Tim Northover [Wed, 19 Feb 2014 12:24:19 +0000 (12:24 +0000)]
X86: move test requiring X86TargetLowering info into its own directory
If LLVM is built without X86 as a supported target then the test would
mysteriously fail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201668
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Wed, 19 Feb 2014 10:32:40 +0000 (10:32 +0000)]
Try addding datalayout in case that's what Hexagon doesn't like.
Just a wild stab in the dark really, but in the absence of any ability to
reproduce the problem...
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201658
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Wed, 19 Feb 2014 10:02:43 +0000 (10:02 +0000)]
X86 CodeGenPrep: sink shufflevectors before shifts
On x86, shifting a vector by a scalar is significantly cheaper than shifting a
vector by another fully general vector. Unfortunately, because SelectionDAG
operates on just one basic block at a time, the shufflevector instruction that
reveals whether the right-hand side of a shift *is* really a scalar is often
not visible to CodeGen when it's needed.
This adds another handler to CodeGenPrepare, to sink any useful shufflevector
instructions down to the basic block where they're used, predicated on a target
hook (since on other architectures, doing so will often just introduce extra
real work).
rdar://problem/
16063505
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201655
91177308-0d34-0410-b5e6-
96231b3b80d8
Evgeniy Stepanov [Wed, 19 Feb 2014 09:51:22 +0000 (09:51 +0000)]
Build PIE binaries when cross-compiling to Android.
This change also removes CMAKE_LINK_FLAGS setting that seems to be ignored by cmake.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201654
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexey Samsonov [Wed, 19 Feb 2014 08:30:55 +0000 (08:30 +0000)]
Try to revive buildbots after r201620
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201651
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Wed, 19 Feb 2014 08:25:02 +0000 (08:25 +0000)]
Remove special FP opcode maps and instead add enough MRM_XX formats to handle all the FP operations. This increases format by 1 bit, but decreases opcode map by 1 bit so the TSFlags size doesn't change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201649
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Wed, 19 Feb 2014 07:29:07 +0000 (07:29 +0000)]
Reduce size of map field in X86 TSFlags since it now requires less bits.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201646
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Wed, 19 Feb 2014 06:59:13 +0000 (06:59 +0000)]
Put some of the X86 formats in a more logical order.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201645
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Wed, 19 Feb 2014 05:34:21 +0000 (05:34 +0000)]
Remove A6/A7 opcode maps. They can all be handled with a TB map, opcode of 0xa6/0xa7, and adding MRM_C0/MRM_E0 forms. Removes 376K from the disassembler tables.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201641
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Wed, 19 Feb 2014 03:53:11 +0000 (03:53 +0000)]
llvm-objdump/COFF: Print load configuration table.
Load Configuration Table may contain a pointer to SEH table. This patch is to
print the offset to the table. Printing SEH table contents is a TODO.
The layout of Layout Configuration Table is described in Microsoft PE/COFF
Object File Format Spec, but the table's offset/size descriptions seems to be
totally wrong, at least in revision 8.3 of the spec. I believe the table in
this patch is the correct one.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201638
91177308-0d34-0410-b5e6-
96231b3b80d8
Mingjie Xing [Wed, 19 Feb 2014 03:20:46 +0000 (03:20 +0000)]
Fix a typo in a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201633
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Wed, 19 Feb 2014 03:00:29 +0000 (03:00 +0000)]
MCAsmParser: support required parameters
This enhances the macro parser to parse and handle parameter qualifications,
which is needed to support required formal parameters in macro definitions. A
required parameter may not be defaulted (though providing a default value is
accepted with a warning). This improves GAS compatibility.
Partially addresses PR9248.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201630
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Wed, 19 Feb 2014 03:00:23 +0000 (03:00 +0000)]
MCAsmParser: change representation of MCAsmMacroParameter
Rather than using std::pair, create a structure to represent the type. This is
a preliminary refactoring to enable required parameter handling. Additional
state is needed to indicate required parameters. This has a minor side effect
of improving readability by providing more accurate names compared to first and
second.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201629
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 19 Feb 2014 02:08:39 +0000 (02:08 +0000)]
Now that llvm always does the right thing with private, use it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201625
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 19 Feb 2014 01:46:57 +0000 (01:46 +0000)]
Use PrivateLinkage now that it is safe.
