Commit Graph

4717 Commits

Author SHA1 Message Date
Arnold Schwaighofer
1bdb320dae BasicAA: Fix value equality and phi cycles
When there are cycles in the value graph we have to be careful interpreting
"Value*" identity as "value" equivalence. We interpret the value of a phi node
as the value of its operands.
When we check for value equivalence now we make sure that the "Value*" dominates
all cycles (phis).

%0 = phi [%noaliasval, %addr2]
%l = load %ptr
%addr1 = gep @a, 0, %l
%addr2 = gep @a, 0, (%l + 1)
store %ptr ...

Before this patch we would return NoAlias for (%0, %addr1) which is wrong
because the value of the load is from different iterations of the loop.

Tested on x86_64 -mavx at O3 and O3 -flto with no performance or compile time
regressions.

PR18068
radar://15653794

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198290 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-02 03:31:36 +00:00
Yuchen Wu
5f2ddd6a0c BlockFrequencyInfo: Readded getEntryFreq.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197839 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-20 22:11:11 +00:00
Michael Gottesman
b68ca3e9c5 [block-freq] Teach branch probability how to return the edge weight in between a BasicBlock and one of its successors.
IMHO At some point BasicBlock should be refactored along the lines of
MachineBasicBlock so that successors/weights are actually embedded within the
block. Now is not that time though.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197303 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-14 02:24:25 +00:00
Matt Arsenault
1e80bef896 Teach MemoryBuiltins about address spaces
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197292 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-14 00:27:48 +00:00
Michael Gottesman
b8c5cfb130 [block-freq] Update BlockFrequencyInfo/MachineBlockFrequencyInfo to use the new print methods.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197289 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-14 00:25:42 +00:00
Michael Gottesman
a73959a988 [block-freq] Add the equivalent methods to MachineBlockFrequencyInfo and BlockFrequencyInfo that were added to BlockFrequencyImpl in r197285 and r197284.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197287 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-14 00:06:03 +00:00
Chandler Carruth
a7e2ef7e6e [inliner] Fix PR18206 by preventing inlining functions that call setjmp
through an invoke instruction.

The original patch for this was written by Mark Seaborn, but I've
reworked his test case into the existing returns_twice test case and
implemented the fix by the prior refactoring to actually run the cost
analysis over invoke instructions, and then here fixing our detection of
the returns_twice attribute to work for both calls and invokes. We never
noticed because we never saw an invoke. =[

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197216 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-13 08:00:01 +00:00
Chandler Carruth
0bdc7cd5de [inliner] Completely change (and fix) how the inline cost analysis
handles terminator instructions.

The inline cost analysis inheritted some pretty rough handling of
terminator insts from the original cost analysis, and then made it much,
much worse by factoring all of the important analyses into a separate
instruction visitor. That instruction visitor never visited the
terminator.

This works fine for things like conditional branches, but for many other
things we simply computed The Wrong Value. First example are
unconditional branches, which should be free but were counted as full
cost. This is most significant for conditional branches where the
condition simplifies and folds during inlining. We paid a 1 instruction
tax on every branch in a straight line specialized path. =[

Oh, we also claimed that the unreachable instruction had cost.

But it gets worse. Let's consider invoke. We never applied the call
penalty. We never accounted for the cost of the arguments. Nope. Worse
still, we didn't handle the *correctness* constraints of not inlining
recursive invokes, or exception throwing returns_twice functions. Oops.
See PR18206. Sadly, PR18206 requires yet another fix, but this
refactoring is at least a huge step in that direction.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197215 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-13 07:59:56 +00:00
Chandler Carruth
8a513abbc8 [cleanup] Remove trailing whitespace before I start changing this file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@197149 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-12 11:59:26 +00:00
Jakub Staszak
7ae72bfd94 Don't #include heavy Dominators.h file in LoopInfo.h. This change reduces
overall time of LLVM compilation by ~1%.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196667 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-07 21:20:17 +00:00
Alp Toker
087ab613f4 Correct word hyphenations
This patch tries to avoid unrelated changes other than fixing a few
hyphen-related ambiguities and contractions in nearby lines.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196471 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-05 05:44:44 +00:00
Eric Christopher
1cf9e7ff96 Fix typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196434 91177308-0d34-0410-b5e6-96231b3b80d8
2013-12-04 23:55:09 +00:00
Chandler Carruth
54fec07ec0 [PM] Split the CallGraph out from the ModulePass which creates the
CallGraph.

