Craig Topper [Sun, 16 Dec 2018 18:35:55 +0000 (18:35 +0000)]
[X86] Remove truncation handling from EmitTest. Replace it with a DAG combine.
I'd like to try to move a lot of the flag matching out of EmitTest and push it to isel or isel preprocessing. This is a step towards that.
The test-shrink-bug.ll changie is an improvement because we are no longer interfering with test shrink handling in isel.
The pr34137.ll change is a regression, but the IR came from -O0 and was not reduced by InstCombine. So it contains a lot of redundancies like duplicate loads that made it combine poorly.
Sanjay Patel [Sun, 16 Dec 2018 15:05:48 +0000 (15:05 +0000)]
[x86] increment/decrement constant vector with min/max in vsetcc lowering (PR39859)
This is part of fixing PR39859:
https://bugs.llvm.org/show_bug.cgi?id=39859
We have a crippled vector ISA, so we have to invert a typical fold and create min/max here.
As discussed in the bug report, we can probably do better by using saturating subtract when
it's available, but we should have this improvement for the min/max patterns regardless.
Sanjay Patel [Sun, 16 Dec 2018 14:57:04 +0000 (14:57 +0000)]
[DAGCombiner] allow hoisting vector bitwise logic ahead of truncates
The transform performs a bitwise logic op in a wider type followed by
truncate when both inputs are truncated from the same source type:
logic_op (truncate x), (truncate y) --> truncate (logic_op x, y)
There are a bunch of other checks that should prevent doing this when
it might be harmful.
We already do this transform for scalars in this spot. The vector
limitation was shared with a check for the case when the operands are
extended. I'm not sure if that limit is needed either, but that would
be a separate patch.
Simon Pilgrim [Sun, 16 Dec 2018 13:33:37 +0000 (13:33 +0000)]
[SelectionDAG] Add FSHL/FSHR support to computeKnownBits
Also exposes an issue in DAGCombiner::visitFunnelShift where we were assuming the shift amount had the result type (after legalization it'll have the targets shift amount type).
Simon Pilgrim [Sat, 15 Dec 2018 19:43:44 +0000 (19:43 +0000)]
[X86] Lower to SHLD/SHRD on slow machines for optsize
Use consistent rules for when to lower to SHLD/SHRD for slow machines - fixes a weird issue where funnel shift gets expanded but then X86ISelLowering's combineOr sees the optsize and combines to SHLD/SHRD, but now with the modulo amount guard......
Summary:
Make machine PHIs optimization to work for single value register taken from
several different copies. This is the first step to fix PR38917. This change
allows to get rid of redundant PHIs (see opt_phis2.mir test) to make
the subsequent optimizations (like CSE) possible and simpler.
For instance, before this patch the code like this:
%b = COPY %z
...
%a = PHI %bb1, %a; %bb2, %b
could be optimized to:
%a = %b
but the code like this:
%c = COPY %z
...
%b = COPY %z
...
%a = PHI %bb1, %a; %bb2, %b; %bb3, %c
would remain unchanged.
With this patch the latter case will be optimized:
%a = %z```.
Committed on behalf of: Anton Afanasyev anton.a.afanasyev@gmail.com
Nikita Popov [Sat, 15 Dec 2018 10:37:01 +0000 (10:37 +0000)]
[InstSimplify] Add tests for saturating add/sub + icmp; NFC
If a saturating add/sub with a constant operand is compared to
another constant, we should be able to determine that the condition
is always true/false in some cases (but currently don't).
Florian Hahn [Sat, 15 Dec 2018 01:32:58 +0000 (01:32 +0000)]
[SILoadStoreOptimizer] Use std::abs to avoid truncation.
Using regular abs() causes the following warning
error: absolute value function 'abs' given an argument of type 'int64_t' (aka 'long') but has parameter of type 'int' which may cause truncation of value [-Werror,-Wabsolute-value]
(uint32_t)abs(Dist) > MaxDist) {
^
lib/Target/AMDGPU/SILoadStoreOptimizer.cpp:1369:19: note: use function 'std::abs' instead
which causes a bot to fail:
http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux/builds/18284/steps/bootstrap%20clang/logs/stdio
Craig Topper [Sat, 15 Dec 2018 01:07:19 +0000 (01:07 +0000)]
[X86] Rename hasNoSignedComparisonUses to hasNoSignFlagUses. Add the instruction that only modify the O flag to the waiver list.
