Commit Graph

51250 Commits

Author SHA1 Message Date
Sebastian Redl 7379577ab0 Lay the groundwork for converting the entire parser-sema chain to smart pointers.
llvm-svn: 60782
2008-12-09 19:36:21 +00:00
Steve Naroff b0498eebc2 Sema::ActOnMethodDeclaration(): Make sure we perform the default function/array conversion for parameter types.
This fixes <rdar://problem/6424064> checker on xcode: (possible bad AST) can the type of a method parameter really have "isFunctionType() == true"?
and http://llvm.org/bugs/show_bug.cgi?id=2997.

llvm-svn: 60781
2008-12-09 19:36:17 +00:00
Chris Lattner b6fc4b8d92 Switch GVN::processNonLocalLoad to using the new
MemDep::getNonLocalPointerDependency method.  There are
some open issues with this (missed optimizations) and
plenty of future work, but this does allow GVN to eliminate
*slightly* more loads (49246 vs 49033).

Switching over now allows simplification of the other code
path in memdep.

llvm-svn: 60780
2008-12-09 19:25:07 +00:00
Chris Lattner 0a5a8d54a9 random cleanups, no functionality change.
llvm-svn: 60779
2008-12-09 19:21:47 +00:00
Chris Lattner 2e41b0e6e7 Fix a serious null termination bug found by David Chisnall!
llvm-svn: 60778
2008-12-09 19:10:54 +00:00
Evan Cheng 058522f1da xfail this for now.
llvm-svn: 60777
2008-12-09 18:43:00 +00:00
Evan Cheng 8f81e4e31b Back out 60748 for now. It's breaking SPASS, 254.gap, and 464.h264ref.
llvm-svn: 60776
2008-12-09 18:21:49 +00:00
Evan Cheng 74a24b260e Cosmetic changes.
llvm-svn: 60771
2008-12-09 17:56:30 +00:00
Nuno Lopes 10ab44a3a2 remove unused var
llvm-svn: 60770
2008-12-09 17:04:06 +00:00
Dan Gohman 7d32974069 Whitespace cleanups.
llvm-svn: 60769
2008-12-09 16:37:48 +00:00
Mikhail Glushenkov e001666100 Remove Clang tests since clang is not installed on the buildbots.
llvm-svn: 60767
2008-12-09 15:11:45 +00:00
Mikhail Glushenkov 5752117a5b Add some rudimentary tests for .
llvm-svn: 60766
2008-12-09 14:41:27 +00:00
Mikhail Glushenkov e5cd77150e Man page typo fix.
llvm-svn: 60765
2008-12-09 14:41:00 +00:00
Mikhail Glushenkov 65b0b2bf28 Support -emit-llvm properly (with -S and -c).
llvm-svn: 60764
2008-12-09 14:40:18 +00:00
Sebastian Redl e0c2affc5e Fill in C++ status table for expressions.
llvm-svn: 60763
2008-12-09 14:40:01 +00:00
Sebastian Redl 17f2c7d251 Consistently use smart pointers for stmt and expr nodes in parser local variables.
llvm-svn: 60761
2008-12-09 13:15:23 +00:00
Steve Naroff 08628db9d2 Fix <rdar://problem/6429113> clang ObjC rewriter: crash rewriting file with Blocks and properties
More fancy footwork to cope with rewriting property 'setters'.

llvm-svn: 60760
2008-12-09 12:56:34 +00:00
Zhongxing Xu 2a46a8ef52 initialize member variable.
llvm-svn: 60759
2008-12-09 12:26:46 +00:00
Zhongxing Xu 2e8123e5bb Add utility method. Remove an unused method.
llvm-svn: 60758
2008-12-09 10:51:19 +00:00
Duncan Sands 3812542a2c Handle a compiler warning.
llvm-svn: 60755
2008-12-09 09:58:11 +00:00
Chris Lattner 4f10733cf3 fix typos gabor noticed
llvm-svn: 60754
2008-12-09 08:38:36 +00:00
Bill Wendling e25d3417f5 Correct my English.
llvm-svn: 60753
2008-12-09 07:55:31 +00:00
Chris Lattner 75510d8d5c restructure the top level non-local ptr dep query to handle
the first block of a query specially.  This makes the "complete query
caching" subsystem more effective, avoiding predecessor queries.  This
speeds up GVN another 4%.

llvm-svn: 60752
2008-12-09 07:52:59 +00:00
Chris Lattner f903fe1df0 rename getNonLocalPointerDepInternal -> getNonLocalPointerDepFromBB
and split its inner loop out into a new GetNonLocalInfoForBlock
function.  No functionality change.

llvm-svn: 60751
2008-12-09 07:47:11 +00:00
Evan Cheng 75900bf5cd DisableGVCompilation should not abort on internal GlobalValue's.
llvm-svn: 60750
2008-12-09 07:31:49 +00:00
Mon P Wang 8a5366332f In LegalizeOp, don't change the result type of CONVERT_RNDSAT when promoting
one of its operand.

