Commit Graph

3389 Commits

Author SHA1 Message Date
Duncan Sands
7af00c0f6e Commit 124487 broke 254.gap. See if disabling the part that might be triggered
by PR9088 fixes things.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124561 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-30 18:24:20 +00:00
Duncan Sands
1895e98ef3 Transform (X/Y)*Y into X if the division is exact. Instcombine already knows how
to do this and more, but would only do it if X/Y had only one use.  Spotted as the
most common missed simplification in SPEC by my auto-simplifier, now that it knows
about nuw/nsw/exact flags.  This removes a bunch of multiplications from 447.dealII
and 483.xalancbmk.  It also removes a lot from tramp3d-v4, which results in much
more inlining.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124560 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-30 18:03:50 +00:00
Nick Lewycky
175e7aec13 Add the select optimization recently added to instcombine to constant folding.
This is the one where one of the branches of the select is another select on
the same condition.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124547 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 20:35:06 +00:00
Frits van Bommel
1fca2c32cc Move InstCombine's knowledge of fdiv to SimplifyInstruction().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124534 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 15:26:31 +00:00
Duncan Sands
9136782d27 Fix typo: should have been testing that X was odd, not V.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124533 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 13:27:00 +00:00
Evan Cheng
60f5ad46c2 Add a test for TCE return duplication.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124527 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-29 04:53:35 +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
Duncan Sands
593faa53fa My auto-simplifier noticed that ((X/Y)*Y)/Y occurs several times in SPEC
benchmarks, and that it can be simplified to X/Y.  (In general you can only
simplify (Z*Y)/Y to Z if the multiplication did not overflow; if Z has the
form "X/Y" then this is the case).  This patch implements that transform and
moves some Div logic out of instcombine and into InstructionSimplify.
Unfortunately instcombine gets in the way somewhat, since it likes to change
(X/Y)*Y into X-(X rem Y), so I had to teach instcombine about this too.
Finally, thanks to the NSW/NUW flags, sometimes we know directly that "Z*Y"
does not overflow, because the flag says so, so I added that logic too.  This
eliminates a bunch of divisions and subtractions in 447.dealII, and has good
effects on some other benchmarks too.  It seems to have quite an effect on
tramp3d-v4 but it's hard to say if it's good or bad because inlining decisions
changed, resulting in massive changes all over.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124487 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-28 16:51:11 +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
Nick Lewycky
26859587fd Clean up the tests a little, make sure we match an instruction in the right
test.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124473 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-28 05:13:17 +00:00
Nick Lewycky
df3bfae151 Fold select + select where both selects are on the same condition.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124469 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-28 03:28:10 +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
Nick Lewycky
25296e25fd Fix surprising missed optimization in mergefunc where we forgot to consider
that relationships like "i8* null" is equivalent to "i32* null".


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124368 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-27 08:38:19 +00:00
Duncan Sands
37f87c7aa9 Fix PR9039, a use-after-free in reassociate. The issue was that the
operand being factorized (and erased) could occur several times in Ops,
resulting in freed memory being used when the next occurrence in Ops was
analyzed.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124287 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-26 10:08:38 +00:00
Duncan Sands
227fba11ca In which I discover that zero+zero is zero, d'oh!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124188 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-25 15:14:15 +00:00
Duncan Sands
dadcf8eedd Turn off this test - the corresponding instsimplify logic has been
disabled.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124185 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-25 12:31:43 +00:00
Duncan Sands
d70d1a5c44 According to my auto-simplifier the most common missed simplifications in
optimized code are:
  (non-negative number)+(power-of-two) != 0 -> true
and
  (x | 1) != 0 -> true
Instcombine knows about the second one of course, but only does it if X|1
has only one use.  These fire thousands of times in the testsuite.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124183 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-25 09:38:29 +00:00
Nick Lewycky
b38824f866 Teach mergefunc how to emit aliases safely again -- but keep it turned it off
for now. It's controlled by the HasGlobalAliases variable which is not attached
to any flag yet.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124182 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-25 08:56:50 +00:00
Chris Lattner
850bdd2fcc merge all the "crash tests" into crash.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124101 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-24 03:37:34 +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
e3357863aa enhance SRoA to promote allocas that are used by PHI nodes. This often
occurs because instcombine sinks loads and inserts phis.  This kicks in 
on such apps as 175.vpr, eon, 403.gcc, xalancbmk and a bunch of times in
spec2006 in some app that uses std::deque.