The only caller of this turns CMP with 0 into TEST. CMP with 0 and TEST both set OF to 0 so we should have no issues with instructions that only use OF.
Though I don't think there's any reason we would read just OF after a compare with 0 anyway. So this probably isn't an observable change.
Craig Topper [Sat, 15 Dec 2018 01:07:16 +0000 (01:07 +0000)]
[X86] Make hasNoCarryFlagUses/hasNoSignedComparisonUses take an SDValue that indicates which result is the flag result. NFCI
hasNoCarryFlagUses hardcoded that the flag result is 1 and used that to filter which uses were of interest. hasNoSignedComparisonUses just assumes the only result is flags and checks whether any user of the node is a CopyToReg instruction.
After this patch we now do a result number check in both and rely on the caller to provide the result number.
This shouldn't change behavior it was just an odd difference between the two functions that I noticed.
Heejin Ahn [Sat, 15 Dec 2018 00:58:12 +0000 (00:58 +0000)]
[WebAssembly] Check if the section order is correct
Summary:
This patch checks if the section order is correct when reading a wasm
object file in `WasmObjectFile` and converting YAML to wasm object in
yaml2wasm. (It is not possible to check when reading YAML because it is
handled exclusively by the YAML reader.)
This checks the ordering of all known sections (core sections + known
custom sections). This also adds section ID DataCount section that will
be scheduled to be added in near future.
Florian Hahn [Sat, 15 Dec 2018 00:32:38 +0000 (00:32 +0000)]
[NewGVN] Update use counts for SSA copies when replacing them by their operands.
The current code relies on LeaderUseCount to determine if we can remove
an SSA copy, but in that the LeaderUseCount does not refer to the SSA
copy. If a SSA copy is a dominating leader, we use the operand as dominating
leader instead. This means we removed a user of a ssa copy and we should
decrement its use count, so we can remove the ssa copy once it becomes dead.
Vedant Kumar [Sat, 15 Dec 2018 00:03:33 +0000 (00:03 +0000)]
[Util] Refer to [s|z]exts of args when converting dbg.declares (fix PR35400)
When converting dbg.declares, if the described value is a [s|z]ext,
refer to the ext directly instead of referring to its operand.
This fixes a narrowing bug (the debugger got the sign of a variable
wrong, see llvm.org/PR35400).
The main reason to refer to the ext's operand was that an optimization
may remove the ext itself, leading to a dropped variable. Now that
InstCombine has been taught to use replaceAllDbgUsesWith (r336451), this
is less of a concern. Other passes can/should adopt this API as needed
to fix dropped variable bugs.
Artem Belevich [Fri, 14 Dec 2018 23:53:06 +0000 (23:53 +0000)]
[NVPTX] Lower instructions that expand into libcalls.
The change is an effort to split and refactor abandoned
D34708 into smaller parts.
Here the behaviour of unsupported instructions is changed
to match the behaviour of explicit intrinsics calls.
Currently LLVM crashes with:
> Assertion getInstruction() && "Not a call or invoke instruction!" failed.
With this patch LLVM produces a more sensible error message:
> Cannot select: ... i32 = ExternalSymbol'__foobar'
David Blaikie [Fri, 14 Dec 2018 22:44:46 +0000 (22:44 +0000)]
DebugInfo: Avoid using split DWARF when the split unit would be empty.
In ThinLTO many split CUs may be effectively empty because of the lack
of support for cross-unit references in split DWARF.
Using a split unit in those cases is just a waste/overhead - and turned
out to be one contributor to a significant symbolizer performance issue
when global variable debug info was being imported (see r348416 for the
primary fix) due to symbolizers seeing CUs with no ranges, assuming
there might still be addresses covered and walking into the split CU to
see if there are any ranges (when that split CU was in a DWP file, that
meant loading the DWP and its index, the index was extra large because
of all these fractured/empty CUs... and so was very expensive to load).
(the 3rd fix which will follow, is to assume that a CU with no ranges is
empty rather than merely missing its CU level range data - and to not
walk into its DIEs (split or otherwise) in search of address information
that is generally not present)
Michael Kruse [Fri, 14 Dec 2018 19:45:43 +0000 (19:45 +0000)]
[TransformWarning] Do not warn missed transformations in optnone functions.