This makes the CallGraph a totally generic analysis object that is the
container for the graph data structure and the primary interface for
querying and manipulating it. The pass logic is separated into its own
class. For compatibility reasons, the pass provides wrapper methods for
most of the methods on CallGraph -- they all just forward.

This will allow the new pass manager infrastructure to provide its own
analysis pass that constructs the same CallGraph object and makes it
available. The idea is that in the new pass manager, the analysis pass's
'run' method returns a concrete analysis 'result'. Here, that result is
a 'CallGraph'. The 'run' method will typically do only minimal work,
deferring much of the work into the implementation of the result object
in order to be lazy about computing things, but when (like DomTree)
there is *some* up-front computation, the analysis does it prior to
handing the result back to the querying pass.

I know some of this is fairly ugly. I'm happy to change it around if
folks can suggest a cleaner interim state, but there is going to be some
amount of unavoidable ugliness during the transition period. The good
thing is that this is very limited and will naturally go away when the
old pass infrastructure goes away. It won't hang around to bother us
later.

Next up is the initial new-PM-style call graph analysis. =]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195722 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-26 04:19:30 +00:00
Chandler Carruth
bdd300b22c [PM] Reformat some code with clang-format as I'm going to be editting as
part of generalizing the call graph infrastructure for the new pass
manager.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195718 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-26 03:45:26 +00:00
Chandler Carruth
0c9fcd2be8 [PM] Rename the 'Mod' member to the more idiomatic 'M'. No functionality
changed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195701 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-26 00:37:23 +00:00
Kostya Serebryany
42a9da35b8 Don't speculate loads under ThreadSanitizer
Summary:
Don't speculate loads under ThreadSanitizer.
This fixes https://code.google.com/p/thread-sanitizer/issues/detail?id=40
Also discussed here: http://lists.cs.uiuc.edu/pipermail/llvmdev/2013-November/067929.html

Reviewers: chandlerc

Reviewed By: chandlerc

CC: llvm-commits, dvyukov

Differential Revision: http://llvm-reviews.chandlerc.com/D2227

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195324 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-21 07:29:28 +00:00
Paul Robinson
fe45fd084d The 'optnone' attribute means don't inline anything into this function
(except functions marked always_inline).
Functions with 'optnone' must also have 'noinline' so they don't get
inlined into any other function.

Based on work by Andrea Di Biagio.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195046 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-18 21:44:03 +00:00
Benjamin Kramer
b69143c6a9 Annotate APInt methods where it's not clear whether they are in place with warn_unused_result.
Fix ScalarEvolution bugs uncovered by this.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194928 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-16 16:25:41 +00:00
Matt Arsenault
74c996cbd1 Use correct size for address space in BasicAA.
The tests just hit this with a different sized
address space since I haven't figured out how
to use this to break it.

I thought I committed this a long time ago,
and I'm not sure why missing this hasn't caused
any problems.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194903 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-16 00:36:43 +00:00
Matt Arsenault
59d3ae6cdc Add addrspacecast instruction.
Patch by Michele Scandale!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194760 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-15 01:34:59 +00:00
Michael Gottesman
e2058ff5bd Added BlockFrequencyInfo::view for displaying the block frequency propagation graph via graphviz.
This is useful for debugging issues in the BlockFrequency implementation since
one can easily visualize where probability mass and other errors occur in the
propagation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194654 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-14 02:27:46 +00:00
Yunzhong Gao
2999b2f2cc Fixing a heisenbug where the memory dependence analysis behaves differently
with and without -g.

Adding a test case to make sure that the threshold used in the memory
dependence analysis is respected. The test case also checks that debug
intrinsics are not counted towards this threshold.

