platform/upstream/llvm.git
15 years agoMinor mod. per Doug's comment.
Fariborz Jahanian [Wed, 22 Jul 2009 00:42:46 +0000 (00:42 +0000)]
Minor mod. per Doug's comment.

llvm-svn: 76708

15 years agoset the ELF "small" flag on objects that end up in .rodata.cst4 consistently,
Chris Lattner [Wed, 22 Jul 2009 00:41:56 +0000 (00:41 +0000)]
set the ELF "small" flag on objects that end up in .rodata.cst4 consistently,
updating a mips testcase to expect it.

llvm-svn: 76707

15 years agodon't set the small flag yet.
Chris Lattner [Wed, 22 Jul 2009 00:30:39 +0000 (00:30 +0000)]
don't set the small flag yet.

llvm-svn: 76706

15 years agoremove the SelectSectionForMachineConst hook, replacing it with
Chris Lattner [Wed, 22 Jul 2009 00:28:43 +0000 (00:28 +0000)]
remove the SelectSectionForMachineConst hook, replacing it with
a new getSectionForMergableConstant hook.  This removes one dependence
of TAI on Type, and provides the hook with enough info to make the
right decision based on whether the global has relocations etc.

llvm-svn: 76705

15 years agoComplain if we're entering the context of a dependent nested-name-specifier but
Douglas Gregor [Wed, 22 Jul 2009 00:28:09 +0000 (00:28 +0000)]
Complain if we're entering the context of a dependent nested-name-specifier but
cannot match that nested-name-specifier to a class template or class template
partial specialization.

llvm-svn: 76704

15 years agoLet each target determines whether a machine instruction is dead. If true, that allow...
Evan Cheng [Wed, 22 Jul 2009 00:25:27 +0000 (00:25 +0000)]
Let each target determines whether a machine instruction is dead. If true, that allows late codeine passes to delete it.

This is considered a workaround. The problem is some targets are not modeling side effects correctly. PPC is apparently one of those. This patch allows ppc llvm-gcc to bootstrap on Darwin. Once we find out which instruction definitions are wrong, we can remove the PPCInstrInfo workaround.

llvm-svn: 76703

15 years agoGet rid of the Pass+Context magic.
Owen Anderson [Wed, 22 Jul 2009 00:24:57 +0000 (00:24 +0000)]
Get rid of the Pass+Context magic.

llvm-svn: 76702

15 years agoreimplement Constant::ContainsRelocations as
Chris Lattner [Wed, 22 Jul 2009 00:05:44 +0000 (00:05 +0000)]
reimplement Constant::ContainsRelocations as
Constant::getRelocationInfo(), which has a much simpler
to use API.  It still should not be part of libvmcore, but
is better than it was.  Also teach it to be smart about
hidden visibility.

llvm-svn: 76700

15 years agoMisc. doc fixes following suggestions from Eli.
Dan Gohman [Wed, 22 Jul 2009 00:04:19 +0000 (00:04 +0000)]
Misc. doc fixes following suggestions from Eli.

llvm-svn: 76699

15 years agoFixing cp island pass. Step 1: Determine whether the constant pool offset can be
Evan Cheng [Tue, 21 Jul 2009 23:56:01 +0000 (23:56 +0000)]
Fixing cp island pass. Step 1: Determine whether the constant pool offset can be
negative on an individual bases rather than basing on whether it's in thumb
mode.

llvm-svn: 76698

15 years agoFix comment.
Evan Cheng [Tue, 21 Jul 2009 23:54:22 +0000 (23:54 +0000)]
Fix comment.

llvm-svn: 76693

15 years agoBasic parsing and semantic analysis for out-of-line definitions of the
Douglas Gregor [Tue, 21 Jul 2009 23:53:31 +0000 (23:53 +0000)]
Basic parsing and semantic analysis for out-of-line definitions of the
member functions of class templates, e.g.,

  template<typename T>
  struct X {
    void f(T);
  };

  template<typename T> X<T>::f(T) { /* ... */ }

llvm-svn: 76692

15 years agoRevert this, we can now avoid error cascades better.
Mike Stump [Tue, 21 Jul 2009 23:52:50 +0000 (23:52 +0000)]
Revert this, we can now avoid error cascades better.