Now that llvm's codegen knows to use an 'l' prefix when needed, we can just
use PrivateLinkage.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201624
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Wed, 19 Feb 2014 01:28:30 +0000 (01:28 +0000)]
Avoid an infinite cycle with private linkage and -f{data|function}-sections.
When outputting an object we check its section to find its name, but when
looking for the section with -ffunction-section we look for the symbol name.
Break the loop by requesting a name with the private prefix when constructing
the section name. This matches the behavior before r201608.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201622
91177308-0d34-0410-b5e6-
96231b3b80d8
Sean Silva [Wed, 19 Feb 2014 00:12:34 +0000 (00:12 +0000)]
[docs] Clean up some more llvm-gcc stuff
Some references to llvm-gcc were so crusty that I wasn't sure how to
proceed and so I've left them intact.
I also slipped in a quick peephole fix to use a :doc: link instead of
raw HTML link.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201619
91177308-0d34-0410-b5e6-
96231b3b80d8
Sean Silva [Tue, 18 Feb 2014 23:56:43 +0000 (23:56 +0000)]
[docs] Nuke some references to llvm-gcc
From a cursory look it seems like all the described commandline options
and such apply to clang just fine, but I'd appreciate a second opinion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201616
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 18 Feb 2014 22:24:57 +0000 (22:24 +0000)]
Fix PR18743.
The IR
@foo = private constant i32 42
is valid, but before this patch we would produce an invalid MachO from it. It
was invalid because it would use an L label in a section where the liker needs
the labels in order to atomize it.
One way of fixing it would be to just reject this IR in the backend, but that
would not be very front end friendly.
What this patch does is use an 'l' prefix in sections that we know the linker
requires symbols for atomizing them. This allows frontends to just use
private and not worry about which sections they go to or how the linker handles
them.
One small issue with this strategy is that now a symbol name depends on the
section, which is not available before codegen. This is not a problem in
practice. The reason is that it only happens with private linkage, which will
be ignored by the non codegen users (llvm-nm and llvm-ar).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201608
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 18 Feb 2014 22:05:46 +0000 (22:05 +0000)]
Rename a DebugLoc variable to DbgLoc and a DataLayout to DL.
This is quiet a bit less confusing now that TargetData was renamed DataLayout.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201606
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Tue, 18 Feb 2014 21:46:39 +0000 (21:46 +0000)]
Consistently check 'IsCode' when allocating sections in RuntimeDyld (via
findOrEmitSection).
Vaidas Gasiunas's patch, r201259, fixed one instance where we were always
allocating sections as text. This patch fixes the remaining buggy call sites.
No test case: This isn't breaking anything that I know of, it's just
inconsistent.
<rdar://problem/
15943542>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201605
91177308-0d34-0410-b5e6-
96231b3b80d8
Ana Pazos [Tue, 18 Feb 2014 20:31:05 +0000 (20:31 +0000)]
[AArch64] Expanded sin, cos, pow with FP vector types inputs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201601
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Sanders [Tue, 18 Feb 2014 15:57:52 +0000 (15:57 +0000)]
[mips] Add support for ELF64-mips and the R_MIPS_32/R_MIPS_64 relocs for it.
Summary:
This fixes several test failures when building LLVM on a MIPS host.
The failures were:
LLVM :: DebugInfo/enum.ll
LLVM :: DebugInfo/inlined-arguments.ll
LLVM :: DebugInfo/member-order.ll
LLVM :: DebugInfo/namespace.ll
LLVM :: DebugInfo/template-recursive-void.ll
LLVM :: DebugInfo/tu-composite.ll
LLVM :: DebugInfo/two-cus-from-same-file.ll
LLVM :: Linker/type-unique-simple-a.ll
LLVM :: Linker/type-unique-simple2.ll
Reviewers: jacksprat, matheusalmeida
Reviewed By: matheusalmeida
Differential Revision: http://llvm-reviews.chandlerc.com/D2721
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201582
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Tue, 18 Feb 2014 15:33:12 +0000 (15:33 +0000)]
Rename some member variables from TD to DL.
TargetData was renamed DataLayout back in r165242.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201581
91177308-0d34-0410-b5e6-
96231b3b80d8
Marshall Clow [Tue, 18 Feb 2014 14:03:17 +0000 (14:03 +0000)]
Add myself as owner for libc++
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201573
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Lytton [Tue, 18 Feb 2014 11:21:59 +0000 (11:21 +0000)]
XCore target: Handle common linkage
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201563
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Lytton [Tue, 18 Feb 2014 11:21:53 +0000 (11:21 +0000)]
XCore target: addMemOperand as necessary
BuildMI instructions were not including MachineMemOperand information.
