Alina Sbirlea [Tue, 29 Jan 2019 22:33:20 +0000 (22:33 +0000)]
Check bool attribute value in getOptionalBoolLoopAttribute.
Summary:
Check the bool value of the attribute in getOptionalBoolLoopAttribute
not just its existance.
Eliminates the warning noise generated when vectorization is explicitly disabled.
Shoaib Meenai [Tue, 29 Jan 2019 22:17:51 +0000 (22:17 +0000)]
[docs] Prevent O0 optnone for opt input
If we just compile with -O0, clang will add optnone attributes
everywhere, so opt won't actually be able to perform any passes.
Instruct clang to not emit the optnone so opt can do its thing.
Sanjay Patel [Tue, 29 Jan 2019 20:02:45 +0000 (20:02 +0000)]
[InstCombine] canonicalize cmp/select form of uadd saturate with constant
I'm circling back around to a loose end from D51929.
The backend (either CGP or DAG) doesn't recognize this pattern, so we end up with different asm for these IR variants.
Regardless of any future changes to canonicalize to saturation/overflow intrinsics, we want to get raw IR variations
into the minimal number of raw IR forms. If/when we can canonicalize to intrinsics, that will make that step easier.
Sanjay Patel [Tue, 29 Jan 2019 19:13:39 +0000 (19:13 +0000)]
[DAGCombiner] fold extract_subvector of extract_subvector
This is the sibling fold for insert-of-insert that was added with D56604.
Now that we have x86 shuffle narrowing (D57156), this change shows improvements for
lots of AVX512 reduction code (not sure that we would ever expect extract-of-extract otherwise).
There's a small regression in some of the partial-permute tests (extracting followed by splat).
That is tracked by PR40500:
https://bugs.llvm.org/show_bug.cgi?id=40500
Andrea Di Biagio [Tue, 29 Jan 2019 16:47:27 +0000 (16:47 +0000)]
[X86][Btver2] Improved latency/throughput model for scalar int-to-float conversions.
Account for bypass delays when computing the latency of scalar int-to-float
conversions.
On Jaguar we need to account for an extra 6cy latency (see AMD fam16h SOG).
This patch also fixes the number of micropcodes for the register-memory variants
of scalar int-to-float conversions.
James Y Knight [Tue, 29 Jan 2019 16:37:27 +0000 (16:37 +0000)]
Adjust documentation for git migration.
This fixes most references to the paths:
llvm.org/svn/
llvm.org/git/
llvm.org/viewvc/
github.com/llvm-mirror/
github.com/llvm-project/
reviews.llvm.org/diffusion/
to instead point to https://github.com/llvm/llvm-project.
This is *not* a trivial substitution, because additionally, all the
checkout instructions had to be migrated to instruct users on how to
use the monorepo layout, setting LLVM_ENABLE_PROJECTS instead of
checking out various projects into various subdirectories.
I've attempted to not change any scripts here, only documentation. The
scripts will have to be addressed separately.
Additionally, I've deleted one document which appeared to be outdated
and unneeded:
lldb/docs/building-with-debug-llvm.txt
Jordan Rupprecht [Tue, 29 Jan 2019 15:05:38 +0000 (15:05 +0000)]
[llvm-objcopy] Implement --set-section-flags.
Summary:
--set-section-flags is used to change the section flags (e.g. SHF_ALLOC) for given sections. The flags allowed are the same from the existing --rename-section=.old=.new[,flags] feature.
Additionally, make sure that --set-section-flag cannot be used with --rename-section (either the source or destination), since --rename-section accepts flags. This avoids ambiguity for something like "--rename-section=.foo=.bar,alloc --set-section-flag=.bar,code".
Neil Henning [Tue, 29 Jan 2019 14:28:17 +0000 (14:28 +0000)]
[AMDGPU] Fix a weird WWM intrinsic issue.
I found a really strange WWM issue through a very convoluted shader that
essentially boils down to a bug in SIInstrInfo where canReadVGPR did not
correctly identify that WWM is like a copy and can have a VGPR as its
source.
Hans Wennborg [Tue, 29 Jan 2019 13:43:22 +0000 (13:43 +0000)]
Revert r351833 and r352250.
They were breaking the Windows build when using MSBuild, see the
discussion on D56781.
r351833: "Use response file when generating LLVM-C.dll"
> Use response file when generating LLVM-C.dll
>
> As discovered in D56774 the command line gets to long, so use a response file to give the script the libs. This change has been tested and is confirmed working for me.
