Evandro Menezes [Fri, 10 Mar 2017 20:20:04 +0000 (20:20 +0000)]
[AArch64, X86] Additional debug information for MacroFusion
In order to make it easier to parse information about the performance of
MacroFusion, this patch adds the function and the instruction names to the
debug output of this pass.
[SLP] Revert everything that has to do with memory access sorting.
This reverts r293386, r294027, r294029 and r296411.
Turns out the SLP tree isn't actually a "tree" and we don't handle
accessing the same packet of loads in several different orders well,
causing miscompiles.
Volkan Keles [Fri, 10 Mar 2017 18:34:57 +0000 (18:34 +0000)]
[GlobalISel] Make LegalizerInfo accessible in LegalizerHelper
Summary:
We don’t actually use LegalizerInfo in Legalizer pass, it’s just passed
as an argument.
In order to check if an instruction is legal or not, we need to get LegalizerInfo
by calling `MI.getParent()->getParent()->getSubtarget().getLegalizerInfo()`.
Instead, make LegalizerInfo accessible in LegalizerHelper.
Zachary Turner [Fri, 10 Mar 2017 18:33:41 +0000 (18:33 +0000)]
[Support] Don't return an error if realPath fails.
In openFileForRead, we would not previously return an error
if real_path resolution failed. After a recent patch, we
started propagating this error up. This caused a failure
in clang when trying to call openFileForRead("nul"). This
patch restores the previous behavior of not propagating this
error up.
Simon Pilgrim [Fri, 10 Mar 2017 18:01:53 +0000 (18:01 +0000)]
[X86][SSE] Added tests showing missed truncations for sitofp conversion
SelectionDAG::ComputeNumSignBits is poor at build_vector handling, meaning that we can't see that all the vXi64 sources are in fact sign extended i32 or smaller.
Zachary Turner [Fri, 10 Mar 2017 17:39:21 +0000 (17:39 +0000)]
Add llvm::sys::fs::real_path.
LLVM already has real_path like functionality, but it is
cumbersome to use and involves clean up after (e.g. you have
to call openFileForRead, then close the resulting FD).
Furthermore, on Windows it doesn't work for directories since
opening a directory and opening a file require slightly
different flags.
So I add a simple function `real_path` which works for all
paths on all platforms and has a simple to use interface.
In doing so, I add the ability to opt in to resolving tilde
expressions (e.g. ~/foo), which are normally handled by
the shell.
Simon Pilgrim [Fri, 10 Mar 2017 17:23:55 +0000 (17:23 +0000)]
[X86][MMX] Add tests showing missed opportunities to use MMX sitofp conversions
If we are transferring MMX registers to XMM for conversion we could use the MMX equivalents (CVTPI2PD + CVTPI2PS) without affecting rounding/exceptions etc.
Simon Pilgrim [Fri, 10 Mar 2017 16:59:43 +0000 (16:59 +0000)]
[X86][MMX] Add tests showing missed opportunities to use MMX fptosi conversions
If we are transferring XMM conversion results to MMX registers we could use the MMX equivalents (CVTPD2PI/CVTTPD2PI + CVTPS2PI/CVTTPS2PI) with affecting rounding/expections etc.
Simon Pilgrim [Fri, 10 Mar 2017 13:44:32 +0000 (13:44 +0000)]
[APInt] Add APInt::insertBits() method to insert an APInt into a larger APInt
We currently have to insert bits via a temporary variable of the same size as the target with various shift/mask stages, resulting in further temporary variables, all of which require the allocation of memory for large APInts (MaskSizeInBits > 64).
This is another of the compile time issues identified in PR32037 (see also D30265).
This patch adds the APInt::insertBits() helper method which avoids the temporary memory allocation and masks/inserts the raw bits directly into the target.
Simon Dardis [Fri, 10 Mar 2017 13:27:14 +0000 (13:27 +0000)]
[mips][msa] Accept more values for constant splats
This patches teaches the MIPS backend to accept more values for constant
splats. Previously, only 10 bit signed immediates or values that could be
loaded using an ldi.[bhwd] instruction would be acceptted. This patch relaxes
that constraint so that any constant value that be splatted is accepted.
