Rafael Espindola [Fri, 24 Jan 2014 02:28:11 +0000 (02:28 +0000)]
Inline trivial functions called only once or twice.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199967
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Fri, 24 Jan 2014 02:24:47 +0000 (02:24 +0000)]
[LPM] Fix a logic error in LICM spotted by inspection.
We completely skipped promotion in LICM if the loop has a preheader or
dedicated exits, but not *both*. We hoist if there is a preheader, and
sink if there are dedicated exits, but either hoisting or sinking can
move loop invariant code out of the loop!
I have no idea if this has a practical consequence. If anyone has ideas
for a test case, let me know.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199966
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 24 Jan 2014 02:18:40 +0000 (02:18 +0000)]
Inline functions that are only called once.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199965
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Fri, 24 Jan 2014 01:59:49 +0000 (01:59 +0000)]
[cleanup] Use the type-based preservation method rather than a string
literal that bakes a pass name and forces parsing it in the pass
manager.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199963
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Thu, 23 Jan 2014 23:14:14 +0000 (23:14 +0000)]
InitToTextSection is redundant with InitSections. Remove it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199955
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 23 Jan 2014 22:55:47 +0000 (22:55 +0000)]
Make the use of DW_AT_ranges in the compile unit depend also upon
the existence of comdat/special sections.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199954
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Thu, 23 Jan 2014 22:49:25 +0000 (22:49 +0000)]
Remove duplicated info on what .text, .data and .bss look like.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199951
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Enderby [Thu, 23 Jan 2014 22:34:42 +0000 (22:34 +0000)]
Update the X86 assembler for .intel_syntax to produce an error for invalid base
registers in memory addresses that do not match the index register. As it does
for .att_syntax.
rdar://
15887380
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199948
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 22:19:45 +0000 (22:19 +0000)]
lli: Factor portable messaging into a new RPCChannel facility
The client and server now use a single unified low-level RPC core built around
LLVM's existing cross-platform abstractions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199947
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Enderby [Thu, 23 Jan 2014 21:52:41 +0000 (21:52 +0000)]
Update the X86 assembler for .intel_syntax to produce an error for invalid
scale factors in memory addresses. As it does for .att_syntax.
It was producing:
Assertion failed: (((Scale == 1 || Scale == 2 || Scale == 4 || Scale == 8)) && "Invalid scale!"), function CreateMem, file /Volumes/SandBox/llvm/lib/Target/X86/AsmParser/X86AsmParser.cpp, line 1133.
rdar://
14967214
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199942
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 23 Jan 2014 21:41:10 +0000 (21:41 +0000)]
Fix out of bounds access to the double regs array. Given the
code this looks correct, but could use review. The previous
was definitely not correct.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199940
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Thu, 23 Jan 2014 21:27:27 +0000 (21:27 +0000)]
Add a few missing cases from r199933. Testcase coming shortly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199938
91177308-0d34-0410-b5e6-
96231b3b80d8
Aaron Ballman [Thu, 23 Jan 2014 20:46:44 +0000 (20:46 +0000)]
Updating the getting started guide for Visual Studio a smidge.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199934
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Thu, 23 Jan 2014 20:23:36 +0000 (20:23 +0000)]
Replace vfmaddxx213 instructions with their 231-type equivalents in accumulator
loops. Writing back to the accumulator (231-type) allows the coalescer to
eliminate an extra copy.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199933
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Thu, 23 Jan 2014 20:17:12 +0000 (20:17 +0000)]
Note the PR number.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199932
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 20:01:21 +0000 (20:01 +0000)]
Remove unused include following r199929
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199930
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 19:57:16 +0000 (19:57 +0000)]
Replace the interim lli build fix with something cleaner
Eliminates the LLI_BUILDING_CHILD build hack from r199885.
Also add a FIXME to remove code that tricks the tests into passing when the
feature fails to work. Please don't do stuff like this, the tests exist for a
reason!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199929
91177308-0d34-0410-b5e6-
96231b3b80d8
Weiming Zhao [Thu, 23 Jan 2014 19:55:33 +0000 (19:55 +0000)]
[Thumbv8] Fix the value of BLXOperandIndex of isV8EligibleForIT
Originally, BLX was passed as operand #0 in MachineInstr and as operand
#2 in MCInst. But now, it's operand #2 in both cases.
