Duncan Sands [Mon, 10 Nov 2008 17:29:56 +0000 (17:29 +0000)]
Small cleanups. No functionality change intended!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58992
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 10 Nov 2008 17:28:30 +0000 (17:28 +0000)]
When promoting the result of fp_to_uint/fp_to_sint,
inform the optimizers that the result must be zero/
sign extended from the smaller type. For example,
if a fp to unsigned i16 is promoted to fp to i32,
then we are allowed to assume that the extra 16 bits
are zero (because the result of fp to i16 is undefined
if the result does not fit in an i16). This is
quite aggressive, but should help the optimizers
produce better code. This requires correcting a
test which thought that fp_to_uint is some kind
of truncation, which it is not: in the testcase
(which does fp to i1), either the fp value converts
to 0 or 1 or the result is undefined, which is
quite different to truncation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58991
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 10 Nov 2008 10:05:09 +0000 (10:05 +0000)]
Work around PR1000.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58984
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Mon, 10 Nov 2008 07:33:13 +0000 (07:33 +0000)]
Proper way of doing llvm canadian-cross compilation.
Patch by Jim Grosbach!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58981
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Mon, 10 Nov 2008 07:30:32 +0000 (07:30 +0000)]
Reenable test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58980
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Mon, 10 Nov 2008 07:16:42 +0000 (07:16 +0000)]
Really fix testb optimization on big-endian.
Fixes ppc32 bootstrap.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58979
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Mon, 10 Nov 2008 06:59:06 +0000 (06:59 +0000)]
Third time's a charm.
The previous patches didn't match correctly. Also, we need to make sure that
the conditional is the same before doing the transformation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58978
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 05:40:34 +0000 (05:40 +0000)]
final edits
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58966
91177308-0d34-0410-b5e6-
96231b3b80d8
Mon P Wang [Mon, 10 Nov 2008 04:46:22 +0000 (04:46 +0000)]
Added support for the following definition of shufflevector
<result> = shufflevector <n x <ty>> <v1>, <n x <ty>> <v2>, <m x i32> <mask>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58964
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 04:35:24 +0000 (04:35 +0000)]
eliminate a couple more uses of utohexstr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58963
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 04:30:26 +0000 (04:30 +0000)]
Use utohex_buffer instead of utohexstr to avoid creating a temporary
string in the .ll and .s printers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58962
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 04:22:46 +0000 (04:22 +0000)]
split out the functionality of utohexstr into a new utohex_buffer
helper. This allows us to convert numbers to hex without necessarily
needing to make a std::string to hold the result.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58961
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 04:10:34 +0000 (04:10 +0000)]
Move getCastToEmpty out of DIDescriptor into DIFactory. It is an
implementation detail of DIFactory anyway, and this allows it to avoid
recomputing the same type over and over.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58960
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 03:48:55 +0000 (03:48 +0000)]
Fix an over aggressive assumption that struct decls will have a pointer to
an array that is empty. Instead of requiring this array, allow a null pointer.
This shrinks all forward references of structs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58959
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 03:11:39 +0000 (03:11 +0000)]
Fix a bug with default arguments that apple gcc doesn't notice that llvmbb does.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58958
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 03:05:41 +0000 (03:05 +0000)]
move some cases around to silence these sorts of warnings in
release-asserts build:
llvm/CodeGen/SelectionDAGNodes.h:1298: warning: control may reach end of non-void function 'unsigned int llvm::MVT::getSizeInBits() const' being inlined
what an unhelpful warning.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58957
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 02:58:28 +0000 (02:58 +0000)]
add new file
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58955
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 10 Nov 2008 02:56:27 +0000 (02:56 +0000)]
Add a new set of helper classes for creating and reading debug
information. This logically replaces the "Desc" classes in
MachineModuleInfo. Nice features of these classes are that they:
1. Are much more efficient than MMI because they don't create a
temporary parallel data structure for debug info that has to be
'serialized' and 'deserialized' into/out of the module.
2. These provide a much cleaner abstraction for debug info than
MMI, which will make it easier to change the implementation in
the future (to be MDNode-based).
