/D79028
We allocated a suitably aligned frame index so we know that all the values
have ABI alignment.
For MIPS this avoids using pair of lwl + lwr instructions instead of a
single lw. I found this when compiling CHERI pure capability code where
we can't use the lwl/lwr unaligned loads/stores and and were to falling
back to a byte load + shift + or sequence.
This should save a few instructions for MIPS and possibly other backends
that don't have fast unaligned loads/stores.
It also improves code generation for CodeGen/X86/pr34653.ll and
CodeGen/WebAssembly/offset.ll since they can now use aligned loads.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D78999
Sanjay Patel