Commit Graph

2109 Commits

Author SHA1 Message Date
Devang Patel 31206b56d5 Cosmetic change
"True Loop" and "False Loop" naming terminology to refer two loops
after loop cloning is confusing. Instead just use A_Loop and B_Loop.

llvm-svn: 41287
2007-08-22 19:33:29 +00:00
Devang Patel 90da534987 Refactor loop condition check in a separate function.
llvm-svn: 41282
2007-08-22 18:27:01 +00:00
Devang Patel cd8beb7645 Fix thinko.
Starting value of second loop's induction variable can not be lower 
then starting value of original loop's induction variable.

llvm-svn: 41280
2007-08-22 18:07:47 +00:00
Devang Patel a12000d572 Rename bunch of variables.
llvm-svn: 41250
2007-08-21 21:12:02 +00:00
Devang Patel f98db5e62a Preserve LCSSA.
llvm-svn: 41246
2007-08-21 19:47:46 +00:00
Devang Patel b5933bbbd5 Use SmallVector instead of std::vector.
llvm-svn: 41207
2007-08-21 00:31:24 +00:00
Devang Patel 8f4228d619 s/ExitBlock/ExitingBlock/g
llvm-svn: 41204
2007-08-20 23:51:18 +00:00
Devang Patel 49c4f9a889 Replace indunction variable with split value in loop body.
This fixes art miscompile.

llvm-svn: 41195
2007-08-20 20:49:01 +00:00
Devang Patel c2e2d15f45 Do not split loops rejected by processOneIterationLoop().
llvm-svn: 41194
2007-08-20 20:24:15 +00:00
Nick Lewycky bfa9499a88 Oops, remove assert that wasn't meant to be committed.
llvm-svn: 41170
2007-08-18 23:21:28 +00:00
Nick Lewycky 5b5b1ab9e0 Never insert duplicate edges.
llvm-svn: 41169
2007-08-18 23:18:03 +00:00
Nick Lewycky a0d49dac26 Typo.
llvm-svn: 41168
2007-08-18 15:08:56 +00:00
Devang Patel 1282b6e181 Avoid spliting loops where two split condition branches are not independent.
llvm-svn: 41148
2007-08-18 00:00:32 +00:00
Devang Patel d1fcfcc76c When one branch of condition is eliminated then head of the other
branch is not necessary immediate dominators of merge blcok in all cases.

llvm-svn: 41144
2007-08-17 21:59:16 +00:00
Owen Anderson f5023a7a84 Factor out some code into a helper function.
llvm-svn: 41131
2007-08-16 22:51:56 +00:00
Owen Anderson 221a43604e Add some more comments to GVN.
llvm-svn: 41129
2007-08-16 22:02:55 +00:00
Devang Patel 3640e78057 Dominance frontier is now required.
llvm-svn: 41096
2007-08-15 03:34:53 +00:00
Devang Patel b81bcbde09 Cleanup removeBlocks.
Use dominance frontier to fixup incoming edges of successor blocks not domianted by DeadBB.
Use df_iterator to walk and delete basic blocks dominated by DeadBB.

llvm-svn: 41095
2007-08-15 03:31:47 +00:00
Reid Spencer 0db035567c Remove unneeded header file.
llvm-svn: 41094
2007-08-15 03:01:04 +00:00
Devang Patel f55b79fa71 Avoid triangle loops.
llvm-svn: 41093
2007-08-15 02:14:55 +00:00
Devang Patel 7cad917160 Avoid nested loops at the moment.
llvm-svn: 41090
2007-08-14 23:53:57 +00:00
Devang Patel 33ba97d747 Fix dominance frontier update while removing blocks.
llvm-svn: 41082
2007-08-14 18:35:57 +00:00
Owen Anderson bc271a02fd Eliminate PHI nodes with constant values during normal GVN processing, even when
they're not related to eliminating a load.

llvm-svn: 41081
2007-08-14 18:33:27 +00:00
Owen Anderson 398602a6eb Be more aggressive in pruning unnecessary PHI nodes when doing PHI construction.
llvm-svn: 41080
2007-08-14 18:16:29 +00:00
Owen Anderson 676070d503 Make GVN iterative.
llvm-svn: 41078
2007-08-14 18:04:11 +00:00
Owen Anderson a7b220f23a Fix a case where GVN was failing to return true when it had, in fact, modified
the function.

