Commit Graph

97 Commits

Author SHA1 Message Date
Eli Friedman c458885c58 loop-rotate shouldn't hoist alloca instructions out of a loop. Patch by Patrik Hägglund, with slightly modified test. Issue reported by Patrik Hägglund on llvmdev.
llvm-svn: 150642
2012-02-16 00:41:10 +00:00
Andrew Trick 10cc45336d Add simplifyLoopLatch to LoopRotate pass.
This folds a simple loop tail into a loop latch. It covers the common (in fortran) case of postincrement loops. It's a "free" way to expose this type of loop to downstream loop optimizations that bail out on non-canonical loops (getLoopLatch is a heavily used check).

llvm-svn: 150439
2012-02-14 00:00:23 +00:00
Andrew Trick a20f198747 whitespace
llvm-svn: 150438
2012-02-14 00:00:19 +00:00
Jay Foad 372ad64b4d Make better use of the PHINode API.
Change various bits of code to make better use of the existing PHINode
API, to insulate them from forthcoming changes in how PHINodes store
their operands.

llvm-svn: 133434
2011-06-20 14:18:48 +00:00
Devang Patel c1f7c1d469 Preserve line number information.
llvm-svn: 130536
2011-04-29 20:38:55 +00:00
Chris Lattner 88974f4625 fix PR9523, a crash in looprotate on a non-canonical loop made out of indirectbr.
llvm-svn: 129203
2011-04-09 07:25:58 +00:00
Devang Patel 3058398655 Do not hoist @llvm.dbg.value. Here, @llvm.dbg.value is "referring" a value that is modified inside loop.
llvm-svn: 125529
2011-02-14 23:03:23 +00:00
Chris Lattner 63fe78de68 remove a bogus assertion: the latch block of a loop is not
neccesarily an uncond branch to the header.  This fixes 
PR8955 (the assertion tripping).

llvm-svn: 123219
2011-01-11 07:47:59 +00:00
Chris Lattner 59c82f850d When loop rotation happens, it is *very* common for the duplicated condbr
to be foldable into an uncond branch.  When this happens, we can make a
much simpler CFG for the loop, which is important for nested loop cases
where we want the outer loop to be aggressively optimized.

Handle this case more aggressively.  For example, previously on
phi-duplicate.ll we would get this:


define void @test(i32 %N, double* %G) nounwind ssp {
entry:
  %cmp1 = icmp slt i64 1, 1000
  br i1 %cmp1, label %bb.nph, label %for.end

bb.nph:                                           ; preds = %entry
  br label %for.body

for.body:                                         ; preds = %bb.nph, %for.cond
  %j.02 = phi i64 [ 1, %bb.nph ], [ %inc, %for.cond ]
  %arrayidx = getelementptr inbounds double* %G, i64 %j.02
  %tmp3 = load double* %arrayidx
  %sub = sub i64 %j.02, 1
  %arrayidx6 = getelementptr inbounds double* %G, i64 %sub
  %tmp7 = load double* %arrayidx6
  %add = fadd double %tmp3, %tmp7
  %arrayidx10 = getelementptr inbounds double* %G, i64 %j.02
  store double %add, double* %arrayidx10
  %inc = add nsw i64 %j.02, 1
  br label %for.cond

for.cond:                                         ; preds = %for.body
  %cmp = icmp slt i64 %inc, 1000
  br i1 %cmp, label %for.body, label %for.cond.for.end_crit_edge

for.cond.for.end_crit_edge:                       ; preds = %for.cond
  br label %for.end

for.end:                                          ; preds = %for.cond.for.end_crit_edge, %entry
  ret void
}