Optimization transformations are intentionally disabled by the 'optnone'
function attribute. Therefore do not warn if transformation metadata is
still present.
Using the legacy pass manager structure, the `skipFunction` method takes
care for the optnone attribute (already called before this patch). For
the new pass manager, there is no equivalent, so we check for the
'optnone' attribute manually.
Michael Kruse [Fri, 14 Dec 2018 18:15:11 +0000 (18:15 +0000)]
[Transforms] Preserve metadata when converting invoke to call.
The `changeToCall` function did not preserve the invoke's metadata.
Currently, there is probably no metadata that depends on being applied
on a CallInst or InvokeInst. Therefore we can replace the instruction's
metadata.
This fixes http://llvm.org/PR39994
Suggested-by: Moritz Kreutzer <moritz.kreutzer@siemens.com>
Differential Revision: https://reviews.llvm.org/D55666
Zachary Turner [Fri, 14 Dec 2018 18:10:13 +0000 (18:10 +0000)]
[MS Demangler] Fail gracefully on invalid pointer types.
Once we detect a 'P', we know we a pointer type is upcoming, so
we make some assumptions about the output that follows. If those
assumptions didn't hold, we would assert. Instead, we should
fail gracefully and propagate the error up.
Daniel Sanders [Fri, 14 Dec 2018 17:50:14 +0000 (17:50 +0000)]
[globalisel][combiner] Make the CombinerChangeObserver a MachineFunction::Delegate
Summary:
This allows us to register it with the MachineFunction delegate and be
notified automatically about erasure and creation of instructions. However,
we still need explicit notification for modifications such as those caused
by setReg() or replaceRegWith().
There is a catch with this though. The notification for creation is
delivered before any operands can be added. While appropriate for
scheduling combiner work. This is unfortunate for debug output since an
opcode by itself doesn't provide sufficient information on what happened.
As a result, the work list remembers the instructions (when debug output is
requested) and emits a more complete dump later.
Another nit is that the MachineFunction::Delegate provides const pointers
which is inconvenient since we want to use it to schedule future
modification. To resolve this GISelWorkList now has an optional pointer to
the MachineFunction which describes the scope of the work it is permitted
to schedule. If a given MachineInstr* is in this function then it is
permitted to schedule work to be performed on the MachineInstr's. An
alternative to this would be to remove the const from the
MachineFunction::Delegate interface, however delegates are not permitted
to modify the MachineInstr's they receive.
In addition to this, the observer has three interface changes.
* erasedInstr() is now erasingInstr() to indicate it is about to be erased
but still exists at the moment.
* changingInstr() and changedInstr() have been added to report changes
before and after they are made. This allows us to trace the changes
in the debug output.
* As a convenience changingAllUsesOfReg() and
finishedChangingAllUsesOfReg() will report changingInstr() and
changedInstr() for each use of a given register. This is primarily useful
for changes caused by MachineRegisterInfo::replaceRegWith()
With this in place, both combine rules have been updated to report their
changes to the observer.
Finally, make some cosmetic changes to the debug output and make Combiner
and CombinerHelp
Implement options in clang to enable recording the driver command-line
in an ELF section.
Implement a new special named metadata, llvm.commandline, to support
frontends embedding their command-line options in IR/ASM/ELF.
This differs from the GCC implementation in some key ways:
* In GCC there is only one command-line possible per compilation-unit,
in LLVM it mirrors llvm.ident and multiple are allowed.
* In GCC individual options are separated by NULL bytes, in LLVM entire
command-lines are separated by NULL bytes. The advantage of the GCC
approach is to clearly delineate options in the face of embedded
spaces. The advantage of the LLVM approach is to support merging
multiple command-lines unambiguously, while handling embedded spaces
with escaping.
It costs nothing to spill an IMPLICIT_DEF value (the only spill code that's
generated is a KILL of the value), so when creating split constraints if the
live-out value is IMPLICIT_DEF the exit constraint should be DontCare instead
of PrefReg.
Simon Tatham [Fri, 14 Dec 2018 11:39:55 +0000 (11:39 +0000)]
[TableGen:AsmWriter] Cope with consecutive tied operands.