llvm-svn: 76691

15 years agoRevert this, was a bug in my new warning code, not the test case.
Mike Stump [Tue, 21 Jul 2009 23:50:15 +0000 (23:50 +0000)]
Revert this, was a bug in my new warning code, not the test case.

llvm-svn: 76690

15 years agosimplify code now that it is inlined.
Chris Lattner [Tue, 21 Jul 2009 23:49:55 +0000 (23:49 +0000)]
simplify code now that it is inlined.

llvm-svn: 76689

15 years agoExposed PHIElimination pass within CodeGen.
Lang Hames [Tue, 21 Jul 2009 23:47:33 +0000 (23:47 +0000)]
Exposed PHIElimination pass within CodeGen.

llvm-svn: 76688

15 years agoRevert this, we have a better way to do this.
Mike Stump [Tue, 21 Jul 2009 23:47:12 +0000 (23:47 +0000)]
Revert this, we have a better way to do this.

llvm-svn: 76687

15 years agoNow that RelocBehaviour() is never overloaded, it doesn't need to be
Chris Lattner [Tue, 21 Jul 2009 23:47:11 +0000 (23:47 +0000)]
Now that RelocBehaviour() is never overloaded, it doesn't need to be
virtual.  Just inline it into its two current call sites in preparation
for simplifying the code.

llvm-svn: 76686

15 years agoRevert this, we have a better way to handle this.
Mike Stump [Tue, 21 Jul 2009 23:45:39 +0000 (23:45 +0000)]
Revert this, we have a better way to handle this.

llvm-svn: 76685

15 years agoAdd intrinsic for __builtin_ia32_vec_set_v4si which turns out to be
Eric Christopher [Tue, 21 Jul 2009 23:43:14 +0000 (23:43 +0000)]
Add intrinsic for __builtin_ia32_vec_set_v4si which turns out to be
pinsrd in sse4.1.

llvm-svn: 76684

15 years agothis doesn't break any of the 4 ia64 tests.
Chris Lattner [Tue, 21 Jul 2009 23:42:24 +0000 (23:42 +0000)]
this doesn't break any of the 4 ia64 tests.

llvm-svn: 76683

15 years agoalpha doesn't need to redefine this: it only supports PIC codegen anyway.
Chris Lattner [Tue, 21 Jul 2009 23:41:35 +0000 (23:41 +0000)]
alpha doesn't need to redefine this: it only supports PIC codegen anyway.

llvm-svn: 76682

15 years agoRevert recent change, I now have a better way to solve this (thanks Chris).
Mike Stump [Tue, 21 Jul 2009 23:41:23 +0000 (23:41 +0000)]
Revert recent change, I now have a better way to solve this (thanks Chris).

llvm-svn: 76681

15 years agoAdd some support for iterative coalescers to calculate a joined live
David Greene [Tue, 21 Jul 2009 23:36:14 +0000 (23:36 +0000)]
Add some support for iterative coalescers to calculate a joined live
range's weight properly.  This is turned off right now in the sense that
you'll get an assert if you get into a situation that can only be caused
by an iterative coalescer.  All other code paths operate exactly as
before so there is no functional change with this patch.  The asserts
should be disabled if/when an iterative coalescer gets added to trunk.

llvm-svn: 76680

15 years agono really, I can spell!
Chris Lattner [Tue, 21 Jul 2009 23:36:01 +0000 (23:36 +0000)]
no really, I can spell!

llvm-svn: 76679

15 years agoadd an API so target-independent codegen can determine if a constant
Chris Lattner [Tue, 21 Jul 2009 23:34:23 +0000 (23:34 +0000)]
add an API so target-independent codegen can determine if a constant
pool entry will require relocations against it.  I implemented this
conservatively for ARM, someone who is knowledgable about it should
see if this can be improved.