This patch also removes unnecessary FileCheck in the test case added by r199127.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199928
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 23 Jan 2014 19:32:19 +0000 (19:32 +0000)]
Move test to x86 directory.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199927
91177308-0d34-0410-b5e6-
96231b3b80d8
Juergen Ributzka [Thu, 23 Jan 2014 19:23:28 +0000 (19:23 +0000)]
Add target analysis passes to the codegen pipeline for MCJIT.
This patch adds the target analysis passes (usually TargetTransformInfo) to the
codgen pipeline. We also expose now the AddAnalysisPasses method through the C
API, because the optimizer passes would also benefit from better target-specific
cost models.
Reviewed by Andrew Kaylor
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199926
91177308-0d34-0410-b5e6-
96231b3b80d8
Ana Pazos [Thu, 23 Jan 2014 19:18:57 +0000 (19:18 +0000)]
[AArch64] Added vselect patterns with float and double types
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199925
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 23 Jan 2014 19:16:28 +0000 (19:16 +0000)]
Avoid emitting a DWARF type attribute for an ObjC property of type
void.
Patch by Scott Talbot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199924
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Thu, 23 Jan 2014 19:01:25 +0000 (19:01 +0000)]
Suppress an annoying "unused variable" warning caused by bug 17897.
Clang says that "flow" is unused when building LLD. This patch suppresses it.
Differential Revision: http://llvm-reviews.chandlerc.com/D2573
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199922
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 23 Jan 2014 18:49:34 +0000 (18:49 +0000)]
R600: Remove successive JUMP in AnalyzeBranch when AllowModify is true
This fixes a crash in the OpenCV OpenCL test suite.
There is no lit test for this, because the test would be very large
and could easily be invalidated by changes to the scheduler
or other parts of the compiler.
Patch by: Vincent Lejeune
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199919
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 23 Jan 2014 18:49:33 +0000 (18:49 +0000)]
R600: Disable the BFE pattern
This pattern uses an SDNodeXForm, which isn't being emitted for some
reason. I can get it to work by attaching the PatLeaf that has the
XForm to the argument in the output pattern, but this results in an
immediate being used in a register operand, which the backend can't
handle yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199918
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 23 Jan 2014 18:49:31 +0000 (18:49 +0000)]
R600: Correctly handle vertex fetch clauses the precede ENDIFs
The control flow finalizer would sometimes use an ALU_POP_AFTER
instruction before the vetex fetch clause instead of using a POP
instruction after it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199917
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 23 Jan 2014 18:49:28 +0000 (18:49 +0000)]
R600: Unconditionally unroll loops that contain GEPs with alloca pointers
Implement the getUnrollingPreferences() function for
AMDGPUTargetTransformInfo so that loops that do address calculations
on pointers derived from alloca are unconditionally unrolled.
Unrolling these loops makes it more likely that SROA will be able to
eliminate the allocas, which is a big win for R600 since memory
allocated by alloca (private memory) is really slow.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199916
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Thu, 23 Jan 2014 17:47:57 +0000 (17:47 +0000)]
Move a unit test into the correct dir. Sorry if it broke Mips-only builds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199911
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Thu, 23 Jan 2014 17:19:42 +0000 (17:19 +0000)]
Remove tail marker when changing an argument to an alloca.
Argument promotion can replace an argument of a call with an alloca. This
requires clearing the tail marker as it is very likely that the callee is now
using an alloca in the caller.
This fixes pr14710.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199909
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 23 Jan 2014 16:18:02 +0000 (16:18 +0000)]
R600: Recommit 199842: Add work-around for the CF stack entry HW bug
The unit test is now disabled on non-asserts builds.
The CF stack can be corrupted if you use CF_ALU_PUSH_BEFORE,
CF_ALU_ELSE_AFTER, CF_ALU_BREAK, or CF_ALU_CONTINUE when the number of
sub-entries on the stack is greater than or equal to the stack entry
size and sub-entries modulo 4 is either 0 or 3 (on cedar the bug is
present when number of sub-entries module 8 is either 7 or 0)
We choose to be conservative and always apply the work-around when the
number of sub-enries is greater than or equal to the stack entry size,
so that we can safely over-allocate the stack when we are unsure of the
stack allocation rules.
