351a3f881f
Avoid fusing bitcasts with dynamic allocas if the amount-to-allocate
...
might overflow. Re-typing the alloca to a larger type (e.g. double)
hoists a shift into the alloca, potentially exposing overflow in the
expression. rdar://problem/9265821
llvm-svn: 132926
2011-06-13 18:48:49 +00:00
c970849ea0
InstCombine: Fold A-b == C --> b == A-C if A and C are constants.
...
The backend already knew this trick.
llvm-svn: 132915
2011-06-13 15:24:24 +00:00
91f914ce21
InstCombine: Shrink ((zext X) & C1) == C2 to fold away the cast if the "zext" and the "and" have one use.
...
llvm-svn: 132897
2011-06-12 22:48:00 +00:00
35159c114c
Simplify code. No functionality changes, name changes aside.
...
llvm-svn: 132896
2011-06-12 22:47:53 +00:00
2380483355
Reapply 132348 with fixes. rdar://problem/6501862
...
llvm-svn: 132402
2011-06-01 16:42:47 +00:00
9d6a06d536
Revert to pacify a buildbot. rdar://problem/6501862
...
llvm-svn: 132351
2011-05-31 19:56:35 +00:00
780f723309
Followup to 132316; accept arbitrary constants, add with a constant,
...
sub with a non-constant. Fix comments, enlarge test case.
rdar://problem/6501862
llvm-svn: 132348
2011-05-31 19:29:55 +00:00
8284374b07
(1 - X) * (-2) -> (x - 1) * 2, for all positive nonzero powers of 2
...
rdar://problem/6501862
llvm-svn: 132316
2011-05-30 20:00:33 +00:00
fd53a27f99
ConstantFoldInstOperands doesn't like compares, hand it off to instsimplify instead.
...
Fixes PR10040.
llvm-svn: 132254
2011-05-28 10:16:58 +00:00
749ef5f420
InstCombine: Make switch folding with equality compares more aggressive by trying instsimplify on the arm where we know the compared value.
...
Stuff like "x == y ? y : x&y" now folds into "x&y".
llvm-svn: 132185
2011-05-27 13:00:16 +00:00
b362884ca9
Renamed llvm.x86.sse42.crc32 intrinsics; crc64 doesn't exist.
...
crc32.[8|16|32] have been renamed to .crc32.32.[8|16|32] and
crc64.[8|16|32] have been renamed to .crc32.64.[8|64].
llvm-svn: 132163
2011-05-26 23:13:19 +00:00
865866e7fe
PR9998: ashr exact %x, 31 is not equivalent to sdiv exact %x, -2147483648.
...
llvm-svn: 132097
2011-05-25 23:26:20 +00:00
8aff4f8efc
Transform any logical shift of a power of two into an exact/NUW shift when
...
in a known-non-zero context.
llvm-svn: 131887
2011-05-23 00:21:50 +00:00
7c99f19d9f
Carve out a place in instcombine to put transformations which work knowing that their
...
result is non-zero. Implement an example optimization (PR9814), which allows us to
transform:
A / ((1 << B) >>u 2)
into:
A >>u (B-2)
which we compile into:
_divu3: ## @divu3
leal -2(%rsi), %ecx
shrl %cl, %edi
movl %edi, %eax
ret
instead of:
_divu3: ## @divu3
movb %sil, %cl
movl $1, %esi
shll %cl, %esi
shrl $2, %esi
movl %edi, %eax
xorl %edx, %edx
divl %esi, %eax
ret
llvm-svn: 131860
2011-05-22 18:18:41 +00:00
fda5dc4968
Revert "InstCombine: Turn mul.with.overflow(X, 2) into the cheaper add.with.overflow(X, X)"
...
It's better to do this in codegen, mul.with.overflow(X, 2) is more canonical because it has only one use on "X".
llvm-svn: 131798
2011-05-21 18:31:42 +00:00
691731eb9c
InstCombine: Turn mul.with.overflow(X, 2) into the cheaper add.with.overflow(X, X)
...
llvm-svn: 131789
2011-05-21 09:22:06 +00:00
e8d2e9eb35
Revert r131664 and fix it in instcombine instead. rdar://9467055
...
llvm-svn: 131708
2011-05-20 00:54:37 +00:00
ae012a7525
Move test to Transforms/InstCombine.
