Commit Graph

1717 Commits

Author SHA1 Message Date
Eli Friedman
2adc5b6a17 PR9420; an instruction before an unreachable is guaranteed not to have any
reachable uses, but there still might be uses in dead blocks.  Use the
standard solution of replacing all the uses with undef.  This is
a rare case because it's very sensitive to phase ordering in SimplifyCFG.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127299 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 00:48:33 +00:00
Jakob Stoklund Olesen
2bc2a08b1b DenseMap<uintptr_t,...> doesn't allow all values as keys.
Avoid colliding with the sentinels, hopefully unbreaking
llvm-gcc-x86_64-linux-selfhost.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126982 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-04 02:48:56 +00:00
Frits van Bommel
f7b2a9d7df Teach SimplifyCFG that (switch (select cond, X, Y)) is better expressed as a branch.
Based on a patch by Alistair Lynn.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126647 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-28 09:44:07 +00:00
Benjamin Kramer
278be783b5 Revert "SimplifyCFG: GEPs with just one non-constant index are also cheap."
Yes, there are other types than i8* and GEPs on them can produce an add+multiply.
We don't consider that cheap enough to be speculatively executed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126481 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-25 10:33:33 +00:00
Benjamin Kramer
9ae59e3444 SimplifyCFG: GEPs with just one non-constant index are also cheap.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126452 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-24 23:26:09 +00:00
Benjamin Kramer
9b61c550c2 SimplifyCFG: GEPs with constant indices are cheap enough to be executed unconditionally.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126445 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-24 22:46:11 +00:00
Devang Patel
afd0d0e8a7 Do not use DIFactory. Use DIBuilder.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126398 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-24 18:49:55 +00:00
Duncan Sands
2cfbf018a9 If the phi node was used by an unreachable instruction that ends up using
itself without going via a phi node then we could return false here in
spite of making a change.  Also, tweak the comment because this method
can (and always could) return true without deleting the original phi node.
For example, if the phi node was used by a read-only invoke instruction
which is used by another phi node phi2 which is only used by and only uses
the invoke, then phi2 would be deleted but not the invoke instruction and
not the original phi node.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126129 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-21 17:32:05 +00:00
Duncan Sands
b4098ba03d Simplify RecursivelyDeleteDeadPHINode. The only functionality change
should be that if the phi is used by a side-effect free instruction with
no uses then the phi and the instruction now get zapped (checked by the
unittest).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126124 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-21 16:27:36 +00:00
Nick Lewycky
eff5e69c82 Make RecursivelyDeleteDeadPHINode delete a phi node that has no users and add a
test for that. With this change, test/CodeGen/X86/codegen-dce.ll no longer finds
any instructions to DCE, so delete the test.

Also renamed J and JP to I and IP in RecursivelyDeleteDeadPHINode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126088 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-20 18:05:56 +00:00
Nick Lewycky
1a4021a2be Teach RecursivelyDeleteDeadPHINodes to handle multiple self-references. Patch
by Andrew Clinton!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126077 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-20 08:38:20 +00:00
Chris Lattner
f5ebfb0d28 Don't unroll loops whose header block's address is taken.
This is part of a futile attempt to not "break" bizzaro
code like this:

 l1:
  printf("l1: %p\n", &&l1);
  ++x;

  if( x < 3 ) goto l1;


Previously we'd fold &&l1 to 1, which is fine per our semantics
but not helpful to the user.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125827 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-18 04:25:21 +00:00
Cameron Zwarich
4a60b932a2 Rename 'loopsimplify' to 'loop-simplify'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125317 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-10 23:38:10 +00:00
Benjamin Kramer
33828bcb24 SimplifyCFG: Track the number of used icmps when turning a icmp chain into a switch. If we used only one icmp, don't turn it into a switch.
Also prevent the switch-to-icmp transform from creating identity adds, noticed by Marius Wachtler.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125056 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-07 22:37:28 +00:00
Benjamin Kramer
042b27f40e SimplifyCFG: Also transform switches that represent a range comparison but are not sorted into sub+icmp.
This transforms another 1000 switches in gcc.c.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124826 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-03 22:51:41 +00:00
Benjamin Kramer
56442dfdcf SimplifyCFG: Turn switches into sub+icmp+branch if possible.
This makes the job of the later optzn passes easier, allowing the vast amount of
icmp transforms to chew on it.