llvm-svn: 41077
2007-08-14 17:59:48 +00:00
Devang Patel dbe8497d45 Handle last value assignments.
llvm-svn: 41063
2007-08-14 01:30:57 +00:00
Devang Patel f74ccbb4e8 StartValue is already calculated.
llvm-svn: 41062
2007-08-14 00:15:45 +00:00
Devang Patel 948653915f Preserve simple analysis.
llvm-svn: 41054
2007-08-13 22:22:13 +00:00
Devang Patel b8a41bb4f1 Preserve dominator info.
llvm-svn: 41053
2007-08-13 22:13:24 +00:00
Devang Patel f258578206 Split loops and do CFG cleanup.
llvm-svn: 41029
2007-08-12 07:02:51 +00:00
Reid Spencer 9f90f965de Remove unused variables.
llvm-svn: 41028
2007-08-12 04:45:36 +00:00
Chris Lattner 99c8ee2977 Transform a load from an undef/zero global into an undef/global even if we
have complex pointer manipulation going on.  This allows us to compile
stuff like this:

__m128i foo(__m128i x){
                static const unsigned int c_0[4] = { 0, 0, 0, 0 };
                __m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
                x  = _mm_unpacklo_epi8(x,  v_Zero);
                return x;
}

into:

_foo:
        xorps   %xmm1, %xmm1
        punpcklbw       %xmm1, %xmm0
        ret

llvm-svn: 41022
2007-08-11 18:48:48 +00:00
Devang Patel f417c2cc34 Clone loop.
llvm-svn: 40998
2007-08-10 18:07:13 +00:00
Devang Patel 9a4761464f Remove unncessary duplication.
llvm-svn: 40979
2007-08-10 00:59:03 +00:00
Devang Patel 7bdf4531bb Calculate exit and start value of true loop and false loop respectively.
llvm-svn: 40978
2007-08-10 00:53:35 +00:00
Devang Patel 67af6cd7ea ExitCondition and Induction variable are loop constraints
not split condition constraints.

llvm-svn: 40977
2007-08-10 00:33:50 +00:00
Chris Lattner a8e4b4bc7b when we see a unaligned load from an insufficiently aligned global or
alloca, increase the alignment of the load, turning it into an aligned load.

This allows us to compile:

#include <xmmintrin.h>
__m128i foo(__m128i x){
 static const unsigned int c_0[4] = { 0, 0, 0, 0 };
	  __m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
  x  = _mm_unpacklo_epi8(x,  v_Zero);
  return x;
}

into:

_foo:
	punpcklbw	_c_0.5944, %xmm0
	ret
	.data
	.lcomm	_c_0.5944,16,4		# c_0.5944

instead of:

_foo:
	movdqu	_c_0.5944, %xmm1
	punpcklbw	%xmm1, %xmm0
	ret
	.data
	.lcomm	_c_0.5944,16,2		# c_0.5944

llvm-svn: 40971
2007-08-09 19:05:49 +00:00
Owen Anderson 9b1cc8cac0 Make NonLocal and None const in the right way. :-)
llvm-svn: 40961
2007-08-09 04:42:44 +00:00
Devang Patel 42e3e5bec1 Traverse loop blocks' terminators to find split candidates.
llvm-svn: 40960
2007-08-09 01:39:01 +00:00
Devang Patel 0183c797c4 Add cost analysis.
llvm-svn: 40952
2007-08-08 22:25:28 +00:00
Devang Patel 0e34ee25ab Preserve dom info while processing one iteration loop.
llvm-svn: 40947
2007-08-08 21:39:47 +00:00
Owen Anderson b84d3b1c92 Change the None and NonLocal markers in memdep to be const.
llvm-svn: 40946
2007-08-08 21:39:39 +00:00
Devang Patel 8abc5c82b7 Clear split info.
llvm-svn: 40944
2007-08-08 21:18:27 +00:00
Devang Patel 593bf9ceb3 Handle multiple split conditions.
llvm-svn: 40941
2007-08-08 21:02:17 +00:00
Owen Anderson 680862880d Global values also don't undead-ify pointers in our dead alloca's set.
llvm-svn: 40936
2007-08-08 19:12:31 +00:00
Owen Anderson ddf4aee543 Make handleEndBlock significantly faster with one trivial improvement,
and one hack to avoid hitting a bad case when the alias analysis is imprecise.

llvm-svn: 40935
2007-08-08 18:38:28 +00:00
Owen Anderson 50df9685b0 Small improvement: if a function doesn't access memory, we don't need to scan
it for potentially undeading pointers.

llvm-svn: 40933
2007-08-08 17:58:56 +00:00
Owen Anderson 52aaabf74d Add some comments, remove a dead argument, and simplify some control flow.
No functionality change.

