oota-llvm.git
10 years agollvm-readobj: address review comments for ARM EHABI printing
Saleem Abdulrasool [Thu, 9 Jan 2014 04:31:18 +0000 (04:31 +0000)]
llvm-readobj: address review comments for ARM EHABI printing

Rename bytecode to opcodes to make it more clear.  Change an impossible case to
llvm_unreachable instead.  Avoid allocation of a buffer by modifying the
PrintOpcodes iteration.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198848 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm-readobj: fix endianness
Saleem Abdulrasool [Thu, 9 Jan 2014 04:31:14 +0000 (04:31 +0000)]
llvm-readobj: fix endianness

Explicitly handle endianness to ensure that bytes are read properly on
big-endian systems.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198847 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDwarfDebug: Refactor out common skeleton construction code to be reused for type...
David Blaikie [Thu, 9 Jan 2014 04:28:46 +0000 (04:28 +0000)]
DwarfDebug: Refactor out common skeleton construction code to be reused for type unit skeletons.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198846 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoExtend llvm::AlignedCharArrayUnion to support up to 10 arguments, as required by...
Richard Smith [Thu, 9 Jan 2014 03:28:55 +0000 (03:28 +0000)]
Extend llvm::AlignedCharArrayUnion to support up to 10 arguments, as required by Clang's APValue.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198844 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReformatting for r198842
David Blaikie [Thu, 9 Jan 2014 03:24:13 +0000 (03:24 +0000)]
Reformatting for r198842

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198843 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDwarfUnit: Rename "Node" to "CUNode" and propagate it through DwarfTypeUnit as well.
David Blaikie [Thu, 9 Jan 2014 03:23:41 +0000 (03:23 +0000)]
DwarfUnit: Rename "Node" to "CUNode" and propagate it through DwarfTypeUnit as well.

Since we'll now also need the split dwarf file name along with the
language in DwarfTypeUnits, just use the whole DICompileUnit rather than
explicitly handling each field needed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198842 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRevert "DwarfUnit: Move the DICompileUnit Node to the DwarfCompileUnit only"
David Blaikie [Thu, 9 Jan 2014 03:03:27 +0000 (03:03 +0000)]
Revert "DwarfUnit: Move the DICompileUnit Node to the DwarfCompileUnit only"

This reverts commit r198830.

Decided to go a different way with this...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198841 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[PM] Rename this source file to something a bit more generic before
Chandler Carruth [Thu, 9 Jan 2014 02:39:45 +0000 (02:39 +0000)]
[PM] Rename this source file to something a bit more generic before
I add support for the new pass manager to it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198838 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoPut the functionality for printing a value to a raw_ostream as an
Chandler Carruth [Thu, 9 Jan 2014 02:29:41 +0000 (02:29 +0000)]
Put the functionality for printing a value to a raw_ostream as an
operand into the Value interface just like the core print method is.
That gives a more conistent organization to the IR printing interfaces
-- they are all attached to the IR objects themselves. Also, update all
the users.

This removes the 'Writer.h' header which contained only a single function
declaration.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198836 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDwarfUnit: Move the DICompileUnit Node to the DwarfCompileUnit only
David Blaikie [Thu, 9 Jan 2014 01:20:14 +0000 (01:20 +0000)]
DwarfUnit: Move the DICompileUnit Node to the DwarfCompileUnit only

It's unused in DwarfTypeUnit, as is expected.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198830 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove the test for endianness in configure.ac and regenerate.
Eric Christopher [Thu, 9 Jan 2014 01:09:57 +0000 (01:09 +0000)]
Remove the test for endianness in configure.ac and regenerate.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198825 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReplace fstream use with raw_fd_ostream.
Lang Hames [Thu, 9 Jan 2014 00:47:54 +0000 (00:47 +0000)]
Replace fstream use with raw_fd_ostream.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198821 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove dead code.
Rafael Espindola [Thu, 9 Jan 2014 00:32:54 +0000 (00:32 +0000)]
Remove dead code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198819 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse the existing typedef to avoid forming a reference to a reference.
Rafael Espindola [Thu, 9 Jan 2014 00:25:25 +0000 (00:25 +0000)]
Use the existing typedef to avoid forming a reference to a reference.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198817 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm.experimental.stackmap: fix encoding of large constants.
Andrew Trick [Thu, 9 Jan 2014 00:22:31 +0000 (00:22 +0000)]
llvm.experimental.stackmap: fix encoding of large constants.

In the stackmap format we advertise the constant field as signed.
However, we were determining whether to promote to a 64-bit constant
pool based on an unsigned comparison.