>
> Commited on behalf of Jakob Bornecrantz
>
> Differential Revision: https://reviews.llvm.org/D56781
r352250: "Build LLVM-C.dll by default on windows and enable in release package"
> Build LLVM-C.dll by default on windows and enable in release package
>
> With the fixes to the building of LLVM-C.dll in D56781 this should now
> be safe to land. This will greatly simplify dealing with LLVM for people
> that just want to use the C API on windows. This is a follow up from
> D35077.
>
> Patch by Jakob Bornecrantz!
>
> Differential revision: https://reviews.llvm.org/D56774
Ayonam Ray [Tue, 29 Jan 2019 12:01:32 +0000 (12:01 +0000)]
[CodeGen] Omit range checks from jump tables when lowering switches with unreachable default
During the lowering of a switch that would result in the generation of a
jump table, a range check is performed before indexing into the jump
table, for the switch value being outside the jump table range and a
conditional branch is inserted to jump to the default block. In case the
default block is unreachable, this conditional jump can be omitted. This
patch implements omitting this conditional branch for unreachable
defaults.
Review ID: D52002
Reviewers: Hans Wennborg, Eli Freidman, Roman Lebedev
Re-enable the code to rewrite main-function signatures into
"int main(int argc, char *argv[])", but limited to only handling
the case of "int main(void)", so that it doesn't silently strip
an argument in the "int main(int argc, char *argv[], char *envp[])"
case.
This allows main to be called by C startup code, since WebAssembly
requires caller and callee signatures to match, so it can't rely
on passing main a different number of arguments than it expects.
David Green [Tue, 29 Jan 2019 10:40:31 +0000 (10:40 +0000)]
[ARM] Use sub for negative offset load/store in thumb1
This attempts to optimise negative values used in load/store operands
a little. We currently try to selct them as rr, materialising the
negative constant using a MOV/MVN pair. This instead selects ri with
an immediate of 0, forcing the add node to become a simpler sub.
Jeremy Morse [Tue, 29 Jan 2019 10:23:43 +0000 (10:23 +0000)]
Remove 'XFAIL: powerpc64' from a debuginfo test
This test started XPASSing with r352467, and the change in behaviour
performed by that patch does appear to fix the cause of the original XFAIL
(missing FrameIndex DBG_VALUE), which I've replicated locally with
-mtriple=powerpc64--.
I'll write this up in PR21881 which documents the XFAIL, and seek
confirmation I haven't overlooked something here.
Bjorn Pettersson [Tue, 29 Jan 2019 10:19:44 +0000 (10:19 +0000)]
[IPCP] Don't crash due to arg count/type mismatch between caller/callee
Summary:
This patch avoids an assert in IPConstantPropagation when
there is a argument count/type mismatch between the caller and
the callee.
While this is actually UB on C-level (clang emits a warning),
the IR verifier seems to accept it. I'm not sure what other
frontends/languages might think about this, so simply bailing out
to avoid hitting an assert (in CallSiteBase<>::getArgOperand or
Value::doRAUW) seems like a simple solution.
The problem is exposed by the fact that AbstractCallSites will look
through a bitcast at the callee position of a call/invoke.
Jeremy Morse [Tue, 29 Jan 2019 09:40:05 +0000 (09:40 +0000)]
[DebugInfo][DAG] Process FrameIndex dbg.values unconditionally
A FrameIndex should be valid throughout a block regardless of what instructions
get selected in that block -- therefore we shouldn't harness dbg.values that
refer to FrameIndexes to an SDNode. There are numerous codegen reasons why
an SDNode never appears or doesn't become a location that a DBG_VALUE can
refer to. None of them actually affect the variable location.
Therefore, before any other tests to encode dbg_values in a SelectionDAG,
identify FrameIndex operands and encode them unattached to any SDNode.
Martin Storsjo [Tue, 29 Jan 2019 09:36:48 +0000 (09:36 +0000)]
[COFF, ARM64] Don't put jump table into a separate COFF section for EK_LabelDifference32
Windows ARM64 has PIC relocation model and uses jump table kind
EK_LabelDifference32. This produces jump table entry as
".word LBB123 - LJTI1_2" which represents the distance between the block
and jump table.
A new relocation type (IMAGE_REL_ARM64_REL32) is needed to do the fixup
correctly if they are in different COFF section.
This change saves the jump table to the same COFF section as the
associated code. An ideal fix could be utilizing IMAGE_REL_ARM64_REL32
relocation type.