This was discovered by 'SingleSource/Benchmarks/Stanford/Oscar' failing
due to a FrameIndex load incorrectly being hoisted by postra-machine-licm.
No other tests have been found to fail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201562
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Lytton [Tue, 18 Feb 2014 11:21:48 +0000 (11:21 +0000)]
XCore target: Fix llvm.eh.return and EH info register handling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201561
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Tue, 18 Feb 2014 11:20:44 +0000 (11:20 +0000)]
Darwin builds: handle different possible form for SDKROOT.
Modifying build_llvm to handle SDKROOT being the name of an SDK rather than a
path. This will still work if SDKROOT is a path.
rdar://problem/
15162322
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201560
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Tue, 18 Feb 2014 11:17:29 +0000 (11:17 +0000)]
GlobalMerge: move "-global-merge" option to the pass itself.
It's rather odd to have the flag enabling and disabling this pass only affect a
single target.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201559
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Northover [Tue, 18 Feb 2014 11:15:32 +0000 (11:15 +0000)]
X86: use vpsllvd (& friends) for 16-bit shifts on Haswell
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201558
91177308-0d34-0410-b5e6-
96231b3b80d8
Justin Bogner [Tue, 18 Feb 2014 09:19:48 +0000 (09:19 +0000)]
llvm-cov: Support gcov's extermely lenient treatment of -o
In gcov, the -o flag can accept either a directory or a file name.
When given a directory, the gcda and gcno files are expected to be in
that directory. When given a file, the gcda and gcno files are
expected to be named based on the stem of that file. Non-existent
paths are treated as files.
This implements compatible behaviour.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201555
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 18 Feb 2014 08:24:22 +0000 (08:24 +0000)]
Add PS prefix to some classes I missed in r201538.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201551
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 18 Feb 2014 08:18:29 +0000 (08:18 +0000)]
Add a bunch of OpSize32 tags to 64-bit mode only instructions to match their 32-bit mode counterparts for cases where there is also a OpSize16 instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201550
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Tue, 18 Feb 2014 07:52:26 +0000 (07:52 +0000)]
AVX-512: Fixed size of mask registers
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201546
91177308-0d34-0410-b5e6-
96231b3b80d8
Jiangning Liu [Tue, 18 Feb 2014 02:37:42 +0000 (02:37 +0000)]
Fix a typo about lowering AArch64 va_copy.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201541
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 18 Feb 2014 00:21:49 +0000 (00:21 +0000)]
Add an x86 prefix encoding for instructions that would decode to a different instruction with 0xf2/f3/66 were in front of them, but don't themselves have a prefix. For now this doesn't change any bbehavior, but plan to use it to fix some bugs in the disassembler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201538
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan P. N. Exon Smith [Mon, 17 Feb 2014 23:22:49 +0000 (23:22 +0000)]
PGO: llvm-profdata: tool for merging profiles
Introducing llvm-profdata, a tool for merging profile data generated by
PGO instrumentation in clang.
- The name indicates a file extension of <name>.profdata. Eventually
profile data output by clang should be changed to that extension.
- llvm-profdata merges two profiles. However, the name is more general,
since it will likely pick up more tasks (such as summarizing a single
profile).
- llvm-profdata parses the current text-based format, but will be
updated once we settle on a binary format.
<rdar://problem/
15949645>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201535
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Enderby [Mon, 17 Feb 2014 21:45:27 +0000 (21:45 +0000)]
Fix the arm assembler so that this malformed instruction:
ldrd r6, r7 [r2, #15]
simply gives an error and does not triggers an assertion.
As Jim points out, the diagnostic is really strange here,
but fixing that would be more complicated. The missing
comma results in the parser expecting a construct like r2[2],
which is the vector index thing the error message is talking
about. That's not what the user intended, though, and there's
nothing else in the instruction that looks at all like a vector.
Yet more fallout from not having a real parser here and trying
to do context-free generic matching for addressing modes.
rdar://
15097243
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201531
91177308-0d34-0410-b5e6-
96231b3b80d8
Anders Waldenborg [Mon, 17 Feb 2014 20:48:32 +0000 (20:48 +0000)]
Add support for assigning to . in AsmParser.