We transform 840 switches in gcc.c, leading to a 16k byte shrink of the resulting
binary on i386-linux.

The testcase from README.txt now compiles into
  decl  %edi
  cmpl  $3, %edi
  sbbl  %eax, %eax
  andl  $1, %eax
  ret

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124724 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-02 15:56:22 +00:00
Evan Cheng
c3f507f98a Re-apply r124518 with fix. Watch out for invalidated iterator.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124526 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 04:46:23 +00:00
Evan Cheng
b0a42fdb36 Revert r124518. It broke Linux self-host.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124522 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 02:43:04 +00:00
Evan Cheng
5e6940788f Re-commit r124462 with fixes. Tail recursion elim will now dup ret into unconditional predecessor to enable TCE on demand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124518 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 01:29:26 +00:00
Evan Cheng
1b5c0cb71d Revert r124462. There are a few big regressions that I need to fix first.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124478 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-28 07:12:38 +00:00
Evan Cheng
40f64cb0de - Stop simplifycfg from duplicating "ret" instructions into unconditional
branches. PR8575, rdar://5134905, rdar://8911460.
- Allow codegen tail duplication to dup small return blocks after register
  allocation is done.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124462 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-28 02:19:21 +00:00
Chris Lattner
867be59684 fix PR9017, a bug where we'd assert when promoting in unreachable
code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124100 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-24 03:29:07 +00:00
Chris Lattner
51e62f0f73 fix PR9015, a crash linking recursive metadata.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124099 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-24 03:18:24 +00:00
Cameron Zwarich
491d8d4370 Convert two std::vectors to SmallVectors for a 3.4% speedup running -scalarrepl
on test-suite + SPEC2000 & SPEC2006.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124068 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-23 08:03:04 +00:00
Cameron Zwarich
59f5319719 Convert a std::map to a DenseMap for another 1.7% speedup on -scalarrepl.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123732 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 04:50:38 +00:00
Cameron Zwarich
443997de8b Make a std::vector a SmallVector<*, 32> like the other vectors in the same
function. This seems to be about a 1.5% speedup of -scalarrepl on test-suite
with SPEC2000 and SPEC2006.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123731 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 04:41:32 +00:00
Cameron Zwarich
301278719b Remove code for updating dominance frontiers and some outdated references to
dominance and post-dominance frontiers.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123725 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 04:11:31 +00:00
Cameron Zwarich
b1686c32fc Remove outdated references to dominance frontiers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123724 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 03:53:26 +00:00
Cameron Zwarich
419e8a6299 Roll r123609 back in with two changes that fix test failures with expensive
checks enabled:

1) Use '<' to compare integers in a comparison function rather than '<='.

2) Use the uniqued set DefBlocks rather than Info.DefiningBlocks to initialize
the priority queue.

The speedup of scalarrepl on test-suite + SPEC2000 + SPEC2006 is a bit less, at
just under 16% rather than 17%.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123662 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-17 17:38:41 +00:00
Cameron Zwarich
b1086a9c6d Roll out r123609 due to failures on the llvm-x86_64-linux-checks bot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123618 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-17 07:26:51 +00:00
Cameron Zwarich
ebed6de7b1 Eliminate the use of dominance frontiers in PromoteMemToReg. In addition to
eliminating a potentially quadratic data structure, this also gives a 17%
speedup when running -scalarrepl on test-suite + SPEC2000 + SPEC2006. My initial
experiment gave a greater speedup around 25%, but I moved the dominator tree
level computation from dominator tree construction to PromoteMemToReg.