Philip Reames [Tue, 29 Jan 2019 06:40:02 +0000 (06:40 +0000)]
Correct contents for r352453
I had a local change I hadn't realized when submitting that auto-update. As such, the auto-update was wrong. This should fix it, and with that, it's clearly time to stop submitting changes and go to bed.
Philip Reames [Tue, 29 Jan 2019 06:34:46 +0000 (06:34 +0000)]
[Tests] Regen to remove future test diffs
This file appears to have been manually editted at some point after being auto-updated. A future change adjusts this file slightly, and all of the updates makes the diff super confusing.
Max Kazantsev [Tue, 29 Jan 2019 05:37:59 +0000 (05:37 +0000)]
[SCEV] Take correct loop in AddRec simplification. PR40420
The code of AddRec simplification is using wrong loop when it creates a new
AddRecExpr. It should be using AddRecLoop which we have saved and against which
all gate checks are made, and not calling AddRec->getLoop() over and over
again because AddRec may change and become an AddRecurrency from outer loop
during the transform iterations.
Considering this change trivial, commiting for postcommit review.
Teresa Johnson [Tue, 29 Jan 2019 02:04:01 +0000 (02:04 +0000)]
Try to make new test more resilient to different orderings
New test added in r352441 getting a bot failure which I believe is
due to different ordering in the dumping which isn't being handled
well. Try to make test more resilient to ordering differences.
Sam Clegg [Tue, 29 Jan 2019 00:30:46 +0000 (00:30 +0000)]
[WebAssembly] Handle more types of uses in WebAssemblyAddMissingPrototypes
Previously we were only handling bitcast operations, however
prototypeless functions can also appear in other places such as
comparisons and as function params.
Switch to using replaceAllUsesWith() to replace the prototype-less
function uses. This new approach results in some redundant bitcasting
but is much simpler and handles all cases.
Reid Kleckner [Tue, 29 Jan 2019 00:30:35 +0000 (00:30 +0000)]
[PPC] Include tablegenerated PPCGenCallingConv.inc once
Move the CC analysis implementation to its own .cpp file instead of
duplicating it and artificually using functions in PPCISelLowering.cpp
and PPCFastISel.cpp. Follow-up to the same change done for X86, ARM, and
AArch64.
Teresa Johnson [Mon, 28 Jan 2019 23:43:26 +0000 (23:43 +0000)]
[ThinLTO] Add option to dump per-module summary dot graph
Summary:
I found that there currently isn't a way to invoke exportToDot from
the command line for a per-module summary index, and therefore no
testing of that case. Add an internal option and use it to test dumping
of per module summary indexes.
In particular, I am looking at fixing the limitation that causes the
aliasee GUID in the per-module summary to be 0, and want to be able to
test that change.
Philip Reames [Mon, 28 Jan 2019 23:24:49 +0000 (23:24 +0000)]
Demanded elements support for vector GEPs
GEPs can produce either scalar or vector results. If we're extracting only a subset of the vector lanes, simplifying the operands is helpful in eliminating redundant computation, and (eventually) allowing further optimizations
Teresa Johnson [Mon, 28 Jan 2019 22:27:05 +0000 (22:27 +0000)]
[ThinLTO] Refine reachability check to fix compile time increase
Summary:
A recent fix to the ThinLTO whole program dead code elimination (D56117)
increased the thin link time on a large MSAN'ed binary by 2x.
It's likely that the time increased elsewhere, but was more noticeable
here since it was already large and ended up timing out.
That change made it so we would repeatedly scan all copies of linkonce
symbols for liveness every time they were encountered during the graph
traversal. This was needed since we only mark one copy of an aliasee as
live when we encounter a live alias. This patch fixes the issue in a
more efficient manner by simply proactively visiting the aliasee (thus
marking all copies live) when we encounter a live alias.
Two notes: One, this requires a hash table lookup (finding the aliasee
summary in the index based on aliasee GUID). However, the impact of this
seems to be small compared to the original pre-D56117 thin link time. It
could be addressed if we keep the aliasee ValueInfo in the alias summary
instead of the aliasee GUID, which I am exploring in a separate patch.
Second, we only populate the aliasee GUID field when reading summaries
from bitcode (whether we are reading individual summaries and merging on
the fly to form the compiled index, or reading in a serialized combined
index). Thankfully, that's currently the only way we can get to this
code as we don't yet support reading summaries from LLVM assembly
directly into a tool that performs the thin link (they must be converted
to bitcode first). I added a FIXME, however I have the fix under test
already. The easiest fix is to simply populate this field always, which
isn't hard, but more likely the change I am exploring to store the
ValueInfo instead as described above will subsume this. I don't want to
hold up the regression fix for this though.