3. These are much easier to use than the MMI interfaces, requiring
a lot less code in the front-ends.
4. These can be used to both create (for frontends) and read (for
codegen) debug information. DebugInfoBuilder can only be used
to create the nodes.
So far, this is implemented just enough to support the debug info
generation needs of clang. This can and should be extended to
support the full set of debug info constructs, and we should switch
llvm-gcc and llc over to using this in the near future.
This code also has a ton of FIXMEs in it, because the way we
currently represent debug info in LLVM IR is basically insane in a
variety of details. This sort of issue should be fixed when we
eventually reimplement debug info on top of MDNodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58954
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Mon, 10 Nov 2008 02:35:55 +0000 (02:35 +0000)]
CMake: Builds a native tblgen when cross-compiling and the user didn't
set LLVM_TABLEGEN.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58953
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 10 Nov 2008 01:52:24 +0000 (01:52 +0000)]
Forgot these.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58952
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Mon, 10 Nov 2008 01:47:07 +0000 (01:47 +0000)]
CMake: Implement magic word `all' for selecting all targets. Check
that specified targets are known.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58951
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Mon, 10 Nov 2008 01:32:14 +0000 (01:32 +0000)]
CMake: Non-working code for auto-building the native tblgen util when
cross-compiling.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58950
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 10 Nov 2008 01:08:07 +0000 (01:08 +0000)]
Rename isGVNonLazyPtr to isIndirectSym to reflect how it will be used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58949
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Sun, 9 Nov 2008 23:37:53 +0000 (23:37 +0000)]
Correction for the last patch. Should match the conditional in the first part
of the select match, not the select instruction itself.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58947
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Sun, 9 Nov 2008 23:17:42 +0000 (23:17 +0000)]
The method of doing the matching with a 'select' instruction was wrong. The
original code was matching like this:
if (match(A, m_Not(m_Value(B))))
B was already matched as a 'select' instruction. However, this isn't matching
what we think it's matching. It would match B as a 'Value', so basically
anything would match to it. In this case, a Constant matched. B was replaced
with a constant representation. And then the wrong value would be used in the
SelectInst::Create statement, causing a crash.
After thinking on this for a moment, and after Nick L. told me how the pattern
matching stuff was supposed to work, the solution was to match NOT an m_Value,
but an m_Select.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58946
91177308-0d34-0410-b5e6-
96231b3b80d8
Gabor Greif [Sun, 9 Nov 2008 21:33:37 +0000 (21:33 +0000)]
suppress build products
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58943
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Sun, 9 Nov 2008 20:26:31 +0000 (20:26 +0000)]
CMake: Corrected detection of `nm'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58941
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Sun, 9 Nov 2008 18:53:19 +0000 (18:53 +0000)]
CMake: Support for cross-compiling. For now, requires a previously
built native tblgen which is passed to cmake in the variable
LLVM_TABLEGEN.
See
http://www.cmake.org/Wiki/CmakeMingw
for a quick example on how to cross-compile with CMake.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58939
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 9 Nov 2008 17:19:14 +0000 (17:19 +0000)]
fix validation problem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58938
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sun, 9 Nov 2008 13:07:47 +0000 (13:07 +0000)]
XFAIL this while waiting for a fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58934
91177308-0d34-0410-b5e6-
96231b3b80d8
Nuno Lopes [Sun, 9 Nov 2008 12:45:23 +0000 (12:45 +0000)]
fix leakage of ValueNumbering
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58933
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Sun, 9 Nov 2008 06:48:10 +0000 (06:48 +0000)]
Temporarily revert 58825, which breaks PPC bootstrap.
xs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58930
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Sun, 9 Nov 2008 04:26:50 +0000 (04:26 +0000)]
If the LHS of the FCMP is coming from a UIToFP instruction, then we don't want
to generate signed ICMP instructions to replace the FCMP. This would violate
the following:
define i1 @test1(i32 %val) {
%1 = uitofp i32 %val to double
%2 = fcmp ole double %1, 0.
