Eric Christopher [Mon, 23 Feb 2015 19:28:45 +0000 (19:28 +0000)]
Rewrite the global merge pass to be subprogram agnostic for now.
It was previously using the subtarget to get values for the global
offset without actually checking each function as it was generating
code. Go ahead and solidify the current behavior and make the
existing FIXMEs more prominent.
As a note the ARM backend previously had a thumb1 and non-thumb1
set of defaults. Only the former was tested so I've changed the
behavior to only use that for now.
Chad Rosier [Mon, 23 Feb 2015 19:15:16 +0000 (19:15 +0000)]
Prevent hoisting fmul from THEN/ELSE to IF if there is fmsub/fmadd opportunity.
This patch adds the isProfitableToHoist API. For AArch64, we want to prevent a
fmul from being hoisted in cases where it is more profitable to form a
fmsub/fmadd.
Phabricator Review: http://reviews.llvm.org/D7299
Patch by Lawrence Hu <lawrence@codeaurora.org>
Daniel Sanders [Mon, 23 Feb 2015 17:22:16 +0000 (17:22 +0000)]
[mips] Honour -mno-odd-spreg for vector insert/extract when MSA is enabled.
Summary:
-mno-odd-spreg prohibits the use of odd-numbered single-precision floating
point registers. However, vector insert/extract was still using them when
manipulating the subregisters of an MSA register. Fixed this by ensuring
that insertion/extraction is only performed on even-numbered vector
registers when -mno-odd-spreg is given.
Bob Wilson [Mon, 23 Feb 2015 16:57:19 +0000 (16:57 +0000)]
Fix incorrect immediate size for AddrModeT2_i8s4 in rewriteT2FrameIndex.
The natural way to handle this addressing mode would be to say that it has
8 bits and gets scaled by 4, but since the MC layer is expecting the scaling
to be already reflected in the immediate value, we have been setting the
Scale to 1. That's fine, but then NumBits needs to be adjusted to reflect
the effective increase in the range of the immediate. That adjustment was
missing.
The consequence is that the register scavenger can fail.
The estimateRSStackSizeLimit() function in ARMFrameLowering.cpp correctly
assumes that the AddrModeT2_i8s4 address mode can handle scaled offsets up to
1020. Under just the right circumstances, we fail to reserve space for the
scavenger because it thinks that nothing will be needed. However, the overly
pessimistic behavior in rewriteT2FrameIndex causes some frame indexes to be
out of range and require scavenged registers, and so the scavenger asserts.
Unfortunately I have not been able to come up with a testcase for this. I
can only reproduce it on an internal branch where the frame layout and
register allocation is slightly different than trunk. We really need a
way to serialize MachineInstr-level IR to write reasonable tests for things
like this.
Benjamin Kramer [Mon, 23 Feb 2015 16:41:36 +0000 (16:41 +0000)]
Sync the __builtin_expects for our 3 quadratically probed hash table implementations.
This assumes that
a) finding the bucket containing the value is LIKELY
b) finding an empty bucket is LIKELY
c) growing the table is UNLIKELY
I also switched the a) and b) cases for SmallPtrSet as we seem to use
the set mostly more for insertion than for checking existence.
In a simple benchmark consisting of 2^21 insertions of 2^20 unique
pointers into a DenseMap or SmallPtrSet a few percent speedup on average,
but nothing statistically significant.
[X86][MMX] Add MMX instructions to foldable tables
Teach the peephole optimizer to work with MMX instructions by adding
entries into the foldable tables. This covers folding opportunities not
handled during isel.
[X86][MMX] Add tests for pslli, psrli and psrai intrinsics
Add tests to cover the RR form of the pslli, psrli and psrai intrinsics.
In the next commit, the loads are going to be folded and the
instructions use the RM form.
Zachary Turner [Mon, 23 Feb 2015 05:58:34 +0000 (05:58 +0000)]
[llvm-pdbdump] Add an option to dump full class definitions.
This adds the --class-definitions flag. If specified, when dumping
types, instead of "class Foo" you will see the full class definition,
with member functions, constructors, access specifiers.
NOTE: Using this option can be very slow, as generating a full class
definition requires accessing many different parts of the PDB.
Zachary Turner [Sun, 22 Feb 2015 22:03:38 +0000 (22:03 +0000)]
[llvm-pdbdump] Rewrite dumper using visitor pattern.
This increases the flexibility of how to dump different
symbol types -- necessary for context-sensitive formatting of
symbol types -- and also improves the modularity by allowing
the dumping to be implemented in the actual dumper, as opposed
to in the PDB library.
