Commit Graph

869 Commits

Author SHA1 Message Date
Dan Gohman
7ba145b0b4 Revert 54147.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54148 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-29 01:02:18 +00:00
Dan Gohman
b1e8cad61e Add x86 isel patterns to match what would be a ZERO_EXTEND_INREG operation,
which is represented in codegen as an 'and' operation. This matches them
with movz instructions, instead of leaving them to be matched by and
instructions with an immediate field.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54147 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-28 22:18:25 +00:00
Dan Gohman
6c4942641f Fix embedded CRLF characters.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54125 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-27 18:37:58 +00:00
Nate Begeman
907f46a81f Fix test RUN line
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54040 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-25 19:08:59 +00:00
Nate Begeman
fb8ead0c20 Disable mov{L, LP, HP, HLP, *DUP} shuffles for mmx
mmx needs its own fancy shuffle logic based on unpack; for now we get correct but awful code.

Also commit Mon Ping's VSETCC patch


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54039 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-25 19:05:58 +00:00
Dan Gohman
1f335e3139 This test needs -aggressive-remat enabled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54015 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-25 15:25:32 +00:00
Dan Gohman
6d69ba8a69 Enable rematerialization of constants using AliasAnalysis::pointsToConstantMemory,
and knowledge of PseudoSourceValues. This unfortunately isn't sufficient to allow
constants to be rematerialized in PIC mode -- the extra indirection is a
complication.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54000 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-25 00:02:30 +00:00
Dan Gohman
5444c30723 Add target triples so these tests behave as expected on non-darwin hosts.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53991 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-24 18:08:01 +00:00
Evan Cheng
2d58bfac59 New test case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53971 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-24 00:22:05 +00:00
Evan Cheng
5e6ebaf4d1 Fix PR2485: do all 4-element SSE shuffles in max. of 2 shuffle instructions.
Based on patch by Nicolas Capens.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53939 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-23 00:22:17 +00:00
Duncan Sands
11e56cb4dc LegalizeTypes support for VSETCC. Fixes PR2575.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53938 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-22 23:54:03 +00:00
Evan Cheng
2925786765 Fix pr2566: incorrect assumption about bit_convert. It doesn't not have to output a vector value. Patch by Nicolas Capens!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53932 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-22 20:42:56 +00:00
Evan Cheng
52672b813e Fix PR2574: implement v2f32 scalar_to_vector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53927 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-22 18:39:19 +00:00
Bill Wendling
3180e20cda Fix for first part of PR2562. Generate the "pinsrw" instruction for inserts
into v4i16 vectors.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53807 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-20 02:32:23 +00:00
Anton Korobeynikov
33500e5882 Testcase for PR2549
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53785 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-19 06:31:12 +00:00
Evan Cheng
76a4d58998 Subreg live interval valno may not have a corresponding def machineinstr since it's less precise.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53734 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-17 19:48:53 +00:00
Evan Cheng
d1718252b5 Add nounwind.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53733 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-17 19:48:04 +00:00
Evan Cheng
79a796c2b1 Fix PR2536: a nasty spiller bug. If a two-address instruction uses a register but the use portion of its live range is not part of its liveinterval, it must be defined by an implicit_def. In that case, do not spill the use. e.g.
8   %reg1024<def> = IMPLICIT_DEF
12  %reg1024<def> = INSERT_SUBREG %reg1024<kill>, %reg1025, 2

The live range [12, 14) are not part of the r1024 live interval since it's defined by an implicit def. It will not conflicts with live interval of r1025. Now suppose both registers are spilled, you can easily see a situation where both registers are reloaded before the INSERT_SUBREG and both target registers that would overlap.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53503 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-12 01:56:02 +00:00
Duncan Sands
edfba7e707 Port a shift-by-1 optimization from LegalizeDAG: it
was presumably added after the rest of the code was
copied to LegalizeTypes.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53459 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-11 16:54:57 +00:00
Bill Wendling
71ca353ae6 The frame address on an x86-64 box needs to be offset by -8, not -4.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53450 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-11 07:18:52 +00:00
Evan Cheng
331e2bd942 Fix for PR2472. Use movss to set lower 32-bits of a zero XMM vector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53386 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-10 01:08:23 +00:00
Anton Korobeynikov
091be59220 Testcase for PR2024
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53327 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-09 14:09:41 +00:00
Evan Cheng
1ce75dcbbc Fix two serious LSR bugs.
1. LSR runOnLoop is always returning false regardless if any transformation is made.
2. AddUsersIfInteresting can create new instructions that are added to DeadInsts. But there is a later early exit which prevents them from being freed.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53193 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-07 19:51:32 +00:00
Dale Johannesen
ecdc82a643 Considering predecessors of exit blocks gets
us a little more tail merging.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52986 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-01 21:50:49 +00:00
Chris Lattner
bd2acd736d test doesn't need eh info
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52811 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-27 03:14:20 +00:00
Dale Johannesen
4e97790682 Allow for rounding up of stack frame.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52751 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-26 01:55:32 +00:00
Chris Lattner
cda8875433 when we know the signbit of an input to uint_to_fp is zero,
change it to sint_to_fp on targets where that is cheaper (and
visaversa of course).  This allows us to compile uint_to_fp to:

_test:
	movl	4(%esp), %eax
	shrl	$23, %eax
	cvtsi2ss	%eax, %xmm0
	movl	8(%esp), %eax
	movss	%xmm0, (%eax)
	ret

instead of:

	.align	3
LCPI1_0:					##  double
	.long	0	## double least significant word 4.5036e+15
	.long	1127219200	## double most significant word 4.5036e+15
	.text
	.align	4,0x90
	.globl	_test
_test:
	subl	$12, %esp
	movl	16(%esp), %eax
	shrl	$23, %eax
	movl	%eax, (%esp)
	movl	$1127219200, 4(%esp)
	movsd	(%esp), %xmm0
	subsd	LCPI1_0, %xmm0
	cvtsd2ss	%xmm0, %xmm0
	movl	20(%esp), %eax
	movss	%xmm0, (%eax)
	addl	$12, %esp
	ret




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52747 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-26 00:16:49 +00:00
Evan Cheng
ab26227c8c - Fix a x86 vector isel bug: illegal transformation of a vector_shuffle into a
shift.
- Add a readme entry for a missing vector_shuffle optimization that results in
  awful codegen.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52740 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-25 20:52:59 +00:00
Mon P Wang
2887310630 Added MemOperands to Atomic operations since Atomics touches memory.
Added abstract class MemSDNode for any Node that have an associated MemOperand
Changed atomic.lcs => atomic.cmp.swap, atomic.las => atomic.load.add, and
atomic.lss => atomic.load.sub