Craig Topper [Mon, 28 Jan 2019 21:38:47 +0000 (21:38 +0000)]
Recommit r352255 "[SelectionDAG][X86] Don't use SEXTLOAD for promoting masked loads in the type legalizer"
This did not cause the buildbot failure it was previously reverted for.
Original commit message:
I'm not sure why we were using SEXTLOAD. EXTLOAD seems more appropriate since we don't care about the upper bits.
This patch changes this and then modifies the X86 post legalization combine to emit a extending shuffle instead of a sign_extend_vector_inreg. Could maybe use an any_extend_vector_inre
On AVX512 targets I think we might be able to use a masked vpmovzx and not have to expand this at all.
Yonghong Song [Mon, 28 Jan 2019 21:35:23 +0000 (21:35 +0000)]
[RuntimeDyld] load all sections with ProcessAllSections
This patch tried to address the following use case.
. bcc (https://github.com/iovisor/bcc) utilizes llvm JIT to
compile for BTF target.
. with -g, .BTF and .BTF.ext sections (BPF debug info)
will be generated by LLVM.
. .BTF does not have relocations and .BTF.ext has some
relocations.
. With ProcessAllSections, .BTF.ext is loaded by JIT dynamic linker
and is available to application. But .BTF is not loaded.
The bcc application needs both .BTF.ext and .BTF for debugging
purpose, and .BTF is not loaded. This patch addressed this issue
by iterating over all sections and loading any missing
sections, after symbol/relocation processing in loadObjectImpl().
Signed-off-by: Yonghong Song <yhs@fb.com>
Differential Revision: https://reviews.llvm.org/D55943
Reid Kleckner [Mon, 28 Jan 2019 21:28:40 +0000 (21:28 +0000)]
[AArch64] Include AArch64GenCallingConv.inc once
Summary:
Avoids duplicating generated static helpers for calling convention
analysis.
This also means you can modify AArch64CallingConv.td without recompiling
the AArch64ISelLowering.cpp monolith, so it provides faster incremental
rebuilds.
Saves 12K in llc.exe, but adds a new object file, which is large.
Matt Arsenault [Mon, 28 Jan 2019 20:14:49 +0000 (20:14 +0000)]
AMDGPU: Add DS append/consume intrinsics
Since these pass the pointer in m0 unlike other DS instructions, these
need to worry about whether the address is uniform or not. This
assumes the address is dynamically uniform, and just uses
readfirstlane to get a copy into an SGPR.
I don't know if these have the same 16-bit add for the addressing mode
offset problem on SI or not, but I've just assumed they do.
Also includes some misc. changes to avoid test differences between the
LDS and GDS versions.
Jessica Paquette [Mon, 28 Jan 2019 19:53:14 +0000 (19:53 +0000)]
[GlobalISel][AArch64] Add instruction selection support for @llvm.log10
This adds instruction selection support for @llvm.log10 in AArch64. It teaches
GISel to lower it to a library call, updates the relevant tests, and adds a
legalizer test for log10.
Nico Weber [Mon, 28 Jan 2019 19:32:52 +0000 (19:32 +0000)]
gn build: Fix `lld-link: unknown flag: -fuse-ld=lld` warnings on Windows
Fixes a minor regression from r351248.
While here, also make it possible to opt out of lld by saying
use_lld=false when clang_base_path is set. (use_lld still defaults to
true if clang_base_path is set.)
Jessica Paquette [Mon, 28 Jan 2019 19:22:29 +0000 (19:22 +0000)]
[GlobalISel] Add ISel support for @llvm.lifetime.start and @llvm.lifetime.end
This adds ISel support for lifetime markers in opt levels above O0.
It also updates the arm64-irtranslator test, and updates some AArch64 tests that
use them for added coverage.
It also adds a testcase taken from the X86 codegen tests which verified a bug
caused by lifetime markers + stack colouring in the past. This is intended to
make sure that GISel doesn't re-introduce the bug.
(This is basically a straight copy from what SelectionDAG does in
SelectionDAGBuilder.cpp)
Jessica Paquette [Mon, 28 Jan 2019 18:34:18 +0000 (18:34 +0000)]
[GlobalISel][AArch64] Add instruction selection support for G_FCOS and G_FSIN
This contains all of the legalizer changes from D57197 necessary to select
G_FCOS and G_FSIN. It also updates several existing IR tests in
test/CodeGen/AArch64 that verify that we correctly lower the G_FCOS and G_FSIN
instructions.