This resolves the last of rdar://7339113.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124090 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-24 01:07:11 +00:00
Chris Lattner
c87c50a39c Enhance SRoA to promote allocas that are used by selects in some
common cases.  This triggers a surprising number of times in SPEC2K6
because min/max idioms end up doing this.  For example, code from the
STL ends up looking like this to SRoA:

  %202 = load i64* %__old_size, align 8, !tbaa !3
  %203 = load i64* %__old_size, align 8, !tbaa !3
  %204 = load i64* %__n, align 8, !tbaa !3
  %205 = icmp ult i64 %203, %204
  %storemerge.i = select i1 %205, i64* %__n, i64* %__old_size
  %206 = load i64* %storemerge.i, align 8, !tbaa !3

We can now promote both the __n and the __old_size allocas.

This addresses another chunk of rdar://7339113, poor codegen on
stringswitch.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124088 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-23 22:04:55 +00:00
Chris Lattner
145c532e68 Enhance SRoA to be more aggressive about scalarization of aggregate allocas
that have PHI or select uses of their element pointers.  This can often happen
when instcombine sinks two loads into a successor, inserting a phi or select.

With this patch, we can scalarize the alloca, but the pinned elements are not
yet promoted.  This is still a win for large aggregates where only one element
is used.  This fixes rdar://8904039 and part of rdar://7339113 (poor codegen
on stringswitch).



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124070 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-23 08:27:54 +00:00
Chris Lattner
85a7c69085 remove an old hack that avoided creating MMX datatypes. The
X86 backend has been fixed.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124064 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-23 06:40:33 +00:00
Dan Gohman
8fb25c53bd Actually check memcpy lengths, instead of just commenting about
how they should be checked.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123999 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21 22:07:57 +00:00
Owen Anderson
5d2e188962 Just because we have determined that an (fcmp | fcmp) is true for A < B,
A == B, and A > B, does not mean we can fold it to true.  We still need to
check for A ? B (A unordered B).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123993 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21 19:39:42 +00:00
Chris Lattner
cd151d2f95 fix PR9013, an infinite loop in instcombine.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123968 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21 05:29:50 +00:00
Nick Lewycky
acf4a7c0e6 Don't try to pull vector bitcasts that change the number of elements through
a select. A vector select is pairwise on each element so we'd need a new
condition with the right number of elements to select on. Fixes PR8994.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123963 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21 02:30:43 +00:00
Nick Lewycky
4c93a0f092 Add a constant folding of casts from zero to zero. Fixes PR9011!
While here, I'd like to complain about how vector is not an aggregate type
according to llvm::Type::isAggregateType(), but they're listed under aggregate
types in the LangRef and zero vectors are stored as ConstantAggregateZero.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123956 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21 01:12:09 +00:00
Duncan Sands
9d32f60a6f At -O123 the early-cse pass is run before instcombine has run. According to my
auto-simplier the transform most missed by early-cse is (zext X) != 0 -> X != 0.
This patch adds this transform and some related logic to InstructionSimplify
and removes some of the logic from instcombine (unfortunately not all because
there are several situations in which instcombine can improve things by making
new instructions, whereas instsimplify is not allowed to do this).  At -O2 this
often results in more than 15% more simplifications by early-cse, and results in
hundreds of lines of bitcode being eliminated from the testsuite.  I did see some
small negative effects in the testsuite, for example a few additional instructions
in three programs.  One program, 483.xalancbmk, got an additional 35 instructions,
which seems to be due to a function getting an additional instruction and then
being inlined all over the place.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123911 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-20 13:21:55 +00:00
Rafael Espindola
c4440e3e30 Add unnamed_addr when we can show that address of a global is not used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123834 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-19 16:32:21 +00:00
Duncan Sands
b2f3c383ec For completeness, generalize the (X + Y) - Y -> X transform and add X - (X + 1) -> -1.
These were not recommended by my auto-simplifier since they don't fire often enough.
However they do fire from time to time, for example they remove one subtraction from
the final bitcode for 483.xalancbmk.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123755 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 11:50:19 +00:00
Duncan Sands
fe02c69f84 Simplify (X<<1)-X into X. According to my auto-simplier this is the most common missed
simplification in fully optimized code.  It occurs sporadically in the testsuite, and
many times in 403.gcc: the final bitcode has 131 fewer subtractions after this change.
The reason that the multiplies are not eliminated is the same reason that instcombine
did not catch this: they are used by other instructions (instcombine catches this with
a more general transform which in general is only profitable if the operands have only
one use).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123754 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-18 09:24:58 +00:00
Nick Lewycky
33d84007b5 Test for lazy value info's ability to prove the absense of NULLs in pointers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123601 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 21:57:20 +00:00
Anders Carlsson
0599c6bb3c Teach DAE to look for functions whose arguments are unused, and change all callers to pass in an undefvalue instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123596 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 21:25:33 +00:00
Rafael Espindola
d6e5cbc842 Don't merge two constants if we care about the address of both.
This fixes the original testcase in PR8927. It also causes a clang
binary built with a patched clang to increase in size by 0.21%.