llvm-svn: 76678

15 years agoFix long-lurking bug in ObjCSummaryCache revealed by Torok's recent memory
Ted Kremenek [Tue, 21 Jul 2009 23:27:57 +0000 (23:27 +0000)]
Fix long-lurking bug in ObjCSummaryCache revealed by Torok's recent memory
poisoning changes to DenseMap. We were using an iterator after it had been
invalidated by an insertion into the DenseMap.

llvm-svn: 76677

15 years agoConvert instcombine from using using getAnalysis<TargetData> to
Dan Gohman [Tue, 21 Jul 2009 23:21:54 +0000 (23:21 +0000)]
Convert instcombine from using using getAnalysis<TargetData> to
getAnalysisIfAvailable<TargetData>.

llvm-svn: 76676

15 years agoPermit the IntPtrTy argument to isEliminableCastPair to be null,
Dan Gohman [Tue, 21 Jul 2009 23:19:40 +0000 (23:19 +0000)]
Permit the IntPtrTy argument to isEliminableCastPair to be null,
to help support use when TargetData is not available.

llvm-svn: 76675

15 years agofix typo
Chris Lattner [Tue, 21 Jul 2009 23:17:26 +0000 (23:17 +0000)]
fix typo

llvm-svn: 76674

15 years agoChange ELFCodeEmitter logic to emit the constant pool and jump tables to
Bruno Cardoso Lopes [Tue, 21 Jul 2009 23:13:26 +0000 (23:13 +0000)]
Change ELFCodeEmitter logic to emit the constant pool and jump tables to
their appropriate sections before the code itself. They need to be emitted
before the function because on some targets (x86 but not x86_64) the later
may reference a JT or CP entry address

llvm-svn: 76672

15 years agoReplace the original ad-hoc code for determining whether (v pred w) implies
Dan Gohman [Tue, 21 Jul 2009 23:03:19 +0000 (23:03 +0000)]
Replace the original ad-hoc code for determining whether (v pred w) implies
(x pred y) with more thorough code that does more complete canonicalization
before resorting to range checks. This helps it find more cases where
the canonicalized expressions match.

llvm-svn: 76671

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 22:54:02 +0000 (22:54 +0000)]
Prep for new warning.

llvm-svn: 76670

15 years agoRemove a big test case.
Evan Cheng [Tue, 21 Jul 2009 22:52:04 +0000 (22:52 +0000)]
Remove a big test case.

llvm-svn: 76669

15 years agoVarious doc updates from Edward O'Callaghan!
Chris Lattner [Tue, 21 Jul 2009 22:47:03 +0000 (22:47 +0000)]
Various doc updates from Edward O'Callaghan!

llvm-svn: 76668

15 years agoremove the last bits of SectionFlagsForGlobal. There is some flag here that
Chris Lattner [Tue, 21 Jul 2009 22:42:37 +0000 (22:42 +0000)]
remove the last bits of SectionFlagsForGlobal.  There is some flag here that
depends on XS1A, but I think the ReadOnlySection is already set up for this
and there is no testcase that this breaks.  If this is really needed, we can
add the appropriate parameterization to TargetAsmInfo in the future to support
this.

llvm-svn: 76667

15 years agodon't mask out the small flag and then reapply it later.
Chris Lattner [Tue, 21 Jul 2009 22:39:28 +0000 (22:39 +0000)]
don't mask out the small flag and then reapply it later.

llvm-svn: 76666

15 years agoif Xcore doesn't support TLS, it doesn't have to worry about thread local LLVM IR...
Chris Lattner [Tue, 21 Jul 2009 22:37:38 +0000 (22:37 +0000)]
if Xcore doesn't support TLS, it doesn't have to worry about thread local LLVM IR, it should be rejected by a front-end.

