Commit Graph

125110 Commits

Author SHA1 Message Date
Bill Wendling f4ec97d9b0 Fix build by passing in the needed variable after r153860.
llvm-svn: 153861
2012-04-02 06:17:37 +00:00
Craig Topper 54bfde79db Make MCInstrInfo available to the MCInstPrinter. This will be used to remove getInstructionName and the static data it contains since the same tables are already in MCInstrInfo.
llvm-svn: 153860
2012-04-02 06:09:36 +00:00
Bill Wendling 3a0bcf06ef It could come about that we parse the inline ASM before we get a potential
definition for it. In that case, we want to wait for the potential definition
before we create a symbol for it.

llvm-svn: 153859
2012-04-02 03:33:31 +00:00
Richard Smith 23d55873cb Basic semantic analysis support for inheriting constructor declarations in
dependent contexts.

llvm-svn: 153858
2012-04-02 01:30:27 +00:00
Craig Topper 7a2cea1814 Use SequenceToOffsetTable to generate instruction name table for AsmWriter.
llvm-svn: 153857
2012-04-02 00:47:39 +00:00
Howard Hinnant 788c9979d9 This is an initial commit of constexpr support as proposed by Richard Smith. This by no means completes constexpr support. Indeed, it hardly scratches the surface. All it does is lay the foundation in <__config> and changes those few places in the library that are already using that foundation.
llvm-svn: 153856
2012-04-02 00:40:41 +00:00
Howard Hinnant 0527c6207a I believe tuple is still under development in the standard. Daniel Krugler is/will be making convincing arguments that a modified form of LWG 2051 (currently NAD Future) is easily acheivable and desirable. He has demonstrated that a tuple<T...> where all of the T are implicitly convertible from U... should have a tuple constructor that is also implicit, instead of explicit. This would support the use cases in LWG 2051 while not undermining T... with explicit conversions from U.... This check-in is an experimental implementation of Daniel's work. I believe this work to be mature enough to warrant inclusion into libc++. If anyone sees real-world problems that this check in causes, please let me know and I will revert it, and provide the feedback to the LWG.
llvm-svn: 153855
2012-04-01 23:10:42 +00:00
Bob Wilson 0090df24d7 Ignore unused --serialize-diagnostics option when using llvm-gcc.
llvm-gcc doesn't handle --serialize-diagnostics so when compiling i386
kernel/kext code with -Werror, you get an error about that option being
unused.  Claim the argument to prevent this from breaking builds.
<rdar://problem/11161933>

llvm-svn: 153854
2012-04-01 23:03:29 +00:00
Chandler Carruth 219173a1be Start cleaning up the InlineCost class. This switches to sentinel values
rather than a bitfield, a great suggestion by Chris during code review.

There is still quite a bit of cruft in the interface, but that requires
sorting out some awkward uses of the cost inside the actual inliner.

No functionality changed intended here.

llvm-svn: 153853
2012-04-01 22:44:09 +00:00
Hal Finkel 3ecfa7b277 Fix some 80-col. violations I introduced with the A2 PPC64 core.
llvm-svn: 153852
2012-04-01 21:20:14 +00:00
Hal Finkel 322e41a914 Enable prefetch generation on PPC64.
llvm-svn: 153851
2012-04-01 20:08:17 +00:00
Hal Finkel 9032344c15 Add LdStSTD* itin. for the PPC64 A2 core.
llvm-svn: 153850
2012-04-01 20:08:08 +00:00
Sebastian Redl 048a6d7976 Properly handle explicit constructors in list-initialization. Fixes PR12120.
llvm-svn: 153849
2012-04-01 19:54:59 +00:00
Nadav Rotem b078350872 This commit contains a few changes that had to go in together.
1. Simplify xor/and/or (bitcast(A), bitcast(B)) -> bitcast(op (A,B))
   (and also scalar_to_vector).