This is implemented by handling assignments to the '.' pseudo symbol
as ".org" directives.
Differential Revision: http://llvm-reviews.chandlerc.com/D2625
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201530
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 17 Feb 2014 10:03:43 +0000 (10:03 +0000)]
Fix diassembler handling of rex.b when mod=00/01/10 and bbb=101. Mod=00 should ignore the base register entirely. Mod=01/10 should treat this as R13 plus displacment. Fixes PR18860.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201507
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Mon, 17 Feb 2014 07:29:33 +0000 (07:29 +0000)]
AVX-512: implemented zext fron i1 to i16
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201502
91177308-0d34-0410-b5e6-
96231b3b80d8
Gerolf Hoflehner [Mon, 17 Feb 2014 03:06:16 +0000 (03:06 +0000)]
fix for null VectorizedValue assertion in the SLP Vectorizer (in function vectorizeTree()). radar://
16064178
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201501
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Mon, 17 Feb 2014 00:40:22 +0000 (00:40 +0000)]
MCAsmParser: add some mixed argument tests
Add some tests to explicitly validate handling of comma and non-comma separated
arguments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201500
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Mon, 17 Feb 2014 00:40:17 +0000 (00:40 +0000)]
MCAsmParser: better handling for named arguments
Until this point only macro definition with named parameters were parsed but the
names were ignored. This adds support for using that information for named
parameter instantiation.
In order to support the full semantics of the keyword arguments, the arguments
are no longer lazily initialised since the keyword arguments can be specified
out of order and partially if they are defaulted. Prepopulate the arguments
with the default value for any defaulted parameters, and then parse the
specified arguments.
This simplies some of the handling of the arguments in the inner loop since
empty arguments simply increment the parameter index and move on.
Note that keyword and positional arguments cannot be mixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201499
91177308-0d34-0410-b5e6-
96231b3b80d8
Mark Seaborn [Sun, 16 Feb 2014 18:59:48 +0000 (18:59 +0000)]
Use 16 byte stack alignment for NaCl on ARM
NaCl's ARM ABI uses 16 byte stack alignment, so set that in
ARMSubtarget.cpp.
Using 16 byte alignment exposes an issue in code generation in which a
varargs function leaves a 4 byte gap between the values of r1-r3 saved
to the stack and the following arguments that were passed on the
stack. (Previously, this code only needed to support 4 byte and 8
byte alignment.)
With this issue, llc generated:
varargs_func:
sub sp, sp, #16
push {lr}
sub sp, sp, #12
add r0, sp, #16 // Should be 20
stm r0, {r1, r2, r3}
ldr r0, .LCPI0_0 // Address of va_list
add r1, sp, #16
str r1, [r0]
bl external_func
Fix the bug by checking for "Align > 4". Also simplify the code by
using OffsetToAlignment(), and update comments.
Differential Revision: http://llvm-reviews.chandlerc.com/D2677
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201497
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sun, 16 Feb 2014 15:49:50 +0000 (15:49 +0000)]
SCEVExpander: Try hard not to create derived induction variables in other loops
During LSR of one loop we can run into a situation where we have to expand the
start of a recurrence of a loop induction variable in this loop. This start
value is a value derived of the induction variable of a preceeding loop. SCEV
has cannonicalized this value to a different recurrence than the recurrence of
the preceeding loop's induction variable (the type and/or step direction) has
changed). When we come to instantiate this SCEV we created a second induction
variable in this preceeding loop. This patch tries to base such derived
induction variables of the preceeding loop's induction variable.
This helps twolf on arm and seems to help scimark2 on x86.
Reapply with a fix for the case of a value derived from a pointer.
radar://
15970709
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201496
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Sun, 16 Feb 2014 14:36:26 +0000 (14:36 +0000)]
Remove dead code, we already require cmake 2.8.8.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201495
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Sun, 16 Feb 2014 14:12:35 +0000 (14:12 +0000)]
Remove unnecessary typename.