llvm-svn: 60749
2008-12-09 07:27:39 +00:00
Nick Lewycky f545749f2b It's easy to handle SLE/SGE when the loop has a unit stride.
llvm-svn: 60748
2008-12-09 07:25:04 +00:00
Chris Lattner aeaec0838b if we have two elements, insert both, don't use std::sort.
This speeds up the new GVN by another 3%

llvm-svn: 60747
2008-12-09 07:05:45 +00:00
Chris Lattner 4d1281cdf2 If we're only adding one new element to 'Cache', insert it into its known
position instead of using a full sort. This speeds up GVN by ~4% with the
new memdep stuff.

llvm-svn: 60746
2008-12-09 06:58:04 +00:00
Chris Lattner e8113a70fa convert a couple other places that use pred_iterator to use the caching
pred iterator.

llvm-svn: 60745
2008-12-09 06:44:17 +00:00
Chris Lattner 18e1da3e80 newline at end of phile
llvm-svn: 60744
2008-12-09 06:32:32 +00:00
Chris Lattner 768e5bcafc use hte new pred cache to speed up the new non-local memdep
queries.  This speeds up GVN using the new queries (not yet
checked in) by just over 10%.

llvm-svn: 60743
2008-12-09 06:28:49 +00:00
Chris Lattner 4c9bb5336a pred_iterator got a lot slower since use-diet landed. This is a
really simple cache class for these queries.  Hopefully this can
be removed if pred_iterator speeds back up.

llvm-svn: 60742
2008-12-09 06:27:07 +00:00
Scott Michel 02e2c2450e CellSPU:
- Fix call.ll and call_indirect.ll expected results, now that it's using a
  different pre-register allocation scheduler.

llvm-svn: 60741
2008-12-09 06:12:03 +00:00
Mon P Wang 4dd832d241 Fix getNode to allow a vector for the shift amount for shifts of vectors.
Fix the shift amount when unrolling a vector shift into scalar shifts.
Fix problem in getShuffleScalarElt where it assumes that the input of
a bit convert must be a vector.

llvm-svn: 60740
2008-12-09 05:46:39 +00:00
Chris Lattner 56b20ffc5f Fix a really subtle off-by-one bug that Duncan noticed with valgrind
on test/CodeGen/Generic/2007-06-06-CriticalEdgeLandingPad.

llvm-svn: 60739
2008-12-09 04:47:21 +00:00
Scott Michel 8deac5db5a CellSPU:
- Change default scheduling preference to list-burr, which produces somewhat
  better code than the default. Could also use list-tdrr, but need to ask
  dev list about the appropriate handy mnemonic before commiting.

llvm-svn: 60738
2008-12-09 03:37:19 +00:00
Bill Wendling 80b34b3f47 Add initial support for fast-isel of the [SU]ADDO intrinsics. It isn't
complete. For instance, it lowers the common case into this less-than-optimal
code:

        addl    %ecx, %eax
        seto    %cl
        testb   %cl, %cl
        jne     LBB1_2  ## overflow

instead of:

        addl    %ecx, %eax
        jo      LBB1_2  ## overflow

That will come in a future commit.

llvm-svn: 60737
2008-12-09 02:42:50 +00:00
Dan Gohman 16516ba7b3 Fix the name of ISD::TokenFactor in a comment. Thanks Gabor!
llvm-svn: 60736
2008-12-09 02:18:52 +00:00
Chris Lattner c3fbda4d19 remove two unneeded forward declarations, pointed out by Gabor.
llvm-svn: 60735
2008-12-09 02:10:10 +00:00
Ted Kremenek c28ce29a12 [static analyzer] Extend VLA size checking to look for undefined sizes.
llvm-svn: 60734
2008-12-09 00:44:16 +00:00
Dan Gohman 37c496979c Don't charge full latency for an anti-dependence, in this simplistic
pipeline model.

llvm-svn: 60733
2008-12-09 00:26:46 +00:00
Ted Kremenek 08134c984b Update Driver to new interface for LiveVariables.
llvm-svn: 60732
2008-12-09 00:17:51 +00:00
Ted Kremenek d7266bf20a Add zero-sized VLA check test case.
llvm-svn: 60731
2008-12-09 00:14:48 +00:00
Ted Kremenek 6ee0a11814 Fixed LiveVariables bug where we didn't consider block-level expressions that functioned as the size of a VLA to be live.
llvm-svn: 60730
2008-12-09 00:14:14 +00:00
Fariborz Jahanian 8e0079c787 Change condition under which 'retain'/'copy' are directly evaluated.
llvm-svn: 60729
2008-12-08 23:56:17 +00:00
Dan Gohman bc55c2a18f Fix a couple of mistaken switch case fall-throughs. Thanks to Bill
for spotting these!

llvm-svn: 60728
2008-12-08 23:50:06 +00:00
Devang Patel 5f769e2d40 Actually test something. Use PR3170 test case.
llvm-svn: 60727
2008-12-08 23:44:46 +00:00
Ted Kremenek 8f7afdd21e Add checking for zero-sized VLAs.
llvm-svn: 60726
2008-12-08 22:47:34 +00:00
Chris Lattner e598370ae9 remove DebugIterations option. Despite the accusations,
jump threading has been shown to only expose problems not
have bugs itself.  I'm sure it's completely bug free! ;-)

llvm-svn: 60725
2008-12-08 22:44:07 +00:00