Commit Graph

2106 Commits

Author SHA1 Message Date
Chris Lattner
7150371097 remove a test that depends on -debug.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66937 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-13 20:31:48 +00:00
John Criswell
090c0a2ffd Do not attempt to do parial redundancy elimination on void values.
Also fixed a punctuation error in the header comment.
This fixes PR3775.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66542 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-10 15:04:53 +00:00
Evan Cheng
8c7848f17f If a function is marked alwaysinline, it must be inlined (possibly for correctness). Do so even if the callee has dynamic alloca and the caller doesn't.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66539 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-10 07:57:50 +00:00
Devang Patel
412a446974 Ignore debug info while evaluating function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66490 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-09 23:04:12 +00:00
Chris Lattner
29e641761e teach SROA to handle promoting vector allocas with a memset into them into
a vector type instead of into an integer type.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66368 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-08 04:17:04 +00:00
Chris Lattner
c570487d45 Enhance SROA to "promote to scalar" allocas which are
memcpy/memmove'd into or out of.  This fixes a serious
perf issue that Nate ran into.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66366 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-08 04:04:21 +00:00
Chris Lattner
6f6923fd4f fix a serious pessimization that Tron on IRC pointed out where we would
"boolify" pointers, generating really awful code because getting the pointer
value requires a load itself.  Before:

_foo:
	movb	$1, _X.b
	ret
_get:
	xorl	%ecx, %ecx
	movb	_X.b, %al
	testb	%al, %al
	movl	$_Y, %eax
	cmove	%ecx, %eax
	ret

With the xform disabled:

_foo:
	movl	$_Y, _X
	ret
_get:
	movl	_X, %eax
	ret



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66351 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-07 23:32:02 +00:00
Duncan Sands
fc5940d2a0 While thinking about the one-definition-rule and trying
to find a tiny mouse hole to squeeze through, it struck
me that globals without a name can be considered internal
since they can't be referenced from outside the current
module.  This patch makes GlobalOpt give them internal
linkage.  Also done for aliases even though they always
have names, since in my opinion anonymous aliases should
be allowed for consistency with global variables and
functions.  So if that happens one day, this code is ready!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66267 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 10:21:56 +00:00
Devang Patel
00e389c8c8 While converting an aggregate to scalare, ignore and remove aggregate's debug info.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66262 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 07:03:54 +00:00
Devang Patel
06b1e67d44 While hoisting instruction to speculatively execute simple bb, ignore dbg intrinsics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66255 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 06:00:17 +00:00
Devang Patel
c64bc16cae Skip DbgInfoIntrinsic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66244 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 02:59:27 +00:00
Devang Patel
771281f180 Revert 66224.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66233 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 01:39:36 +00:00
Devang Patel
743cdf8edf Revert rev. 66167.
We are still not out of woods yet.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66232 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 01:37:41 +00:00
Devang Patel
f70bda2f81 Do not let debug info prevert globalopt from shriking a global vars to boolean.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66224 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-06 00:21:00 +00:00
Devang Patel
5049600672 GlobalOpt only process non constant local GVs while optimizing global vars.
If non constant local GV named A is used by a constant local GV named B (e.g. llvm.dbg.variable) and B is not used by anyone else then eliminate A as well as B.

In other words, debug info should not interfere in removal of unused GV.
--This life, and those below, will be ignored--

M    test/Transforms/GlobalOpt/2009-03-03-dbg.ll
M    lib/Transforms/IPO/GlobalOpt.cpp