Alina Sbirlea [Mon, 28 Jan 2019 17:48:45 +0000 (17:48 +0000)]
[SimpleLoopUnswitch] Early check exit for trivial unswitch with MemorySSA.
Summary:
If MemorySSA is avaiable, we can skip checking all instructions if block has any Defs.
(volatile loads are also Defs).
We still need to check all instructions for "canThrow", even if no Defs are found.
George Rimar [Mon, 28 Jan 2019 16:36:12 +0000 (16:36 +0000)]
[llvm-objdump] - Restore a piece of code removed by mistake in r352366.
Seems when committed the r352366
("[llvm-objdump] - Print LMAs when dumping section headers.")
I resolved merge conflict incorrectly and removed this piece by mistake.
Bots did not catch this yet, seems they are slow today,
but the `X86/adjust-vma.test` test case fails locally for me without that.
Sanjay Patel [Mon, 28 Jan 2019 15:51:34 +0000 (15:51 +0000)]
[x86] allow more shuffle splitting to avoid vpermps (PR40434)
This is tricky to make optimal: sometimes we're better off using
a single wider op, but other times it makes more sense to combine
a narrow ops to achieve the same result.
This solves the case from:
https://bugs.llvm.org/show_bug.cgi?id=40434
There's potentially a similar change for vectors with 64-bit elements,
but it needs adjustments similar to rL352333 to avoid creating infinite
loops.
Ranjeet Singh [Mon, 28 Jan 2019 15:48:07 +0000 (15:48 +0000)]
VERSION_GREATER_EQUAL not supported in llvm cmake.
Patch https://reviews.llvm.org/D56329 caused build failures for me when
building on Windows because of the use of cmake operator
'VERSION_GREATER_EQUAL' which isn't supported in older versions of cmake. The
llvm website states that minimum required version of cmake for building llvm is
3.4.3 https://llvm.org/docs/CMake.html
Remove no longer needed Arm specific LICENSE.TXT file.
As the codebase is now under the Apache 2.0 license with LLVM
Exceptions, and all Arm's contributions, past or future, are under that
new license, this Arm specific LICENSE.TXT is no longer needed, thus
removing it.
Michal Gorny [Mon, 28 Jan 2019 15:16:03 +0000 (15:16 +0000)]
[cmake] Fix get_llvm_lit_path() to respect LLVM_EXTERNAL_LIT always
Refactor the get_llvm_lit_path() logic to respect LLVM_EXTERNAL_LIT,
and require the fallback to be defined explicitly
as LLVM_DEFAULT_EXTERNAL_LIT. This fixes building libcxx standalone
after r346888.
The old logic was using LLVM_EXTERNAL_LIT both as user-defined cache
variable and an optional pre-definition of default value from caller
(e.g. libcxx). It included a hack to make this work by assigning
the value back and forth but it was fragile and stopped working
in libcxx.
The new logic is simpler and more transparent. Default value is
provided in a separate variable, and used only when user-specified
variable is empty (i.e. not overriden).
Jordan Rupprecht [Mon, 28 Jan 2019 15:02:40 +0000 (15:02 +0000)]
[llvm-objcopy] Fix crash when writing empty binary output
Summary: When using llvm-objcopy -O binary and the resulting file will be empty (e.g. removing the only section that would be written, or using --only-keep with a section that doesn't exist/isn't SHF_ALLOC), we crash because FileOutputBuffer expects Size > 0. Add a regression test, and change Buffer to open/truncate the output file in this case.
Instruction abs.[ds] is not generating correct result when working
with NaNs for revisions prior mips32r6 and mips64r6.
To generate a sequence which always produce a correct result, but also
to allow user more control on how his code is compiled, attribute
+abs2008 is added, so user can choose legacy or 2008.
By default legacy mode is used on revisions prior R6. Mips32r6 and
mips64r6 use abs2008 mode by default.
George Rimar [Mon, 28 Jan 2019 14:11:35 +0000 (14:11 +0000)]
[llvm-objdump] - Print LMAs when dumping section headers.
When --section-headers is used, GNU objdump prints both LMA and VMA for sections.
llvm-objdump does not do that what makes it's output be slightly inconsistent.
Patch teaches llvm-objdump to print LMA/VMA for ELF file formats.
The behavior for other formats remains unchanged.
projects / llvm / log