When you define an instruction alias as a subclass of InstAlias, you
specify all the MC operands for the instruction it expands to, except
for operands that are tied to a previous one, which you leave out in
the expectation that the Tablegen output code will fill them in
automatically.
But the code in Tablegen's AsmWriter backend that skips over a tied
operand was doing it using 'if' instead of 'while', because it wasn't
expecting to find two tied operands in sequence.
So if an instruction updates a pair of registers in place, so that its
MC representation has two input operands tied to the output ones (for
example, Arm's UMLAL instruction), then any alias which wants to
expand to a special case of that instruction is likely to fail to
match, because the indices of subsequent operands will be off by one
in the generated printAliasInstr function.
This patch re-indents some existing code, so it's clearest when
viewed as a diff with whitespace changes ignored.
Craig Topper [Fri, 14 Dec 2018 08:28:24 +0000 (08:28 +0000)]
[DAGCombiner][X86] Prevent visitSIGN_EXTEND from returning N when (sext (setcc)) already has the target desired type for the setcc
Summary:
If the setcc already has the target desired type we can reach the getSetCC/getSExtOrTrunc after the MatchingVecType check with the exact same types as the nodes we started with. This causes those causes VsetCC to be CSEd to N0 and the getSExtOrTrunc will CSE to N. When we return N, the caller will think that meant we called CombineTo and did our own worklist management. But that's not what happened. This prevents target hooks from being called for the node.
To fix this, I've now returned SDValue if the setcc is already the desired type. But to avoid some regressions in X86 I've had to disable one of the target combines that wasn't being reached before in the case of a (sext (setcc)). If we get vector widening legalization enabled that entire function will be deleted anyway so hopefully this is only for the short term.
Fangrui Song [Fri, 14 Dec 2018 07:46:58 +0000 (07:46 +0000)]
[Object] Rename getRelrRelocationType to getRelativeRelocationType
Summary:
The two utility functions were added in D47919 to support SHT_RELR.
However, these are just relative relocations types and are't
necessarily be named Relr.
Petr Hosek [Fri, 14 Dec 2018 05:56:20 +0000 (05:56 +0000)]
[llvm-xray] Store offset pointers in temporaries
DataExtractor::getU64 modifies the OffsetPtr which also pass to
RelocateOrElse which breaks on Windows. This addresses the issue
introduced in r349120.
Petr Hosek [Fri, 14 Dec 2018 01:37:56 +0000 (01:37 +0000)]
[llvm-xray] Support for PIE
When the instrumented binary is linked as PIE, we need to apply the
relative relocations to sleds. This is handled by the dynamic linker
at runtime, but when processing the file we have to do it ourselves.
Alex Lorenz [Fri, 14 Dec 2018 01:14:10 +0000 (01:14 +0000)]
[macho] save the SDK version stored in module metadata into the version min and
build version load commands in the object file
This commit introduces a new metadata node called "SDK Version". It will be set
by the frontend to mark the platform SDK (macOS/iOS/etc) version which was used
during that particular compilation.
This node is used when machine code is emitted, by either saving the SDK version
into the appropriate macho load command (version min/build version), or by
emitting the assembly for these load commands with the SDK version specified as
well.
The assembly for both load commands is extended by allowing it to contain the
sdk_version X, Y [, Z] trailing directive to represent the SDK version
respectively.
Sanjay Patel [Fri, 14 Dec 2018 00:09:08 +0000 (00:09 +0000)]
[DAGCombiner] clean up visitEXTRACT_VECTOR_ELT
This isn't quite NFC, but I don't know how to expose
any outward diffs from these changes. Mostly, this
was confusing because it used 'VT' to refer to the
operand type rather the usual type of the input node.
There's also a large block at the end that is dedicated
solely to matching loads, but that wasn't obvious. This
could probably be split up into separate functions to
make it easier to see.
It's still not clear to me when we make certain transforms
because the legality and constant conditions are
intertwined in a way that might be improved.
Craig Topper [Thu, 13 Dec 2018 23:55:30 +0000 (23:55 +0000)]
[X86] Demote EmitTest to a helper function of EmitCmp. Route all callers except EmitCmp through EmitCmp.