llvm-svn: 76665

15 years agoremove the Xcore implementation of SelectSectionForGlobal. While you have
Chris Lattner [Tue, 21 Jul 2009 22:36:53 +0000 (22:36 +0000)]
remove the Xcore implementation of SelectSectionForGlobal.  While you have
to twist your brain to see it, I believe it is the same as ELFTargetAsmInfo::SelectSectionForGlobal.

llvm-svn: 76664

15 years agoPatch to accomodate Doug's comment on default
Fariborz Jahanian [Tue, 21 Jul 2009 22:36:06 +0000 (22:36 +0000)]
Patch to accomodate Doug's comment on default
destruction of base/members for each destructor AST.

llvm-svn: 76663

15 years agosimplify based on the fact that darwin always uses L/l.
Chris Lattner [Tue, 21 Jul 2009 22:32:55 +0000 (22:32 +0000)]
simplify based on the fact that darwin always uses L/l.

llvm-svn: 76662

15 years agomake some stuff private.
Chris Lattner [Tue, 21 Jul 2009 22:30:39 +0000 (22:30 +0000)]
make some stuff private.

llvm-svn: 76661

15 years agoimprove comments.
Chris Lattner [Tue, 21 Jul 2009 22:30:25 +0000 (22:30 +0000)]
improve comments.

llvm-svn: 76660

15 years agoRemove the XCore custom implementation of MergeableConstSection, relying on
Chris Lattner [Tue, 21 Jul 2009 22:25:52 +0000 (22:25 +0000)]
Remove the XCore custom implementation of MergeableConstSection, relying on
the generic ELF version instead.  This will result in its mergable constant
sections getting named ".rodata.cst4" instead of ".cp.const4", but the
linker looks at the section flags, not the name of the section AFAICT.

llvm-svn: 76659

15 years agoFix ocaml tests for 64-bit MacOS systems. LLVM is currently built
Bob Wilson [Tue, 21 Jul 2009 21:56:46 +0000 (21:56 +0000)]
Fix ocaml tests for 64-bit MacOS systems.  LLVM is currently built
as 32-bit code by default, and if gcc defaults to 64-bit code then ocamlc
requires a -cc "gcc -arch i386" option.  We were hardcoding -cc g++
and throwing away any other compiler options that were determined when
ocamlc was configured and built.

llvm-svn: 76658

15 years agoReorder the "Metadata" entry to match the C bindings.
Bob Wilson [Tue, 21 Jul 2009 21:52:57 +0000 (21:52 +0000)]
Reorder the "Metadata" entry to match the C bindings.
This fixes a regression in the vmcore.ml dejagnu test.

llvm-svn: 76657

15 years agoRemove stale comment and fix RUN line.
Ted Kremenek [Tue, 21 Jul 2009 21:48:25 +0000 (21:48 +0000)]
Remove stale comment and fix RUN line.

llvm-svn: 76656

15 years agoAdd missing include
Daniel Dunbar [Tue, 21 Jul 2009 21:33:58 +0000 (21:33 +0000)]
Add missing include

llvm-svn: 76655

15 years agoinline a trivial method into its only call site and fix indentation of cases
Chris Lattner [Tue, 21 Jul 2009 21:29:08 +0000 (21:29 +0000)]
inline a trivial method into its only call site and fix indentation of cases

llvm-svn: 76654

15 years agoRemove some overridden functions in XCoreTargetAsmInfo that are
Chris Lattner [Tue, 21 Jul 2009 21:26:32 +0000 (21:26 +0000)]
Remove some overridden functions in XCoreTargetAsmInfo that are
implemented exactly the same way as its ELFTargetAsmInfo subclass
has them.

llvm-svn: 76653

15 years agoAdd test case for PR 4596, which is already fixed due to Steve Naroff's overhaul...
Ted Kremenek [Tue, 21 Jul 2009 21:21:04 +0000 (21:21 +0000)]
Add test case for PR 4596, which is already fixed due to Steve Naroff's overhaul of the Objective-C type system, but isn't in a checker build yet.

llvm-svn: 76648

15 years agoPatch by Stefan Bühler: Escape text in macro expansion when emitting HTML in the...
Ted Kremenek [Tue, 21 Jul 2009 21:16:46 +0000 (21:16 +0000)]
Patch by Stefan Bühler: Escape text in macro expansion when emitting HTML in the HTMLRewriter.