2. Xor/and/or are indifferent to the swizzle operation (shuffle of one src).
   Simplify xor/and/or (shuff(A), shuff(B)) -> shuff(op (A, B))

3. Optimize swizzles of shuffles:  shuff(shuff(x, y), undef) -> shuff(x, y).

4. Fix an X86ISelLowering optimization which was very bitcast-sensitive.

Code which was previously compiled to this:

movd    (%rsi), %xmm0
movdqa  .LCPI0_0(%rip), %xmm2
pshufb  %xmm2, %xmm0
movd    (%rdi), %xmm1
pshufb  %xmm2, %xmm1
pxor    %xmm0, %xmm1
pshufb  .LCPI0_1(%rip), %xmm1
movd    %xmm1, (%rdi)
ret

Now compiles to this:

movl    (%rsi), %eax
xorl    %eax, (%rdi)
ret

llvm-svn: 153848
2012-04-01 19:31:22 +00:00
Benjamin Kramer ac19edd2b0 Analyzer: Store BugReports directly in a ilist instead of adding another layer of inderection with std::list
llvm-svn: 153847
2012-04-01 19:30:51 +00:00
Lang Hames 652f21274f Fix typo.
llvm-svn: 153846
2012-04-01 19:27:25 +00:00
Hal Finkel 88ed4e3b15 Set the default PPC node scheduling preference to ILP (for the embedded cores).
The 440 and A2 cores have detailed itineraries, and this allows them to be
fully used to maximize throughput.

llvm-svn: 153845
2012-04-01 19:23:08 +00:00
Hal Finkel b9845f5758 Add ppc440 itin. entries for LdStSTD*
llvm-svn: 153844
2012-04-01 19:23:04 +00:00
Hal Finkel ec5a1e3669 Use full anti-dep. breaking with post-ra sched. on the embedded ppc cores.
Post-RA scheduling gives a significant performance improvement on
the embedded cores, so turn it on. Using full anti-dep. breaking is
important for FP-intensive blocks, so turn it on (just on the
embedded cores for now; this should also be good on the 970s because
post-ra scheduling is all that we have for now, but that should have
more testing first).

llvm-svn: 153843
2012-04-01 19:22:57 +00:00
Hal Finkel 9f9f8929ee Add instruction itinerary for the PPC64 A2 core.
This adds a full itinerary for IBM's PPC64 A2 embedded core. These
cores form the basis for the CPUs in the new IBM BG/Q supercomputer.

llvm-svn: 153842
2012-04-01 19:22:40 +00:00
Craig Topper 91773ab2ca Use SequenceToOffsetTable to create instruction name table. Saves space particularly on X86 where AVX instructions just add a 'v' to the front of other instructions.
llvm-svn: 153841
2012-04-01 18:14:14 +00:00
Tobias Grosser d6d7d7e128 CodeGen: Remove unused variable
llvm-svn: 153840
2012-04-01 16:55:30 +00:00
Tobias Grosser de49ef76f6 Remove unneeded alias analysis
llvm-svn: 153839
2012-04-01 16:49:48 +00:00
Tobias Grosser 89339067b0 CodeGen: Allow function parameters to be rewritten in getNewValue()
When deriving new values for the statements of a SCoP, we assumed that parameter
values are constant within the SCoP and consquently do not need to be rewritten.
For OpenMP code generation this assumption is wrong, as such values are not
available in the OpenMP subfunction and consequently also may need to be
rewritten.

Committed with some changes.

Contributed-By: Johannes Doerfert  <s9jodoer@stud.uni-saarland.de>
llvm-svn: 153838
2012-04-01 16:49:45 +00:00
Benjamin Kramer 12af4285d1 Emit the LLVM<->DWARF register mapping as a sorted table and use binary search to do the lookup.
This also avoids emitting the information twice, which led to code bloat. On i386-linux-Release+Asserts
with all targets built this change shaves a whopping 1.3 MB off clang. The number is probably exaggerated
by recent inliner changes but the methods were already enormous with the old inline cost computation.

The DWARF reg -> LLVM reg mapping doesn't seem to have holes in it, so it could be a simple lookup table.
I didn't implement that optimization yet to avoid potentially changing functionality.

There is still some duplication both in tablegen and the generated code that should be cleaned up eventually.

llvm-svn: 153837
2012-04-01 14:23:58 +00:00
Chandler Carruth 45ae88f5fc Belatedly address some code review from Chris.
As a side note, I really dislike array_pod_sort... Do we really still
care about any STL implementations that get this so wrong? Does libc++?

llvm-svn: 153834
2012-04-01 10:41:24 +00:00
Chandler Carruth cdb1f8cff1 Add some more testing to cover the remaining two cases where
always-inlining is disabled: recursive functions and indirectbr.

llvm-svn: 153833
2012-04-01 10:36:17 +00:00
Chandler Carruth c5bfb3c0f5 Fix a pretty scary bug I introduced into the always inliner with
a single missing character. Somehow, this had gone untested. I've added
tests for returns-twice logic specifically with the always-inliner that
would have caught this, and fixed the bug.