llvm-svn: 40932
2007-08-08 17:50:09 +00:00
Owen Anderson b17ab03081 A few more small cleanups.
llvm-svn: 40922
2007-08-08 06:06:02 +00:00
Owen Anderson 0aecf0ebef First round of cleanups from Chris' feedback.
llvm-svn: 40919
2007-08-08 04:52:29 +00:00
Devang Patel 68de1ae816 Embrace patch review feedback.
llvm-svn: 40915
2007-08-08 01:51:27 +00:00
Devang Patel c7e53bdcfd Fix new compare instruction's signness. Caught by Chris during review.
llvm-svn: 40912
2007-08-07 23:17:52 +00:00
Owen Anderson 0cc1a76283 Don't insert nearly as many redundant phi nodes.
llvm-svn: 40909
2007-08-07 23:12:31 +00:00
Devang Patel 19211b6528 Use eraseFromParent().
llvm-svn: 40903
2007-08-07 17:45:35 +00:00
David Greene bacdbaa0da Fix comment typo
llvm-svn: 40898
2007-08-07 16:52:03 +00:00
David Greene 816a190cdf Fix GLIBCXX_DEBUG error triggered by incrementing erased iterator.
llvm-svn: 40897
2007-08-07 16:44:38 +00:00
Devang Patel c70106cb30 Begin loop index split pass.
llvm-svn: 40883
2007-08-07 00:25:56 +00:00
Nick Lewycky 8052019a20 It's safe to fold not of fcmp.
llvm-svn: 40870
2007-08-06 20:04:16 +00:00
David Greene 77b2accbca Make this code more efficient.
llvm-svn: 40861
2007-08-06 15:09:17 +00:00
Reid Spencer d959cfc882 Silence some warnings from doxygen about @param argument name not matching the
actual argument name of the documented function.

llvm-svn: 40851
2007-08-05 19:35:22 +00:00
Chris Lattner f0da7975ea at the end of instcombine, explicitly clear WorklistMap.
This shrinks it down to something small.  On the testcase
from PR1432, this speeds up instcombine from 0.7959s to 0.5000s,
(59%)

llvm-svn: 40840
2007-08-05 08:47:58 +00:00
Nick Lewycky 20f0811fc0 Clean up comments, fix up some confusing code logic.
Predsimplify fails llvm-gcc bootstrap.

llvm-svn: 40815
2007-08-04 18:45:32 +00:00
Chandler Carruth 7132e00de7 This is the patch to provide clean intrinsic function overloading support in LLVM. It cleans up the intrinsic definitions and generally smooths the process for more complicated intrinsic writing. It will be used by the upcoming atomic intrinsics as well as vector and float intrinsics in the future.
This also changes the syntax for llvm.bswap, llvm.part.set, llvm.part.select, and llvm.ct* intrinsics. They are automatically upgraded by both the LLVM ASM reader and the bitcode reader. The test cases have been updated, with special tests added to ensure the automatic upgrading is supported.

llvm-svn: 40807
2007-08-04 01:51:18 +00:00
Owen Anderson 2d19aae4ca Fix a subtle miscompilation. This allows 197.parser to be compiled correctly.
llvm-svn: 40791
2007-08-03 19:59:35 +00:00
Owen Anderson 774761c503 Fix a subtle iterator invalidation bug in a recursive algorithm.
llvm-svn: 40776
2007-08-03 11:03:26 +00:00
Chris Lattner 1f70816c73 Fix an accidental commit.
llvm-svn: 40758
2007-08-02 21:33:36 +00:00
Owen Anderson a8ba659976 Fix 80 col. violations.
llvm-svn: 40751
2007-08-02 18:20:52 +00:00
Owen Anderson 9699a6ea03 Fix 80 col. violations.
llvm-svn: 40750
2007-08-02 18:16:06 +00:00
Owen Anderson e3590584b9 Fix 80 col. violations.
llvm-svn: 40749
2007-08-02 18:11:11 +00:00
Owen Anderson 0ac1fc8ac1 Fix a bug that was causing several miscompilations on SPEC.
llvm-svn: 40746
2007-08-02 17:56:05 +00:00
Chris Lattner dc2cf228ce Replacing a cast with another one does not reduce the number of
casts in the input.

llvm-svn: 40741
2007-08-02 17:23:38 +00:00
Chris Lattner 222b214be7 Disable an xform that causes an infinite loop. This fixes PR1594
llvm-svn: 40739
2007-08-02 16:56:32 +00:00
Chris Lattner 2740694450 wrap some long lines. Major offenders that are left include
gvn, gvnpre, dse, and predsimplify.  To see these, use:

  make check-line-length

llvm-svn: 40738
2007-08-02 16:53:43 +00:00
Devang Patel a882328e61 Update dominator info for the middle blocks created while spliting
exit edge to preserve LCSSA.