000000e+00
ret i1 %2
}
would be transformed into:
define i1 @test1(i32 %val) {
%1 = icmp slt i33 %val, 1
ret i1 %1
}
which is obviously wrong. This patch modifes InstCombiner::FoldFCmp_IntToFP_Cst
to handle when the LHS comes from UIToFP.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58929
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Sun, 9 Nov 2008 02:54:13 +0000 (02:54 +0000)]
Typo fix
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58928
91177308-0d34-0410-b5e6-
96231b3b80d8
Scott Michel [Sun, 9 Nov 2008 01:03:41 +0000 (01:03 +0000)]
CellSPU: Update expected counts on expected patterns
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58927
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Sat, 8 Nov 2008 23:05:05 +0000 (23:05 +0000)]
Temporary revert my last commit: it seems it's triggering some subtle bug in backend
and breaks llvm-gcc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58926
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Sat, 8 Nov 2008 21:23:15 +0000 (21:23 +0000)]
CMake: corrected library target name for dependency: LLVMCellSPU ->
LLVMCellSPUCodeGen.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58925
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Sat, 8 Nov 2008 20:37:19 +0000 (20:37 +0000)]
CMake: Reflected changes on the CellSPU target build. May require a
clean start.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58924
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Sat, 8 Nov 2008 20:34:18 +0000 (20:34 +0000)]
Fixed a pasto.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58923
91177308-0d34-0410-b5e6-
96231b3b80d8
Mikhail Glushenkov [Sat, 8 Nov 2008 19:43:32 +0000 (19:43 +0000)]
Allow $CALL and $ENV in command names. Fixes #3025.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58922
91177308-0d34-0410-b5e6-
96231b3b80d8
Scott Michel [Sat, 8 Nov 2008 18:59:02 +0000 (18:59 +0000)]
CellSPU: Bring SPU's assembly printer more in-line with current LLVM code
structure. Assembly printer now outputs the correct section for strings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58921
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sat, 8 Nov 2008 18:26:48 +0000 (18:26 +0000)]
Try to produce better code when scalarizing VSETCC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58920
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Sat, 8 Nov 2008 17:21:38 +0000 (17:21 +0000)]
Factor out offset printing code into generic AsmPrinter.
FIXME: it seems, that most of targets don't support
offsets wrt CPI/GlobalAddress', was it intentional?
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58917
91177308-0d34-0410-b5e6-
96231b3b80d8
Nicolas Geoffray [Sat, 8 Nov 2008 15:36:01 +0000 (15:36 +0000)]
The Index field of an AttributeWithIndex is of type unsigned, not uint16_t.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58908
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Sat, 8 Nov 2008 12:58:07 +0000 (12:58 +0000)]
StoreInst does not produce any result thus it's useless to create new
variable for it. This greatly reduces amount of unused variables in
llvm2cpp-generated code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58905
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Sat, 8 Nov 2008 10:16:21 +0000 (10:16 +0000)]
Properly escape dashes in TableGen's LLVMC2 emitter.
Patch by Patrick Walton!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58901
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Sat, 8 Nov 2008 08:20:49 +0000 (08:20 +0000)]
Fix syntax of iterate_complex example. Noticed by Martin!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58900
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 08:16:49 +0000 (08:16 +0000)]
Moved InvalidateInstructionCache to ARMJITInfo::emitFunctionStub which knows size of stub.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58899
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 08:15:39 +0000 (08:15 +0000)]
Remove a InvalidateInstructionCache call with incorrect size.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58898
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 08:02:53 +0000 (08:02 +0000)]
Rename startFunctionStub to startGVStub since it's also used for GV non-lazy ptr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58897
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 07:38:22 +0000 (07:38 +0000)]
Tell ARMJITInfo if codegen relocation is PIC. It changes how function stubs are generated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58896
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 07:37:34 +0000 (07:37 +0000)]
Rename isString -> isExternalSymbol; getString -> getExternalSymbol since these work on externsym machine relocations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58895
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 07:22:53 +0000 (07:22 +0000)]
More debug output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58894
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 07:22:33 +0000 (07:22 +0000)]
Fix relocation for calls to external symbols.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58893
91177308-0d34-0410-b5e6-
96231b3b80d8
Scott Michel [Sat, 8 Nov 2008 05:16:20 +0000 (05:16 +0000)]
CellSPU: Fix prologue/epilogue emission when function contains calls but
theframe size is 0; the prologue and epilogue should be emitted in this case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58890
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Sat, 8 Nov 2008 04:12:17 +0000 (04:12 +0000)]
Rework r58829, allowing removal of dbg info intrinsics during alloca
promotion.