Thanks to Matt for the careful review and spotting this!!! =D

llvm-svn: 153832
2012-04-01 10:21:05 +00:00
Chandler Carruth 1989bb9c43 Replace four tiny tests with various uses of grep and not with a single
test and FileCheck.

llvm-svn: 153831
2012-04-01 10:11:17 +00:00
Andrew Trick 779b32a44e misched: Add finalizeScheduler to complete the target interface.
llvm-svn: 153827
2012-04-01 07:24:23 +00:00
Eli Bendersky f5becf617f Removing a file that's no longer being used after the recent refactorings
llvm-svn: 153825
2012-04-01 06:50:01 +00:00
Sean Callanan 139656580f Updating Xcode project version numbers for lldb-137 and debugserver-186
llvm-svn: 153823
2012-04-01 06:36:59 +00:00
Sean Callanan fe630c51ba Updated LLVM to take a comprehensive fix to the
JIT that enables paired relocations in the i386
Mach-O JIT.  This eliminates crashes in the
testsuite when running under i386.

llvm-svn: 153822
2012-04-01 06:35:00 +00:00
Hal Finkel 59607e63cb Split the LdStGeneral PPC itin. class into LdStLoad and LdStStore.
Loads and stores can have different pipeline behavior, especially on
embedded chips. This change allows those differences to be expressed.
Except for the 440 scheduler, there are no functionality changes.
On the 440, the latency adjustment is only by one cycle, and so this
probably does not affect much. Nevertheless, it will make a larger
difference in the future and this removes a FIXME from the 440 itin.

llvm-svn: 153821
2012-04-01 04:44:16 +00:00
Rafael Espindola 1eaae50734 Add a workaround for building with old versions of clang.
llvm-svn: 153820
2012-03-31 21:54:20 +00:00
Abramo Bagnara c9e48c0a3f Fixed MacroInfo definition range for function like macros with empty body.
llvm-svn: 153819
2012-03-31 20:17:27 +00:00
Rafael Espindola 77242fa79e Add a triple to the test.
llvm-svn: 153818
2012-03-31 18:59:07 +00:00
Rafael Espindola 80c540e656 Teach CodeGen's version of computeMaskedBits to understand the range metadata.
This is the CodeGen equivalent of r153747. I tested that there is not noticeable
performance difference with any combination of -O0/-O2 /-g when compiling
gcc as a single compilation unit.

llvm-svn: 153817
2012-03-31 18:14:00 +00:00
Hal Finkel 51861b4855 Fix dynamic linking on PPC64.
Dynamic linking on PPC64 has had problems since we had to move the top-down
hazard-detection logic post-ra. For dynamic linking to work there needs to be
a nop placed after every call. It turns out that it is really hard to guarantee
that nothing will be placed in between the call (bl) and the nop during post-ra
scheduling. Previous attempts at fixing this by placing logic inside the
hazard detector only partially worked.

This is now fixed in a different way: call+nop codegen-only instructions. As far
as CodeGen is concerned the pair is now a single instruction and cannot be split.
This solution works much better than previous attempts.

The scoreboard hazard detector is also renamed to be more generic, there is currently
no cpu-specific logic in it.

llvm-svn: 153816
2012-03-31 14:45:15 +00:00
Chandler Carruth 1a4cc6cc9f Fix a typo reported in IRC by someone reviewing this code.
llvm-svn: 153815
2012-03-31 13:18:09 +00:00
Chandler Carruth a88a0faaa3 Give the always-inliner its own custom filter. It shouldn't have to pay
the very high overhead of the complex inline cost analysis when all it
wants to do is detect three patterns which must not be inlined. Comment
the code, clean it up, and leave some hints about possible performance
improvements if this ever shows up on a profile.

Moving this off of the (now more expensive) inline cost analysis is
particularly important because we have to run this inliner even at -O0.

llvm-svn: 153814
2012-03-31 13:17:18 +00:00
Chandler Carruth edd2826f3e Remove a bunch of empty, dead, and no-op methods from all of these
interfaces. These methods were used in the old inline cost system where
there was a persistent cache that had to be updated, invalidated, and
cleared. We're now doing more direct computations that don't require
this intricate dance. Even if we resume some level of caching, it would
almost certainly have a simpler and more narrow interface than this.

llvm-svn: 153813
2012-03-31 12:48:08 +00:00
Chandler Carruth 0539c071ea Initial commit for the rewrite of the inline cost analysis to operate
on a per-callsite walk of the called function's instructions, in
breadth-first order over the potentially reachable set of basic blocks.