This fixes PR 4602.

llvm-svn: 76647

15 years agorevert r76602, 76603, and r76615, pending design discussions.
Chris Lattner [Tue, 21 Jul 2009 21:12:58 +0000 (21:12 +0000)]
revert r76602, 76603, and r76615, pending design discussions.

llvm-svn: 76646

15 years agominor cleanups.
Chris Lattner [Tue, 21 Jul 2009 21:09:35 +0000 (21:09 +0000)]
minor cleanups.

llvm-svn: 76645

15 years agoUpdate stale comment.
Ted Kremenek [Tue, 21 Jul 2009 21:09:31 +0000 (21:09 +0000)]
Update stale comment.

llvm-svn: 76644

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 21:06:08 +0000 (21:06 +0000)]
Prep for new warning.

llvm-svn: 76643

15 years agoReorder files.
Ted Kremenek [Tue, 21 Jul 2009 21:03:50 +0000 (21:03 +0000)]
Reorder files.

llvm-svn: 76642

15 years agoFix PR 4594 by refactoring almost all casting logic from GRExprEngine::VisitCast
Ted Kremenek [Tue, 21 Jul 2009 21:03:30 +0000 (21:03 +0000)]
Fix PR 4594 by refactoring almost all casting logic from GRExprEngine::VisitCast
to SValuator::EvalCast. In the process, the StoreManagers now use this new cast
machinery, and the hack in GRExprEngine::EvalBind to handle implicit casts
involving OSAtomicCompareAndSwap and friends has been removed (and replaced with
logic closer to the logic specific to those functions).

llvm-svn: 76641

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 21:03:09 +0000 (21:03 +0000)]
Prep for new warning.

llvm-svn: 76640

15 years agoPrivatize the ConstantArray table.
Owen Anderson [Tue, 21 Jul 2009 20:55:28 +0000 (20:55 +0000)]
Privatize the ConstantArray table.

llvm-svn: 76639

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 20:52:43 +0000 (20:52 +0000)]
Prep for new warning.

llvm-svn: 76638

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 20:50:41 +0000 (20:50 +0000)]
Prep for new warning.

llvm-svn: 76637

15 years agoAdd a few fairly obvious API changes I noticed while porting some old code.
Torok Edwin [Tue, 21 Jul 2009 20:27:10 +0000 (20:27 +0000)]
Add a few fairly obvious API changes I noticed while porting some old code.

llvm-svn: 76636

15 years agoMissed a piece of the commit to remove the shift flavor.
Eli Friedman [Tue, 21 Jul 2009 20:15:24 +0000 (20:15 +0000)]
Missed a piece of the commit to remove the shift flavor.

llvm-svn: 76635

15 years agoPrivatize the first of the value maps.
Owen Anderson [Tue, 21 Jul 2009 20:13:12 +0000 (20:13 +0000)]
Privatize the first of the value maps.

llvm-svn: 76634

15 years agoRemove shift amount flavor. It isn't actually complete enough to
Eli Friedman [Tue, 21 Jul 2009 20:12:16 +0000 (20:12 +0000)]
Remove shift amount flavor.  It isn't actually complete enough to
be useful, and it's currently unused.  (Some issues: it isn't actually
rich enough to capture the semantics on many architectures, and
semantics can vary depending on the type being shifted.)

llvm-svn: 76633

15 years agodriver: Print --version on stdout, to match gcc.
Daniel Dunbar [Tue, 21 Jul 2009 20:06:58 +0000 (20:06 +0000)]
driver: Print --version on stdout, to match gcc.
 - Patch by Jean-Daniel Dupas

llvm-svn: 76632

15 years agoRemove a couple of already-implemented notes.
Eli Friedman [Tue, 21 Jul 2009 20:05:43 +0000 (20:05 +0000)]
Remove a couple of already-implemented notes.