Thanks to Elena Demikhovsky for noticing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201494
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sun, 16 Feb 2014 13:28:39 +0000 (13:28 +0000)]
Fix more broken CHECK lines
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201493
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sun, 16 Feb 2014 13:28:15 +0000 (13:28 +0000)]
Add extra CHECK prefix to tests with explicit prefix
These tests mistakenly assume that CHECK is still available even if an
explicit prefix is specified.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201492
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sun, 16 Feb 2014 13:27:39 +0000 (13:27 +0000)]
Actually call FileCheck in tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201491
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Sun, 16 Feb 2014 12:14:24 +0000 (12:14 +0000)]
llvmbuild: Exclude disabled targets from LLVMExports.cmake.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201490
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Sun, 16 Feb 2014 12:14:13 +0000 (12:14 +0000)]
LLVMExports.cmake: Add System libs $(LIBS) to LLVMSupport, corresponding to r201077.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201489
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Sun, 16 Feb 2014 12:14:03 +0000 (12:14 +0000)]
LLVMExports.cmake: Exclude gtest since they are not installed.
FIXME: gtest might be included if external project could refer to the build tree.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201488
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Sun, 16 Feb 2014 11:34:23 +0000 (11:34 +0000)]
AVX-512: simpyfied BUILD_VECTOR for masks; fixed cmp/test sequence
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201487
91177308-0d34-0410-b5e6-
96231b3b80d8
Gerolf Hoflehner [Sun, 16 Feb 2014 10:43:25 +0000 (10:43 +0000)]
fixed typo in comment as my test commit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201486
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Sun, 16 Feb 2014 08:46:55 +0000 (08:46 +0000)]
Add a DIELoc class to cover the DW_FORM_exprloc set of expressions
alongside DIEBlock and replace uses accordingly. Use DW_FORM_exprloc
in DWARF4 and later code. Update testcases.
Adding a DIELoc instead of using extra forms inside DIEBlock so
that we can keep location expressions separate from other uses. No
direct use at the moment, however, it's not a lot of code and
using a separately named class keeps it somewhat more obvious
what's going on in various locations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201481
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sun, 16 Feb 2014 07:31:05 +0000 (07:31 +0000)]
Fix broken CHECK lines
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201479
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Sun, 16 Feb 2014 04:56:31 +0000 (04:56 +0000)]
MCAsmParser: relax declaration parsing
The Linux kernel defines empty macros for compatibility with ARM UAL syntax.
The comma after the name is optional, and if present can be safely lexed. This
improves compatibility with the GNU assembler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201474
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Sun, 16 Feb 2014 00:16:41 +0000 (00:16 +0000)]
ARM IAS: (partially) support .arch_extension directive
This adds a partial implementation of the .arch_extension directive to the
integrated ARM assembler. There are a number of limitations to this
implementation arising from the target backend support rather than the
implementation itself. Namely, iWMMXT (v1 and v2), Maverick, and XScale support
is not present in the ARM backend. Currently, there is no check for A-class
only (needed for virt), and no ARMv6k detection (needed for os and sec). The
remainder of the extensions are fully supported.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201471
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Sat, 15 Feb 2014 19:34:03 +0000 (19:34 +0000)]
DebugInfo: Deduplicate entries in the fission address table
This broke in r185459 while TLS support was being generalized to handle
non-symbol TLS representations.
I thought about/tried having an enum rather than a bool to track the
TLS-ness of the address table entry, but namespaces and naming seemed
more hassle than it was worth for only one caller that needed to specify
this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201469
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Sat, 15 Feb 2014 18:33:11 +0000 (18:33 +0000)]
DwarfDebug: Remove dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201467
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sat, 15 Feb 2014 18:16:56 +0000 (18:16 +0000)]
Revert "SCEVExpander: Try hard not to create derived induction variables in other loops"
This reverts commit r201465. It broke an arm bot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201466
91177308-0d34-0410-b5e6-
96231b3b80d8
Arnold Schwaighofer [Sat, 15 Feb 2014 17:11:56 +0000 (17:11 +0000)]
SCEVExpander: Try hard not to create derived induction variables in other loops
During LSR of one loop we can run into a situation where we have to expand the
start of a recurrence of a loop induction variable in this loop. This start
value is a value derived of the induction variable of a preceeding loop. SCEV
has cannonicalized this value to a different recurrence than the recurrence of
the preceeding loop's induction variable (the type and/or step direction) has
changed). When we come to instantiate this SCEV we created a second induction
variable in this preceeding loop. This patch tries to base such derived
induction variables of the preceeding loop's induction variable.