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199905
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Atanasyan [Thu, 23 Jan 2014 15:05:45 +0000 (15:05 +0000)]
[Object][ELF][Mips] Print symbol name for MIPS ELF relocations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199898
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Thu, 23 Jan 2014 14:27:26 +0000 (14:27 +0000)]
AVX-512: added VPERM2D VPERM2Q VPERM2PS VPERM2PD instructions,
they give better sequences than VPERMI
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199893
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Thu, 23 Jan 2014 13:43:47 +0000 (13:43 +0000)]
ARM: use litpools for normal i32 imms when compiling minsize.
With constant-sharing, litpool loads consume 4 + N*2 bytes of code, but
movw/movt pairs consume 8*N. This means litpools are better than movw/movt even
with just one use. Other materialisation strategies can still be better though,
so the logic is a little odd.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199891
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexander Kornienko [Thu, 23 Jan 2014 13:37:15 +0000 (13:37 +0000)]
Remove empty directories left after r190748 and r198688.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199890
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Thu, 23 Jan 2014 12:55:05 +0000 (12:55 +0000)]
Windows/ChildTarget.inc: LLIChildTarget::allocate() has gone since r199881.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199889
91177308-0d34-0410-b5e6-
96231b3b80d8
Artyom Skrobov [Thu, 23 Jan 2014 11:31:38 +0000 (11:31 +0000)]
Prevent repetitive warnings for unrecognized processors and features
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199886
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 11:26:37 +0000 (11:26 +0000)]
Interim build fix for Makefiles
Looks like some parts still need detangling. Let's see if this holds for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199885
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Thu, 23 Jan 2014 11:23:19 +0000 (11:23 +0000)]
[LPM] Make LoopSimplify no longer a LoopPass and instead both a utility
function and a FunctionPass.
This has many benefits. The motivating use case was to be able to
compute function analysis passes *after* running LoopSimplify (to avoid
invalidating them) and then to run other passes which require
LoopSimplify. Specifically passes like unrolling and vectorization are
critical to wire up to BranchProbabilityInfo and BlockFrequencyInfo so
that they can be profile aware. For the LoopVectorize pass the only
things in the way are LoopSimplify and LCSSA. This fixes LoopSimplify
and LCSSA is next on my list.
There are also a bunch of other benefits of doing this:
- It is now very feasible to make more passes *preserve* LoopSimplify
because they can simply run it after changing a loop. Because
subsequence passes can assume LoopSimplify is preserved we can reduce
the runs of this pass to the times when we actually mutate a loop
structure.
- The new pass manager should be able to more easily support loop passes
factored in this way.
- We can at long, long last observe that LoopSimplify is preserved
across SCEV. This *halves* the number of times we run LoopSimplify!!!
Now, getting here wasn't trivial. First off, the interfaces used by
LoopSimplify are all over the map regarding how analysis are updated. We
end up with weird "pass" parameters as a consequence. I'll try to clean
at least some of this up later -- I'll have to have it all clean for the
new pass manager.
Next up I discovered a really frustrating bug. LoopUnroll *claims* to
preserve LoopSimplify. That's actually a lie. But the way the
LoopPassManager ends up running the passes, it always ran LoopSimplify
on the unrolled-into loop, rectifying this oversight before any
verification could kick in and point out that in fact nothing was
preserved. So I've added code to the unroller to *actually* simplify the
surrounding loop when it succeeds at unrolling.
The only functional change in the test suite is that we now catch a case
that was previously missed because SCEV and other loop transforms see
their containing loops as simplified and thus don't miss some
opportunities. One test case has been converted to check that we catch
this case rather than checking that we miss it but at least don't get
the wrong answer.
