Commit Graph

59762 Commits

Author SHA1 Message Date
Dan Gohman 4d273f4519 Commit this, which should have accompanied 101531.
llvm-svn: 101532
2010-04-16 20:22:43 +00:00
Dan Gohman 148c69a3f6 Eliminate an unnecessary SelectionDAG dependency in getOptimalMemOpType.
llvm-svn: 101531
2010-04-16 20:11:05 +00:00
Dan Gohman 9dbe61456a Fix this code to avoid implicit assumptions about the length of the array.
llvm-svn: 101530
2010-04-16 20:08:45 +00:00
Johnny Chen 2b7aba10c2 In the same spirit of r101524, which removed the assert() from printAddrMode2OffsetOperand(),
this patch removes the assert() from printAddrMode3OffsetOperand() and adds a test case.

llvm-svn: 101529
2010-04-16 19:57:21 +00:00
Dan Gohman 0591a050ff Revert 101520, which depended on 101503, which was reverted.
llvm-svn: 101528
2010-04-16 19:36:41 +00:00
Johnny Chen 807e1748fc Multiclass LdStCop was using pre-UAL syntax LDC<c>L for the L fragment. Changed
to the UAL syntax of LDCL<c>, instead.

Add a test case for this change which also tests the removal of assert() from
printAddrMode2OffsetOperand().

llvm-svn: 101527
2010-04-16 19:33:23 +00:00
Duncan Sands 7fc8123b37 Revert commit 101503 (johannes), in the hope of fixing the dragonegg build,
see http://google1.osuosl.org:8011/builders/dragonegg-x86_64-linux/builds/693
Original commit text:
Use a ValueMap not a std::map for the reason indicated
in the comment.  This was causing nondeterministic changes
in inlining decisions.

llvm-svn: 101525
2010-04-16 19:28:59 +00:00
Johnny Chen 88599a42bb Remove the assert() from printAddrMode2OffsetOperand(). "#0 and #-0" are
considered legal instructions.

Refs: A8.6.51 LDC, LDC2 (immediate) -- page A8-107, A8.6.58 LDR (immediate, ARM)
-- page A8-121, and A8.6.194 STR (immediate, ARM) -- page A8-395.

llvm-svn: 101524
2010-04-16 19:10:52 +00:00
Dan Gohman 076a3278a1 Trim a #include.
llvm-svn: 101520
2010-04-16 18:45:11 +00:00
Dan Gohman c1ce91603c Revert r101455, which fails on the llvm-arm-linux buildbot.
llvm-svn: 101515
2010-04-16 18:37:31 +00:00
Dale Johannesen 3104bf1618 Use a ValueMap not a std::map for the reason indicated
in the comment.  This was causing nondeterministic changes
in inlining decisions.

llvm-svn: 101503
2010-04-16 18:16:08 +00:00
Evan Cheng 954bd598dd 80 col.
llvm-svn: 101501
2010-04-16 17:58:41 +00:00
Evan Cheng d6b0a7c075 80 col.
llvm-svn: 101500
2010-04-16 17:57:59 +00:00
Dan Gohman 3a7ee8eead Avoid creating virtual registers for unused values.
llvm-svn: 101480
2010-04-16 17:15:02 +00:00
Dan Gohman 5664b9f1a9 Fix an assertion string.
llvm-svn: 101478
2010-04-16 16:55:18 +00:00
Dan Gohman 4572a9f479 Fix a comment.
llvm-svn: 101477
2010-04-16 16:52:37 +00:00
Dan Gohman f13f69f296 Disable inlining of recursive calls. It can complicate tailcallelim and
dependent analyses, and increase code size, so doing it profitably would
require more complex heuristics.

llvm-svn: 101471
2010-04-16 16:01:18 +00:00
Dan Gohman 99e5327bfd Refine the detection of seemingly infinitely recursive calls where the
callee is expected to be expanded to something else by codegen, so that
normal infinitely recursive calls are still transformed.

llvm-svn: 101468
2010-04-16 15:57:50 +00:00
Gabor Greif f375520f7b reapply r101434
with a fix for self-hosting

rotate CallInst operands, i.e. move callee to the back
of the operand array

the motivation for this patch are laid out in my mail to llvm-commits:
more efficient access to operands and callee, faster callgraph-construction,
smaller compiler binary

llvm-svn: 101465
2010-04-16 15:33:14 +00:00
Dan Gohman b3862ecd48 Make callIsSmall accessible as a utility function.
llvm-svn: 101463
2010-04-16 15:14:50 +00:00
Dan Gohman 12293815de Fix SCEVCommutativeExpr::print to be robust in the case of improper
expression canonicalization. Its job is to print what's there, not to
make judgements about it.

llvm-svn: 101461
2010-04-16 15:03:25 +00:00
Dan Gohman aa4d9ea465 Delete a blank line.
llvm-svn: 101459
2010-04-16 13:32:55 +00:00
Bill Wendling ae4541d758 Add JIT exception handling test.
llvm-svn: 101455
2010-04-16 09:04:28 +00:00
Bill Wendling 929f3c0927 The JIT calls TidyLandingPads to tidy up the landing pads. However, because the
JIT doesn't use the MC back-end asm printer to emit labels that it uses, the
section for the MCSymbol is never set. And thus the MCSymbol for the EH label
isn't marked as "defined". Because of that, TidyLandingPads removes the needed
landing pads from the JIT output. This breaks EH for every JIT program.