Now we get the much nicer:

define void @test(i32 %N, double* %G) nounwind ssp {
entry:
  br label %for.body

for.body:                                         ; preds = %entry, %for.body
  %j.01 = phi i64 [ 1, %entry ], [ %inc, %for.body ]
  %arrayidx = getelementptr inbounds double* %G, i64 %j.01
  %tmp3 = load double* %arrayidx
  %sub = sub i64 %j.01, 1
  %arrayidx6 = getelementptr inbounds double* %G, i64 %sub
  %tmp7 = load double* %arrayidx6
  %add = fadd double %tmp3, %tmp7
  %arrayidx10 = getelementptr inbounds double* %G, i64 %j.01
  store double %add, double* %arrayidx10
  %inc = add nsw i64 %j.01, 1
  %cmp = icmp slt i64 %inc, 1000
  br i1 %cmp, label %for.body, label %for.end

for.end:                                          ; preds = %for.body
  ret void
}

With all of these recent changes, we are now able to compile:

void foo(char *X) {
 for (int i = 0; i != 100; ++i) 
   for (int j = 0; j != 100; ++j)
     X[j+i*100] = 0;
}

into a single memset of 10000 bytes.  This series of changes
should also be helpful for other nested loop scenarios as well.

llvm-svn: 123079
2011-01-08 19:59:06 +00:00
Chris Lattner 30f318e5d1 split ssa updating code out to its own helper function. Don't bother
moving the OrigHeader block anymore: we just merge it away anyway so
its code layout doesn't matter.

llvm-svn: 123077
2011-01-08 19:26:33 +00:00
Chris Lattner 2615130e1d Implement a TODO: Enhance loopinfo to merge away the unconditional branch
that it was leaving in loops after rotation (between the original latch
block and the original header.

With this change, it is possible for rotated loops to have just a single
basic block, which is useful.

llvm-svn: 123075
2011-01-08 19:10:28 +00:00
Chris Lattner fee37c5fa3 inline preserveCanonicalLoopForm now that it is simple.
llvm-svn: 123073
2011-01-08 18:55:50 +00:00
Chris Lattner 063dca0f6a Three major changes:
1. Rip out LoopRotate's domfrontier updating code.  It isn't
   needed now that LICM doesn't use DF and it is super complex
   and gross.
2. Make DomTree updating code a lot simpler and faster.  The 
   old loop over all the blocks was just to find a block??
3. Change the code that inserts the new preheader to just use
   SplitCriticalEdge instead of doing an overcomplex 
   reimplementation of it.

No behavior change, except for the name of the inserted preheader.

llvm-svn: 123072
2011-01-08 18:52:51 +00:00
Chris Lattner 7fab23bc1d LoopRotate requires canonical loop form, so it always has preheaders
and latch blocks.  Reorder entry conditions to make hte pass faster
and more logical.

llvm-svn: 123069
2011-01-08 18:06:22 +00:00
Chris Lattner d62691f4e8 use the LI ivar.
llvm-svn: 123068
2011-01-08 17:49:51 +00:00
Chris Lattner 385f2ec6d8 some cleanups: remove dead arguments and eliminate ivars
that are just passed to one function.

llvm-svn: 123067
2011-01-08 17:48:33 +00:00
Chris Lattner 25ba40a0cc fix an issue duncan pointed out, which could cause loop rotate
to violate LCSSA form

llvm-svn: 123066
2011-01-08 17:38:45 +00:00
Chris Lattner 8c5defd0b0 Have loop-rotate simplify instructions (yay instsimplify!) as it clones
them into the loop preheader, eliminating silly instructions like
"icmp i32 0, 100" in fixed tripcount loops.  This also better exposes the 
bigger problem with loop rotate that I'd like to fix: once this has been
folded, the duplicated conditional branch *often* turns into an uncond branch.