This is a major shift in how inline cost analysis works to improve the
accuracy and rationality of inlining decisions. A brief outline of the
algorithm this moves to:

- Build a simplification mapping based on the callsite arguments to the
  function arguments.
- Push the entry block onto a worklist of potentially-live basic blocks.
- Pop the first block off of the *front* of the worklist (for
  breadth-first ordering) and walk its instructions using a custom
  InstVisitor.
- For each instruction's operands, re-map them based on the
  simplification mappings available for the given callsite.
- Compute any simplification possible of the instruction after
  re-mapping, and store that back int othe simplification mapping.
- Compute any bonuses, costs, or other impacts of the instruction on the
  cost metric.
- When the terminator is reached, replace any conditional value in the
  terminator with any simplifications from the mapping we have, and add
  any successors which are not proven to be dead from these
  simplifications to the worklist.
- Pop the next block off of the front of the worklist, and repeat.
- As soon as the cost of inlining exceeds the threshold for the
  callsite, stop analyzing the function in order to bound cost.

The primary goal of this algorithm is to perfectly handle dead code
paths. We do not want any code in trivially dead code paths to impact
inlining decisions. The previous metric was *extremely* flawed here, and
would always subtract the average cost of two successors of
a conditional branch when it was proven to become an unconditional
branch at the callsite. There was no handling of wildly different costs
between the two successors, which would cause inlining when the path
actually taken was too large, and no inlining when the path actually
taken was trivially simple. There was also no handling of the code
*path*, only the immediate successors. These problems vanish completely
now. See the added regression tests for the shiny new features -- we
skip recursive function calls, SROA-killing instructions, and high cost
complex CFG structures when dead at the callsite being analyzed.

Switching to this algorithm required refactoring the inline cost
interface to accept the actual threshold rather than simply returning
a single cost. The resulting interface is pretty bad, and I'm planning
to do lots of interface cleanup after this patch.

Several other refactorings fell out of this, but I've tried to minimize
them for this patch. =/ There is still more cleanup that can be done
here. Please point out anything that you see in review.

I've worked really hard to try to mirror at least the spirit of all of
the previous heuristics in the new model. It's not clear that they are
all correct any more, but I wanted to minimize the change in this single
patch, it's already a bit ridiculous. One heuristic that is *not* yet
mirrored is to allow inlining of functions with a dynamic alloca *if*
the caller has a dynamic alloca. I will add this back, but I think the
most reasonable way requires changes to the inliner itself rather than
just the cost metric, and so I've deferred this for a subsequent patch.
The test case is XFAIL-ed until then.

As mentioned in the review mail, this seems to make Clang run about 1%
to 2% faster in -O0, but makes its binary size grow by just under 4%.
I've looked into the 4% growth, and it can be fixed, but requires
changes to other parts of the inliner.

llvm-svn: 153812
2012-03-31 12:42:41 +00:00
Chandler Carruth 056b460917 Add support to the InstVisitor for visiting a generic callsite. The
visitor will now visit a CallInst and an InvokeInst with
instruction-specific visitors, then visit a generic CallSite visitor,
then delegate back to the Instruction visitor and the TerminatorInst
visitors depending on whether a call or an invoke originally. This will
be used in the soon-to-land inline cost rewrite.

llvm-svn: 153811
2012-03-31 11:31:24 +00:00
Bill Wendling 62152e7389 Move trivial functions into the class definition.
llvm-svn: 153810
2012-03-31 11:25:18 +00:00
Bill Wendling 4f87c4fdff Trim headers.
llvm-svn: 153809
2012-03-31 11:22:30 +00:00
Bill Wendling 0e1824cb9e Indent according to LLVM's style guide.
llvm-svn: 153808
2012-03-31 11:15:43 +00:00
Bill Wendling dbc02d84ce Cleanup whitespace and trim some of the #includes.
llvm-svn: 153807
2012-03-31 11:10:35 +00:00
Benjamin Kramer 53dc873342 Internalize: Remove reference of @llvm.noinline, it was replaced with the noinline attribute a long time ago.
llvm-svn: 153806
2012-03-31 11:03:47 +00:00
Bill Wendling 5c15044f47 These strings aren't 'const char *' but 'char *'.
llvm-svn: 153805
2012-03-31 10:51:45 +00:00