This requires the two callers to manifest a 0 to make EmitCmp call EmitTest.
I'm looking into changing how we combine TEST and flag setting instructions to not be part of lowering. And instead be part of DAG combine or isel. Which will mean EmitTest will probably become gutted and maybe disappear entirely.
Evandro Menezes [Thu, 13 Dec 2018 23:19:46 +0000 (23:19 +0000)]
[AArch64] Fix Exynos predicates (NFC)
Fix the logic in the definition of the `ExynosShiftExPred` as a more
specific version of `ExynosShiftPred`. But, since `ExynosShiftExPred` is
not used yet, this change has NFC.
Wei Mi [Thu, 13 Dec 2018 21:51:42 +0000 (21:51 +0000)]
[SampleFDO] handle ProfileSampleAccurate when initializing function entry count
ProfileSampleAccurate is used to indicate the profile has exact match to the
code to be optimized.
Previously ProfileSampleAccurate is handled in ProfileSummaryInfo::isColdCallSite
and ProfileSummaryInfo::isColdBlock. A better solution is to initialize function
entry count to 0 when ProfileSampleAccurate is true, so we don't have to handle
ProfileSampleAccurate in multiple places.
Nikita Popov [Thu, 13 Dec 2018 20:04:27 +0000 (20:04 +0000)]
Reapply "[MemCpyOpt] memset->memcpy forwarding with undef tail"
Currently memcpyopt optimizes cases like
memset(a, byte, N);
memcpy(b, a, M);
to
memset(a, byte, N);
memset(b, byte, M);
if M <= N. Often this allows further simplifications down the line,
which drop the first memset entirely.
This patch extends this optimization for the case where M > N, but we
know that the bytes a[N..M] are undef due to alloca/lifetime.start.
This situation arises relatively often for Rust code, because Rust does
not initialize trailing structure padding and loves to insert redundant
memcpys. This also fixes https://bugs.llvm.org/show_bug.cgi?id=39844.
The previous version of this patch did not perform dependency checking
properly: While the dependency is checked at the position of the memset,
the used size must be that of the memcpy. Previously the size of the
memset was used, which missed modification in the region
MemSetSize..CopySize, resulting in miscompiles. The added tests cover
variations of this issue.
Easwaran Raman [Thu, 13 Dec 2018 19:54:27 +0000 (19:54 +0000)]
[ThinLTO] Compute synthetic function entry count
Summary:
This patch computes the synthetic function entry count on the whole
program callgraph (based on module summary) and writes the entry counts
to the summary. After function importing, this count gets attached to
the IR as metadata. Since it adds a new field to the summary, this bumps
up the version.
Mircea Trofin [Thu, 13 Dec 2018 19:40:59 +0000 (19:40 +0000)]
[llvm] Address base discriminator overflow in X86DiscriminateMemOps
Summary:
Macros are expanded on a single line. In case of large expansions,
with sufficiently many instructions with memory operands (and when
-fdebug-info-for-profiling is requested), we may be unable to generate
new base discriminator values - new values overflow (base
discriminators may not be larger than 2^12).
This CL warns instead of asserting in such a case. A subsequent CL
will add APIs to check for overflow before creating new debug info.
Jordan Rupprecht [Thu, 13 Dec 2018 19:40:12 +0000 (19:40 +0000)]
[llvm-size][libobject] Add explicit "inTextSegment" methods similar to "isText" section methods to calculate size correctly.
Summary:
llvm-size uses "isText()" etc. which seem to indicate whether the section contains code-like things, not whether or not it will actually go in the text segment when in a fully linked executable.
The unit test added (elf-sizes.test) shows some types of sections that cause discrepencies versus the GNU size tool. llvm-size is not correctly reporting sizes of things mapping to text/data segments, at least for ELF files.
Craig Topper [Thu, 13 Dec 2018 19:08:10 +0000 (19:08 +0000)]
[CostModel][X86] Don't count 2 shuffles on the last level of a pairwise arithmetic or min/max reduction
This is split from D55452 with the correct patch this time.
Pairwise reductions require two shuffles on every level but the last. On the last level the two shuffles are <1, u, u, u...> and <0, u, u, u...>, but <0, u, u, u...> will be dropped by InstCombine/DAGCombine as being an identity shuffle.
projects / llvm / log