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52706 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-25 08:15:39 +00:00
Evan Cheng
601ca4b434 Enable two-address remat by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52701 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-25 01:16:38 +00:00
Dale Johannesen
4a103e6d7e v2f32 is now a valid (MMX) type which breaks this
test (doesn't work for any MMX vector types, it's
not me).  Rewritten to use v2i16 which is generic
and going to stay that way; I think that preserves
the point of the test.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52692 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-24 22:03:36 +00:00
Evan Cheng
9ef4ca2e81 If it's determined safe, remat MOV32r0 (i.e. xor r, r) and others as it is instead of using the longer MOV32ri instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52670 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-24 07:10:51 +00:00
Bill Wendling
04eeefb32a Make test work on non-x86 machines (like my G4 PPC).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52619 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-23 06:16:31 +00:00
Evan Cheng
c3417609ae Undo spill weight tweak. Need to investigate the performance regressions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52572 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-21 06:45:54 +00:00
Eli Friedman
2c8e64d3e4 Fix a bug with <8 x i16> shuffle lowering on X86 where parts of the
shuffle could be skipped.  The check is invalid because the loop index i 
doesn't correspond to the element actually inserted. The correct check is
already done a few lines earlier, for whether the element is already in 
the right spot, so this shouldn't have any effect on the codegen for 
code that was already correct.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52486 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-19 06:09:51 +00:00
Evan Cheng
0d46d753f9 New test case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52483 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-19 01:50:24 +00:00
Evan Cheng
72d255aff9 This also got better (55 - 51 instructions). But doing one more re-materialization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52482 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-19 01:50:13 +00:00
Evan Cheng
81d39e63e2 This got better.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52481 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-19 01:46:43 +00:00
Evan Cheng
c3de802ad9 Do not issue identity copies.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52373 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-16 22:52:53 +00:00
Evan Cheng
6bd9567a6a - Add "Commutative" property to intrinsics. This allows tblgen to generate the commuted variants for dagisel matching code.
- Mark lots of X86 intrinsics as "Commutative" to allow load folding.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52353 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-16 20:29:38 +00:00
Evan Cheng
87bb991aa8 Teach the spiller to commute instructions in order to fold a reload. This hits 410 times on 444.namd and 122 times on 252.eon.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52266 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-13 23:58:02 +00:00
Duncan Sands
d4b9c17fb7 Disable some DAG combiner optimizations that may be
wrong for volatile loads and stores.  In fact this
is almost all of them!  There are three types of
problems: (1) it is wrong to change the width of
a volatile memory access.  These may be used to
do memory mapped i/o, in which case a load can have
an effect even if the result is not used.  Consider
loading an i32 but only using the lower 8 bits.  It
is wrong to change this into a load of an i8, because
you are no longer tickling the other three bytes.  It
is also unwise to make a load/store wider.  For
example, changing an i16 load into an i32 load is
wrong no matter how aligned things are, since the
fact of loading an additional 2 bytes can have
i/o side-effects.  (2) it is wrong to change the
number of volatile load/stores: they may be counted
by the hardware.  (3) it is wrong to change a volatile
load/store that requires one memory access into one
that requires several.  For example on x86-32, you
can store a double in one processor operation, but to
store an i64 requires two (two i32 stores).  In a
multi-threaded program you may want to bitcast an i64
to a double and store as a double because that will
occur atomically, and be indivisible to other threads.
So it would be wrong to convert the store-of-double
into a store of an i64, because this will become two
i32 stores - no longer atomic.  My policy here is
to say that the number of processor operations for
an illegal operation is undefined.  So it is alright
to change a store of an i64 (requires at least two
stores; but could be validly lowered to memcpy for
example) into a store of double (one processor op).
In short, if the new store is legal and has the same
size then I say that the transform is ok.  It would
also be possible to say that transforms are always
ok if before they were illegal, whether after they
are illegal or not, but that's more awkward to do
and I doubt it buys us anything much.
However this exposed an interesting thing - on x86-32
a store of i64 is considered legal!  That is because
operations are marked legal by default, regardless of
whether the type is legal or not.  In some ways this
is clever: before type legalization this means that
operations on illegal types are considered legal;
after type legalization there are no illegal types
so now operations are only legal if they really are.
But I consider this to be too cunning for mere mortals.
Better to do things explicitly by testing AfterLegalize.
So I have changed things so that operations with illegal
types are considered illegal - indeed they can never
map to a machine operation.  However this means that
the DAG combiner is more conservative because before
it was "accidentally" performing transforms where the
type was illegal because the operation was nonetheless
marked legal.  So in a few such places I added a check
on AfterLegalize, which I suppose was actually just
forgotten before.  This causes the DAG combiner to do
slightly more than it used to, which resulted in the X86
backend blowing up because it got a slightly surprising
node it wasn't expecting, so I tweaked it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52254 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-13 19:07:40 +00:00
Evan Cheng
933b5065e5 Fix some tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52245 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-12 21:23:38 +00:00
Dale Johannesen
287abdb357 Fix parameter spelling: sse not sse1
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52185 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-10 17:57:58 +00:00
Matthijs Kooijman
5efb967052 Fix some more quoting issues in RUN lines, this time regarding unintended
variable expansions involving the $ character.

This fixes 4 tests that were not running properly before.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52183 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-10 16:10:32 +00:00
Matthijs Kooijman
7f88d9c62b Remove double pipes in RUN commandlines.
This fixes 5 testcases that were not being run properly before.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52180 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-10 15:11:36 +00:00
Dan Gohman
c5b822b5b6 Convert several tests to use temporary files instead of redundantly
executing the test commands.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52163 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-10 00:36:41 +00:00
Rafael Espindola
d674b4e87d add support for PIC on linux x86-64
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52139 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-09 09:52:31 +00:00
Evan Cheng
fb4db316d8 Fix a memcpy lowering bug. Even though the memcpy alignment is smaller than the desired alignment, the frame destination alignment may still be larger than the desired alignment. Don't change its alignment to something smaller.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51970 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-04 23:37:54 +00:00
Dan Gohman
37cdad3e59 Fix the position of MemOperands in nodes that use variadic_ops
in DAGISelEmitter output. This bug was recently uncovered by the
addition of patterns for CALL32m and CALL64m, which are nodes
that now have both MemOperands and variadic_ops.

This bug was especially visible with PIC in various configurations,
because the new patterns are matching the indirect call code used
in many PIC configurations.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51877 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-02 17:40:38 +00:00