Fix dominance frontier update during loop unswitch. This fixes PR 1589, again

llvm-svn: 40737
2007-08-02 15:25:57 +00:00
Chris Lattner b0418fc607 Enhance instcombine to be more aggressive about folding casts of
operations of casts.  This implements InstCombine/zext-fold.ll

llvm-svn: 40726
2007-08-02 06:11:14 +00:00
Chris Lattner d7cb625a9e Fix PR1575 and test/Transforms/CondProp/2007-08-01-InvalidRead.ll
llvm-svn: 40720
2007-08-02 04:47:05 +00:00
Devang Patel 34890b2f27 Undo previous check-in.
llvm-svn: 40698
2007-08-01 23:24:50 +00:00
Devang Patel 561b0c29a3 Update dominator info for the middle blocks created while spliting
exit edge to preserve LCSSA.

Fix dominance frontier update during loop unswitch. This fixes PR 1589.

llvm-svn: 40695
2007-08-01 22:23:50 +00:00
Owen Anderson c321e5e272 Make non-local memdep not be recursive, and fix a bug on 403.gcc that this exposed.
llvm-svn: 40692
2007-08-01 22:01:54 +00:00
Dan Gohman 34d442f274 More explicit keywords.
llvm-svn: 40673
2007-08-01 15:32:29 +00:00
Owen Anderson 10e52eddb3 Rename FastDSE to just DSE.
llvm-svn: 40668
2007-08-01 06:36:51 +00:00
Owen Anderson e4a374812b Move FastDSE in to DeadStoreElimination.
llvm-svn: 40667
2007-08-01 06:30:51 +00:00
Owen Anderson 4894e6d8bc Remove old DSE.
llvm-svn: 40666
2007-08-01 06:30:10 +00:00
David Greene 17a5dfe6f7 New CallInst interface to address GLIBCXX_DEBUG errors caused by
indexing an empty std::vector.

Updates to all clients.

llvm-svn: 40660
2007-08-01 03:43:44 +00:00
Owen Anderson 10ffa860d8 Don't let the memory allocator outsmart GVN. ;-)
llvm-svn: 40655
2007-07-31 23:27:13 +00:00
Owen Anderson 2464f4f048 Fix a failure I accidentally caused in my last commit by mishandling the
removal of redundant phis.

llvm-svn: 40650
2007-07-31 20:18:28 +00:00
Lauro Ramos Venancio 549e775e67 Fix a bug in GetKnownAlignment of packed structs.
llvm-svn: 40649
2007-07-31 20:13:21 +00:00
Owen Anderson d58fa6b09f Fix a misoptimization in aha.
llvm-svn: 40642
2007-07-31 17:43:14 +00:00
Dan Gohman 8c4da37b1f Use SCEVExpander::InsertCastOfTo instead of calling new IntToPtrInst
directly, because the insert point used by the SCEVExpander may vary
from what LSR originally computes.

llvm-svn: 40641
2007-07-31 17:22:27 +00:00
Devang Patel d8b1ceb5b4 Add note.
llvm-svn: 40638
2007-07-31 16:52:25 +00:00
Devang Patel d491198000 Loop unswitch preserves dom info.
Use simple analysis interface to preserve analysis info maintained by other loop passes.

llvm-svn: 40627
2007-07-31 08:03:26 +00:00
Devang Patel b98a097ae9 Implement Simple Analysis interfaces - cloneBasicBlockAnalysis and deleteAnalysisValue.
llvm-svn: 40626
2007-07-31 08:01:41 +00:00
Devang Patel 7d165e1d84 If loop can be unswitched again, then do it yourself.
llvm-svn: 40609
2007-07-30 23:07:10 +00:00
Owen Anderson 850138157e Avoid potential iterator invalidation problems.
llvm-svn: 40607
2007-07-30 21:26:39 +00:00
Devang Patel 14fae50666 Remove dead code.
llvm-svn: 40606
2007-07-30 21:10:44 +00:00
Devang Patel 698852561c Loop Rotation pass preserves dominator tree and frontier.
llvm-svn: 40603
2007-07-30 20:22:53 +00:00
Devang Patel bb97ac4dce LICM preserves scalar evolution and dom frontier.
llvm-svn: 40602
2007-07-30 20:19:59 +00:00
Reid Spencer dff9d69cfb Fix a typo/thinko.
llvm-svn: 40599
2007-07-30 19:53:57 +00:00
Owen Anderson 212d5c27f6 Use more caching when computing non-local dependence. This makes bzip2 not
use up the entire 32-bit address space.

llvm-svn: 40596
2007-07-30 17:29:24 +00:00