This fix allows -1 to be encoded as a small constant.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198816 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoSimplify/collapse/denest a conditions/blocks.
David Blaikie [Thu, 9 Jan 2014 00:13:35 +0000 (00:13 +0000)]
Simplify/collapse/denest a conditions/blocks.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198813 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm-dwarfdump: reorder dwo sections to immediately proceed their non-dwo equivalents
David Blaikie [Wed, 8 Jan 2014 23:29:59 +0000 (23:29 +0000)]
llvm-dwarfdump: reorder dwo sections to immediately proceed their non-dwo equivalents

This makes it easier to write a test that's mostly shared between
fission and non-fission (using FileCheck's multiple prefix support).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198806 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix the C++03 build.
Rafael Espindola [Wed, 8 Jan 2014 22:27:04 +0000 (22:27 +0000)]
Fix the C++03 build.

With c++11 we never instantiate the copy constructor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198803 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse getError and remove the error_code operator.
Rafael Espindola [Wed, 8 Jan 2014 22:03:39 +0000 (22:03 +0000)]
Use getError and remove the error_code operator.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198799 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove vestigal bits of MC from the mangler. It no longer uses this, and
Chandler Carruth [Wed, 8 Jan 2014 21:59:22 +0000 (21:59 +0000)]
Remove vestigal bits of MC from the mangler. It no longer uses this, and
having the include could cause weird layering problems between the IR
and MC libraries.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198796 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoConservatively handle multiple MMOs in MIsNeedChainEdge
Hal Finkel [Wed, 8 Jan 2014 21:52:02 +0000 (21:52 +0000)]
Conservatively handle multiple MMOs in MIsNeedChainEdge

MIsNeedChainEdge, which is used by -enable-aa-sched-mi (AA in misched), had an
llvm_unreachable when -enable-aa-sched-mi is enabled and we reach an
instruction with multiple MMOs. Instead, return a conservative answer. This
allows testing -enable-aa-sched-mi on x86.

Also, this moves the check above the isUnsafeMemoryObject checks.
isUnsafeMemoryObject is currently correct only for instructions with one MMO
(as noted in the comment in isUnsafeMemoryObject):

  // We purposefully do no check for hasOneMemOperand() here
  // in hope to trigger an assert downstream in order to
  // finish implementation.

The problem with this is that, had the candidate edge passed the
"!MIa->mayStore() && !MIb->mayStore()" check, the hoped-for assert would never
happen (which could, in theory, lead to incorrect behavior if one of these
secondary MMOs was volatile, for example).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198795 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMove declaration of variables down to first use.
Matt Arsenault [Wed, 8 Jan 2014 21:47:14 +0000 (21:47 +0000)]
Move declaration of variables down to first use.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198794 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd missing definitions of key_type and value_type to DenseSet.
Matt Arsenault [Wed, 8 Jan 2014 21:38:04 +0000 (21:38 +0000)]
Add missing definitions of key_type and value_type to DenseSet.

This matches std::set and allows using DenseSet with the functions
in SetOperations.h

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198793 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd get and getError methods to ErrorOr.
Rafael Espindola [Wed, 8 Jan 2014 21:17:09 +0000 (21:17 +0000)]
Add get and getError methods to ErrorOr.

ErrorOr is modeled after boost::optional which has a get method.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198792 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[AArch64][NEON] Added UXTL and UXTL2 instruction aliases
Ana Pazos [Wed, 8 Jan 2014 21:02:13 +0000 (21:02 +0000)]
[AArch64][NEON] Added UXTL and UXTL2 instruction aliases

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198791 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoForce emit a relocation for @gnu_indirect_function symbols so that the indirect
Roman Divacky [Wed, 8 Jan 2014 18:50:32 +0000 (18:50 +0000)]
Force emit a relocation for @gnu_indirect_function symbols so that the indirect
resolution works.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198780 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Remove OpSize16 flag from MOV32r0
David Woodhouse [Wed, 8 Jan 2014 18:38:26 +0000 (18:38 +0000)]
[x86] Remove OpSize16 flag from MOV32r0

It's not a real instruction any more and doesn't need encoding information.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198778 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoTeach the DAGCombiner how to fold 'vselect' dag nodes according
Andrea Di Biagio [Wed, 8 Jan 2014 18:33:04 +0000 (18:33 +0000)]
Teach the DAGCombiner how to fold 'vselect' dag nodes according
to the following two rules:
  1) fold (vselect (build_vector AllOnes), A, B) -> A
  2) fold (vselect (build_vector AllZeros), A, B) -> B

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198777 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd missing rename from the previous commit.
Rafael Espindola [Wed, 8 Jan 2014 17:56:46 +0000 (17:56 +0000)]
Add missing rename from the previous commit.

