1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 11:42:57 +01:00
Commit Graph

57788 Commits

Author SHA1 Message Date
Chris Lattner
1918d5b0aa fix an odd thinko in an assertion, all arm tests pass now.
llvm-svn: 96758
2010-02-21 20:57:28 +00:00
Chris Lattner
691dcb90a4 Always emit register class id references as i32 like
DAGISelEmitter does.  This fixes 11 arm failures (8 
left).

llvm-svn: 96757
2010-02-21 20:53:45 +00:00
Chris Lattner
37f20c29c8 add some no-unwinds, other minor cleanups.
llvm-svn: 96756
2010-02-21 20:33:20 +00:00
Anton Korobeynikov
0b402e1a61 It turned out that we failed to emit proper symbol stubs on non-x86/darwin for ages (we emitted a reference to a stub, but no stub was emitted). The code inside x86-32/macho target objfile lowering should actually be the generic one - move it there.
This (I really, really hope) should fix EH issues on ppc/darwin
and arm/darwin.

llvm-svn: 96755
2010-02-21 20:28:15 +00:00
Chris Lattner
4473066b55 speculatively teach OPC_CheckValueType and OPC_EmitNode to handle
MVT::iPTR.

llvm-svn: 96753
2010-02-21 20:15:25 +00:00
Chris Lattner
11ea2e1578 teach OPC_CheckType to handle MVT::iPTR, down to 2 ppc failures.
llvm-svn: 96752
2010-02-21 20:10:43 +00:00
Chris Lattner
26215565bf make this check a bit more generous, it may be outliving its
utility.  Down to 6 ppc failures.

llvm-svn: 96751
2010-02-21 20:02:15 +00:00
Chris Lattner
3d7cc59f2c Relax an assertion a bit. We allow replacing things like
<4 x i32> with <4 x float> values if they end up the same
register class.  This gets us up to 231 passes on the ppc
tests (only 7 fails).

llvm-svn: 96750
2010-02-21 19:35:07 +00:00
Chris Lattner
fa1fdcf146 add a triple so that this doesn't fail due to linux/ppc register printing
syntax.

llvm-svn: 96748
2010-02-21 19:27:38 +00:00
Chris Lattner
fc00a4e6b8 Sort the patterns before adding them to the FA so that we get the
least cost matches.  This gets us from 195 -> 208 passes on the ppc codegen tests.

llvm-svn: 96747
2010-02-21 19:22:06 +00:00
Duncan Sands
23f728e995 Remove a bunch of duplicated code, where there was one version taking a std::ostream
and another taking a raw_ostream, but otherwise identical.  Use raw_ostream everywhere.

llvm-svn: 96746
2010-02-21 19:15:19 +00:00
Chris Lattner
654f38165b filecheckize and add nouwinds.
llvm-svn: 96745
2010-02-21 18:53:28 +00:00
Jakob Stoklund Olesen
3a434bf60d Enable assertion to detect cyclic valno references.
This changes the stack overflow in PR6363 to an assertion failure.

llvm-svn: 96744
2010-02-21 18:51:48 +00:00
Anton Korobeynikov
fe0d6453ec IT turns out that during jumpless setcc lowering eq and ne were swapped.
This fixes PR6348

llvm-svn: 96734
2010-02-21 12:28:58 +00:00
Chris Lattner
c7fac8c8b2 when a match fails and we have to rollback, make sure to keep 'N' in
sync with the top of stack.  This fixes a bunch of failures on larger
testcases.

llvm-svn: 96732
2010-02-21 07:55:48 +00:00
Chris Lattner
e8842eb260 handle optional in flags that aren't present.
llvm-svn: 96731
2010-02-21 07:19:06 +00:00
Chris Lattner
4a9c3615f8 emit table indexes before each row so that it is debuggable.
llvm-svn: 96730
2010-02-21 07:16:41 +00:00
Chris Lattner
8a969fb824 fix two bugs in OPC_EmitRegister, which makes ppc happier.
llvm-svn: 96728
2010-02-21 06:58:27 +00:00
Chris Lattner
594803ba50 fix a table size miscomputation, target opcodes are 2 bytes.
With this, the matcher actually works reasonably well, but
crashes on larger examples in the scheduler.

llvm-svn: 96727
2010-02-21 06:44:29 +00:00
Chris Lattner
98d6a38236 emit to the right streams, to avoid emitting the push
body before the push.

llvm-svn: 96726
2010-02-21 06:30:04 +00:00
Chris Lattner
359c75b60c oops don't turn this on for everyone yet.
llvm-svn: 96725
2010-02-21 06:03:56 +00:00
Chris Lattner
0164635dad implement the last known missing feature: updating uses of results
of the matched pattern to use the newly created node results.  Onto
the "making it actually work" phase!

llvm-svn: 96724
2010-02-21 06:03:07 +00:00
Chris Lattner
b328c57aa4 fix and un-xfail X86/vec_ss_load_fold.ll
llvm-svn: 96720
2010-02-21 04:53:34 +00:00
Johnny Chen
abea29dc85 Undo r96654. The printing of ARM shift instructions in canonical forms can be
handled in ARMInstPrinter.cpp.