llvm-svn: 76631

15 years agoMisc fixes for -Wreorder:
Eli Friedman [Tue, 21 Jul 2009 19:28:10 +0000 (19:28 +0000)]
Misc fixes for -Wreorder:

1. Make it work correctly with anonymous unions.
2. Don't compute it if the warning isn't enabled.
3. Optimize the algorithm slightly to make it linear time in the
case where we don't produce any warnings.

llvm-svn: 76630

15 years ago80 col violation.
Evan Cheng [Tue, 21 Jul 2009 19:25:09 +0000 (19:25 +0000)]
80 col violation.

llvm-svn: 76629

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:03:43 +0000 (19:03 +0000)]
Prep for new warning.

llvm-svn: 76628

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:02:55 +0000 (19:02 +0000)]
Prep for new warning.

llvm-svn: 76627

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:01:48 +0000 (19:01 +0000)]
Prep for new warning.

llvm-svn: 76626

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:01:31 +0000 (19:01 +0000)]
Prep for new warning.

llvm-svn: 76625

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:01:01 +0000 (19:01 +0000)]
Prep for new warning.

llvm-svn: 76624

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 19:00:12 +0000 (19:00 +0000)]
Prep for new warning.

llvm-svn: 76623

15 years agoMake Sema::ActOnCXXEnterDeclaratorScope robust against failures to compute
Douglas Gregor [Tue, 21 Jul 2009 18:59:28 +0000 (18:59 +0000)]
Make Sema::ActOnCXXEnterDeclaratorScope robust against failures to compute
the declaration context, as occurs with out-of-line class template member
definitions.

llvm-svn: 76622

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:59:16 +0000 (18:59 +0000)]
Prep for new warning.

llvm-svn: 76621

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:58:50 +0000 (18:58 +0000)]
Prep for new warning.

llvm-svn: 76620

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:58:15 +0000 (18:58 +0000)]
Prep for new warning.

llvm-svn: 76619

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:57:14 +0000 (18:57 +0000)]
Prep for new warning.

llvm-svn: 76617

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:56:34 +0000 (18:56 +0000)]
Prep for new warning.

llvm-svn: 76616

15 years agoPrefix IR dumps with LiveInterval indices when possible. This turns
David Greene [Tue, 21 Jul 2009 18:56:32 +0000 (18:56 +0000)]
Prefix IR dumps with LiveInterval indices when possible.  This turns
this:

%ESI<def> = MOV32rr %EDI<kill>
ADJCALLSTACKDOWN64 0, %RSP<imp-def>, %EFLAGS<imp-def,dead>, %RSP<imp-use>
%reg1027<def> = MOVZX64rr32 %ESI
%reg1027<def> = ADD64ri8 %reg1027, 15, %EFLAGS<imp-def,dead>
%reg1027<def> = AND64ri8 %reg1027, -16, %EFLAGS<imp-def,dead>
%RDI<def> = MOV64rr %RSP
%RDI<def> = SUB64rr %RDI, %reg1027<kill>, %EFLAGS<imp-def,dead>
%RSP<def> = MOV64rr %RDI

into this:

4 %reg1024<def> = MOV32rr %EDI<kill>
12 ADJCALLSTACKDOWN64 0, %RSP<imp-def>, %EFLAGS<imp-def,dead>, %RSP<imp-use>
20 %reg1025<def> = MOVZX64rr32 %reg1024
28 %reg1026<def> = MOV64rr %reg1025<kill>
36 %reg1026<def> = ADD64ri8 %reg1026, 15, %EFLAGS<imp-def,dead>
44 %reg1027<def> = MOV64rr %reg1026<kill>
52 %reg1027<def> = AND64ri8 %reg1027, -16, %EFLAGS<imp-def,dead>
60 %reg1028<def> = MOV64rr %RSP
68 %reg1029<def> = MOV64rr %reg1028<kill>
76 %reg1029<def> = SUB64rr %reg1029, %reg1027<kill>, %EFLAGS<imp-def,dead>
84 %RSP<def> = MOV64rr %reg1029