...
llvm-svn: 131634
2011-05-19 05:53:22 +00:00
41e509a33d
More instcombine cleanup, towards improving debug line info.
...
llvm-svn: 131604
2011-05-18 23:58:37 +00:00
49346010f8
More instcombine cleanup aimed towards improving debug line info.
...
llvm-svn: 131559
2011-05-18 19:57:14 +00:00
96254a0d53
Start trying to make InstCombine preserve more debug info. The idea here is to set the debug location on the IRBuilder, which will be then right location in most cases. This should magically give many transformations debug locations, and fixing places which are missing a debug location will usually just means changing the code creating it to use the IRBuilder.
...
As an example, the change to InstCombineCalls catches a common case where a call to a bitcast of a function is rewritten.
Chris, does this approach look reasonable?
llvm-svn: 131516
2011-05-18 01:28:27 +00:00
a7ae4552af
Drop lli, revise test.
...
llvm-svn: 131452
2011-05-17 02:38:59 +00:00
cb7e56e592
Disable test harder.
...
llvm-svn: 131363
2011-05-14 19:30:39 +00:00
3c2fd1cf62
Disable this test while I revise it. rdar://problem/9267970
...
llvm-svn: 131350
2011-05-14 18:39:05 +00:00
66a82b966e
Avoid combining GEPs that might overflow at runtime.
...
rdar://problem/9267970
Patch by Julien Lerouge!
llvm-svn: 131339
2011-05-14 05:55:10 +00:00
8a20e66926
PR9838: Fix transform introduced in r127064 to not trigger when only one side of the icmp is an exact shift.
...
llvm-svn: 130954
2011-05-05 21:59:18 +00:00
f91c5ab341
Fix PR9579: when simplifying a compare to "true" or "false", and it was
...
a vector compare, generate a vector result rather than i1 (and crashing).
llvm-svn: 130706
2011-05-02 18:51:41 +00:00
9aa91b1f4e
InstCombine: Turn (zext A) udiv (zext B) into (zext (A udiv B)). Same for urem or constant B.
...
This obviously helps a lot if the division would be turned into a libcall
(think i64 udiv on i386), but div is also one of the few remaining instructions
on modern CPUs that become more expensive when the bitwidth gets bigger.
This also helps register pressure on i386 when dividing chars, divb needs
two 8-bit parts of a 16 bit register as input where divl uses two registers.
int foo(unsigned char a) { return a/10; }
int bar(unsigned char a, unsigned char b) { return a/b; }
compiles into (x86_64)
_foo:
imull $205, %edi, %eax
shrl $11, %eax
ret
_bar:
movzbl %dil, %eax
divb %sil, %al
movzbl %al, %eax
ret
llvm-svn: 130615
2011-04-30 18:16:07 +00:00
57b3df59b9
Use SimplifyDemandedBits on div instructions.
...
This folds away silly stuff like (a&255)/1000 -> 0.
llvm-svn: 130614
2011-04-30 18:16:00 +00:00
6a50bbd284
FileCheckize.
...
llvm-svn: 130613
2011-04-30 18:15:53 +00:00
16f18ed7b5
InstCombine: turn (C1 << A) << C2) into (C1 << C2) << A)
...
Fixes PR9809.
llvm-svn: 130485
2011-04-29 08:15:41 +00:00
4145c0d3b1
InstCombine: Merge "(trunc x) == C1 & (and x, CA) == C2" into a single and+icmp.
...
This happens when GVN widens loads. Part of PR6627.
llvm-svn: 130405
2011-04-28 16:58:40 +00:00
1b06c71668
Transform: "icmp eq (trunc (lshr(X, cst1)), cst" to "icmp (and X, mask), cst"
...
when X has multiple uses. This is useful for exposing secondary optimizations,
but the X86 backend isn't ready for this when X has a single use. For example,
this can disable load folding.
This is inching towards resolving PR6627.
llvm-svn: 130238
2011-04-26 20:18:20 +00:00
d097212a08
Add test cases for Jay's r129641 and fix a 32-bit-centric testcase in a file with a 64-bit datalayout.