No idea how this was compiling locally. Found by the bots.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198775 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRename get to getStorage and getError to getErrorStorage.
Rafael Espindola [Wed, 8 Jan 2014 17:43:26 +0000 (17:43 +0000)]
Rename get to getStorage and getError to getErrorStorage.

These private functions return pointers to the internal storage.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198774 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd missing test case for r198737.
Lang Hames [Wed, 8 Jan 2014 16:31:16 +0000 (16:31 +0000)]
Add missing test case for r198737.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198772 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove mention of old deleted test scripts from testing guide
Nico Rieck [Wed, 8 Jan 2014 16:30:03 +0000 (16:30 +0000)]
Remove mention of old deleted test scripts from testing guide

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198771 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[DAGCombiner] Factor duplicated rotate code into a separate function
Richard Sandiford [Wed, 8 Jan 2014 15:40:47 +0000 (15:40 +0000)]
[DAGCombiner] Factor duplicated rotate code into a separate function

No functional change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198768 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agolit: Provide file location in cfg error messages
Alp Toker [Wed, 8 Jan 2014 14:20:59 +0000 (14:20 +0000)]
lit: Provide file location in cfg error messages

Python doesn't do a good job at diagnosing string exec() so use execfile()
where available.

This should be a timesaver when trying to get to the bottom of build bot
failures.

Before:

    File "llvm/utils/lit/lit/TestingConfig.py", line 93, in load_from_path
      exec("exec data in cfg_globals")
    File "<string>", line 1, in <module>
    File "<string>", line 194, in <module>
  NameError: name 'typo' is not defined

After:

    File "llvm/utils/lit/lit/TestingConfig.py", line 95, in load_from_path
    execfile(path, cfg_globals)
    File "clang/test/lit.cfg", line 194, in <module>
      typo
      ^~~~
  NameError: name 'typo' is not defined

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198766 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Support R_386_PC8, R_386_PC16 and R_X86_64_PC8
David Woodhouse [Wed, 8 Jan 2014 12:58:40 +0000 (12:58 +0000)]
[x86] Support R_386_PC8, R_386_PC16 and R_X86_64_PC8

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198763 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Add JMP_2 and other 16-bit PC-relative branch instructions
David Woodhouse [Wed, 8 Jan 2014 12:58:36 +0000 (12:58 +0000)]
[x86] Add JMP_2 and other 16-bit PC-relative branch instructions

Mark them as requiring 16-bit mode for now, since we don't yet have
relaxation support for FK_Data_2.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198762 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Do not relax PUSHi16 to PUSHi32 (PR18414)
David Woodhouse [Wed, 8 Jan 2014 12:58:32 +0000 (12:58 +0000)]
[x86] Do not relax PUSHi16 to PUSHi32 (PR18414)

They do *different* things to %esp, so they are not equivalent.

Rename PUSHi8 to PUSH32i8 and add the missing PUSH16i8.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198761 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Make AsmParser validate registers for memory operands a bit better
David Woodhouse [Wed, 8 Jan 2014 12:58:28 +0000 (12:58 +0000)]
[x86] Make AsmParser validate registers for memory operands a bit better

We can't do a perfect job here. We *have* to allow (%dx) even in 64-bit
mode, for example, because it might be used for an unofficial form of
the in/out instructions. We actually want to do a better job of validation
*later*. Perhaps *instead* of doing it where we are at the moment.

But for now, doing what validation we *can* do in the place that the code
already has its validation, is an improvement.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198760 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Fix MOV8ao8 et al for 16-bit mode, fix up disassembler to understand
David Woodhouse [Wed, 8 Jan 2014 12:58:24 +0000 (12:58 +0000)]
[x86] Fix MOV8ao8 et al for 16-bit mode, fix up disassembler to understand

It seems there is no separate instruction class for having AdSize *and*
OpSize bits set, which is required in order to disambiguate between all
these instructions. So add that to the disassembler.

Hm, perhaps we do need an AdSize16 bit after all?

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198759 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Use 16-bit addressing where possible in 16-bit mode
David Woodhouse [Wed, 8 Jan 2014 12:58:18 +0000 (12:58 +0000)]
[x86] Use 16-bit addressing where possible in 16-bit mode

Where "where possible" means that it's an immediate value and it's below
0x10000. In fact GAS will either truncate or error with larger values,
and will insist on using the addr32 prefix to get 32-bit addressing. So
perhaps we should do that, in a later patch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198758 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Fix JCXZ,JECXZ_32 for 16-bit mode
David Woodhouse [Wed, 8 Jan 2014 12:58:12 +0000 (12:58 +0000)]
[x86] Fix JCXZ,JECXZ_32 for 16-bit mode

JCXZ should have the 0x67 prefix only if we're in 32-bit mode, so make that
appropriately conditional. And JECXZ needs the prefix instead.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198757 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Disambiguate RET[QL] and fix aliases for 16-bit mode
David Woodhouse [Wed, 8 Jan 2014 12:58:07 +0000 (12:58 +0000)]
[x86] Disambiguate RET[QL] and fix aliases for 16-bit mode

I couldn't see how to do this sanely without splitting RETQ from RETL.