This helps greatly when debugging register allocation and coalescing
problems.

llvm-svn: 76615

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:56:04 +0000 (18:56 +0000)]
Prep for new warning.

llvm-svn: 76614

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:54:29 +0000 (18:54 +0000)]
Prep for new warning.

llvm-svn: 76613

15 years agoAdd fake v7 itineraries for now.
Evan Cheng [Tue, 21 Jul 2009 18:54:14 +0000 (18:54 +0000)]
Add fake v7 itineraries for now.

llvm-svn: 76612

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:52:41 +0000 (18:52 +0000)]
Prep for new warning.

llvm-svn: 76611

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:51:31 +0000 (18:51 +0000)]
Prep for new warning.

llvm-svn: 76610

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:48:43 +0000 (18:48 +0000)]
Prep for new warning.

llvm-svn: 76609

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:46:15 +0000 (18:46 +0000)]
Prep for new warning.

llvm-svn: 76608

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:45:53 +0000 (18:45 +0000)]
Prep for new warning.

llvm-svn: 76607

15 years agoPrep for new warning.
Mike Stump [Tue, 21 Jul 2009 18:45:22 +0000 (18:45 +0000)]
Prep for new warning.

llvm-svn: 76606

15 years agoPrep for new warnings about control flow falling off the ends of
Mike Stump [Tue, 21 Jul 2009 18:44:24 +0000 (18:44 +0000)]
Prep for new warnings about control flow falling off the ends of
functions that return a value.  I was going to buffer the whole lot
up, but it should be easier to review if I check them in
incrementally.  Most of the forth coming changes either add a return
value, or make it impossible to return, or alter the return type.

llvm-svn: 76605

15 years agomake AsmPrinter::doFinalization iterate over the global variables
Chris Lattner [Tue, 21 Jul 2009 18:38:57 +0000 (18:38 +0000)]
make AsmPrinter::doFinalization iterate over the global variables
and call PrintGlobalVariable, allowing elimination and simplification
of various targets.

llvm-svn: 76604

15 years agoAdd PrefixPrinter arguments to the dump routines for MachineFunction and
David Greene [Tue, 21 Jul 2009 18:27:05 +0000 (18:27 +0000)]
Add PrefixPrinter arguments to the dump routines for MachineFunction and
MachineBasicBlock.  We'll use these shortly.

llvm-svn: 76603

15 years agoAdd a small utility class to configure IR printers. This will allow
David Greene [Tue, 21 Jul 2009 18:21:46 +0000 (18:21 +0000)]
Add a small utility class to configure IR printers.  This will allow
printers to do neat and wonderful things when printing debug
information.  The ideas is to allow passes to configer printers to emit
pass-specific information when dumping IR.

llvm-svn: 76602

15 years agoMake a bunch of changes suggested by Chris and others to improve
David Greene [Tue, 21 Jul 2009 18:16:24 +0000 (18:16 +0000)]
Make a bunch of changes suggested by Chris and others to improve
efficiency.  Fix a few formatting bugs along the way.

llvm-svn: 76601

15 years agoDo not select tSXTB / tSXTH in thumb2 mode.
Evan Cheng [Tue, 21 Jul 2009 18:15:26 +0000 (18:15 +0000)]
Do not select tSXTB / tSXTH in thumb2 mode.

llvm-svn: 76600

15 years agoUpdate for LLVM API change.
Owen Anderson [Tue, 21 Jul 2009 18:06:41 +0000 (18:06 +0000)]
Update for LLVM API change.

llvm-svn: 76599

15 years agoRename getConstantInt{True|False} to get{True|False} at Chris' behest.
Owen Anderson [Tue, 21 Jul 2009 18:03:38 +0000 (18:03 +0000)]
Rename getConstantInt{True|False} to get{True|False} at Chris' behest.

llvm-svn: 76598