Differential Revision: http://llvm-reviews.chandlerc.com/D2141



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194646 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-14 01:10:52 +00:00
Michael Gottesman
006806267a Fixed 80+ violations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194634 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-14 00:05:07 +00:00
Sebastian Pop
f44941d81d add more comments around the delinearization of arrays
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194612 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-13 22:37:58 +00:00
Jakub Staszak
dc9a217d05 Simplify code. No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194602 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-13 20:18:38 +00:00
Benjamin Kramer
1a362619f8 Move Delinearization pass into an anonymous namespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194582 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-13 15:35:17 +00:00
Sebastian Pop
5230ad61fd delinearization of arrays
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194527 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-12 22:47:20 +00:00
Wan Xiaofei
3cda2d3885 Change data structure to memorize computed result in ScalarEvolution
Replace std::map with SmallVector to memorize the cached result since SCEV usually belongs to little Loop/BB
Linear scan on SmallVector is faster than std::map.

Code reviewer : Andrew Trick.
Test result   : Pass Unit Test & LLVM Test Suite

401.bzip2	0.425721	0.419981	101.37%
403.gcc		24.53855	24.2667		101.12%
429.mcf		0.060847	0.059944	101.51%
433.milc	0.646009	0.636119	101.55%
444.namd	1.383928	1.370614	100.97%
445.gobmk	5.836575	5.800225	100.63%
450.soplex	1.911257	1.895963	100.81%
456.hmmer	1.039565	1.032534	100.68%
458.sjeng	0.897401	0.885567	101.34%
464.h264ref	3.645908	3.577991	101.90%
470.lbm		0.049456	0.048398	102.19%
471.omnetpp	5.638575	5.60435		100.61%
bitmnp01	0.045738	0.045291	100.99%
cjpegv2data	0.304359	0.302833	100.50%
idctrn01	0.046433	0.045763	101.46%
quake2		4.534416	4.4952		100.87%
quake		2.688566	2.659208	101.10%
xcsoar		12.42545	12.30385	100.99%
linpack		0.038739	0.03803		101.86%
matrix01	0.053564	0.0528		101.45%
nbench		0.402867	0.395803	101.78%
tblook01	0.021265	0.021015	101.19%
ttsprk01	0.066384	0.065566	101.25%

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194459 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-12 09:40:41 +00:00
Matt Arsenault
20f1fe5c50 Use size function instead of manually calculating it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194345 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-10 03:18:50 +00:00
Chandler Carruth
49837ef811 Move the old pass manager infrastructure into a legacy namespace and
give the files a legacy prefix in the right directory. Use forwarding
headers in the old locations to paper over the name change for most
clients during the transitional period.

No functionality changed here! This is just clearing some space to
reduce renaming churn later on with a new system.

Even when the new stuff starts to go in, it is going to be hidden behind
a flag and off-by-default as it is still WIP and under development.

This patch is specifically designed so that very little out-of-tree code
has to change. I'm going to work as hard as I can to keep that the case.
Only direct forward declarations of the PassManager class are impacted
by this change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194324 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-09 12:26:54 +00:00
Andrew Trick
10bb82e54f Rewrite SCEV's backedge taken count computation.
Patch by Michele Scandale!

Rewrite of the functions used to compute the backedge taken count of a
loop on LT and GT comparisons.

I decided to split the handling of LT and GT cases becasue the trick
"a > b == -a < -b" in some cases prevents the trip count computation
due to the multiplication by -1 on the two operands of the
comparison. This issue comes from the conservative computation of
value range of SCEVs: taking the negative SCEV of an expression that
have a small positive range (e.g. [0,31]), we would have a SCEV with a
fullset as value range.

Indeed, in the new rewritten function I tried to better handle the
maximum backedge taken count computation when MAX/MIN expression are
used to handle the cases where no entry guard is found.

Some test have been modified in order to check the new value correctly
(I manually check them and reasoning on possible overflow the new
values seem correct).