Eric says: "sad about the inability to roundtrip them now, but...".
I have no idea what that means, but perhaps it wants preserving in the
commit comment.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198756 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Disambiguate [LS][IG]DT{32,64}m and add 16-bit versions, fix aliases
David Woodhouse [Wed, 8 Jan 2014 12:57:55 +0000 (12:57 +0000)]
[x86] Disambiguate [LS][IG]DT{32,64}m and add 16-bit versions, fix aliases

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198755 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Add JMP16[rm],CALL16[rm] instructions, and fix up aliases
David Woodhouse [Wed, 8 Jan 2014 12:57:49 +0000 (12:57 +0000)]
[x86] Add JMP16[rm],CALL16[rm] instructions, and fix up aliases

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198754 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Add PUSHA16,POPA16 instructions, and fix aliases for 16-bit mode
David Woodhouse [Wed, 8 Jan 2014 12:57:45 +0000 (12:57 +0000)]
[x86] Add PUSHA16,POPA16 instructions, and fix aliases for 16-bit mode

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198753 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Add OpSize16 to instructions that need it
David Woodhouse [Wed, 8 Jan 2014 12:57:40 +0000 (12:57 +0000)]
[x86] Add OpSize16 to instructions that need it

This fixes the bulk of 16-bit output, and the corresponding test case
x86-16.s now looks mostly like the x86-32.s test case that it was
originally based on. A few irrelevant instructions have been dropped,
and there are still some corner cases to be fixed in subsequent patches.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198752 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoUse -std=gnu99 in tools/llvm-c-test/CMakeLists.txt
Rafael Espindola [Wed, 8 Jan 2014 11:48:19 +0000 (11:48 +0000)]
Use -std=gnu99 in tools/llvm-c-test/CMakeLists.txt

With a current mingw (gcc 4.8.1) it looks like we hit some variation of

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40278

The end result is that off_t is not defined and the build fails without this
patch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198749 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoCMake: Provide LLVM_PLUGIN_EXT definition
Alp Toker [Wed, 8 Jan 2014 11:10:24 +0000 (11:10 +0000)]
CMake: Provide LLVM_PLUGIN_EXT definition

This is needed to support the addition of tests for clang loadable plugins.

In clang, plugins are built as modules (bundles on OS X) rather than dynamic
libraries (dylib) so the build system needs to inform lit of the actual
file extension in use, typically '.so' on Unix and '.dll' on Windows.

(LLVM itself should probably switch to this scheme to fix PR14903 once and for
all.)

No change in build output or functionality intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198746 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAVX-512: Added more intrinsics for pmin/pmax, pabs, blend, pmuldq.
Elena Demikhovsky [Wed, 8 Jan 2014 10:54:22 +0000 (10:54 +0000)]
AVX-512: Added more intrinsics for pmin/pmax, pabs, blend, pmuldq.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198745 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[patch] Adjust behavior of FDE cross-section relocs for targets that don't support...
Iain Sandoe [Wed, 8 Jan 2014 10:22:54 +0000 (10:22 +0000)]
[patch] Adjust behavior of FDE cross-section relocs for targets that don't support abs-differences.

Modern versions of OSX/Darwin's ld (ld64 > 97.17) have an optimisation present that allows the back end to omit relocations (and replace them with an absolute difference) for FDE some text section refs.

This patch allows a backend to opt-in to this behaviour by setting "DwarfFDESymbolsUseAbsDiff".  At present, this is only enabled for modern x86 OSX ports.

test changes by David Fang.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198744 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[AArch64 NEON] Fix generating incorrect value type of NEON_VDUPLANE
Kevin Qin [Wed, 8 Jan 2014 08:06:14 +0000 (08:06 +0000)]
[AArch64 NEON] Fix generating incorrect value type of NEON_VDUPLANE
when lower build_vector if result value type mismatch with operand
value type.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198743 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[SparcV9] Rename operands in some sparc64 instructions so that TableGen can encode...
Venkatraman Govindaraju [Wed, 8 Jan 2014 07:47:57 +0000 (07:47 +0000)]
[SparcV9] Rename operands in some sparc64 instructions so that TableGen can encode them correctly.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198740 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[Sparc] Correct the mask for fixup_sparc_br19.
Venkatraman Govindaraju [Wed, 8 Jan 2014 06:46:51 +0000 (06:46 +0000)]
[Sparc] Correct the mask for fixup_sparc_br19.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198739 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[Sparc] Add support for parsing branch instructions and conditional moves.
Venkatraman Govindaraju [Wed, 8 Jan 2014 06:14:52 +0000 (06:14 +0000)]
[Sparc] Add support for parsing branch instructions and conditional moves.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198738 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRe-apply r196639: Add support for archives and object file caching under MCJIT.
Lang Hames [Wed, 8 Jan 2014 04:09:09 +0000 (04:09 +0000)]
Re-apply r196639: Add support for archives and object file caching under MCJIT.