As a result, the constant pool is used less for vector operations, and the
suite of bit manipulation instructions b(clr|set|neg)i can now be used with
the full range of their immediate operand.
Sanne Wouda [Fri, 10 Mar 2017 13:08:20 +0000 (13:08 +0000)]
[Assembler] Add location info to unary expressions.
Summary:
This is a continuation of D28861. Add an SMLoc to MCUnaryExpr such that
a better diagnostic can be given in case of an error in later stages of
assembling.
Artyom Skrobov [Fri, 10 Mar 2017 12:41:33 +0000 (12:41 +0000)]
Refactor the multiply-accumulate combines to act on
ARMISD::ADD[CE] nodes, instead of the generic ISD::ADD[CE].
Summary:
This allows for some simplification because the combines
are no longer limited to just one go at the node before
it gets legalized into an ARM target-specific one.
George Rimar [Fri, 10 Mar 2017 10:31:56 +0000 (10:31 +0000)]
WholeProgramDevirt: Fixed compilation error under MSVS2015.
It was introduced in:
r296945
WholeProgramDevirt: Implement exporting for single-impl devirtualization.
---------------------
r296939
WholeProgramDevirt: Add any unsuccessful llvm.type.checked.load devirtualizations to the list of llvm.type.test users.
---------------------
Microsoft Visual Studio Community 2015
Version 14.0.23107.0 D14REL
Does not compile that code without additional brackets, showing multiple error like below:
WholeProgramDevirt.cpp(1216): error C2958: the left bracket '[' found at 'c:\access_softek\llvm\lib\transforms\ipo\wholeprogramdevirt.cpp(1216)' was not matched correctly
WholeProgramDevirt.cpp(1216): error C2143: syntax error: missing ']' before '}'
WholeProgramDevirt.cpp(1216): error C2143: syntax error: missing ';' before '}'
WholeProgramDevirt.cpp(1216): error C2059: syntax error: ']'
Simon Atanasyan [Fri, 10 Mar 2017 08:22:20 +0000 (08:22 +0000)]
[MC] Set SHT_MIPS_DWARF section type for all .debug_* sections on MIPS
All MIPS .debug_* sections should be marked with ELF type SHT_MIPS_DWARF
accordingly the specification [1]. Also the same section type is assigned
to these sections by GNU tools.
Simon Atanasyan [Fri, 10 Mar 2017 08:22:13 +0000 (08:22 +0000)]
[MC] Accept a numeric value as an ELF section header's type
GAS supports specification of section header's type using a numeric
value [1]. This patch brings the same functionality to LLVM. That allows
to setup some target-specific section types belong to the SHT_LOPROC -
SHT_HIPROC range. If we attempt to print unknown section type, MCSectionELF
class shows an error message. It's better than print sole '@' sign
without any section type name.
In case of MIPS, example of such section's type is SHT_MIPS_DWARF.
Without the patch we will have to implement some workarounds
in probably not-MIPS-specific part of code base to convert SHT_MIPS_DWARF
to the @progbits while printing assembly and to assign SHT_MIPS_DWARF for
@progbits sections named .debug_* if we encounter such section in
an input assembly.
Daniel Berlin [Fri, 10 Mar 2017 04:54:10 +0000 (04:54 +0000)]
Move memory coercion functions from GVN.cpp to VNCoercion.cpp so they can be shared between GVN and NewGVN.
Summary:
These are the functions used to determine when values of loads can be
extracted from stores, etc, and to perform the necessary insertions to
do this. There are no changes to the functions themselves except
reformatting, and one case where memdep was informed of a removed load
(which was pushed into the caller).
- Fix the insertion point, which occasionally could have been incorrect.
- Avoid creating multiple bitsplits with the same operands, if an old one
could be reused.
Tim Northover [Thu, 9 Mar 2017 21:12:06 +0000 (21:12 +0000)]
GlobalISel: put debug info for static allocas in the MachineFunction.