And added PLD/PLDW/PLI (Preload Data/Instruction) for disassembly only.

llvm-svn: 96719
2010-02-21 04:42:01 +00:00
Charles Davis
9dbf8d7899 Reduce size of 'StackAlignment' field from 5 to 3 bits. Seriously, who needs a
2GB-aligned stack anyway? 256 bytes is plenty. Requested by Chris.

llvm-svn: 96718
2010-02-21 04:26:06 +00:00
Chris Lattner
b28cf9f8d9 temporarily disable this.
llvm-svn: 96717
2010-02-21 03:24:41 +00:00
Chris Lattner
8983077c87 Lots of improvements to the new dagisel emitter. This gets it to
the point where it is to the 95% feature complete mark, it just
needs result updating to be done (then testing, optimization 
etc).

More specificallly, this adds support for chain and flag handling
on the result nodes, support for sdnodexforms, support for variadic
nodes, memrefs, pinned physreg inputs, and probably lots of other
stuff.

In the old DAGISelEmitter, this deletes the dead code related to
OperatorMap, cleans up a variety of dead stuff handling "implicit
remapping" from things like globaladdr -> targetglobaladdr (which
is no longer used because globaladdr always needs to be legalized),
and some minor formatting fixes.

llvm-svn: 96716
2010-02-21 03:22:59 +00:00
Chris Lattner
fe81834e6e rename SelectScalarSSELoad -> SelectScalarSSELoadXXX and rewrite
it to follow the mode needed by the new isel.  Instead of returning
the input and output chains, it just returns the (currently only one,
which is a silly limitation) node that has input and output chains.

Since we want the old thing to still work, add a new 
SelectScalarSSELoad to emulate the old interface.  The XXX suffix
and the wrapper will eventually go away.

llvm-svn: 96715
2010-02-21 03:17:59 +00:00
Chris Lattner
bb364c9b46 add a hook so that the new isel can run SDNodeXForms.
llvm-svn: 96714
2010-02-21 03:15:11 +00:00
Chris Lattner
ad9fa2075b Eliminate some uses of immAllOnes, just use -1, it does
the same thing and is more efficient for the matcher.

llvm-svn: 96713
2010-02-21 03:13:10 +00:00
Chris Lattner
ce7be2638a Eliminate some uses of immAllOnes, just use -1, it does
the same thing and is more efficient for the matcher.

llvm-svn: 96712
2010-02-21 03:12:16 +00:00
Tanya Lattner
8664bf0d0b Test commit.
llvm-svn: 96711
2010-02-21 03:01:05 +00:00
Tanya Lattner
24f73d5814 Test commit
llvm-svn: 96710
2010-02-21 02:59:05 +00:00
Dan Gohman
220c2212c1 Rename getSDiv to getExactSDiv to reflect its behavior in cases where
the division would have a remainder.

llvm-svn: 96693
2010-02-19 19:35:48 +00:00
Dan Gohman
9db0689627 Check for overflow when scaling up an add or an addrec for
scaled reuse.

llvm-svn: 96692
2010-02-19 19:32:49 +00:00
Bob Wilson
df432a30b2 Revert 96634. It causes assertion failures for 126.gcc and 176.gcc in
the armv6 nightly tests.

llvm-svn: 96691
2010-02-19 18:59:53 +00:00
Dan Gohman
39b7e5961e Add a comment.
llvm-svn: 96688
2010-02-19 18:49:22 +00:00
Charles Davis
a64fc8c41b Add support for the 'alignstack' attribute to the x86 backend. Fixes PR5254.
Also, FileCheck'ize a test.

llvm-svn: 96686
2010-02-19 18:17:13 +00:00
Dan Gohman
9268d67079 Teach ScalarEvolution how to compute a tripcount for a loop with
true or false as its exit condition. These are usually eliminated by
SimplifyCFG, but the may be left around during a pass which wishes
to preserve the CFG.

llvm-svn: 96683
2010-02-19 18:12:07 +00:00
Bob Wilson
c03fbf812d Revert Anton's most recent EH patch (r96637), since it breaks a lot of
ARM and Thumb tests.