This helps twolf on arm and seems to help scimark2 on x86.
radar://
15970709
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201465
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sat, 15 Feb 2014 08:35:56 +0000 (08:35 +0000)]
Cleanup docs about lit substitutions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201464
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sat, 15 Feb 2014 07:29:18 +0000 (07:29 +0000)]
Add opcode extension forms of MOV8ri/MOV16ri/MOV32ri.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201463
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Rieck [Sat, 15 Feb 2014 06:02:36 +0000 (06:02 +0000)]
Fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201461
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Sat, 15 Feb 2014 00:45:14 +0000 (00:45 +0000)]
Remove tautological test line (unsigneds are always >=0).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201451
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Sat, 15 Feb 2014 00:34:43 +0000 (00:34 +0000)]
Remove unnecessary typename introduced in r201440.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201450
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 23:58:13 +0000 (23:58 +0000)]
DebugInfo: Implement DW_AT_stmt_list for type units
Type units will share the statement list of their defining compile unit.
This is a tradeoff that reduces .o debug info size at the cost of some
linked debug info size (since the contents of those string tables won't
be deduplicated along with the type unit) which seems right for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201445
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 22:50:59 +0000 (22:50 +0000)]
DwarfUnit: Remove unnecessarily explicit/out of line virtual dtors.
These types have an out of line virtual function each (emitHeader at
least) so they won't have weak vtables - no need for more than that.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201444
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 14 Feb 2014 22:48:49 +0000 (22:48 +0000)]
Hopefully fix the build with gcc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201443
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 22:47:55 +0000 (22:47 +0000)]
DwarfUnit: Remove unnecessary (void)t; that was previously used to suppress -Wunused-member-variable
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201442
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 22:41:51 +0000 (22:41 +0000)]
DwarfUnit: Refactor out DW_AT_stmt_list creation into common function for fission and non-fission cases
This probably also addresses the FIXME in the fission case regarding
multiple compile units, though I haven't tested that.
This code still confuses me (the literal zero offset makes little sense,
the limitations surrounding asm output I'm not sure about either - but
perhaps we should just always emit one line table? Or should we not rely
on .loc/.file even in assembly so we can produce the same output between
asm and object output?) but this maintains the existing functionality.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201441
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 14 Feb 2014 22:36:16 +0000 (22:36 +0000)]
Add extern template instantiations of llvm::Calculate.
This should be a small build time improvement in general and fixes
the build on OS X with -DBUILD_SHARED_LIBS=ON.
The issue is that not all users are including GenericDomTreeConstruction.h,
causing undefined references when ld64 managed to hide the
linkonce_odr symbols.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201440
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Fri, 14 Feb 2014 22:23:22 +0000 (22:23 +0000)]
[CodeGenPrepare][AddressingModeMatcher] Give up on type promotion if the
transformation does not bring any immediate benefits and introduce an illegal
operation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201439
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Fri, 14 Feb 2014 21:10:34 +0000 (21:10 +0000)]
TargetLowering: n * r where n > 2 should be an illegal addressing mode
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201433
91177308-0d34-0410-b5e6-
96231b3b80d8
Howard Hinnant [Fri, 14 Feb 2014 21:09:01 +0000 (21:09 +0000)]
Remove myself as owner of libc++
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201432
91177308-0d34-0410-b5e6-
96231b3b80d8
Eli Bendersky [Fri, 14 Feb 2014 20:15:56 +0000 (20:15 +0000)]
Remove unused #include
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201430
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 19:51:35 +0000 (19:51 +0000)]
DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded
Recommitting r201380 (reverted in r201389)
Recommitting r201351 and r201355 (reverted in r201351 and r201355)
We weren't emitting the an empty (header only) line table when the line
table was empty - this made the DWARF invalid (the compile unit would
point to the zero-size debug_lines section where there should've been an
empty line table but there was nothing at all). Fix that, and as a
consequence this works around/addresses PR18809.
Also, we emit a non-empty line table to workaround a darwin linker bug,
so XFAILing on darwin too.
Also, mark the test as 'REQUIRES: object-emission' because it does.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201429
91177308-0d34-0410-b5e6-
96231b3b80d8
Diego Novillo [Fri, 14 Feb 2014 19:27:53 +0000 (19:27 +0000)]
Support DWARF discriminators in object streamer.
Summary:
This adds support for emitting DWARF path discriminator values in
the object streamer. It also changes the DWARF dumper to show
discriminator values in the line table output.