I believe the bot failures on linux systems were due to overestimating the
alignment of object-files within archives, which are only guaranteed to be
two-byte aligned. I have reduced the alignment in
RuntimeDyldELF::createObjectImageFromFile accordingly.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198737 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agotests: disable ARM unwinding tests if ARM is unavailable
Saleem Abdulrasool [Wed, 8 Jan 2014 03:44:01 +0000 (03:44 +0000)]
tests: disable ARM unwinding tests if ARM is unavailable

Appease the buildbots for targets which do not build the ARM support by moving
the ARM specific test into a subdirectory and use the lit configuration to
disable them appropriately.

Thanks to chapuni and thakis for explaining how to do this!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198736 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM IAS: properly handle expression operands
Saleem Abdulrasool [Wed, 8 Jan 2014 03:28:14 +0000 (03:28 +0000)]
ARM IAS: properly handle expression operands

Operands which involved label arithemetic would previously fail to parse.  This
corrects that by adding the additional case for the shift operand validation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198735 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agollvm-readobj: add support for ARM EHABI unwind info
Saleem Abdulrasool [Wed, 8 Jan 2014 03:28:09 +0000 (03:28 +0000)]
llvm-readobj: add support for ARM EHABI unwind info

This adds some preliminary support for decoding ARM EHABI unwinding information.
The major functionality that remains from complete support is bytecode
translation.

Each Unwind Index Table is printed out as a separate entity along with its
section index, name, offset, and entries.

Each entry lists the function address, and if possible, the name, of the
function to which it corresponds.  The encoding model, personality routine or
index, and byte code is also listed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198734 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix a bug about generating undef operand when optimising shuffle vector and insert...
Hao Liu [Wed, 8 Jan 2014 03:06:15 +0000 (03:06 +0000)]
Fix a bug about generating undef operand when optimising shuffle vector and insert element in instruction combine.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198730 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[x86] Kill gratuitous X86_{32,64}TargetMachine subclasses, use X86TargetMachine
David Woodhouse [Wed, 8 Jan 2014 00:08:50 +0000 (00:08 +0000)]
[x86] Kill gratuitous X86_{32,64}TargetMachine subclasses, use X86TargetMachine

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198720 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMove the llvm mangler to lib/IR.
Rafael Espindola [Tue, 7 Jan 2014 21:19:40 +0000 (21:19 +0000)]
Move the llvm mangler to lib/IR.

This makes it available to tools that don't link with target (like llvm-ar).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198708 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoIn the ELFWriter when writing aliased (.set) symbols dont blindly
Roman Divacky [Tue, 7 Jan 2014 20:17:03 +0000 (20:17 +0000)]
In the ELFWriter when writing aliased (.set) symbols dont blindly
take type from the new symbol but merge them so that the type
is never "downgraded".

This is probably quite rare, except for IFUNC symbols which
we used to misassemble, losing the IFUNC type.

Fixes #18372.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198706 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix uninitialized variable warning in DataLayout.
Cameron McInally [Tue, 7 Jan 2014 19:51:38 +0000 (19:51 +0000)]
Fix uninitialized variable warning in DataLayout.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198702 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoDon't assert with private type info variables.
Rafael Espindola [Tue, 7 Jan 2014 19:38:47 +0000 (19:38 +0000)]
Don't assert with private type info variables.

With the gnu objc runtime private strings are used. Since we only need to
produce a unique label, the fix is to just drop the asserts.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198701 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoEmit arange padding with a single directive.
Benjamin Kramer [Tue, 7 Jan 2014 19:28:14 +0000 (19:28 +0000)]
Emit arange padding with a single directive.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198700 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoAdd ARM fconsts/fconstd aliases for vmov.f32/vmov.f64
David Peixotto [Tue, 7 Jan 2014 18:19:23 +0000 (18:19 +0000)]
Add ARM fconsts/fconstd aliases for vmov.f32/vmov.f64

This commit adds the pre-UAL aliases of fconsts and fconstd for
vmov.f32 and vmov.f64. They use an InstAlias rather than a
MnemonicAlias to properly support the predicate operand.