Note that I have #if-ed out all of the verification logic in
LoopSimplify! This is a temporary workaround while extracting these bits
from the LoopPassManager. Currently, there is no way to have a pass in
the LoopPassManager which preserves LoopSimplify along with one which
does not. The LPM will try to verify on each loop in the nest that
LoopSimplify holds but the now-Function-pass cannot distinguish what
loop is being verified and so must try to verify all of them. The inner
most loop is clearly no longer simplified as there is a pass which
didn't even *attempt* to preserve it. =/ Once I get LCSSA out (and maybe
LoopVectorize and some other fixes) I'll be able to re-enable this check
and catch any places where we are still failing to preserve
LoopSimplify. If this causes problems I can back this out and try to
commit *all* of this at once, but so far this seems to work and allow
much more incremental progress.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199884
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 11:14:00 +0000 (11:14 +0000)]
Prospective Makefile build fix
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199882
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Thu, 23 Jan 2014 11:04:42 +0000 (11:04 +0000)]
Refactor lli-child-target to remove duplicated code
Eliminate the copies LLVM's System mmap and cache invalidation code. These were
slowly drifting away from the original version, and moreover the copied code
was a dead end in terms of portability.
We now statically link to Support but in practice with stripping this adds next
to no weight to the resultant binary.
Also avoid installing lli-child-target to the user's $PATH. It's not meant to
be run directly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199881
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Sanders [Thu, 23 Jan 2014 10:31:31 +0000 (10:31 +0000)]
[mips][sched] Split IIStore into II_S[BHWD], II_S[WD][LR], and II_SAVE
No functional change since the InstrItinData's have been duplicated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199876
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 23 Jan 2014 06:47:25 +0000 (06:47 +0000)]
Add a variable to track whether or not we've used a unique section,
e.g. linkonce, to TargetMachine and set it when we've done so
for ELF targets currently. This involved making TargetMachine
non-const in a TLOF use and propagating that change around - I'm
open to other ideas.
This will be used in a future commit to handle emitting debug
information with ranges.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199871
91177308-0d34-0410-b5e6-
96231b3b80d8
Hao Liu [Thu, 23 Jan 2014 02:09:30 +0000 (02:09 +0000)]
[AArch64]Add CHECK for two test cases testing scalar_to_vector committed in r199461.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199861
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Qin [Thu, 23 Jan 2014 01:35:13 +0000 (01:35 +0000)]
fix some spell mistakes around 'ConcatVector' and 'ShuffleVector' in AArch64 backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199858
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Thu, 23 Jan 2014 00:37:25 +0000 (00:37 +0000)]
X86Disassembler.cpp: Fix @param introduced in r199804. [-Wdocumentation]
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199855
91177308-0d34-0410-b5e6-
96231b3b80d8
Jack Carter [Wed, 22 Jan 2014 23:31:38 +0000 (23:31 +0000)]
[Mips] formatting through clang-format
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199853
91177308-0d34-0410-b5e6-
96231b3b80d8
Jack Carter [Wed, 22 Jan 2014 23:08:42 +0000 (23:08 +0000)]
[Mips] TargetStreamer Support for .set mips16.
This patch updates .set mips16 support which
affects the ELF ABI and its flags. In addition the patch uses
a common interface for both the MipsTargetSteamer and
MipsObjectStreamer that the assembler uses for
both ELF and ASCII output for these directives.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199851
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Wed, 22 Jan 2014 22:34:17 +0000 (22:34 +0000)]
Revert r162101 and replace it with a solution that works for targets where the pointer type is illegal.
This is a horrible bit of code. We're calling a simplification routine *in the middle* of type legalization. We tell the
simplification routine that it's running after legalization, but some of the types it will encounter will be illegal! The
fix is only to invoke the simplification if the types in question were legal, so that none of its invariants will be violated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199847
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Wed, 22 Jan 2014 22:32:58 +0000 (22:32 +0000)]
Add CHECK-LABELs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199846
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 22:20:54 +0000 (22:20 +0000)]
Revert "R600: Add work-around for the CF stack entry HW bug"
This reverts commit
35b8331cad6eb512a2506adbc394201181da94ba.
The -debug-only flag for llc doesn't appear to be available in
all build configurations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199845
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Wed, 22 Jan 2014 22:17:51 +0000 (22:17 +0000)]
Tweak r199835 to use can_execute() instead of exists()
The execution code path crashes if it can't execute the binary so we might as
well take precautions here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199844
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 22 Jan 2014 22:11:14 +0000 (22:11 +0000)]
Provide a dummy section to fix a crash with inline assembly in LTO.