- Eliminate uses after free and simplify tests.
Devang: Please check that this is still doing what you intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58887
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Sat, 8 Nov 2008 03:25:47 +0000 (03:25 +0000)]
Add LLVMC2 tool definitions for Objective-C and Objective-C++.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58885
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Sat, 8 Nov 2008 03:10:19 +0000 (03:10 +0000)]
Add svn:ignore properties for Clang llvmc2 plugin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58884
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 01:44:13 +0000 (01:44 +0000)]
Skip over two-address use operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58883
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 01:31:27 +0000 (01:31 +0000)]
Handle ARM machine constantpool entry with non-lazy ptr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58882
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 01:30:20 +0000 (01:30 +0000)]
Indentation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58881
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Sat, 8 Nov 2008 01:02:44 +0000 (01:02 +0000)]
remove unavailable clang project
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58879
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 8 Nov 2008 00:51:41 +0000 (00:51 +0000)]
Use ARMFunctionInfo to track number of constpool entries and jumptables.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58877
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Sat, 8 Nov 2008 00:49:19 +0000 (00:49 +0000)]
Generated code for generic expansion of SETUGT etc.
is noticeably worse than previous PPC-specific code.
Since the latter was also wrong in some cases and
correctness is more important than efficiency, I'm
disabling this test temporarily while I fix it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58876
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Sat, 8 Nov 2008 00:40:24 +0000 (00:40 +0000)]
Xfail an incorrect test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58875
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Sat, 8 Nov 2008 00:01:16 +0000 (00:01 +0000)]
Make testb optimization work on big-endian targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58874
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 22:57:53 +0000 (22:57 +0000)]
More code clean up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58872
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Fri, 7 Nov 2008 22:54:33 +0000 (22:54 +0000)]
Make FP tests requiring two compares work on PPC (PR 642).
This is Chris' patch from the PR, modified to realize that
SETUGT/SETULT occur legitimately with integers, plus
two fixes in LegalizeDAG to pass a valid result type into
LegalizeSetCC. The argument of TLI.getSetCCResultType is
ignored on PPC, but I think I'm following usage elsewhere.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58871
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 22:30:53 +0000 (22:30 +0000)]
Get PIC jump table working.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58869
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 22:30:29 +0000 (22:30 +0000)]
More debug output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58868
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 7 Nov 2008 21:01:13 +0000 (21:01 +0000)]
Make tablegen print out a nice error message for a const char*
exception, like it does for a std::string exception.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58865
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 7 Nov 2008 20:29:17 +0000 (20:29 +0000)]
Document the acronym RAUW. Patch by Jonathan Brandmeyer!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58863
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Fri, 7 Nov 2008 20:13:04 +0000 (20:13 +0000)]
Sign-extend rather than zero-extend when promoting
the condition for a BRCOND, according to what is
returned by getSetCCResultContents. Since all
targets return the same thing (ZeroOrOneSetCCResult),
this should be harmless! The point is that all over
the place the result of SETCC is fed directly into
BRCOND. On machines for which getSetCCResultContents
returns ZeroOrNegativeOneSetCCResult, this is a
sign-extended boolean. So it seems dangerous to
also feed BRCOND zero-extended booleans in some
circumstances - for example, when promoting the
condition.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58861
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 7 Nov 2008 19:49:17 +0000 (19:49 +0000)]
Flush the raw_ostream after emitting the assembly for a function.