Not aggressively handling this is pessimizing later loop optimizations 
somethin' fierce by making "dominates all exit blocks" checks fail.

llvm-svn: 123060
2011-01-08 08:24:46 +00:00
Chris Lattner 43f8d16482 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.

llvm-svn: 123058
2011-01-08 08:15:20 +00:00
Chris Lattner 2b3f20e6ec two minor changes: switch to the standard ValueToValueMapTy
map from ValueMapper.h (giving us access to its utilities)
and add a fastpath in the loop rotation code, avoiding expensive
ssa updator manipulation for values with nothing to update.

llvm-svn: 123057
2011-01-08 07:21:31 +00:00
Chris Lattner bf0aa927cc split dom frontier handling stuff out to its own DominanceFrontier header,
so that Dominators.h is *just* domtree.  Also prune #includes a bit.

llvm-svn: 122714
2011-01-02 22:09:33 +00:00
Chris Lattner 679572e584 improve loop rotation to use CodeMetrics to analyze the
size of a loop header instead of its own code size estimator.
This allows it to handle bitcasts etc more precisely.

llvm-svn: 122681
2011-01-02 07:35:53 +00:00
Owen Anderson a4fefc1949 Passes do not need to recursively initialize passes that they preserve, if
they do not also require them.  This allows us to reduce inter-pass linkage
dependencies.

llvm-svn: 116854
2010-10-19 20:08:44 +00:00
Owen Anderson 6c18d1aac0 Get rid of static constructors for pass registration. Instead, every pass exposes an initializeMyPassFunction(), which
must be called in the pass's constructor.  This function uses static dependency declarations to recursively initialize
the pass's dependencies.

Clients that only create passes through the createFooPass() APIs will require no changes.  Clients that want to use the
CommandLine options for passes will need to manually call the appropriate initialization functions in PassInitialization.h
before parsing commandline arguments.

I have tested this with all standard configurations of clang and llvm-gcc on Darwin.  It is possible that there are problems
with the static dependencies that will only be visible with non-standard options.  If you encounter any crash in pass
registration/creation, please send the testcase to me directly.

llvm-svn: 116820
2010-10-19 17:21:58 +00:00
Owen Anderson 8ac477ffb5 Begin adding static dependence information to passes, which will allow us to
perform initialization without static constructors AND without explicit initialization
by the client.  For the moment, passes are required to initialize both their
(potential) dependencies and any passes they preserve.  I hope to be able to relax
the latter requirement in the future.

llvm-svn: 116334
2010-10-12 19:48:12 +00:00
Owen Anderson df7a4f2515 Now with fewer extraneous semicolons!
llvm-svn: 115996
2010-10-07 22:25:06 +00:00
Chris Lattner b01c24a945 Teach loop rotate to hoist trivially invariant instructions
in the duplicated block instead of duplicating them.  

Duplicating them into the end of the loop and the preheader 
means that we got a phi node in the header of the loop, 
which prevented LICM from hoisting them.  GVN would
usually come around later and merge the duplicated 
instructions so we'd get reasonable output... except that
anything dependent on the shoulda-been-hoisted value can't
be hoisted.  In PR5319 (which this fixes), a memory value
didn't get promoted.

llvm-svn: 113134
2010-09-06 01:10:22 +00:00
Duncan Sands 6778149f7e Reapply commit 112699, speculatively reverted by echristo, since
I'm sure it is harmless.  Original commit message:
If PrototypeValue is erased in the middle of using the SSAUpdator
then the SSAUpdator may access freed memory.  Instead, simply pass
in the type and name explicitly, which is all that was used anyway.

llvm-svn: 112810
2010-09-02 08:14:03 +00:00
Eric Christopher a5d315c665 Speculatively revert 112699 and 112702, they seem to be causing
self host errors on clang-x86-64.

llvm-svn: 112719
2010-09-01 17:29:10 +00:00
Duncan Sands f7b18437b5 If PrototypeValue is erased in the middle of using the SSAUpdator
then the SSAUpdator may access freed memory.  Instead, simply pass
in the type and name explicitly, which is all that was used anyway.

llvm-svn: 112699
2010-09-01 10:29:33 +00:00
Dan Gohman 5047ca0c02 When rotating loops, put the original header at the bottom of the
loop, making the resulting loop significantly less ugly.  Also, zap
its trivial PHI nodes, since it's easy.