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66167 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-05 18:12:02 +00:00
Bill Wendling
85435046ac Should have XFAILed this test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66086 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-04 22:29:34 +00:00
Bill Wendling
380c3ca093 Temporarily revert r65994. It was causing rdar://6646455.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66083 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-04 22:02:09 +00:00
Chris Lattner
1541e0f7da Fix PR3720 by properly propagating alignment information from memcpy/memmove
onto element accesses.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66053 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-04 19:20:50 +00:00
Devang Patel
7f6179d1d8 If a global constant is dead then global's debug info should not prevent the optimizer in deleting the global. And while deleting global, delete global's debug info also.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65994 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-04 01:22:23 +00:00
Devang Patel
cf42ee42b1 If branch conditions' one successor is dominating another non-latch successor then this loop's iteration space can not be restricted. In this example block bb5 is always executed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65902 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-02 23:39:14 +00:00
Duncan Sands
5b7cfb02f7 Fix PR3694: add an instcombine micro-optimization that helps
clean up when using variable length arrays in llvm-gcc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65832 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-02 09:18:21 +00:00
Chris Lattner
585cfb6821 adjust for asmprinter change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65741 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-01 00:26:51 +00:00
Chris Lattner
3aa8f6bc11 adjust for asmprinter change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65740 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-01 00:25:46 +00:00
Chris Lattner
13397b97ee adjust test to make it more robust
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65739 91177308-0d34-0410-b5e6-96231b3b80d8
2009-03-01 00:24:40 +00:00
Zhou Sheng
a8d57fe96b Ignore dbg info intrinsics when folding conditional branch to
conditional branch predecessors.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65509 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-26 06:56:37 +00:00
Chris Lattner
58b1ac76d4 Fix PR3667
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65464 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-25 18:20:01 +00:00
Zhou Sheng
9a7c743fc0 Don't block basic block with only SwitchInst to fold into predecessors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65456 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-25 15:34:27 +00:00
Dan Gohman
85937de26b Add a testcase for the problem fixed in r65289.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65365 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-24 02:17:42 +00:00
Dan Gohman
0001e56f15 Fix a ValueTracking rule: RHS means operand 1, not 0. Add a simple
ashr instcombine to help expose this code. And apply the fix to
SelectionDAG's copy of this code too.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65364 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-24 02:00:40 +00:00
Devang Patel
5622f07a21 While folding unconditional return move DbgRegionEndInst into the predecessor, instead of removing it. This fixes following tests from llvmgcc42 testsuite.
gcc.c-torture/execute/20000605-3.c
gcc.c-torture/execute/20020619-1.c
gcc.c-torture/execute/20030920-1.c
gcc.c-torture/execute/loop-ivopts-1.c



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65353 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-24 00:05:16 +00:00
Dan Gohman
15cab2817b Back out the change in 64918 that used sign-extensions when promoting
trip counts that use signed comparisons. It's not obviously the best
approach for preserving trip count information, and at any rate there
isn't anything in the tree right now that makes use of that, so for
now always using zero-extensions is preferable.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65347 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-23 23:20:35 +00:00
Nick Lewycky
2ec0dbf961 Don't sign extend the char when expanding char -> int during
load(bitcast(char[4] to i32*)) evaluation.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65246 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-21 20:50:42 +00:00
Chris Lattner
36d3e326df rename a function to indicate that it checks for profitability as well
as legality.  Make load sinking and gep sinking more careful: we only
do it when it won't pessimize loads from the stack.  This has the added
benefit of not producing code that is unanalyzable to SROA.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65209 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-21 00:46:50 +00:00
Dan Gohman
c17e0cf6c0 Implement "superhero" strength reduction, or full strength
reduction of address calculations down to basic pointer arithmetic.
This is currently off by default, as it needs a few other features
before it becomes generally useful. And even when enabled, full
strength reduction is only performed when it doesn't increase
register pressure, and when several other conditions are true.

This also factors out a bunch of exisiting LSR code out of
StrengthReduceStridedIVUsers into separate functions, and tidies
up IV insertion. This actually decreases register pressure even
in non-superhero mode. The change in iv-users-in-other-loops.ll
is an example of this; there are two more adds because there are
two fewer leas, and there is less spilling.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65108 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-20 04:17:46 +00:00
Dan Gohman
f5a309e989 Use a sign-extend instead of a zero-extend when promoting a
trip count value when the original loop iteration condition is
signed and the canonical induction variable won't undergo signed
overflow. This isn't required for correctness; it just preserves
more information about original loop iteration values.