We can probably get some of the size back by writing a pass that
detects that a global never has its pointer compared and adds
unnamed_addr to it (maybe extend global opt). It is also possible that
there are some other cases clang could add unnamed_addr to.

I will investigate extending globalopt next.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123584 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 17:05:09 +00:00
Owen Anderson
29c8419f91 Reduce and merge testcases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123579 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 09:13:31 +00:00
Chris Lattner
28252b6f0a fix PR8932, a case where arg promotion could infinitely promote.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123574 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 08:09:24 +00:00
Chris Lattner
7e9b427c87 if an alloca is only ever accessed as a unit, and is accessed with load/store instructions,
then don't try to decimate it into its individual pieces.  This will just make a mess of the
IR and is pointless if none of the elements are individually accessed.  This was generating
really terrible code for std::bitset (PR8980) because it happens to be lowered by clang
as an {[8 x i8]} structure instead of {i64}.

The testcase now is optimized to:

define i64 @test2(i64 %X) {
  br label %L2

L2:                                               ; preds = %0
  ret i64 %X
}

before we generated:

define i64 @test2(i64 %X) {
  %sroa.store.elt = lshr i64 %X, 56
  %1 = trunc i64 %sroa.store.elt to i8
  %sroa.store.elt8 = lshr i64 %X, 48
  %2 = trunc i64 %sroa.store.elt8 to i8
  %sroa.store.elt9 = lshr i64 %X, 40
  %3 = trunc i64 %sroa.store.elt9 to i8
  %sroa.store.elt10 = lshr i64 %X, 32
  %4 = trunc i64 %sroa.store.elt10 to i8
  %sroa.store.elt11 = lshr i64 %X, 24
  %5 = trunc i64 %sroa.store.elt11 to i8
  %sroa.store.elt12 = lshr i64 %X, 16
  %6 = trunc i64 %sroa.store.elt12 to i8
  %sroa.store.elt13 = lshr i64 %X, 8
  %7 = trunc i64 %sroa.store.elt13 to i8
  %8 = trunc i64 %X to i8
  br label %L2

L2:                                               ; preds = %0
  %9 = zext i8 %1 to i64
  %10 = shl i64 %9, 56
  %11 = zext i8 %2 to i64
  %12 = shl i64 %11, 48
  %13 = or i64 %12, %10
  %14 = zext i8 %3 to i64
  %15 = shl i64 %14, 40
  %16 = or i64 %15, %13
  %17 = zext i8 %4 to i64
  %18 = shl i64 %17, 32
  %19 = or i64 %18, %16
  %20 = zext i8 %5 to i64
  %21 = shl i64 %20, 24
  %22 = or i64 %21, %19
  %23 = zext i8 %6 to i64
  %24 = shl i64 %23, 16
  %25 = or i64 %24, %22
  %26 = zext i8 %7 to i64
  %27 = shl i64 %26, 8
  %28 = or i64 %27, %25
  %29 = zext i8 %8 to i64
  %30 = or i64 %29, %28
  ret i64 %30
}