We need to support encoded 8-bit constants in order to implement the
pre-UAL fconsts/fconstd aliases for vmov.f32/vmov.f64, so this
commit also fixes parsing of encoded floating point constants used
in vmov.f32/vmov.f64 instructions. Now we can support assembly code
like this:

  fconsts s0, #0x70

which is equivalent to vmov.f32 s0, #1.0.

Most of the code was already in place to support this feature.
Previously the code was trying to accept encoded 8-bit float
constants for the vmov.f32/vmov.f64 instructions.  It looks like the
support for parsing encoded floats was lost in a refactoring in
commit r148556 and we did not have any tests in place to catch it.

The change in this commit is to keep the parsed value as a 32-bit
float instead of a 64-bit double because that is what the isFPImm()
function expects to find. There is no loss of precision by using a
32-bit float here because we are still limited to an 8-bit encoded
value in the end.

Additionally, we explicitly reject encoded 8-bit floats for
vmovf.32/64. This is the same as the current behavior, but we now do
it explicitly rather than accidently.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198697 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[arm] Fix an incorrect comment in ARMUnwindOpAsm.h.
Logan Chien [Tue, 7 Jan 2014 17:47:25 +0000 (17:47 +0000)]
[arm] Fix an incorrect comment in ARMUnwindOpAsm.h.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198696 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoTry to fix the windows build. The comments in other files don't seem to
Chandler Carruth [Tue, 7 Jan 2014 12:37:13 +0000 (12:37 +0000)]
Try to fix the windows build. The comments in other files don't seem to
be quite accurate. =]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198690 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMove the LLVM IR asm writer header files into the IR directory, as they
Chandler Carruth [Tue, 7 Jan 2014 12:34:26 +0000 (12:34 +0000)]
Move the LLVM IR asm writer header files into the IR directory, as they
are part of the core IR library in order to support dumping and other
basic functionality.

Rename the 'Assembly' include directory to 'AsmParser' to match the
library name and the only functionality left their -- printing has been
in the core IR library for quite some time.

Update all of the #includes to match.

All of this started because I wanted to have the layering in good shape
before I started adding support for printing LLVM IR using the new pass
infrastructure, and commandline support for the new pass infrastructure.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198688 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRe-sort all of the includes with ./utils/sort_includes.py so that
Chandler Carruth [Tue, 7 Jan 2014 11:48:04 +0000 (11:48 +0000)]
Re-sort all of the includes with ./utils/sort_includes.py so that
subsequent changes are easier to review. About to fix some layering
issues, and wanted to separate out the necessary churn.

Also comment and sink the include of "Windows.h" in three .inc files to
match the usage in Memory.inc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198685 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[AArch64]Add support to spill/fill D tuples such as DPair/DTriple/DQuad. There is...
Hao Liu [Tue, 7 Jan 2014 10:50:43 +0000 (10:50 +0000)]
[AArch64]Add support to spill/fill D tuples such as DPair/DTriple/DQuad. There is no test cases for D tuple as the original test cases are too large. As the spill/fill of the D tuple is similar to the Q tuple, the correctness can be guaranteed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198684 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[CMake] Introduce llvm_update_compile_flags(target_name) to update compile flags...
NAKAMURA Takumi [Tue, 7 Jan 2014 10:24:14 +0000 (10:24 +0000)]
[CMake] Introduce llvm_update_compile_flags(target_name) to update compile flags in target properties.

FIXME: Just add_unittest() is using it.
FIXME: Cooperate with source properties.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198683 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[AArch64]Add support to copy D tuples such as DPair/DTriple/DQuad and Q tuples such...
Hao Liu [Tue, 7 Jan 2014 10:00:03 +0000 (10:00 +0000)]
[AArch64]Add support to copy D tuples such as DPair/DTriple/DQuad and Q tuples such as QPair/QTriple/QQuad. There is no test case for D tuple as the original test cases are too large. As the copy of the D tuple is similar to the Q tuple, the correctness can be guaranteed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198682 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[Sparc] Add support for parsing sparc asm modifiers such as %hi, %lo etc.,
Venkatraman Govindaraju [Tue, 7 Jan 2014 08:00:49 +0000 (08:00 +0000)]
[Sparc] Add support for parsing sparc asm modifiers such as %hi, %lo etc.,
Also, correct the offsets for FixupsKindInfo.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198681 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix for PR18396: Assertion: MO->isDead "Cannot fold physreg def".
Andrew Trick [Tue, 7 Jan 2014 07:31:10 +0000 (07:31 +0000)]
Fix for PR18396: Assertion: MO->isDead "Cannot fold physreg def".