llvm-svn: 111255
2010-08-17 17:39:21 +00:00
Owen Anderson a7aed18624 Reapply r110396, with fixes to appease the Linux buildbot gods.
llvm-svn: 110460
2010-08-06 18:33:48 +00:00
Owen Anderson bda59bd247 Revert r110396 to fix buildbots.
llvm-svn: 110410
2010-08-06 00:23:35 +00:00
Owen Anderson 755aceb5d0 Don't use PassInfo* as a type identifier for passes. Instead, use the address of the static
ID member as the sole unique type identifier.  Clean up APIs related to this change.

llvm-svn: 110396
2010-08-05 23:42:04 +00:00
Owen Anderson a57b97e7e7 Fix batch of converting RegisterPass<> to INTIALIZE_PASS().
llvm-svn: 109045
2010-07-21 22:09:45 +00:00
Dan Gohman efd7f9c360 Reorder the contents of various getAnalysisUsage functions, eliminating
a redundant loopsimplify run from the default -O2 sequence.

llvm-svn: 108539
2010-07-16 17:58:45 +00:00
Dan Gohman d2d1ae105d Use pre-increment instead of post-increment when the result is not used.
llvm-svn: 106542
2010-06-22 15:08:57 +00:00
Dan Gohman a83ac2d9e7 Update various Loop optimization passes to cope with the possibility that
LoopSimplify form may not be available.

llvm-svn: 86175
2009-11-05 21:11:53 +00:00
Dan Gohman d9fa1c9c1e Call getAnalysis<LoopInfo> the normal way, instead of asking passed-in
LoopPassManager for it.

llvm-svn: 86163
2009-11-05 19:43:25 +00:00
Dan Gohman 880c92ac1c Rename forgetLoopBackedgeTakenCount to forgetLoop, because it
clears out more information than just the stored backedge taken count.

llvm-svn: 85664
2009-10-31 15:04:55 +00:00
Dan Gohman 6a10d5ebd3 Fix a typo in a comment.
llvm-svn: 85120
2009-10-26 15:55:24 +00:00
Dan Gohman 8f4078ba39 Rename isLoopExit to isLoopExiting, for consistency with the wording
used elsewhere - an exit block is a block outside the loop branched to
from within the loop. An exiting block is a block inside the loop that
branches out.

llvm-svn: 85019
2009-10-24 23:34:26 +00:00
Dan Gohman b979794e4b Rewrite LoopRotation's SSA updating code using SSAUpdater.
llvm-svn: 85016
2009-10-24 23:19:52 +00:00
Dan Gohman fc20b67e80 Tell ScalarEvolution to forget everything it knows about a loop before
rotating the loop, since loop rotation is a very significant change.

llvm-svn: 82901
2009-09-27 15:37:03 +00:00
Nick Lewycky 42fb7452df Instruction::clone does not need to take an LLVMContext&. Remove that and
update all the callers.

llvm-svn: 82889
2009-09-27 07:38:41 +00:00
Dan Gohman ec4557f324 Fix SplitCriticalEdge to properly update LCSSA form when splitting a
loop exit edge -- new PHIs may be needed not only for the additional
splits that are made to preserve LoopSimplify form, but also for the
original split. Factor out the code that inserts new PHIs so that it
can be used for both. Remove LoopRotation.cpp's code for manually
updating LCSSA form, as it is now redundant. This fixes PR4934.

llvm-svn: 81363
2009-09-09 18:18:18 +00:00
Chris Lattner 2dd09dbdf7 eliminate VISIBILITY_HIDDEN from Transforms/Scalar. PR4861
llvm-svn: 80766
2009-09-02 06:11:42 +00:00
Owen Anderson 55f1c09e31 Push LLVMContexts through the IntegerType APIs.
llvm-svn: 78948
2009-08-13 21:58:54 +00:00
Devang Patel b1106fbdbc Fix dom frontier update. This fixes PR4667.
Patch by Jakub Staszak.

llvm-svn: 78388
2009-08-07 17:16:44 +00:00
Owen Anderson 47db941fd3 Get rid of the Pass+Context magic.
llvm-svn: 76702
2009-07-22 00:24:57 +00:00