Zachary Turner [Sun, 22 Feb 2015 21:45:38 +0000 (21:45 +0000)]
[llvm-pdbdump] Simplify options and output.
This removes a wealth of options, and instead now only provides
three options. -symbols, -types, and -compilands. This greatly
simplifies use of the tool, and makes it easier to understand
what you're going to see when you run the tool.
JF Bastien [Sun, 22 Feb 2015 19:32:03 +0000 (19:32 +0000)]
Use common parse routine to read alignment values from bitcode
While fuzzing LLVM bitcode files, I discovered that (1) the bitcode reader doesn't check that alignments are no larger than 2**29; (2) downstream code doesn't check the range; and (3) for values out of range, corresponding large memory requests (based on alignment size) will fail. This code fixes the bitcode reader to check for valid alignments, fixing this problem.
This CL fixes alignment value on global variables, functions, and instructions: alloca, load, load atomic, store, store atomic.
Hal Finkel [Sun, 22 Feb 2015 18:35:32 +0000 (18:35 +0000)]
[LICM] Refactor to expose functionality as utility functions
This refactors the core functionality of LICM: HoistRegion, SinkRegion and
PromoteAliasSet (renamed to promoteLoopAccessesToScalars) as utility functions
in LoopUtils. This will enable other transformations to make use of them
directly.
Hal Finkel [Sun, 22 Feb 2015 16:10:22 +0000 (16:10 +0000)]
[DAGCombine] Don't assume integer-type legailty in reduceBuildVecConvertToConvertBuildVec
DAGCombine will rewrite an BUILD_VECTOR where all non-undef inputs some from
[US]INT_TO_FP, as a BUILD_VECTOR of integers with the conversion applied as a
vector operation. We check operation legality of the conversion, but fail to
check legality of the integer vector type itself. Because targets don't
normally override operation legality defaults for illegal types, we need to
check this also.
This came up in the context of the QPX vector entensions for PowerPC (which can
have legal floating-point vector types without corresponding legal integer
vector types). No in-tree test case for this yes, but one can be added once
the QPX support has been committed.
Hal Finkel [Sun, 22 Feb 2015 15:58:04 +0000 (15:58 +0000)]
[SDAG] Use correct alignments on expanded vector trunc-store/ext-loads
When expanding a truncating store or extending load using vector extracts or
inserts and scalar stores and loads, we were giving each of these scalar stores
or loads the same alignment as the original vector operation. While this will
often be right (most vector operations, especially those produced by
autovectorization, have the alignment of the underlying scalar type), the
vector operation could certainly have a larger alignment.
Zachary Turner [Sun, 22 Feb 2015 07:01:41 +0000 (07:01 +0000)]
Attempt to fix the builders.
The issue was that the test Makefile had not been updated to
provide a value for HAVE_DIA_SDK, so it was being initialized
incorrectly. Hopefully this brings everything back to green.
Zachary Turner [Sun, 22 Feb 2015 06:47:32 +0000 (06:47 +0000)]
[llvm-pdbdump] Resubmit "Add some tests for llvm-pdbdump".
NOTE: This patch intentionally breaks the build. It attempts
to resubmit r230083, but with some debug logging in the CMake
and lit config files to determine why certain bots do not
correctly disable the DIA tests when DIA is not available.
After a sufficient number of bots fail, this patch will either
be reverted or, if the cause of the failure becomes obvious,
a fix submitted with the log statements removed.
Sanjoy Das [Sat, 21 Feb 2015 22:20:22 +0000 (22:20 +0000)]
IRCE: generalize InductiveRangeCheck::computeSafeIterationSpace to
work with a non-canonical induction variable.
This is currently a non-functional change because we only ever call
computeSafeIterationSpace on a canonical induction variable; but the
generalization will be useful in a later commit.
Eric Christopher [Sat, 21 Feb 2015 09:09:15 +0000 (09:09 +0000)]
Unconditionally create a new MCInstrInfo in the asm printer for
asm parsing since it's not subtarget dependent and we can't depend
upon the one hanging off the MachineFunction's subtarget still
being around.
Eric Christopher [Sat, 21 Feb 2015 08:48:22 +0000 (08:48 +0000)]
Have the MipsAsmPrinter fp stub emission code take a custom
MCSubtargetInfo as the MachineFunction has gone away and we need
to emit code at the module level.
David Majnemer [Sat, 21 Feb 2015 05:49:45 +0000 (05:49 +0000)]
X86: Call __main using the SelectionDAG
Synthesizing a call directly using the MI layer would confuse the frame
lowering code. This is problematic as frame lowering is highly
sensitive the particularities of calls, etc.