InlineSpiller::foldMemoryOperand needs to handle undef call operands.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198679 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReapply r198654 "indvars: sink truncates outside the loop."
Andrew Trick [Tue, 7 Jan 2014 06:59:12 +0000 (06:59 +0000)]
Reapply r198654 "indvars: sink truncates outside the loop."

This doesn't seem to have actually broken anything. It was paranoia
on my part. Trying again now that bots are more stable.

This is a follow up of the r198338 commit that added truncates for
lcssa phi nodes. Sinking the truncates below the phis cleans up the
loop and simplifies subsequent analysis within the indvars pass.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198678 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[AArch64 NEON] Fixed incorrect immediate used in BIC instruction.
Kevin Qin [Tue, 7 Jan 2014 05:10:47 +0000 (05:10 +0000)]
[AArch64 NEON] Fixed incorrect immediate used in BIC instruction.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198675 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM IAS: allow more depth in contextual diagnostics
Saleem Abdulrasool [Tue, 7 Jan 2014 02:29:00 +0000 (02:29 +0000)]
ARM IAS: allow more depth in contextual diagnostics

Switch the context to be SmallVectors.  This allows for saving additional
context when providing previous emission sites.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198665 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM IAS: refactor unwind context
Saleem Abdulrasool [Tue, 7 Jan 2014 02:28:55 +0000 (02:28 +0000)]
ARM IAS: refactor unwind context

Move the unwinding context for the ARM IAS into a helper class.  This is purely
a structural refactoring.  A follow up change allows for recording additional
depth to improve diagnostics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198664 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM Streamer: print out tag names
Saleem Abdulrasool [Tue, 7 Jan 2014 02:28:50 +0000 (02:28 +0000)]
ARM Streamer: print out tag names

If using verbose asm, print out the friendly name of the tag if possible.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198663 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM IAS: improve .eabi_attribute handling
Saleem Abdulrasool [Tue, 7 Jan 2014 02:28:42 +0000 (02:28 +0000)]
ARM IAS: improve .eabi_attribute handling

Parse tag names as well as expressions.  The former is part of the
specification, the latter is for improved compatibility with the GNU assembler.
Fix attribute value handling to be comformant to the specification.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198662 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoMCParser: introduce Note and use it for ARM AsmParser
Saleem Abdulrasool [Tue, 7 Jan 2014 02:28:31 +0000 (02:28 +0000)]
MCParser: introduce Note and use it for ARM AsmParser

Introduce a new virtual method Note into the AsmParser.  This completements the
existing Warning and Error methods.  Use the new method to clean up the output
of the unwind routines in the ARM AsmParser.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198661 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix comment of findGCD.
Mingjie Xing [Tue, 7 Jan 2014 01:54:16 +0000 (01:54 +0000)]
Fix comment of findGCD.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198660 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRevert "indvars: sink truncates outside the loop."
Andrew Trick [Tue, 7 Jan 2014 01:50:58 +0000 (01:50 +0000)]
Revert "indvars: sink truncates outside the loop."

This reverts commit r198654.

One of the bots reported a SciMark failure.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198659 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[Sparc] Add support for parsing memory operands in sparc AsmParser.
Venkatraman Govindaraju [Tue, 7 Jan 2014 01:49:11 +0000 (01:49 +0000)]
[Sparc] Add support for parsing memory operands in sparc AsmParser.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198658 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoindvars: sink truncates outside the loop.
Andrew Trick [Tue, 7 Jan 2014 01:02:55 +0000 (01:02 +0000)]
indvars: sink truncates outside the loop.

This is a follow up of the r198338 commit that added truncates for
lcssa phi nodes. Sinking the truncates below the phis cleans up the
loop and simplifies subsequent analysis within the indvars pass.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198654 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago80 col. comment.
Andrew Trick [Tue, 7 Jan 2014 01:02:52 +0000 (01:02 +0000)]
80 col. comment.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198653 91177308-0d34-0410-b5e6-96231b3b80d8

10 years ago[Mips] TargetStreamer Support for .abicalls and .set pic0.
Jack Carter [Mon, 6 Jan 2014 23:27:31 +0000 (23:27 +0000)]
[Mips] TargetStreamer Support for .abicalls and .set pic0.

This patch adds .abicalls and .set pic0 support which
affects the ELF ABI and its flags. In addition the patch uses
a common interface for both the MipsTargetSteamer and
MipsObjectStreamer that both the integrated and standalone
assemblers will use for the output for these directives.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198646 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFor the 'C' disassembler API, add a new ReferenceType for the
Kevin Enderby [Mon, 6 Jan 2014 22:08:08 +0000 (22:08 +0000)]
For the 'C' disassembler API, add a new ReferenceType for the
SymbolLookUp() call back to return a demangled C++ name to
be used as a comment.