Fixes pr18508.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199843
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:46 +0000 (21:55 +0000)]
R600: Add work-around for the CF stack entry HW bug
The CF stack can be corrupted if you use CF_ALU_PUSH_BEFORE,
CF_ALU_ELSE_AFTER, CF_ALU_BREAK, or CF_ALU_CONTINUE when the number of
sub-entries on the stack is greater than or equal to the stack entry
size and sub-entries modulo 4 is either 0 or 3 (on cedar the bug is
present when number of sub-entries module 8 is either 7 or 0)
We choose to be conservative and always apply the work-around when the
number of sub-enries is greater than or equal to the stack entry size,
so that we can safely over-allocate the stack when we are unsure of the
stack allocation rules.
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199842
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:44 +0000 (21:55 +0000)]
R600: Add some missing CF instruction definitions to the .td files.
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199841
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:43 +0000 (21:55 +0000)]
R600: Refactor stack size calculation
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199840
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:41 +0000 (21:55 +0000)]
R600: CF_PUSH is the same on Evergreen and Cayman
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199839
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:40 +0000 (21:55 +0000)]
R600: Add wavefront size property to the subtargets v2
v2:
- Initialize wavefront size to 0
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199838
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 21:55:35 +0000 (21:55 +0000)]
R600: Add stack size to .AMDGPUcsdata section
reviewed-by: Vincent Lejeune <vljn at ovi.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199837
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Wed, 22 Jan 2014 21:53:19 +0000 (21:53 +0000)]
Handle an addrspacecast case in memcpyopt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199836
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Wed, 22 Jan 2014 21:52:35 +0000 (21:52 +0000)]
Eliminate inappropriate use of FindProgramByName() from lli
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199835
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Wed, 22 Jan 2014 20:30:16 +0000 (20:30 +0000)]
Get right cost for addrspacecast in cost model
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199833
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 22 Jan 2014 20:20:52 +0000 (20:20 +0000)]
Fix pr18515.
My understanding (from reading just the llvm code) is that
* most ppc cpus have a "sync n" instruction and an msync alias that is "sync 0".
* "book e" cpus instead have a msync instruction and not the more
general "sync n"
This patch reflects that in the .td files, allowing a single codepath for
asm ond obj streamer and incidentelly fixes a crash when EmitRawText was
called on a obj streamer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199832
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 22 Jan 2014 20:11:50 +0000 (20:11 +0000)]
Add a testcase for r199430.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199831
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 19:24:24 +0000 (19:24 +0000)]
R600: MOVA is vector only
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199827
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 19:24:23 +0000 (19:24 +0000)]
R600: Take alignment into account when calculating the stack offset
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199826
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 19:24:21 +0000 (19:24 +0000)]
R600: Add support for global addresses with constant initializers
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199825
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 19:24:19 +0000 (19:24 +0000)]
R600: Begin private memory at the second GPR.
This way private memory does not over-write work group information
stored in GPRs 0 and 1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199824
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Wed, 22 Jan 2014 19:24:14 +0000 (19:24 +0000)]
R600/SI: Add support for i8 and i16 private loads/stores
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199823
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Wed, 22 Jan 2014 19:21:33 +0000 (19:21 +0000)]
Bug 18228 - Fix accepting bitcasts between vectors of pointers with a
different number of elements.
Bitcasts were passing with vectors of pointers with different number of
elements since the number of elements was checking
SrcTy->getVectorNumElements() == SrcTy->getVectorNumElements() which
isn't helpful. The addrspacecast was also wrong, but that case at least
is caught by the verifier. Refactor bitcast and addrspacecast handling
in castIsValid to be more readable and fix this problem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199821
91177308-0d34-0410-b5e6-
96231b3b80d8
Greg Fitzgerald [Wed, 22 Jan 2014 18:32:35 +0000 (18:32 +0000)]
Fix inline assembly that switches between ARM and Thumb modes
This patch restores the ARM mode if the user's inline assembly
does not. In the object streamer, it ensures that instructions
following the inline assembly are encoded correctly and that
correct mapping symbols are emitted. For the asm streamer, it
emits a .arm or .thumb directive.
This patch does not ensure that the inline assembly contains
the ADR instruction to switch modes at runtime.