Tim Northover [Sat, 21 Feb 2015 02:11:17 +0000 (02:11 +0000)]
CodeGen: convert CCState interface to using ArrayRefs
Everyone except R600 was manually passing the length of a static array
at each callsite, calculated in a variety of interesting ways. Far
easier to let ArrayRef handle that.
There should be no functional change, but out of tree targets may have
to tweak their calls as with these examples.
David Majnemer [Sat, 21 Feb 2015 01:04:47 +0000 (01:04 +0000)]
Win64: Stack alignment constraints aren't applied during SET_FPREG
Stack realignment occurs after the prolog, not during, for Win64.
Because of this, don't factor in the maximum stack alignment when
establishing a frame pointer.
AsmParser/Writer: Handle symbolic constants in DI 'flags:'
Parse (and write) symbolic constants in debug info `flags:` fields.
This prevents a readability (and CHECK-ability) regression with the new
debug info hierarchy.
Old (well, current) assembly, with pretty-printing:
As discussed in the review thread, this isn't a final state. Most of
these flags correspond to `DW_AT_` symbolic constants, and we might
eventually want to support arbitrary attributes in some form. However,
as it stands now, some of the flags correspond to other concepts (like
`FlagStaticMember`); until things are refactored this is the simplest
way to move forward without regressing assembly.
Split debug info 'flags' bitfield over a vector so the current flags can
be iterated over. This API (in combination with r230107) will be used
for assembly support for symbolic constants.
Add `DIDescriptor::getFlag(StringRef)` and
`DIDescriptor::getFlagString(unsigned)`. The latter only converts exact
matches; I'll add separate API for breaking the flags bitfield up into
parts.
AsmParser: Use StringRef for keyword comparisons, NFC
Leverage `StringRef` inside keyword comparison macros. There's no
reason to be so low-level here, and I'm about to add another
`startswith()` use, so let's make it easy to read.
Philip Reames [Sat, 21 Feb 2015 00:09:09 +0000 (00:09 +0000)]
[PlaceSafepoints] Adjust enablement logic to default to off and be GC configurable per GC
Previously, this pass ran over every function in the Module if added to the pass order. With this change, it runs only over those with a GC attribute where the GC explicitly opts in. A GC can also choose which of entry safepoint polls, backedge safepoint polls, and call safepoints it wants. I hope to get these exposed as checks on the GCStrategy at some point, but for now, the checks are manual string comparisons.
`do { ... } while (false)` is standard macro etiquette for forcing
instantiations into a single statement and requiring a `;` afterwards,
making statement-like macros easier to reason about (and harder to use
incorrectly).
I'm about to modify the macros in `LexIdentifier()`. I noticed that the
`KEYWORD` macro *does* follow the rule, so I thought I'd clean up the
other macros to match (otherwise might not be worth changing, since the
benefits of this pattern are fairly irrelevant here).
Matthias Braun [Fri, 20 Feb 2015 23:43:14 +0000 (23:43 +0000)]
LiveRangeCalc: Don't start liveranges of PHI instruction at the block begin.
Summary:
Letting them begin at the PHI instruction slightly simplifies the code
but more importantly avoids breaking the assumption that live ranges
starting at the block begin are also live at the end of the predecessor
blocks. The MachineVerifier checks that but was apparently never run in
the few instances where liveranges are calculated for machine-SSA
functions.
Philip Reames [Fri, 20 Feb 2015 23:32:03 +0000 (23:32 +0000)]
Hide a bunch of advanced testing options in default opt --help output
These are internal options. I need to go through, evaluate which are worth keeping and which not. Many of them should probably be renamed as well. Until I have time to do that, we can at least stop poluting the standard opt -help output.
Philip Reames [Fri, 20 Feb 2015 23:16:52 +0000 (23:16 +0000)]
[RewriteStatepointsForGC] Use DenseSet in place of std::set [NFC]
This should be the last cleanup on non-llvm preferred data structures. I left one use of std::set in an assertion; DenseSet didn't seem to have a tombstone for CallSite defined. That might be worth fixing, but wasn't worth it for a debug only use.
Philip Reames [Fri, 20 Feb 2015 22:48:20 +0000 (22:48 +0000)]
[RewriteStatepointsForGC] Replace std::map with DenseMap
I'd done the work of extracting the typedef in a previous commit, but didn't actually change it. Hopefully this will make any subtle changes easier to isolate.
projects / llvm / log