I finally added a new test case related to the multiplication by -1
issue on GT comparisons.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194116 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-06 02:08:26 +00:00
Matt Arsenault
896a885f21 Fix another constant folding address space place I missed.
This fixes an assertion failure with a different sized address space.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194014 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-04 20:46:52 +00:00
Hal Finkel
e14fb07357 Consider (x == -1) unlikely in BranchProbabilityInfo
This adds another heuristic to BPI, similar to the existing heuristic that
considers (x == 0) unlikely to be true. As suggested in the PACT'98 paper by
Deitrich, Cheng, and Hwu, -1 is often used to indicate an invalid index, and
equality comparisons with -1 are also unlikely to succeed. Local
experimentation supports this hypothesis: This yields a 1-2% speedup in the
test-suite sqlite benchmark on the PPC A2 core, with no significant
regressions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193855 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-01 10:58:22 +00:00
Rafael Espindola
c143c7573b Merge CallGraph and BasicCallGraph.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193734 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-31 03:03:55 +00:00
Benjamin Kramer
19ea37059a SCEV: Make the final add of an inbounds GEP nuw if we know that the index is positive.
We can't do this for the general case as saying a GEP with a negative index
doesn't have unsigned wrap isn't valid for negative indices.
  %gep = getelementptr inbounds i32* %p, i64 -1

But an inbounds GEP cannot run past the end of address space. So we check for
the very common case of a positive index and make GEPs derived from that NUW.
Together with Andy's recent non-unit stride work this lets us analyze loops
like

  void foo3(int *a, int *b) {
    for (; a < b; a++) {}
  }

PR12375, PR12376.

Differential Revision: http://llvm-reviews.chandlerc.com/D2033

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193514 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-28 07:30:06 +00:00
Shuxin Yang
69bd41dfe3 Revert r193251 : Use address-taken to disambiguate global variable and indirect memops.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193489 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-27 03:08:44 +00:00
Wan Xiaofei
887f9c5ec1 Quick look-up for block in loop.
This patch implements quick look-up for block in loop by maintaining a hash set for blocks.
It improves the efficiency of loop analysis a lot, the biggest improvement could be 5-6%(458.sjeng).
Below are the compilation time for our benchmark in llc before & after the patch.

Benchmark	llc - trunk		llc - patched	
401.bzip2	0.339081	100.00%	0.329657	102.86%
403.gcc		19.853966	100.00%	19.605466	101.27%
429.mcf		0.049823	100.00%	0.048451	102.83%
433.milc	0.514898	100.00%	0.510217	100.92%
444.namd	1.109328	100.00%	1.103481	100.53%
445.gobmk	4.988028	100.00%	4.929114	101.20%
456.hmmer	0.843871	100.00%	0.825865	102.18%
458.sjeng	0.754238	100.00%	0.714095	105.62%
464.h264ref	2.9668		100.00%	2.90612		102.09%
471.omnetpp	4.556533	100.00%	4.511886	100.99%
bitmnp01	0.038168	100.00%	0.0357		106.91%
idctrn01	0.037745	100.00%	0.037332	101.11%
libquake2	3.78689		100.00%	3.76209		100.66%
libquake_	2.251525	100.00%	2.234104	100.78%
linpack		0.033159	100.00%	0.032788	101.13%
matrix01	0.045319	100.00%	0.043497	104.19%
nbench		0.333161	100.00%	0.329799	101.02%
tblook01	0.017863	100.00%	0.017666	101.12%
ttsprk01	0.054337	100.00%	0.053057	102.41%

Reviewer	: Andrew Trick <atrick@apple.com>, Hal Finkel <hfinkel@anl.gov>
Approver	: Andrew Trick <atrick@apple.com>
Test		: Pass make check-all & llvm test-suite


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193460 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-26 03:08:02 +00:00
Andrew Trick
4d4bbaf997 Fix SCEVExpander: don't try to expand quadratic recurrences outside a loop.
Partial fix for PR17459: wrong code at -O3 on x86_64-linux-gnu
(affecting trunk and 3.3)

When SCEV expands a recurrence outside of a loop it attempts to scale
by the stride of the recurrence. Chained recurrences don't work that
way. We could compute binomial coefficients, but would hve to
guarantee that the chained AddRec's are in a perfectly reduced form.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193438 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-25 21:35:56 +00:00
Andrew Trick
8aa8cea3e9 Fix LSR: don't normalize quadratic recurrences.
Partial fix for PR17459: wrong code at -O3 on x86_64-linux-gnu
(affecting trunk and 3.3)

ScalarEvolutionNormalization was attempting to normalize by adding and
subtracting strides. Chained recurrences don't work that way.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193437 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-25 21:35:52 +00:00
Rafael Espindola
6e1c511aba Call destroy from ~BasicCallGraph.
This fix a memory leak found by valgrind.