llvm-svn: 96680
2010-02-19 17:10:59 +00:00
Duncan Sands
5d5cce2e19 Revert commits 96556 and 96640, because commit 96556 breaks the
dragonegg self-host build.  I reverted 96640 in order to revert
96556 (96640 goes on top of 96556), but it also looks like with
both of them applied the breakage happens even earlier.  The
symptom of the 96556 miscompile is the following crash:

  llvm[3]: Compiling AlphaISelLowering.cpp for Release build
  cc1plus: /home/duncan/tmp/tmp/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:4982: void llvm::SelectionDAG::ReplaceAllUsesWith(llvm::SDNode*, llvm::SDNode*, llvm::SelectionDAG::DAGUpdateListener*): Assertion `(!From->hasAnyUseOfValue(i) || From->getValueType(i) == To->getValueType(i)) && "Cannot use this version of ReplaceAllUsesWith!"' failed.
  Stack dump:
  0.	Running pass 'X86 DAG->DAG Instruction Selection' on function '@_ZN4llvm19AlphaTargetLowering14LowerOperationENS_7SDValueERNS_12SelectionDAGE'
  g++: Internal error: Aborted (program cc1plus)

This occurs when building LLVM using LLVM built by LLVM (via
dragonegg).  Probably LLVM has miscompiled itself, though it
may have miscompiled GCC and/or dragonegg itself: at this point
of the self-host build, all of GCC, LLVM and dragonegg were built
using LLVM.  Unfortunately this kind of thing is extremely hard
to debug, and while I did rummage around a bit I didn't find any
smoking guns, aka obviously miscompiled code.

Found by bisection.

r96556 | evancheng | 2010-02-18 03:13:50 +0100 (Thu, 18 Feb 2010) | 5 lines

Some dag combiner goodness:
Transform br (xor (x, y)) -> br (x != y)
Transform br (xor (xor (x,y), 1)) -> br (x == y)
Also normalize (and (X, 1) == / != 1 -> (and (X, 1)) != / == 0 to match to "test on x86" and "tst on arm"

r96640 | evancheng | 2010-02-19 01:34:39 +0100 (Fri, 19 Feb 2010) | 16 lines

Transform (xor (setcc), (setcc)) == / != 1 to
(xor (setcc), (setcc)) != / == 1.

e.g. On x86_64
  %0 = icmp eq i32 %x, 0
  %1 = icmp eq i32 %y, 0
  %2 = xor i1 %1, %0
  br i1 %2, label %bb, label %return
=>
	testl   %edi, %edi
	sete    %al
	testl   %esi, %esi
	sete    %cl
	cmpb    %al, %cl
	je      LBB1_2

llvm-svn: 96672
2010-02-19 11:30:41 +00:00
Duncan Sands
88c859c900 Correct LastPrimitiveTyID: MetadataType is a primitive type.
This change probably has no functional effect.

llvm-svn: 96669
2010-02-19 09:18:53 +00:00
Chris Lattner
e6302b3565 add emitter support for integer constants and simple physreg references.
llvm-svn: 96663
2010-02-19 07:49:56 +00:00
Dale Johannesen
0f58c73229 recommit 96626, evidence that it broke things appears
to be spurious

llvm-svn: 96662
2010-02-19 07:14:22 +00:00
Chris Lattner
80491f4c52 fix this to work more determinstically, patch by Thomas Veith!
llvm-svn: 96661
2010-02-19 07:02:20 +00:00
Devang Patel
1ea0277ca2 Test case for r96656.
llvm-svn: 96657
2010-02-19 02:58:33 +00:00
Johnny Chen
2a3db5ee3d Added entries for ASR, LSL, LSR, ROR, and RRX so that the disassembler prints
out the canonical form (A8.6.98) instead of the pseudo-instruction as provided
via MOVs.

DBG_ARM_DISASM=YES llvm-mc -triple=arm-unknown-unknown --disassemble
0xc0 0x00 0xa0 0xe1
Opcode=29 Name=ASR Format=ARM_FORMAT_LDMISCFRM
 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10  9  8  7  6  5  4  3  2  1  0 
-------------------------------------------------------------------------------------------------
| 1: 1: 1: 0| 0: 0: 0: 1| 1: 0: 1: 0| 0: 0: 0: 0| 0: 0: 0: 0| 0: 0: 0: 0| 1: 1: 0: 0| 0: 0: 0: 0|
-------------------------------------------------------------------------------------------------

	asr	r0, r0, #1

llvm-svn: 96654
2010-02-19 02:12:06 +00:00
Dale Johannesen
d73b81bf12 Revert 96626, which causes build failure on ppc Darwin.
llvm-svn: 96653
2010-02-19 01:54:37 +00:00
Evan Cheng
32031f7404 Transform (xor (setcc), (setcc)) == / != 1 to
(xor (setcc), (setcc)) != / == 1.

e.g. On x86_64
  %0 = icmp eq i32 %x, 0
  %1 = icmp eq i32 %y, 0
  %2 = xor i1 %1, %0
  br i1 %2, label %bb, label %return
=>
	testl   %edi, %edi
	sete    %al
	testl   %esi, %esi
	sete    %cl
	cmpb    %al, %cl
	je      LBB1_2

llvm-svn: 96640
2010-02-19 00:34:39 +00:00
Chris Lattner
28cc2f3f44 I confused myself, temporaries will be recorded right along with other inputs.
llvm-svn: 96639
2010-02-19 00:33:13 +00:00