Add a getTruncateOrSignExtend method to ScalarEvolution,
following getTruncateOrZeroExtend.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64918 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-18 17:22:41 +00:00
Owen Anderson
a052fad720 Add a test for r61358, which I forgot to add way back when.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64904 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-18 07:50:22 +00:00
Dan Gohman
38ad0191e9 Change the argument type in this test to something less convoluted,
since it isn't actually used. 


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64883 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-18 04:25:04 +00:00
Dan Gohman
d2067fd730 Fix a corner case in the new indvars promotion logic: if there
are multiple IV's in a loop, some of them may under go signed
or unsigned wrapping even if the IV that's used in the loop
exit condition doesn't. Restrict sign-extension-elimination
and zero-extension-elimination to only those that operate on
the original loop-controlling IV.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64866 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-18 00:52:00 +00:00
Duncan Sands
cdf5ffb7fb If an alias is dead and so is its aliasee, then globaldce would
crash because the alias would still be using the aliasee when the
aliasee was deleted.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64844 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-17 23:05:26 +00:00
Chris Lattner
a84f47c3e7 commit a tweaked version of Daniel's patch for PR3599. We now
eliminate all the extensions and all but the one required truncate
from the testcase, but the or/and/shift stuff still isn't zapped.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64809 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-17 20:47:23 +00:00
Evan Cheng
5a6c1a840a Strengthen the "non-constant stride must dominate loop preheader" check.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64703 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-17 00:13:06 +00:00
Dan Gohman
ecd0fb51cb Fix EnforceKnownAlignment so that it doesn't ever reduce the alignment
of an alloca or global variable.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64693 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-16 23:02:21 +00:00
Dan Gohman
74786c0f43 Rename IndVarsSimplify to IndVarSimplify, to be consistent with
the name used in the code that these tests are for.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64624 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-16 00:56:15 +00:00
Dan Gohman
926b0a27b9 Change these tests to use regular loads instead of llvm.x86.sse2.loadu.dq.
Enhance instcombine to use the preferred field of
GetOrEnforceKnownAlignment in more cases, so that regular IR operations are
optimized in the same way that the intrinsics currently are.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64623 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-16 00:44:23 +00:00
Duncan Sands
4782b305c2 If the target of an alias has internal linkage, then the
alias can be morphed into the target.  Implement this
transform, and fix a crash in the existing transform at
the same time.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64583 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-15 09:56:08 +00:00
Evan Cheng
8f40afe62d Fix pr3571: If stride is a value defined by an instruction, make sure it dominates the loop preheader. When IV users are strength reduced, the stride is inserted into the preheader. It could create a use before def situation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64579 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-15 06:06:15 +00:00
Dan Gohman
aa03649af2 Extend the IndVarSimplify support for promoting induction variables:
- Test for signed and unsigned wrapping conditions, instead of just
   testing for non-negative induction ranges. 
 - Handle loops with GT comparisons, in addition to LT comparisons.
 - Support more cases of induction variables that don't start at 0.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64532 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-14 02:31:09 +00:00
Nick Lewycky
02b6a6a75f Mark strto* as readonly when the endptr is null.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64460 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-13 17:08:33 +00:00
Nick Lewycky
4c49841f34 On strtod and friends, mark 'endptr' nocapture in the function prototype, and
mark the first argument nocapture if endptr=NULL for each particular call.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64453 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-13 15:31:46 +00:00
Nick Lewycky
8968a07768 Reapply r64300:
Make sure the SCC pass manager initializes any contained
function pass managers.  Without this, simplify-libcalls
would add nocapture attributes when run on its own, but
not when run as part of -std-compile-opts or similar.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64443 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-13 07:15:53 +00:00