Calling it from the base class destructor would not destroy the BasicCallGraph
bits.

FIXME: BasicCallGraph is the only thing that inherits from CallGraph. Can
we merge the two?

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193412 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-25 15:01:34 +00:00
Nuno Lopes
5e1d0d39db fix PR17635: false positive with packed structures
LLVM optimizers may widen accesses to packed structures that overflow the structure itself, but should be in bounds up to the alignment of the object

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193317 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-24 09:17:24 +00:00
Shuxin Yang
8e3851a6eb Use address-taken to disambiguate global variable and indirect memops.
Major steps include:
 1). introduces a not-addr-taken bit-field in GlobalVariable
 2). GlobalOpt pass sets "not-address-taken" if it proves a global varirable 
    dosen't have its address taken.
 3). AA use this info for disambiguation. 


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193251 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-23 17:28:19 +00:00
Andrew Trick
78fe4d62fb Clarify SCEV comments.
We handle for(i=n; i>0; i -= s) by canonicalizing within SCEV to for(i=-n; i<0; i += s).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193147 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-22 05:09:40 +00:00
Manman Ren
11d78777d5 TBAA: fix PR17620.
We can have a struct type with a single field and the field does not start
with 0. In that case, we should correctly update the offset.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193137 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-22 01:40:25 +00:00
Matt Arsenault
244d245974 Use more type helper functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193109 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-21 19:43:56 +00:00
Matt Arsenault
4784bb6f44 Fix creating bitcasts between address spaces in SCEV.
The test before wasn't successfully testing this
since it was missing the datalayout piece to change
the size of the second address space.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193102 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-21 18:41:10 +00:00
Matt Arsenault
325ee6e115 Remove unused SCEV functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193097 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-21 18:08:09 +00:00
Andrew Trick
a5c5bc9948 SCEV should use NSW to get trip count for positive nonunit stride loops.
SCEV currently fails to compute loop counts for nonunit stride
loops. This comes up frequently. It prevents loop optimization and
forces vectorization to insert extra loop checks.

For example:
void foo(int n, int *x) {
 for (int i = 0; i < n; i += 3) {
   x[i] = i;
   x[i+1] = i+1;
   x[i+2] = i+2;
 }
}

We need to properly handle the case in which limit > INT_MAX-stride. In
the above case: n > INT_MAX-3. In this case the loop counter will step
beyond the limit and overflow at the same time. However, knowing that
signed integer overlow in undefined, we can assume the loop test
behavior is arbitrary after overflow. This obeys both C undefined
behavior rules, and the more strict LLVM poison value rules.

I'm finally fixing this in response to Hal Finkel's persistence.
The most probable reason that we never optimized this before is that
we were being careful to handle case where the developer expected a
side-effect free infinite loop relying on overflow:

for (int i = 0; i < n; i += s) {
  ++j;
}
return j;

If INT_MAX+1 is a multiple of s and n > INT_MAX-s, then we might
expect an infinite loop. However there are plenty of ways to achieve
this effect without relying on undefined behavior of signed overflow.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193015 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-18 23:43:53 +00:00
Craig Topper
390ff499f0 Remove x86_sse42_crc32_64_8 intrinsic. It has no functional difference from x86_sse42_crc32_32_8 and was not mapped to a clang builtin. I'm not even sure why this form of the instruction is even called out explicitly in the docs. Also add AutoUpgrade support to convert it into the other intrinsic with appropriate trunc and zext.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@192672 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-15 05:20:47 +00:00
Matt Arsenault
cee51c4803 Rename DataLayout variables TD -> DL
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191927 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-03 19:50:01 +00:00