The problem we need to solve is code like this:
int foo(int a, int b) {
int r = a + b;
asm volatile(
".align 2 \n"
".arm \n"
"add r0,r0,r0 \n"
: : "r"(r));
return r+1;
}
If we compile this function in thumb mode then the inline assembly
will switch to arm mode. We need to make sure that we switch back to
thumb mode after emitting the inline assembly or we will incorrectly
encode the instructions that follow (i.e. the assembly instructions
for return r+1).
Based on patch by David Peixotto
Change-Id: Ib57f6d2d78a22afad5de8693fba6230ff56ba48b
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199818
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 22 Jan 2014 16:43:45 +0000 (16:43 +0000)]
Don't open or fstat files twice in llvm-ar.
We still read/mmap them twice, but the fix for that is a bit more complex.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199815
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Wed, 22 Jan 2014 16:22:17 +0000 (16:22 +0000)]
Remove param doxygen comment for non-existing parameter.
Found by -Wdocumentation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199814
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 22 Jan 2014 16:04:52 +0000 (16:04 +0000)]
Pass the computed magic to createBinary and createObjectFile if available.
identify_magic is not free, so we should avoid calling it twice. The argument
also makes it cheap for createBinary to just forward to createObjectFile.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199813
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:31:32 +0000 (15:31 +0000)]
[x86] Silence unused diReg variable warning in non-asserting builds
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199812
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:31:29 +0000 (15:31 +0000)]
[x86] Fix uninitialized variable warning in translate{Src,Dst}Index
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199811
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:58 +0000 (15:08 +0000)]
[x86] Remove now-unused isSrcOp() and isDstOp() from X86AsmParser
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199810
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:55 +0000 (15:08 +0000)]
[x86] Allow segment and address-size overrides for INS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199809
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:49 +0000 (15:08 +0000)]
[x86] Allow segment and address-size overrides for OUTS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199808
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:42 +0000 (15:08 +0000)]
[x86] Allow segment and address-size overrides for MOVS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199807
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:36 +0000 (15:08 +0000)]
]x86] Allow segment and address-size overrides for CMPS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199806
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:27 +0000 (15:08 +0000)]
[x86] Allow address-size overrides for SCAS{8,16,32,64} (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199805
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:21 +0000 (15:08 +0000)]
[x86] Allow address-size overrides for STOS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199804
91177308-0d34-0410-b5e6-
96231b3b80d8
David Woodhouse [Wed, 22 Jan 2014 15:08:08 +0000 (15:08 +0000)]
[x86] Allow segment and address-size overrides for LODS[BWLQ] (PR9385)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199803
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Wed, 22 Jan 2014 13:27:00 +0000 (13:27 +0000)]
Loop strength reduce: fix function name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199801
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Wed, 22 Jan 2014 12:26:19 +0000 (12:26 +0000)]
AVX512: combining setcc and zext is wrong on AVX512
because vector compare instruction puts result in mask register.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199798
91177308-0d34-0410-b5e6-
96231b3b80d8
James Molloy [Wed, 22 Jan 2014 09:12:27 +0000 (09:12 +0000)]
MachineCopyPropagation has special logic for removing COPY instructions. It will remove plain COPYs using eraseFromParent(), but if the COPY has imp-defs/imp-uses it will convert it to a KILL, to keep the imp-def around.
This actually totally breaks and causes the machine verifier to cry in several cases, one of which being:
%RAX<def> = COPY %RCX<kill>
%ECX<def> = COPY %EAX<kill>, %RAX<imp-use,kill>
These subregister copies are together identified as noops, so are both removed. However, the second one as it has an imp-use gets converted into a kill:
%ECX<def> = KILL %EAX<kill>, %RAX<imp-use,kill>
As the original COPY has been removed, the verifier goes into tears at the use of undefined EAX and RAX.
There are several hacky solutions to this hacky problem (which is all to do with imp-use/def weirdnesses), but the least hacky I've come up with is to *always* remove COPYs by converting to KILLs. KILLs are no-ops to the code generator so the generated code doesn't change (which is why they were partially used in the first place), but using them also keeps the def/use and imp-def/imp-use chains alive:
%RAX<def> = KILL %RCX<kill>
%ECX<def> = KILL %EAX<kill>, %RAX<imp-use,kill>
The patch passes all test cases including the ones that check the removal of MOVs in this circumstance, along with an extra test I added to check subregister behaviour (which made the machine verifier fall over before my patch).