This is a work-around for this limitation. We pass in the label locations
map. If the label has a non-zero value, then it was "emitted" by the JIT and
TidyLandingPads shouldn't remove that label.

A nicer solution would be to mark the MCSymbol as "used" by the JIT and not rely
upon the section being set to determine if it's defined or not.

llvm-svn: 101453
2010-04-16 08:46:10 +00:00
Chris Lattner 408f79197c completed my pass over all 6+ months of commits, next step, format and make comprehendable.
llvm-svn: 101449
2010-04-16 06:20:22 +00:00
Evan Cheng af56facacd Adding support for dag combiner to promote operations for profit. This requires target specific queries. For example, x86 should promote i16 to i32 when it does not impact load folding.
x86 support is off by default. It can be enabled with -promote-16bit.

Work in progress.

llvm-svn: 101448
2010-04-16 06:14:10 +00:00
Evan Cheng 3da64f7672 Use getAL() rather than a major constant.
llvm-svn: 101446
2010-04-16 05:46:06 +00:00
Eric Christopher dea3bda860 Add a check for posix_spawn.
Regenerate configure and other autoconf files.

llvm-svn: 101444
2010-04-16 05:14:21 +00:00
Dan Gohman 5563473062 Refine further the scope where the global DebugLoc value is active.
llvm-svn: 101443
2010-04-16 05:06:56 +00:00
Nick Lewycky bb2ec7de9c Revert r64616 which worked around http://gcc.gnu.org/PR42757 , we just didn't
know it at the time.

llvm-svn: 101439
2010-04-16 04:32:20 +00:00
Eric Christopher 206d7ce582 Silence an unused variable warning.
llvm-svn: 101438
2010-04-16 04:02:04 +00:00
Chris Lattner bd2d9430d6 fix comment noticed by Bob
llvm-svn: 101437
2010-04-16 02:32:17 +00:00
Gabor Greif 403e9694f9 back out r101423 and r101397, they break llvm-gcc self-host on darwin10
llvm-svn: 101434
2010-04-16 01:16:20 +00:00
Chris Lattner 393e08536d move comment.
llvm-svn: 101433
2010-04-16 01:05:52 +00:00
Chris Lattner 1146d326a7 fix PR6832: we were using the alignment of a pointer when we
wanted the alignment of the pointee.

llvm-svn: 101432
2010-04-16 01:05:38 +00:00
Dan Gohman 0977358509 Create an exports file, so that the plugin only exports the onload symbol.
llvm-svn: 101431
2010-04-16 00:43:25 +00:00
Dan Gohman ebb4ae0912 Make things static that don't need to be referenced from outside the file.
llvm-svn: 101430
2010-04-16 00:42:57 +00:00
Chris Lattner b73552908e improve comments.
llvm-svn: 101429
2010-04-16 00:38:19 +00:00
Chris Lattner 78d7dbbc30 pull all the ConvertToScalarInfo code together into one
place.

llvm-svn: 101427
2010-04-16 00:24:57 +00:00
Chris Lattner d69c3ee958 more refactoring: suck some stuff out of SRoA into
ConvertToScalarInfo.

llvm-svn: 101425
2010-04-16 00:20:00 +00:00
Gabor Greif 6af0ad846e shift intrinsic operand
llvm-svn: 101423
2010-04-16 00:06:45 +00:00
Chris Lattner 9ef4eae6e6 introduce a new ConvertToScalarInfo struct to simplify
CanConvertToScalar/MergeInType.  Eliminate a pointless
LLVMContext argument to MergeInType.

llvm-svn: 101422
2010-04-15 23:50:26 +00:00
Jakob Stoklund Olesen cdc3df4888 Avoid sinking machine instructions into a loop.
MachineLoopInfo is already available when MachineSinking runs, so the check is
free.

There is no test case because it would require a critical edge into a loop, and
CodeGenPrepare splits those. This check is just to be extra careful.

llvm-svn: 101420
2010-04-15 23:41:02 +00:00
Johnny Chen 1d3ee607b3 Added another test case for am3offset operand, testing Rn, #+/-imm8.
Previous checkin tested Rn, #+/-Rm.

llvm-svn: 101418
2010-04-15 23:23:40 +00:00
Jakob Stoklund Olesen dc6d42dbf8 Add test case for machine-sink on critical edges
llvm-svn: 101416
2010-04-15 23:19:16 +00:00
Johnny Chen acbc06c2a3 Fixed a bug in ARM disassembly where LDRSBT should have am3offset operand, not
am2offset.  Modified the instruction table entry and added a new test case.

llvm-svn: 101415
2010-04-15 23:12:47 +00:00
Dan Gohman ff9ecb2824 Make the export files absolute paths, and change Makefile.rules
to expect them this way, to fix srcdir!=objdir builds.

llvm-svn: 101414
2010-04-15 23:08:00 +00:00
Dan Gohman 1639265421 Add a dependency on the .dir file to make sure that the directory
is created before the native exports file is built in a parallel build.

llvm-svn: 101413
2010-04-15 22:46:27 +00:00
Evan Cheng f7f97b4bbd Use default lowering of DYNAMIC_STACKALLOC. As far as I can tell, ARM isle is doing the right thing and codegen looks correct for both Thumb and Thumb2.
llvm-svn: 101410
2010-04-15 22:20:34 +00:00
Chris Lattner 9c1172d848 tidy interface to isOnlyCopiedFromConstantGlobal
llvm-svn: 101405
2010-04-15 21:59:20 +00:00