This is a temporary fix for the -print-emitted-asm option, where
errs() is used as the stream, in the case where other code is
using stderr without using errs()' buffer. Hopefully soon we'll
fix errs() to be non-buffered instead. Patch by Preston Gurd.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58859
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Fri, 7 Nov 2008 19:11:43 +0000 (19:11 +0000)]
Fix unsigned->ppcf128 conversion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58856
91177308-0d34-0410-b5e6-
96231b3b80d8
Nick Lewycky [Fri, 7 Nov 2008 14:52:51 +0000 (14:52 +0000)]
Update to add newer bitcodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58852
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 12:44:36 +0000 (12:44 +0000)]
Keep CREDITS.TXT sorted by name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58845
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 12:41:14 +0000 (12:41 +0000)]
Add contributer information for XCore backend to CREDITS.TXT
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58844
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 12:37:45 +0000 (12:37 +0000)]
Add XCore backend to CMake build.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58843
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 11:24:12 +0000 (11:24 +0000)]
Add basic test for XCore backend
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58841
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 11:21:09 +0000 (11:21 +0000)]
Fix compile warnings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58840
91177308-0d34-0410-b5e6-
96231b3b80d8
Scott Michel [Fri, 7 Nov 2008 11:06:44 +0000 (11:06 +0000)]
CellSPU: Ensure that C strings are always put in the .rodata section
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58839
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Osborne [Fri, 7 Nov 2008 10:59:00 +0000 (10:59 +0000)]
Add XCore backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58838
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 09:06:08 +0000 (09:06 +0000)]
Jump table JIT support. Work in progress.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58836
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 09:02:17 +0000 (09:02 +0000)]
Jump tables may be emitted by target.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58835
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 09:01:15 +0000 (09:01 +0000)]
Jump table relocation addresses may be resolved by target.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58834
91177308-0d34-0410-b5e6-
96231b3b80d8
Scott Michel [Fri, 7 Nov 2008 04:36:25 +0000 (04:36 +0000)]
Teach CellSPU about ELF sections and new section emitter classes.
NB: This is likely to need more work.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58832
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 7 Nov 2008 01:59:41 +0000 (01:59 +0000)]
BCUI + 1 doesn't work. Use next instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58830
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 7 Nov 2008 01:48:58 +0000 (01:48 +0000)]
Refactor code that adjusts the offsets of stack objects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58829
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 7 Nov 2008 01:41:35 +0000 (01:41 +0000)]
Encode misc arithmetic instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58828
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Fri, 7 Nov 2008 01:30:18 +0000 (01:30 +0000)]
Testcase for testb optimization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58827
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Fri, 7 Nov 2008 01:30:07 +0000 (01:30 +0000)]
Handle (delete) dbg intrinsics while promoting alloca.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58826
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Fri, 7 Nov 2008 01:28:02 +0000 (01:28 +0000)]
When we're doing a compare of load-AND-constant to 0
(e.g. a bitfield test) narrow the load as much as possible.
The has the potential to avoid unnecessary partial-word
load-after-store conflicts, which cause stalls on several targets.
Also a size win on x86 (testb vs testl).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58825
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 7 Nov 2008 01:23:58 +0000 (01:23 +0000)]
- Modify the stack protector algorithm so that the stack slot is allocated in
LLVM IR code and not in the selection DAG ISel. This is a cleaner solution.
- Fix the heuristic for determining if protectors are necessary. The previous
one wasn't checking the proper type size.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58824
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Thu, 6 Nov 2008 23:56:59 +0000 (23:56 +0000)]
Remove unneeded header file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58823
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Thu, 6 Nov 2008 23:55:49 +0000 (23:55 +0000)]
Don't build a vector of returns. Just modify the Function in the loop.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58822
91177308-0d34-0410-b5e6-
96231b3b80d8
Mon P Wang [Thu, 6 Nov 2008 22:52:21 +0000 (22:52 +0000)]
Fixed scalarizing an extract subvector and prevent an infinite loop
when simplify a vector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58820
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Thu, 6 Nov 2008 22:18:44 +0000 (22:18 +0000)]
The size limit is for individual arrays. So if any array has more than 8 bytes
in it, then emit stack protectors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58819
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Thu, 6 Nov 2008 22:15:19 +0000 (22:15 +0000)]
Encode extend instructions; more clean up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58818
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 6 Nov 2008 21:57:17 +0000 (21:57 +0000)]
Fix a use of an invalid iterator when -debug-pass=Details is used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58816
91177308-0d34-0410-b5e6-
96231b3b80d8