The good reason to do this is that static allocas are pretty simple to handle
(especially at -O0) and avoiding tracking DBG_VALUEs throughout the pipeline
should give some kind of performance benefit.
The bad reason is that the debug pipeline is an unholy mess of implicit
contracts, where determining whether "DBG_VALUE %reg, imm" actually implies a
load or not involves the services of at least 3 soothsayers and the sacrifice
of at least one chicken. And it still gets it wrong if the variable is at SP
directly.
Tom Stellard [Thu, 9 Mar 2017 19:24:07 +0000 (19:24 +0000)]
CMake: Don't install llvm-tblgen twice
Summary:
The add_tablegen macros defines its own install target, and it was also calling
add_llvm_utility which adds another install target.
Configuring with -DLLVM_TOOLS_INSTALL_DIR set to something other than
'bin' along with -DLLVM_INSTALL_UTILS=ON was causing llvm-tablgen
to be installed to two separate directories.
Artem Belevich [Thu, 9 Mar 2017 17:59:04 +0000 (17:59 +0000)]
[FileCheck] Added --enable-var-scope option to enable scope for regex variables.
If `--enable-var-scope` is in effect, variables with names that
start with `$` are considered to be global. All other variables are
local. All local variables get undefined at the beginning of each
CHECK-LABEL block. Global variables are not affected by CHECK-LABEL.
This makes it easier to ensure that individual tests are not affected
by variables set in preceding tests.
Sanjay Patel [Thu, 9 Mar 2017 16:20:52 +0000 (16:20 +0000)]
[InstSimplify] vector div/rem with any zero element in divisor is undef
This was suggested as a DAG simplification in the review for rL297026 :
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20170306/435253.html
...but let's start with IR since we have actual docs for IR (LangRef).
Sanjay Patel [Thu, 9 Mar 2017 15:02:25 +0000 (15:02 +0000)]
[DAG] recognize div/rem by 0 as undef before trying constant folding
As discussed in the review thread for rL297026, this is actually 2 changes that
would independently fix all of the test cases in the patch:
1. Return undef in FoldConstantArithmetic for div/rem by 0.
2. Move basic undef simplifications for div/rem (simplifyDivRem()) before
foldBinopIntoSelect() as a matter of efficiency.
I will handle the case of vectors with any zero element as a follow-up. That change
is the DAG sibling for D30665 + adding a check of vector elements to FoldConstantVectorArithmetic().
I'm deleting the test for PR30693 because it does not test for the actual bug any more
(dangers of using bugpoint).
Joey Gouly [Thu, 9 Mar 2017 13:38:06 +0000 (13:38 +0000)]
[SelectionDAG] Make SelectCode return void
SelectCode has been returning nullptr since 182dac0 ("SDAG: Make
SelectCodeCommon return void", 2016-05-10). Make SelectCode also
return void instead, as all callers have been updated.
[PM/Inliner] Make the new PM's inliner process call edges across an
entire SCC before iterating on newly-introduced call edges resulting
from any inlined function bodies.
This more closely matches the behavior of the old PM's inliner. While it
wasn't really clear to me initially, this behavior is actually essential
to the inliner behaving reasonably in its current design.
Because the inliner is fundamentally a bottom-up inliner and all of its
cost modeling is designed around that it often runs into trouble within
an SCC where we don't have any meaningful bottom-up ordering to use. In
addition to potentially cyclic, infinite inlining that we block with the
inline history mechanism, it can also take seemingly simple call graph
patterns within an SCC and turn them into *insanely* large functions by
accidentally working top-down across the SCC without any of the
threshold limitations that traditional top-down inliners use.
Consider this diabolical monster.cpp file that Richard Smith came up
with to help demonstrate this issue:
```
template <int N> extern const char *str;
When compiled with '-DMAX=N' for various values of N, this will create an SCC
with a reasonably large number of functions. Previously, the inliner would try
to exhaust the inlining candidates in a single function before moving on. This,
unfortunately, turns it into a top-down inliner within the SCC. Because our
thresholds were never built for that, we will incrementally decide that it is
always worth inlining and proceed to flatten the entire SCC into that one
function.