For example darwin's otool(1) program the uses the llvm
disassembler now can produce disassembly like:

callq   __ZNK4llvm6Target20createMCDisassemblerERKNS_15MCSubtargetInfoE ## llvm::Target::createMCDisassembler(llvm::MCSubtargetInfo const&) const

Also fix a bug in LLVMDisasmInstruction() that was not flushing
the raw_svector_ostream for the disassembled instruction string
before copying it to the output buffer that was causing truncation
of the output.

rdar://10173828

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198637 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoImprove documentation of the 'a' specifier and the '<abi>:<pref>' align pair.
Rafael Espindola [Mon, 6 Jan 2014 21:40:24 +0000 (21:40 +0000)]
Improve documentation of the 'a' specifier and the '<abi>:<pref>' align pair.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198636 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoFix c++ mode comments
Matt Arsenault [Mon, 6 Jan 2014 19:52:42 +0000 (19:52 +0000)]
Fix c++ mode comments

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198632 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoReapply r198478 "Fix PR18361: Invalidate LoopDispositions after LoopSimplify hoists...
Andrew Trick [Mon, 6 Jan 2014 19:43:14 +0000 (19:43 +0000)]
Reapply r198478 "Fix PR18361: Invalidate LoopDispositions after LoopSimplify hoists things."

Now with a fix for PR18384: ValueHandleBase::ValueIsDeleted.

We need to invalidate SCEV's loop info when we delete a block, even if no values are hoisted.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198631 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoIntroduce a cmake LLVM_ENABLE_LIBCXX build parameter to compile using libc++ instead...
Jean-Daniel Dupas [Mon, 6 Jan 2014 18:27:27 +0000 (18:27 +0000)]
Introduce a cmake LLVM_ENABLE_LIBCXX build parameter to compile using libc++ instead of the system default

Summary:
This parameter is required to build C++11 projects (like lld or lldb) on OS X as the default STL does not provide c++ classes.

CC: llvm-commits, triton
Differential Revision: http://llvm-reviews.chandlerc.com/D2381

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198625 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoRemove dead code.
Rafael Espindola [Mon, 6 Jan 2014 18:14:34 +0000 (18:14 +0000)]
Remove dead code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198624 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoARM MachO: sort out isTargetDarwin/isTargetIOS/... checks.
Tim Northover [Mon, 6 Jan 2014 14:28:05 +0000 (14:28 +0000)]
ARM MachO: sort out isTargetDarwin/isTargetIOS/... checks.

The ARM backend has been using most of the MachO related subtarget
checks almost interchangeably, and since the only target it's had to
run on has been IOS (which is all three of MachO, Darwin and IOS) it's
worked out OK so far.

But we'd like to support embedded targets under the "*-*-none-macho"
triple, which means everything starts falling apart and inconsistent
behaviours emerge.

This patch should pick a reasonably sensible set of behaviours for the
new triple (and any others that come along, with luck). Some choices
were debatable (notably FP == r7 or r11), but we can revisit those
later when deficiencies become apparent.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198617 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoXCore Target: correct callee save register spilling when callsUnwindInit is true.
Robert Lytton [Mon, 6 Jan 2014 14:21:12 +0000 (14:21 +0000)]
XCore Target: correct callee save register spilling when callsUnwindInit is true.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198616 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoXCore target: Lower EH_RETURN
Robert Lytton [Mon, 6 Jan 2014 14:21:07 +0000 (14:21 +0000)]
XCore target: Lower EH_RETURN

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198615 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoXCore target: Lower FRAME_TO_ARGS_OFFSET
Robert Lytton [Mon, 6 Jan 2014 14:21:00 +0000 (14:21 +0000)]
XCore target: Lower FRAME_TO_ARGS_OFFSET

This requires a knowledge of the stack size which is not known until
the frame is complete, hence the need for the XCoreFTAOElim pass
which lowers the XCoreISD::FRAME_TO_ARGS_OFFSET instrution into its
final form.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198614 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoXCore target: Lower RETURNADDR
Robert Lytton [Mon, 6 Jan 2014 14:20:53 +0000 (14:20 +0000)]
XCore target: Lower RETURNADDR

Only handles a depth of zero (the same as FRAMEADDR)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198613 91177308-0d34-0410-b5e6-96231b3b80d8

10 years agoXCore target: Optimise entsp / retsp selection
Robert Lytton [Mon, 6 Jan 2014 14:20:47 +0000 (14:20 +0000)]
XCore target: Optimise entsp / retsp selection

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198612 91177308-0d34-0410-b5e6-96231b3b80d8