The patch also adds some DEBUG() statements because the file hadn't got any.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199797
91177308-0d34-0410-b5e6-
96231b3b80d8
Alp Toker [Wed, 22 Jan 2014 07:28:49 +0000 (07:28 +0000)]
Add unused result attr to the casting templates
This helped catch a couple of bugs locally.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199793
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Qin [Wed, 22 Jan 2014 06:11:03 +0000 (06:11 +0000)]
[AArch64 NEON] Try to generate CONCAT_VECTOR when lowering BUILD_VECTOR or SHUFFLE_VECTOR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199791
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Wed, 22 Jan 2014 03:38:55 +0000 (03:38 +0000)]
Reformat a loop for basic hygeine. Self review.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199788
91177308-0d34-0410-b5e6-
96231b3b80d8
Venkatraman Govindaraju [Wed, 22 Jan 2014 03:18:42 +0000 (03:18 +0000)]
[Sparc] Add support for inline assembly constraints which specify registers by their aliases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199786
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Wed, 22 Jan 2014 03:12:43 +0000 (03:12 +0000)]
Whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199785
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Wed, 22 Jan 2014 02:38:23 +0000 (02:38 +0000)]
Fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199784
91177308-0d34-0410-b5e6-
96231b3b80d8
Venkatraman Govindaraju [Wed, 22 Jan 2014 01:29:51 +0000 (01:29 +0000)]
[Sparc] Add support for inline assembly constraint 'I'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199781
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 22 Jan 2014 00:14:49 +0000 (00:14 +0000)]
Change createObjectFile to return an ErrorOr.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199776
91177308-0d34-0410-b5e6-
96231b3b80d8
Venkatraman Govindaraju [Wed, 22 Jan 2014 00:13:18 +0000 (00:13 +0000)]
[Sparc] Do not add PC to _GLOBAL_OFFSET_TABLE_ address to access GOT in absolute code.
Fixes PR#18521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199775
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Tue, 21 Jan 2014 23:16:05 +0000 (23:16 +0000)]
[SROA] Fix a bug which could cause the common type finding to return
inconsistent results for different orderings of alloca slices. The
fundamental issue is that it is just always a mistake to return early
from this function. There is no effective early exit to leverage. This
patch stops trynig to do so and simplifies the code a bit as
a consequence.
Original diagnosis and patch by James Molloy with some name tweaks by me
in part reflecting feedback from Duncan Smith on the mailing list.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199771
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 21 Jan 2014 23:06:54 +0000 (23:06 +0000)]
Be a bit more consistent about using ErrorOr when constructing Binary objects.
The constructors of classes deriving from Binary normally take an error_code
as an argument to the constructor. My original intent was to change them
to have a trivial constructor and move the initial parsing logic to a static
method returning an ErrorOr. I changed my mind because:
* A constructor with an error_code out parameter is extremely convenient from
the implementation side. We can incrementally construct the object and give
up when we find an error.
* It is very efficient when constructing on the stack or when there is no
error. The only inefficient case is where heap allocating and an error is
found (we have to free the memory).
The result is that this is a much smaller patch. It just standardizes the
create* helpers to return an ErrorOr.
Almost no functionality change: The only difference is that this found that
we were trying to read past the end of COFF import library but ignoring the
error.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199770
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan P. N. Exon Smith [Tue, 21 Jan 2014 22:46:46 +0000 (22:46 +0000)]
CodeGen: Stop treating vectors as aggregates
Fix a crash in SjLjEHPrepare::lowerIncomingArguments caused by treating
VectorType like an aggregate. It's first-class!
<rdar://problem/
15854596>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199768
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Tue, 21 Jan 2014 22:39:19 +0000 (22:39 +0000)]
Tweak the spelling of the asserts requirement a bit more. This makes it
match the (reasonably prevelant) usage in Clang's test suite and so
seems more "canonical".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199767
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Tue, 21 Jan 2014 21:27:37 +0000 (21:27 +0000)]
Fix PR18572 - llc crash during GenericScheduler::initPolicy().
Generalized the heuristic that looks at the (very rough) size of the
register file before enabling regpressure tracking.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@199766
91177308-0d34-0410-b5e6-
96231b3b80d8