What's worse, we'll then proceed to the next function, and do the exact same
thing except we'll skip the first function, and so on. And at each step, we'll
also make some of the constant factors larger, which is awesome.
The fix in this patch is the obvious one which makes the new PM's inliner use
the same technique used by the old PM: consider all the call edges across the
entire SCC before beginning to process call edges introduced by inlining. The
result of this is essentially to distribute the inlining across the SCC so that
every function incrementally grows toward the inline thresholds rather than
allowing the inliner to grow one of the functions vastly beyond the threshold.
The code for this is a bit awkward, but it works out OK.
We could consider in the future doing something more powerful here such as
prioritized order (via lowest cost and/or profile info) and/or a code-growth
budget per SCC. However, both of those would require really substantial work
both to design the system in a way that wouldn't break really useful
abstraction decomposition properties of the current inliner and to be tuned
across a reasonably diverse set of code and workloads. It also seems really
risky in many ways. I have only found a single real-world file that triggers
the bad behavior here and it is generated code that has a pretty pathological
pattern. I'm not worried about the inliner not doing an *awesome* job here as
long as it does *ok*. On the other hand, the cases that will be tricky to get
right in a prioritized scheme with a budget will be more common and idiomatic
for at least some frontends (C++ and Rust at least). So while these approaches
are still really interesting, I'm not in a huge rush to go after them. Staying
even closer to the existing PM's behavior, especially when this easy to do,
seems like the right short to medium term approach.
I don't really have a test case that makes sense yet... I'll try to find a
variant of the IR produced by the monster template metaprogram that is both
small enough to be sane and large enough to clearly show when we get this wrong
in the future. But I'm not confident this exists. And the behavior change here
*should* be unobservable without snooping on debug logging. So there isn't
really much to test.
The test case updates come from two incidental changes:
1) We now visit functions in an SCC in the opposite order. I don't think there
really is a "right" order here, so I just update the test cases.
2) We no longer compute some analyses when an SCC has no call instructions that
we consider for inlining.
Simon Dardis [Thu, 9 Mar 2017 11:19:48 +0000 (11:19 +0000)]
[mips] Fix return lowering
Fix a machine verifier issue where a instruction was using a invalid
register. The return pseudo is expanded and has the return address
register added to it. The return register may have been spuriously
mark as killed earlier.
This partially resolves PR/27458
Thanks to Quentin Colombet for reporting the issue!
Teresa Johnson [Thu, 9 Mar 2017 00:19:49 +0000 (00:19 +0000)]
Perform symbol binding for .symver versioned symbols
Summary:
In a .symver assembler directive like:
.symver name, name2@@nodename
"name2@@nodename" should get the same symbol binding as "name".
While the ELF object writer is updating the symbol binding for .symver
aliases before emitting the object file, not doing so when the module
inline assembly is handled by the RecordStreamer is causing the wrong
behavior in *LTO mode.
E.g. when "name" is global, "name2@@nodename" must also be marked as
global. Otherwise, the symbol is skipped when iterating over the LTO
InputFile symbols (InputFile::Symbol::shouldSkip). So, for example,
when performing any *LTO via the gold-plugin, the versioned symbol
definition is not recorded by the plugin and passed back to the
linker. If the object was in an archive, and there were no other symbols
needed from that object, the object would not be included in the final
link and references to the versioned symbol are undefined.
The llvm-lto2 tests added will give an error about an unused symbol
resolution without the fix.
Don't merge global constants with non-dbg metadata.
!type metadata can not be dropped. An alternative to this is adding
!type metadata from the replaced globals to the replacement, but that
may weaken type tests and make them slower at the same time.
The merged global gets !dbg metadata from replaced globals, and can
end up with multiple debug locations.
This commit changes the BumpPtrAllocator for suffix tree nodes to a SpecificBumpPtrAllocator.
Before, node construction was leaking memory because of the DenseMap in SuffixTreeNodes.
Changing this to a SpecificBumpPtrAllocator allows this memory to properly be released.
projects / llvm / log