Commit Graph

63001 Commits

Author SHA1 Message Date
Nate Begeman
51409214d7 Implement a vectorized algorithm for <16 x i8> << <16 x i8>
This is about 4x faster and smaller than the existing scalarization.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109566 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-28 00:21:48 +00:00
Dan Gohman
f374ba2bcd Add a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109565 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-28 00:21:18 +00:00
Owen Anderson
9a65dc9968 Rearrange several datastructures in LazyValueInfo to improve compile time.
This is still not perfect, but better than it was before.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109563 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 23:58:11 +00:00
Owen Anderson
cf85c96129 Fill out the interface of DenseSet a bit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109562 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 23:55:47 +00:00
Stuart Hastings
3c73a6b650 Testcase for r109556. Radar 8198362.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109557 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 23:15:25 +00:00
Gabor Greif
b344a346f1 reintroduce original (asserting) semantics of CallSite(Instruction *II)
add instead a CallSite(Value* V) constructor that is consistent with ImmutableCallSize
and use that one in client code

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109553 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 22:53:28 +00:00
Nate Begeman
bdcb5afb77 ~40% faster vector shl <4 x i32> on SSE 4.1 Larger improvements for smaller types coming in future patches.
For:

define <2 x i64> @shl(<4 x i32> %r, <4 x i32> %a) nounwind readnone ssp {
entry:
  %shl = shl <4 x i32> %r, %a                     ; <<4 x i32>> [#uses=1]
  %tmp2 = bitcast <4 x i32> %shl to <2 x i64>     ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %tmp2
}

We get:

_shl:                                   ## @shl
	pslld	$23, %xmm1
	paddd	LCPI0_0, %xmm1
	cvttps2dq	%xmm1, %xmm1
	pmulld	%xmm1, %xmm0
	ret

Instead of:

_shl:                                   ## @shl
	pshufd	$3, %xmm0, %xmm2
	movd	%xmm2, %eax
	pshufd	$3, %xmm1, %xmm2
	movd	%xmm2, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm2
	pshufd	$1, %xmm0, %xmm3
	movd	%xmm3, %eax
	pshufd	$1, %xmm1, %xmm3
	movd	%xmm3, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm3
	punpckldq	%xmm2, %xmm3
	movd	%xmm0, %eax
	movd	%xmm1, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm2
	movhlps	%xmm0, %xmm0
	movd	%xmm0, %eax
	movhlps	%xmm1, %xmm1
	movd	%xmm1, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm0
	punpckldq	%xmm0, %xmm2
	movdqa	%xmm2, %xmm0
	punpckldq	%xmm3, %xmm0
	ret


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109549 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 22:37:06 +00:00
Gabor Greif
622b7cf147 recommit simplification (originally r109504, backed out in r109508) now that problem in CallSiteBase is fixed
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109547 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 22:02:00 +00:00
Gabor Greif
a6758446b7 remove bogus assert, use static_cast for additional checking
left two new asserts commented out, because they would fire in clang, have to hunt those down first

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109544 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 21:46:11 +00:00
Devang Patel
37032350c8 It is FE's responsibility to emit proper directory name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109538 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 20:51:15 +00:00
Chris Lattner
90f8795e1c make lookup failures not fatal.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109530 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 18:59:50 +00:00
Jim Grosbach
977fa34df8 Grammar
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109525 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 18:36:27 +00:00
Devang Patel
8b5749da2a Update tests to not rely on input file's absolute path.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109521 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 18:13:53 +00:00
Nate Begeman
e0efc216eb Fix a crash in the dag combiner caused by ConstantFoldBIT_CONVERTofBUILD_VECTOR calling itself
recursively and returning a SCALAR_TO_VECTOR node, but assuming the input was always a BUILD_VECTOR.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109519 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 18:02:18 +00:00
Jim Grosbach
57cb4f8be1 80 column
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109513 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 17:38:47 +00:00
Jim Grosbach
ae64eedae2 fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109511 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 17:14:29 +00:00
Gabor Greif
a292b2f49f recommit simplification (r109502, backed out r109509); seems to innocent
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109510 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 16:44:23 +00:00
Gabor Greif
5b0c25313e back out this too to restore the bots
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109509 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 15:56:07 +00:00
Gabor Greif
49d0bca86c back out r109504, breaks the bots
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109508 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 15:18:11 +00:00
Gabor Greif
86099345db simplify: CallSite::get --> CallSite constructor
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109506 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 15:02:37 +00:00
Gabor Greif
c0d4aed6f7 simplify
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109504 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 14:38:38 +00:00
Gabor Greif
0de11e0e66 use ImmutableCallSite for const-corrgoodness
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109503 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 14:15:29 +00:00
Gabor Greif
50da18171e simplify
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109502 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 13:31:22 +00:00
Mikhail Glushenkov
b482f23c6b Fix silent failure with no input files.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109500 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 11:19:40 +00:00
Mikhail Glushenkov
67d985ffb2 Return -1 only on failure to execute a program.
Also fix some comments.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109499 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 11:19:36 +00:00
Duncan Sands
978bb87f88 In commit 91421, isPod was changed from false to true for these value handles.
Change it back again: destructors and constructors need to be run.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109498 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 09:09:05 +00:00
Tobias Grosser
56029271f9 Make coff-dump.py executable and add python as executable for this script.
This fixes the MC/COFF/basic-coff.ll test case.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109497 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 09:01:26 +00:00
Tobias Grosser
0e6fcf4be3 RegionInfo: Add getMaxRegionExit()
getMaxRegionExit returns the exit of the maximal refined region starting
at a specific basic block.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109496 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 08:39:43 +00:00
Duncan Sands
2c11046fa1 After updating value handles for RAUW, check that no weak or tracking handles
are still on the list.  This might happen if a CallbackVH created some new value
handles for the old value when doing RAUW.  Barf if it occurs, since it is almost
certainly a mistake.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109495 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 06:53:14 +00:00
Michael J. Spencer
dfd30187c6 Make MC use Windows COFF on Windows and add tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109494 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 06:46:15 +00:00
Tobias Grosser
082d587d35 Add function to query RegionInfo about loops.
* contains(Loop), * getOutermostLoop()
* Improve getNameStr() to return a sensible name, if basic blocks are not named.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109490 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 04:17:13 +00:00
Jakob Stoklund Olesen
81c7b19f04 The isLoadFromStackSlot and isStoreToStackSlot have no way of reporting
subregister operands like this:

%reg1040:sub_32bit<def> = MOV32rm <fi#-2>, 1, %reg0, 0, %reg0, %reg1040<imp-def>; mem:LD4[FixedStack-2](align=8)

Make them return false when subreg operands are present. VirtRegRewriter is
making bad assumptions otherwise.

This fixes PR7713.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109489 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 04:17:01 +00:00
Jakob Stoklund Olesen
516cd4575e Add assertions that expose the PR7713 miscompilation: Accessing a stack slot
with a too-big register class.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109488 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 04:16:58 +00:00
Bill Wendling
dfc2c51d12 It's better to have the arrays, which would trigger the creation of stack
protectors, to be near the stack protectors on the stack. Accomplish this by
tagging the stack object with a predicate that indicates that it would trigger
this. In the prolog-epilog inserter, assign these objects to the stack after the
stack protector but before the other objects.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109481 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 01:55:19 +00:00
Dan Gohman
ab28928fe2 Use AssertingVH for InsertedValues and InsertedPostIncValues, to verify
that the values they refer to aren't being deleted underneath them.

Make sure these containters get cleared by clear(), which IndVarSimplify
and LSR both use before deleting instructions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109478 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 01:19:06 +00:00
Owen Anderson
05cfd335bb (Hopefully) One last test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109473 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 00:04:55 +00:00
Owen Anderson
3debf271ff Another test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109472 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-27 00:00:47 +00:00
Owen Anderson
3e12033ef6 Another test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109471 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 23:58:46 +00:00
Jakob Stoklund Olesen
f0179004e9 Add SplitEditor to SplitKit. This class will be used to edit live intervals and
rewrite instructions for live range splitting.

Still work in progress.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109469 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 23:44:11 +00:00
Dan Gohman
3eca15bdb5 Fix a use-after-free.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109468 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 23:40:24 +00:00
Bill Wendling
23295cc1dd Using llvm.eh.catch.all.value instead of .llvm.eh.catch.all.value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109462 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 22:36:52 +00:00
Eli Friedman
a10213e934 And a bit more non-ASCII stuff.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109458 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 22:28:18 +00:00
Anton Korobeynikov
4e084e93a7 Drop some non-ascii stuff
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109456 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 22:23:07 +00:00
Bob Wilson
e4373b0da6 Don't call __register_frame from the JIT on systems that use setjmp/longjmp
exception handling.  Also fix an extra underscore typo in one instance of
"__ARM_EABI__".  Radar 8236264.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109451 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:58:00 +00:00
Evan Cheng
dee81010eb On x86, f32 / f64 nodes share the same registers as 128-bit vector values.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109450 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:50:05 +00:00
Evan Cheng
e0e925efb3 The "excess register pressure" returned by HighRegPressure() is not accurate enough to factor into scheduling priority. Eliminate it and add early exits to speed up scheduling.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109449 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:49:07 +00:00
Anton Korobeynikov
25a6ab0144 Add a note
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109448 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:48:35 +00:00
Dan Gohman
72eb300715 Add a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109435 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:05:37 +00:00
Bruno Cardoso Lopes
3c45734274 Temporary hack to let codegen assert or generate poor code in case
we are using AVX and no AVX version of the desired intruction is present,
this is better for incremental dev (without fallbacks it's easier to spot
what's missing). Not sure this is the best hack thought (we can also disable
all HasSSE* predicates by dinamically marking them 'false' if AVX is present)



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109434 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 21:01:18 +00:00
Owen Anderson
c2b13f919d Testing some more.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109433 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26 20:51:45 +00:00