In this case, instcombine was able to eliminate the nonsense, but in PR8980 enough
PHIs are in play that instcombine backs off.  It's better to not generate this stuff
in the first place.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123571 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 06:18:28 +00:00
Chris Lattner
192228edb1 enhance FoldOpIntoPhi in instcombine to try harder when a phi has
multiple uses.  In some cases, all the uses are the same operation,
so instcombine can go ahead and promote the phi.  In the testcase
this pushes an add out of the loop.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123568 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 05:28:59 +00:00
Owen Anderson
66f708f7e5 Improve the safety of my globalopt enhancement by ensuring that the bitcast
of the stored value to the new store type is always.  Also, add a testcase.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123563 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 04:33:33 +00:00
Chris Lattner
156eb0a569 fix PR8983, a broken assertion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123562 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 03:43:53 +00:00
Chris Lattner
0092b1142f remove the partial specialization pass. It is unmaintained and has bugs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123554 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-16 00:27:10 +00:00
Nick Lewycky
2820c25e84 Make constmerge a two-pass algorithm so that it won't miss merging
opporuntities. Fixes PR8978.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123541 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 18:14:21 +00:00
Chris Lattner
6ccb5ef1b5 temporarily revert r123526. While working on a follow-on patch I
realize that ConstantFoldTerminator doesn't preserve dominfo.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123527 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 07:51:19 +00:00
Chris Lattner
eeba3f5695 fix rdar://8785296 - -fcatch-undefined-behavior generates inefficient code
The basic issue is that isel (very reasonably!) expects conditional branches
to be folded, so CGP leaving around a bunch dead computation feeding
conditional branches isn't such a good idea.  Just fold branches on constants
into unconditional branches.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123526 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 07:36:13 +00:00
Chris Lattner
94e8e0cfbe Now that instruction optzns can update the iterator as they go, we can
have objectsize folding recursively simplify away their result when it
folds.  It is important to catch this here, because otherwise we won't
eliminate the cross-block values at isel and other times.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123524 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 07:25:29 +00:00
Chris Lattner
62fe406dc2 implement an instcombine xform that canonicalizes casts outside of and-with-constant operations.
This fixes rdar://8808586 which observed that we used to compile:


union xy {
        struct x { _Bool b[15]; } x;
        __attribute__((packed))
        struct y {
                __attribute__((packed)) unsigned long b0to7;
                __attribute__((packed)) unsigned int b8to11;
                __attribute__((packed)) unsigned short b12to13;
                __attribute__((packed)) unsigned char b14;
        } y;
};

struct x
foo(union xy *xy)
{
        return xy->x;
}

into:

_foo:                                   ## @foo
	movq	(%rdi), %rax
	movabsq	$1095216660480, %rcx    ## imm = 0xFF00000000
	andq	%rax, %rcx
	movabsq	$-72057594037927936, %rdx ## imm = 0xFF00000000000000
	andq	%rax, %rdx
	movzbl	%al, %esi
	orq	%rdx, %rsi
	movq	%rax, %rdx
	andq	$65280, %rdx            ## imm = 0xFF00
	orq	%rsi, %rdx
	movq	%rax, %rsi
	andq	$16711680, %rsi         ## imm = 0xFF0000
	orq	%rdx, %rsi
	movl	%eax, %edx
	andl	$-16777216, %edx        ## imm = 0xFFFFFFFFFF000000
	orq	%rsi, %rdx
	orq	%rcx, %rdx
	movabsq	$280375465082880, %rcx  ## imm = 0xFF0000000000
	movq	%rax, %rsi
	andq	%rcx, %rsi
	orq	%rdx, %rsi
	movabsq	$71776119061217280, %r8 ## imm = 0xFF000000000000
	andq	%r8, %rax
	orq	%rsi, %rax
	movzwl	12(%rdi), %edx
	movzbl	14(%rdi), %esi
	shlq	$16, %rsi
	orl	%edx, %esi
	movq	%rsi, %r9
	shlq	$32, %r9
	movl	8(%rdi), %edx
	orq	%r9, %rdx
	andq	%rdx, %rcx
	movzbl	%sil, %esi
	shlq	$32, %rsi
	orq	%rcx, %rsi
	movl	%edx, %ecx
	andl	$-16777216, %ecx        ## imm = 0xFFFFFFFFFF000000
	orq	%rsi, %rcx
	movq	%rdx, %rsi
	andq	$16711680, %rsi         ## imm = 0xFF0000
	orq	%rcx, %rsi
	movq	%rdx, %rcx
	andq	$65280, %rcx            ## imm = 0xFF00
	orq	%rsi, %rcx
	movzbl	%dl, %esi
	orq	%rcx, %rsi
	andq	%r8, %rdx
	orq	%rsi, %rdx
	ret