Since this approach to computing IDFs has a much lower overhead than the old
code using precomputed DFs, it is worth looking at using this new code for the
second scalarrepl pass as well.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123609 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-17 01:08:59 +00:00
Chris Lattner
deaf55f698 Generalize LoadAndStorePromoter a bit and switch LICM
to use it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123501 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 00:12:35 +00:00
Chris Lattner
a2d845a3ff Add a new LoadAndStorePromoter class, which implements the general
"promote a bunch of load and stores" logic, allowing the code to
be shared and reused.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123456 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 19:36:13 +00:00
Chris Lattner
70e5e222db indentation
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123426 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 04:23:53 +00:00
Jakob Stoklund Olesen
fbbd4abfe5 Fix a non-deterministic loop in llvm::MergeBlockIntoPredecessor.
DT->changeImmediateDominator() trivially ignores identity updates, so there is
really no need for the uniqueing provided by SmallPtrSet.

I expect this to fix PR8954.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123286 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 22:54:38 +00:00
Chris Lattner
f73b99ab43 this pass claims to preserve scev, make sure to tell it about deletions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123247 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 18:14:50 +00:00
Frits van Bommel
65fdded319 Factor the actual simplification out of SimplifyIndirectBrOnSelect and into a new helper function so it can be reused in e.g. an upcoming SimplifySwitchOnSelect.
No functional change.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123234 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 12:52:11 +00:00
Chris Lattner
b6810991a9 when MergeBlockIntoPredecessor merges two blocks, update MemDep if it
is floating around in the ether.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123223 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 08:16:49 +00:00
Chris Lattner
b5b7997fd0 Fix FoldSingleEntryPHINodes to update memdep and AA when it deletes
phi nodes.  It is called from MergeBlockIntoPredecessor which is 
called from GVN, which claims to preserve these.

I'm skeptical that this is the actual problem behind PR8954, but
this is a stab in the right direction.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123222 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 08:13:40 +00:00
Chris Lattner
b298db7253 random cleanups
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123221 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 08:00:40 +00:00
Chris Lattner
dc85f8ab80 various code cleanups, enhance MergeBlockIntoPredecessor to preserve
loop info.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123074 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 19:08:40 +00:00
Chris Lattner
b0433d4b2f reduce nesting.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123071 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 18:47:43 +00:00
Cameron Zwarich
80f6a507d4 Make more passes preserve dominators (or state that they preserve dominators if
they all ready do). This removes two dominator recomputations prior to isel,
which is a 1% improvement in total llc time for 403.gcc.

The only potentially suspect thing is making GCStrategy recompute dominators if
it used a custom lowering strategy.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123064 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 17:01:52 +00:00
Chris Lattner
b5fa5fcecc Revamp the ValueMapper interfaces in a couple ways:
1. Take a flags argument instead of a bool.  This makes
   it more clear to the reader what it is used for.
2. Add a flag that says that "remapping a value not in the
   map is ok".
3. Reimplement MapValue to share a bunch of code and be a lot
   more efficient.  For lookup failures, don't drop null values
   into the map.
4. Using the new flag a bunch of code can vaporize in LinkModules
   and LoopUnswitch, kill it.

No functionality change.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123058 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 08:15:20 +00:00
Jay Foad
8f9ffbd056 Remove all uses of the "ugly" method BranchInst::setUnconditionalDest().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123025 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07 20:25:56 +00:00
Duncan Sands
dc615e41b5 Speed up instsimplify by about 10-15% by not bothering to retry
InstructionSimplify on instructions that didn't change since the
last time round the loop.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122745 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 10:50:04 +00:00
Chris Lattner
9fc5cdf77c split dom frontier handling stuff out to its own DominanceFrontier header,
so that Dominators.h is *just* domtree.  Also prune #includes a bit.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122714 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 22:09:33 +00:00
Duncan Sands
67fb341f8b Fix PR8702 by not having LoopSimplify claim to preserve LCSSA form. As described
in the PR, the pass could break LCSSA form when inserting preheaders.  It probably
would be easy enough to fix this, but since currently we always go into LCSSA form
after running this pass, doing so is not urgent.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122695 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 13:38:21 +00:00
Duncan Sands
8a3d29cf10 Simplify this pass by using a depth-first iterator to ensure that all
operands are visited before the instructions themselves.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122647 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-31 17:49:05 +00:00
Duncan Sands
fea0b8b666 Zap dead instructions harder.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122645 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-31 16:17:54 +00:00