This patch defines an interleaved-load-combine pass. The pass searches
for ShuffleVector instructions that represent interleaved loads. Matches are
converted such that they will be captured by the InterleavedAccessPass.
The pass extends LLVMs capabilities to use target specific instruction
selection of interleaved load patterns (e.g.: ld4 on Aarch64
architectures).
Differential Revision: https://reviews.llvm.org/D52653
llvm-svn: 347208
Every Analysis pass has a get method that returns a reference of the Result of
the Analysis, for example, BlockFrequencyInfo
&BlockFrequencyInfoWrapperPass::getBFI(). I believe that
ProfileSummaryInfo::getPSI() is the only exception to that, as it was returning
a pointer.
Another change is renaming isHotBB and isColdBB to isHotBlock and isColdBlock,
respectively. Most methods use BB as the argument of variable names while
methods usually refer to Basic Blocks as Blocks, instead of BB. For example,
Function::getEntryBlock, Loop:getExitBlock, etc.
I also fixed one of the comments.
Patch by Rodrigo Caetano Rocha!
Differential Revision: https://reviews.llvm.org/D54669
llvm-svn: 347182
Sadly, this duplicates (twice) the logic from InstSimplify. There
might be some way to at least share the DAG versions of the code,
but copying the folds seems to be the standard method to ensure
that we don't miss these folds.
Unlike in IR, we don't run DAGCombiner to fixpoint, so there's no
way to ensure that we do these kinds of simplifications unless the
code is repeated at node creation time and during combines.
There were other tests that would become worthless with this
improvement that I changed as pre-commits:
rL347161
rL347164
rL347165
rL347166
rL347167
I'm not sure how to salvage the remaining tests (diffs in this patch).
So the x86 tests verify that the new code is working as intended.
The AMDGPU test is actually similar to my motivating case: we have
some undef value that has survived to machine IR in an x86 test, and
then it gets folded in some weird way, or we crash if we don't transfer
the undef flag. But we would have been better off never getting to that
point by doing these simplifications.
This will lead back to PR32023 someday...
https://bugs.llvm.org/show_bug.cgi?id=32023
llvm-svn: 347170
For example, on X86 we emit a sign_extend_vector_inreg from LowerLoad and without sse4.1 this node will need further legalization. Previously this sign_extend_vector_inreg was being custom lowered during DAG legalization instead of vector op legalization.
Unfortunately, this doesn't seem to matter for the output of any existing lit tests.
llvm-svn: 347094
Summary:
Experience has shown that the functionality is useful. It makes linking
optimized clang with debug info for me a lot faster, 20s to 13s. The
type merging phase of PDB writing goes from 10s to 3s.
This removes the LLVM cl::opt and replaces it with a metadata flag.
After this change, users can do the following to use ghash:
- add -gcodeview-ghash to compiler flags
- replace /DEBUG with /DEBUG:GHASH in linker flags
Reviewers: zturner, hans, thakis, takuto.ikuta
Subscribers: aprantl, hiraditya, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D54370
llvm-svn: 347072
Summary:
`throw` instruction is a terminator in wasm, but BBs were not splitted
after `throw` instructions, causing machine instruction verifier to
fail.
This patch
- Splits BBs after `throw` instructions in WasmEHPrepare and adding an
unreachable instruction after `throw`, which will be deleted in
LateEHPrepare pass
- Refactors WasmEHPrepare into two member functions
- Changes the semantics of `eraseBBsAndChildren` in LateEHPrepare pass
to match that of WasmEHPrepare pass, which is newly added. Now
`eraseBBsAndChildren` does not delete BBs with remaining predecessors.
- Fixes style nits, making static function names conform to clang-tidy
- Re-enables the test temporarily disabled by rL346840 && rL346845
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D54571
llvm-svn: 347003
The machine scheduler currently biases register copies to/from
physical registers to be closer to their point of use / def to
minimize their live ranges. This change extends this to also physical
register assignments from immediate values.
This causes a reduction in reduction in overall register pressure and
minor reduction in spills and indirectly fixes an out-of-registers
assertion (PR39391).
Most test changes are from minor instruction reorderings and register
name selection changes and direct consequences of that.
Reviewers: MatzeB, qcolombet, myatsina, pcc
Subscribers: nemanjai, jvesely, nhaehnle, eraman, hiraditya,
javed.absar, arphaman, jfb, jsji, llvm-commits
Differential Revision: https://reviews.llvm.org/D54218
llvm-svn: 346894
Summary:
This adds support for the 'event section' specified in the exception
handling proposal. (This was named 'exception section' first, but later
renamed to 'event section' to take possibilities of other kinds of
events into consideration. But currently we only store exception info in
this section.)
The event section is added between the global section and the export
section. This is for ease of validation per request of the V8 team.