We now compile this into:

_foo:                                   ## @foo
## BB#0:                                ## %entry
	movzwl	12(%rdi), %eax
	movzbl	14(%rdi), %ecx
	shlq	$16, %rcx
	orl	%eax, %ecx
	shlq	$32, %rcx
	movl	8(%rdi), %edx
	orq	%rcx, %rdx
	movq	(%rdi), %rax
	ret

A small improvement :-)



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123520 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-15 06:32:33 +00:00
Duncan Sands
c087e20331 Turn X-(X-Y) into Y. According to my auto-simplifier this is the most common
simplification present in fully optimized code (I think instcombine fails to
transform some of these when "X-Y" has more than one use).  Fires here and
there all over the test-suite, for example it eliminates 8 subtractions in
the final IR for 445.gobmk, 2 subs in 447.dealII, 2 in paq8p etc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123442 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 15:26:10 +00:00
Duncan Sands
cf80bc1d4a Factorize common code out of the InstructionSimplify shift logic. Add in
threading of shifts over selects and phis while there.  This fires here and
there in the testsuite, to not much effect.  For example when compiling spirit
it fires 5 times, during early-cse, resulting in 6 more cse simplifications,
and 3 more terminators being folded by jump threading, but the final bitcode
doesn't change in any interesting way: other optimizations would have caught
the opportunity anyway, only later.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123441 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 14:44:12 +00:00
Duncan Sands
62becca681 Rename this test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123440 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 14:16:33 +00:00
Chris Lattner
bdb6b7f9c7 relax testcase a bit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123433 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 07:46:33 +00:00
Duncan Sands
c43cee3fbb Move some shift transforms out of instcombine and into InstructionSimplify.
While there, I noticed that the transform "undef >>a X -> undef" was wrong.
For example if X is 2 then the top two bits must be equal, so the result can
not be anything.  I fixed this in the constant folder as well.  Also, I made
the transform for "X << undef" stronger: it now folds to undef always, even
though X might be zero.  This is in accordance with the LangRef, but I must
admit that it is fairly aggressive.  Also, I added "i32 X << 32 -> undef"
following the LangRef and the constant folder, likewise fairly aggressive.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123417 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14 00:37:45 +00:00
Bob Wilson
704d1347c5 Extend SROA to handle arrays accessed as homogeneous structs and vice versa.
This is a minor extension of SROA to handle a special case that is
important for some ARM NEON operations.  Some of the NEON intrinsics
return multiple values, which are handled as struct types containing
multiple elements of the same vector type.  The corresponding return
types declared in the arm_neon.h header have equivalent arrays.  We
need SROA to recognize that it can split up those arrays and structs
into separate vectors, even though they are not always accessed with
the same type.  SROA already handles loads and stores of an entire
alloca by using insertvalue/extractvalue to access the individual
pieces, and that code works the same regardless of whether the type
is a struct or an array.  So, all that needs to be done is to check
for compatible arrays and homogeneous structs.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123381 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-13 17:45:11 +00:00
Bob Wilson
694a10e7d8 Make SROA more aggressive with allocas containing padding.
SROA only split up structs and arrays one level at a time, so padding can
only cause trouble if it is located in between the struct or array elements.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123380 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-13 17:45:08 +00:00
Duncan Sands
6dc91253ab The most common simplification missed by instsimplify in unoptimized bitcode
is "X != 0 -> X" when X is a boolean.  This occurs a lot because of the way
llvm-gcc converts gcc's conditional expressions.  Add this, and a few other
similar transforms for completeness.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123372 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-13 08:56:29 +00:00
Chris Lattner
d318fc2ceb revert 123144, reenabling the rest of memset formation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123302 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-12 03:25:15 +00:00
Chris Lattner
d2e905027b revert r123146 which disabled code that wasn't the root cause
of the bootstrap miscompare issue.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123299 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-12 01:52:23 +00:00
Chris Lattner
86099ba2b5 merge tests into one crash.ll test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123220 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 07:50:07 +00:00
Chris Lattner
93767fdb61 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).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123219 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 07:47:59 +00:00
Chandler Carruth
15ed90c859 Teach constant folding to perform conversions from constant floating
point values to their integer representation through the SSE intrinsic
calls. This is the last part of a README.txt entry for which I have real
world examples.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123206 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 01:07:24 +00:00
Chandler Carruth
f7b0047f5f FileCheck-ize a test, and move a no-longer calling test case to another
file and make it actually test something...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123205 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 01:07:20 +00:00
Owen Anderson
da1c122da5 Fix a random missed optimization by making InstCombine more aggressive when determining which bits are demanded by
a comparison against a constant.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123203 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-11 00:36:45 +00:00
Chandler Carruth
9cc9f50abc Teach instcombine about the rest of the SSE and SSE2 conversion
intrinsics element dependencies. Reviewed by Nick.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123161 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10 07:19:37 +00:00
Chandler Carruth
fdc8f2d260 Fold two related tests into the newly FileCheck-ized test, migrating
them to FileCheck as well.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123154 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10 02:53:58 +00:00
Chandler Carruth
548e581dcb Clean up and FileCheck-ize a test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123153 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10 02:53:54 +00:00
Chris Lattner
f86c75da4d fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123148 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10 02:33:34 +00:00
Chris Lattner
a806be66c1 another (more) aggressive attempt to bring llvm-gcc-i386-linux-selfhost
back to life.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123146 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10 00:47:34 +00:00
Chris Lattner
d8408270f3 temporarily disable memset formation from memsets in an effort to restore buildbot stability.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123144 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-09 23:52:48 +00:00
Tobias Grosser
aa2be84356 Instcombine: Fix pattern where the sext did not dominate the icmp using it
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123121 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-09 16:00:11 +00:00
Chris Lattner
d90a192279 Merge memsets followed by neighboring memsets and other stores into
larger memsets.  Among other things, this fixes rdar://8760394 and
allows us to handle "Example 2" from http://blog.regehr.org/archives/320,
compiling it into a single 4096-byte memset:

_mad_synth_mute:                        ## @mad_synth_mute
## BB#0:                                ## %entry
	pushq	%rax
	movl	$4096, %esi             ## imm = 0x1000
	callq	___bzero
	popq	%rax
	ret



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123089 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 21:19:19 +00:00
Chris Lattner
9fa11e94b5 fix an issue in IsPointerOffset that prevented us from recognizing that
P and P+1 are relative to the same base pointer.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123087 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 21:07:56 +00:00
Chris Lattner
06511264f8 enhance memcpyopt to merge a store and a subsequent
memset into a single larger memset.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123086 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 20:54:51 +00:00
Chris Lattner
355f5778aa merge two tests and filecheckify
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123082 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 20:27:22 +00:00
Chris Lattner
5d37370a6f 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.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123079 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 19:59:06 +00:00
Chris Lattner
0e4a1543ab 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123072 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 18:52:51 +00:00
Frits van Bommel
b686eb9186 Fix a bug in r123034 (trying to sext/zext non-integers) and clean up a little.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123061 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 10:51:36 +00:00
Chris Lattner
d9ec3572f3 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123060 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08 08:24:46 +00:00
Tobias Grosser
46431d7a93 InstCombine: Match min/max hidden by sext/zext
X = sext x; x >s c ? X : C+1 --> X = sext x; X <s C+1 ? C+1 : X
X = sext x; x <s c ? X : C-1 --> X = sext x; X >s C-1 ? C-1 : X
X = zext x; x >u c ? X : C+1 --> X = zext x; X <u C+1 ? C+1 : X
X = zext x; x <u c ? X : C-1 --> X = zext x; X >u C-1 ? C-1 : X
X = sext x; x >u c ? X : C+1 --> X = sext x; X <u C+1 ? C+1 : X
X = sext x; x <u c ? X : C-1 --> X = sext x; X >u C-1 ? C-1 : X

Instead of calculating this with mixed types promote all to the
larger type. This enables scalar evolution to analyze this
expression. PR8866