...
llvm-svn: 129643
2011-04-16 14:31:50 +00:00
0ab5e2cded
Fix a ton of comment typos found by codespell. Patch by
...
Luis Felipe Strano Moraes!
llvm-svn: 129558
2011-04-15 05:18:47 +00:00
2395626605
Add an instcombine for constructs like a | -(b != c); a select is more
...
canonical, and generally leads to better code. Found while looking at
an article about saturating arithmetic.
llvm-svn: 129545
2011-04-14 22:41:27 +00:00
cc771acd77
This testcase passed even without the fix. Added the target info to make the
...
test fail (without the fix). Thanks Dan.
llvm-svn: 128999
2011-04-06 11:18:29 +00:00
a069c6ce05
InstCombine optimizes gep(bitcast(x)) even when the bitcasts casts away address
...
space info. We crash with an assert in this case. This change checks that the
address space of the bitcasted pointer is the same as the gep ptr.
llvm-svn: 128884
2011-04-05 14:29:52 +00:00
d121765e64
InstCombine: Turn icmp + sext into bitwise/integer ops when the input has only one unknown bit.
...
int test1(unsigned x) { return (x&8) ? 0 : -1; }
int test3(unsigned x) { return (x&8) ? -1 : 0; }
before (x86_64):
_test1:
andl $8, %edi
cmpl $1, %edi
sbbl %eax, %eax
ret
_test3:
andl $8, %edi
cmpl $1, %edi
sbbl %eax, %eax
notl %eax
ret
after:
_test1:
shrl $3, %edi
andl $1, %edi
leal -1(%rdi), %eax
ret
_test3:
shll $28, %edi
movl %edi, %eax
sarl $31, %eax
ret
llvm-svn: 128732
2011-04-01 20:09:10 +00:00
d74b72b8a9
Instcombile optimization: extractelement(cast) -> cast(extractelement)
...
llvm-svn: 128683
2011-03-31 22:57:29 +00:00
5291054ef1
InstCombine: APFloat can't perform arithmetic on PPC double doubles, don't even try.
...
Thanks Eli!
llvm-svn: 128676
2011-03-31 21:35:49 +00:00
be209ab8a2
InstCombine: Fix transform to use the swapped predicate.
...
Thanks Frits!
llvm-svn: 128628
2011-03-31 10:46:03 +00:00
d159d94644
InstCombine: fold fcmp (fneg x), (fneg y) -> fcmp x, y
...
llvm-svn: 128627
2011-03-31 10:12:22 +00:00
a8c5d0872d
InstCombine: fold fcmp pred (fneg x), C -> fcmp swap(pred) x, -C
...
llvm-svn: 128626
2011-03-31 10:12:15 +00:00
cbb18e91a8
InstCombine: Shrink "fcmp (fpext x), C" to "fcmp x, C" if C can be losslessly converted to the type of x.
...
Fixes PR9592.
llvm-svn: 128625
2011-03-31 10:12:07 +00:00
2ccfbc8b71
InstCombine: fold fcmp (fpext x), (fpext y) -> fcmp x, y.
...
llvm-svn: 128624
2011-03-31 10:11:58 +00:00
af0ed953c5
Avoid turning a floating point division with a constant power of two into a denormal multiplication.
...
Some platforms may treat denormals as zero, on other platforms multiplication
with a subnormal is slower than dividing by a normal.
llvm-svn: 128555
2011-03-30 17:02:54 +00:00
8564e0de96
InstCombine: If the divisor of an fdiv has an exact inverse, turn it into an fmul.
...
Fixes PR9587.
llvm-svn: 128546
2011-03-30 15:42:35 +00:00
272f2b0044
InstCombine: Add a few missing combines for ANDs and ORs of sign bit tests.
...
On x86 we now compile "if (a < 0 && b < 0)" into
testl %edi, %esi
js IF.THEN
llvm-svn: 128496
2011-03-29 22:06:41 +00:00
8544228d5a
Teach the transformation that moves binary operators around selects to preserve
...
the subclass optional data.
llvm-svn: 128388
2011-03-27 19:51:23 +00:00
Stuart Hastings