This patch:
- Creates the event symbol type, which is a weak symbol
- Makes 'throw' instruction take the event symbol '__cpp_exception'
- Adds relocation support for events
- Adds WasmObjectWriter / WasmObjectFile (Reader) support
- Adds obj2yaml / yaml2obj support
- Adds '.eventtype' printing support
Reviewers: dschuff, sbc100, aardappel
Subscribers: jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D54096
llvm-svn: 346825
The scan was incorrectly skipping the first instruction, so a register
could appear to be dead when it was actually live. This eventually leads
to a machine verifier failure and miscompile in arm-ldst-opt.
Differential Revision: https://reviews.llvm.org/D54491
llvm-svn: 346821
We already determine a bunch of information about an MBB in
getMachineOutlinerMBBFlags. We can reuse that information to avoid calculating
things that must be false/true.
The first thing we can easily check is if an outlined sequence could ever
contain calls. There's no reason to walk over the outlined range, checking for
calls, if we already know that there are no calls in the block containing the
sequence.
llvm-svn: 346809
Since we never outline anything with fewer than 2 occurrences, there's no
reason to compute cost model information if there's less than that.
llvm-svn: 346803
Legalizer used to request an ext load from i8 to i1 when promoting
vector element type to i8. Fixed.
Differential Revision: https://reviews.llvm.org/D54440
llvm-svn: 346795
Summary:
The comment refers to the field as "Kind:". However, in gdb,
https://sourceware.org/gdb//onlinedocs/gdb/Index-Section-Format.html names it "attributes",
gdb/dwarf2read.c:dw2_symtab_iter_next refers to the whole value as "cu_index_and_attrs"
Change it to `Attributes:` for consistency.
Reviewers: dblaikie
Reviewed By: dblaikie
Subscribers: aprantl, JDevlieghere, arphaman, llvm-commits
Differential Revision: https://reviews.llvm.org/D54480
llvm-svn: 346790
Summary:
Ranges base address specifiers can save a lot of object size in
relocation records especially in optimized builds.
For an optimized self-host build of Clang with split DWARF and debug
info compression in object files, but uncompressed debug info in the
executable, this change produces about 18% smaller object files and 6%
larger executable.
While it would've been nice to turn this on by default, gold's 32 bit
gdb-index support crashes on this input & I don't think there's any
perfect heuristic to implement solely in LLVM that would suffice - so
we'll need a flag one way or another (also possible people might want to
aggressively optimized for executable size that contains debug info
(even with compression this would still come at some cost to executable
size)) - so let's plumb it through.
Differential Revision: https://reviews.llvm.org/D54242
llvm-svn: 346788
Previously, the extend_vector_inreg opcode required their input register to be the same total width as their output. But this doesn't match up with how the X86 instructions are defined. For X86 the input just needs to be a legal type with at least enough elements to cover the output.
This patch weakens the check on these nodes and allows them to be used as long as they have more input elements than output elements. I haven't changed type legalization behavior so it will still create them with matching input and output sizes.
X86 will custom legalize these nodes by shrinking the input to be a 128 bit vector and once we've done that we treat them as legal operations. We still have one case during type legalization where we must custom handle v64i8 on avx512f targets without avx512bw where v64i8 isn't a legal type. In this case we will custom type legalize to a *extend_vector_inreg with a v16i8 input. After that the input is a legal type so type legalization should ignore the node and doesn't need to know about the relaxed restriction. We are no longer allowed to use the default expansion for these nodes during vector op legalization since the default expansion uses a shuffle which required the widths to match. Custom legalization for all types will prevent us from reaching the default expansion code.
I believe DAG combine works correctly with the released restriction because it doesn't check the number of input elements.
The rest of the patch is changing X86 to use either the vector_inreg nodes or the regular zero_extend/sign_extend nodes. I had to add additional isel patterns to handle any_extend during isel since simplifydemandedbits can create them at any time so we can't legalize to zero_extend before isel. We don't yet create any_extend_vector_inreg in simplifydemandedbits.
Differential Revision: https://reviews.llvm.org/D54346
llvm-svn: 346784
The IEEE-754 Standard makes it clear that fneg(x) and
fsub(-0.0, x) are two different operations. The former is a bitwise
operation, while the latter is an arithmetic operation. This patch
creates a dedicated FNeg IR Instruction to model that behavior.
Differential Revision: https://reviews.llvm.org/D53877
llvm-svn: 346774
Flags variable was not initialized and later used (both isMBBSafeToOutlineFrom
implementations assume it's initialized), which breaks
test/CodeGen/AArch64/machine-outliner.mir. under memory sanitizer:
MemorySanitizer: use-of-uninitialized-value
#0 in llvm::AArch64InstrInfo::getOutliningType(llvm::MachineInstrBundleIterator<llvm::MachineInstr, false>&, unsigned int) const llvm/lib/Target/AArch64/AArch64InstrInfo.cpp:5494:9
#1 in (anonymous namespace)::InstructionMapper::convertToUnsignedVec(llvm::MachineBasicBlock&, llvm::TargetInstrInfo const&) llvm/lib/CodeGen/MachineOutliner.cpp:772:19
#2 in (anonymous namespace)::MachineOutliner::populateMapper((anonymous namespace)::InstructionMapper&, llvm::Module&, llvm::MachineModuleInfo&) llvm/lib/CodeGen/MachineOutliner.cpp:1543:14
#3 in (anonymous namespace)::MachineOutliner::runOnModule(llvm::Module&) llvm/lib/CodeGen/MachineOutliner.cpp:1645:3
#4 in (anonymous namespace)::MPPassManager::runOnModule(llvm::Module&) llvm/lib/IR/LegacyPassManager.cpp:1744:27
#5 in llvm::legacy::PassManagerImpl::run(llvm::Module&) llvm/lib/IR/LegacyPassManager.cpp:1857:44
#6 in compileModule(char**, llvm::LLVMContext&) llvm/tools/llc/llc.cpp:597:8
llvm-svn: 346761
It should be ok to create a new build_vector after legal operations so long as it doesn't cause an infinite loop in DAG combiner.