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123034 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07 21:33:14 +00:00
Benjamin Kramer
eaff66a895 Revert 122959, it needs more thought. Add it back to README.txt with additional notes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123030 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07 20:42:20 +00:00
Benjamin Kramer
8143a84c46 InstCombine: Turn _chk functions into the "unsafe" variant if length and max langth are equal.
This happens when we take the (non-constant) length from a malloc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122961 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-06 14:22:52 +00:00
Benjamin Kramer
240d42d185 InstCombine: If we call llvm.objectsize on a malloc call we can replace it with the size passed to malloc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122959 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-06 13:11:05 +00:00
Benjamin Kramer
783a5c2b69 InstCombine: Teach llvm.objectsize folding to look through GEPs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122958 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-06 13:07:49 +00:00
Chris Lattner
8cd4efb6a5 implement constant folding support for an exotic constant expr:
ret i64 ptrtoint (i8* getelementptr ([1000 x i8]* @X, i64 1, i64 sub (i64 0, i64 ptrtoint ([1000 x i8]* @X to i64))) to i64)

to "ret i64 1000".  This allows us to correctly compute the trip count
on a loop in PR8883, which occurs with std::fill on a char array.  This
allows us to transform it into a memset with a constant size.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122950 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-06 06:19:46 +00:00
Chris Lattner
43b40a4620 fix an off-by-one bug that caused a crash analyzing
ashr's with huge shift amounts, PR8896


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122814 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-04 18:19:15 +00:00
Chris Lattner
e41d3c015c Teach loop-idiom to turn a loop containing a memset into a larger memset
when safe.

The testcase is basically this nested loop:
void foo(char *X) {
  for (int i = 0; i != 100; ++i) 
    for (int j = 0; j != 100; ++j)
      X[j+i*100] = 0;
}

which gets turned into a single memset now.  clang -O3 doesn't optimize
this yet though due to a phase ordering issue I haven't analyzed yet.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122806 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-04 07:46:33 +00:00
Chris Lattner
e508dd4c75 Duncan deftly points out that readnone functions aren't
invalidated by stores, so they can be handled as 'simple'
operations.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122785 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 23:38:13 +00:00
Chris Lattner
75637154c3 earlycse can do trivial with-a-block dead store
elimination as well.  This deletes 60 stores in 176.gcc
that largely come from bitfield code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122736 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 04:17:24 +00:00
Chris Lattner
ef87fc2e0a now that loads are in their own table, we can implement
store->load forwarding.  This allows EarlyCSE to zap 600 more
loads from 176.gcc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122732 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 03:46:34 +00:00
Chris Lattner
03d49e955e add a testcase for readonly call CSE
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122730 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 03:33:47 +00:00
Chris Lattner
8e7f0d70c7 Teach EarlyCSE to do trivial CSE of loads and read-only calls.
On 176.gcc, this catches 13090 loads and calls, and increases the
number of simple instructions CSE'd from 29658 to 36208.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122727 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-03 03:18:43 +00:00
Chris Lattner
91139ccd99 add DEBUG and -stats output to earlycse.
Teach it to CSE the rest of the non-side-effecting instructions.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122716 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 23:19:45 +00:00
Chris Lattner
cc9eab26b3 Enhance earlycse to do CSE of casts, instsimplify and die.
Add a testcase.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122715 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 23:04:14 +00:00
Chris Lattner
63f9c3c49a fix a miscompilation of tramp3d-v4: when forming a memcpy, we have to make
sure that the loop we're promoting into a memcpy doesn't mutate the input
of the memcpy.  Before we were just checking that the dest of the memcpy
wasn't mod/ref'd by the loop.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122712 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 21:14:18 +00:00
Chris Lattner
8e08e73f0e If a loop iterates exactly once (has backedge count = 0) then don't
mess with it.  We'd rather peel/unroll it than convert all of its 
stores into memsets.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122711 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 20:24:21 +00:00
Chris Lattner
62c50fdf69 enhance loop idiom recognition to scan *all* unconditionally executed
blocks in a loop, instead of just the header block.  This makes it more
aggressive, able to handle Duncan's Ada examples.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122704 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 19:01:03 +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
Chris Lattner
cf078f2b20 Allow loop-idiom to run on multiple BB loops, but still only scan the loop
header for now for memset/memcpy opportunities.  It turns out that loop-rotate
is successfully rotating loops, but *DOESN'T MERGE THE BLOCKS*, turning "for 
loops" into 2 basic block loops that loop-idiom was ignoring.

With this fix, we form many *many* more memcpy and memsets than before, including
on the "history" loops in the viterbi benchmark, which look like this:

        for (j=0; j<MAX_history; ++j) {
          history_new[i][j+1] = history[2*i][j];
        }

Transforming these loops into memcpy's speeds up the viterbi benchmark from
11.98s to 3.55s on my machine.  Woo.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122685 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-02 07:58:36 +00:00