Unfortunately, X86's custom constant folding in combineVSZext is hiding any test changes from this. But I'm trying to get to a point where that X86 specific code isn't necessary at all.
Differential Revision: https://reviews.llvm.org/D54285
llvm-svn: 346728
Instead of returning Flags, return true if the MBB is safe to outline from.
This lets us check for unsafe situations, like say, in AArch64, X17 is live
across a MBB without being defined in that MBB. In that case, there's no point
in performing an instruction mapping.
llvm-svn: 346718
Summary:
D44571 changed SimplificationTracker to use SmallSetVector to keep phi nodes. As a result, when the number of phi nodes is large, the build time performance suffers badly. When building for power pc, we have a case where there are more than 600.000 nodes, and it takes too long to compile.
In this change, I partially revert D44571 to use SmallPtrSet, which does an acceptable job with any number of elements. In the original patch, having a deterministic iteration order was mentioned as a motivation, however I think it only applies to the nodes already matched in MatchPhiSet method, which I did not touch.
Reviewers: bjope, skatkov
Reviewed By: bjope, skatkov
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54007
llvm-svn: 346710
Remove another bit of unused configuration potential from GCStrategy. It's not entirely clear what the intention here was, but from the docs, it sounds like this may have been subsumed by patchable call support.
Note: This change is deliberately small to make it clear that while implemented, there's nothing using the option. A following NFC will do most of the simplifications.
llvm-svn: 346701
The function only checks that instruction reads a super-register
containing requested physical register. In case if a sub-register
if being read that is also a use of a super-reg, so added the check.
In particular MI->readsRegister() is broken because of the missing
check. The resulting check is essentially regsOverlap().
Differential Revision: https://reviews.llvm.org/D54128
llvm-svn: 346686
There's no way they can overlap in this case.
This can save a few iterations when the candidate is close to the beginning
of a MachineBasicBlock. It's particularly useful when the average length of
a MachineBasicBlock in the program is small.
llvm-svn: 346682
Summary:
Handle extra output from index loads in cases where we wish to
forward a load value directly from a preceeding store.
Fixes PR39571.
Reviewers: peter.smith, rengolin
Subscribers: javed.absar, hiraditya, arphaman, llvm-commits
Differential Revision: https://reviews.llvm.org/D54265
llvm-svn: 346654
The custom root mechanism didn't actually do anything. ShadowStackGC, the only one which used it, just removed the gcroots before they reached the normal lowering in SelectionDAG. As a result, the state flag had no value.
llvm-svn: 346632
The GCStrategy provides three configuration options were are largely redundant.
1) Support for conditionally lowering gcread and gcwrite to loads and stores. This is redundant since any GC which wished to use these abstractions would lower them out of existance before the built in lowering anyways. As such, there's no need to have the lowering being conditional.
2) Conditional initialization for allocas marked via gcroot. Semantically, roots have to be initialized before first potential use. Arguably, the frontend really should have responsibility for that, but the old API allowed the frontend to ignore this detail. Only one builtin GC used the non-initializing mode. Since no one to my knowledge actually uses the ErlangGC strategy, I decide the slight pessimization was worth the simplicity. If that turns out to be problematic, we can always improve the insertion algorithm to detect more existing initializing stores.
llvm-svn: 346621
This is a long-awaited follow-up suggested in D33578. Since then, we've picked up even more
opportunities for vector narrowing from changes like D53784, so there are a lot of test diffs.
Apart from 2-3 strange cases, these are all wins.
I've structured this to be no-functional-change-intended for any target except for x86
because I couldn't tell if AArch64, ARM, and AMDGPU would improve or not. All of those
targets have existing regression tests (4, 4, 10 files respectively) that would be
affected. Also, Hexagon overrides the shouldReduceLoadWidth() hook, but doesn't show
any regression test diffs. The trade-off is deciding if an extra vector load is better
than a single wide load + extract_subvector.
For x86, this is almost always better (on paper at least) because we often can fold
loads into subsequent ops and not increase the official instruction count. There's also
some unknown -- but potentially large -- benefit from using narrower vector ops if wide
ops are implemented with multiple uops and/or frequency throttling is avoided.
Differential Revision: https://reviews.llvm.org/D54073
llvm-svn: 346595
Martin Elshuber