Reviewers: echristo
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2794
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201427
91177308-0d34-0410-b5e6-
96231b3b80d8
Reed Kotler [Fri, 14 Feb 2014 19:16:39 +0000 (19:16 +0000)]
This patch has two main functions:
1) Fix a specific bug when certain conversion functions are called in a program compiled as mips16 with hard float and
the program is linked as c++. There are two libraries that are reversed in the link order with gcc/g++ and clang/clang++ for
mips16 in this case and the proper stubs will then not be called. These stubs are normally handled in the Mips16HardFloat pass
but in this case we don't know at that time that we need to generate the stubs. This must all be handled later in code generation
and we have moved this functionality to MipsAsmPrinter. When linked as C (gcc or clang) the proper stubs are linked in from libc.
2) Set up the infrastructure to handle 90% of what is in the Mips16HardFloat pass in this new area of MipsAsmPrinter. This is a more
logical place to handle this and we have known for some time that we needed to move the code later and not implement it using
inline asm as we do now but it was not clear exactly where to do this and what mechanism should be used. Now it's clear to us
how to do this and this patch contains the infrastructure to move most of this to MipsAsmPrinter but the actual moving will be done
in a follow on patch. The same infrastructure is used to fix this current bug as described in #1. This change was requested by the list
during the original putback of the Mips16HardFloat pass but was not practical for us do at that time.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201426
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Fri, 14 Feb 2014 19:02:01 +0000 (19:02 +0000)]
Trivial cleanup: reuse existing variable.
Extracted while trying to understand http://llvm-reviews.chandlerc.com/D1764.
Patch by Matt Arsenault.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201425
91177308-0d34-0410-b5e6-
96231b3b80d8
Artyom Skrobov [Fri, 14 Feb 2014 17:19:07 +0000 (17:19 +0000)]
Generate the DWARF stack frame decode operations in the function prologue for ARM/Thumb functions.
Patch by Keith Walker!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201423
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Qin [Fri, 14 Feb 2014 09:41:15 +0000 (09:41 +0000)]
[AArch64 NEON] Fix a bug to avoid using floating type as condition type in lowering SELECT_CC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201395
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Fri, 14 Feb 2014 05:33:16 +0000 (05:33 +0000)]
Revert "DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded"
This reverts commit r201380 for now while we investigate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201389
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Fri, 14 Feb 2014 05:26:49 +0000 (05:26 +0000)]
llvm/test/DebugInfo/empty.ll: Mark it as XFAIL:win32 lacking of line table.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201388
91177308-0d34-0410-b5e6-
96231b3b80d8
Jiangning Liu [Fri, 14 Feb 2014 04:38:09 +0000 (04:38 +0000)]
Enable AArch64 NEON by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201385
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Fri, 14 Feb 2014 03:59:43 +0000 (03:59 +0000)]
[PR18809] Remove XFAIL from DebugInfo/empty.ll.
I added it in r201211.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201383
91177308-0d34-0410-b5e6-
96231b3b80d8
Hao Liu [Fri, 14 Feb 2014 02:21:56 +0000 (02:21 +0000)]
[AArch64]Fix the assertion failure caused by "v1i1 SETCC" DAG node.
As v1i1 is illegal, the type legalizer tries to scalarize such node. But if the type operands of SETCC is legal, the scalarization algorithm will cause an assertion failure.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201381
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Fri, 14 Feb 2014 01:57:59 +0000 (01:57 +0000)]
DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded
Recommitting r201351 and r201355 (reverted in r201351 and r201355)
We weren't emitting the an empty (header only) line table when the line
table was empty - this made the DWARF invalid (the compile unit would
point to the zero-size debug_lines section where there should've been an
empty line table but there was nothing at all). Fix that, and as a
consequence this works around/addresses PR18809.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201380
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Fri, 14 Feb 2014 01:26:55 +0000 (01:26 +0000)]
Disable emission of aranges by default and add a command line
option to enable again that will be matched with a commit to enable
in clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201378
91177308-0d34-0410-b5e6-
96231b3b80d8
Juergen Ributzka [Fri, 14 Feb 2014 00:51:13 +0000 (00:51 +0000)]
[X86] Don't mark movabsq as cheap-as-move - it isn't that cheap.
A simple register copy on X86 is just 3 bytes, whereas movabsq is a 10 byte
instruction. Marking movabsq as not beeing cheap will allow LICM to move it
out of the loop and it also prevents unnecessary rematerializations if the
value is needed in more than one register.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201377
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Fri, 14 Feb 2014 00:49:12 +0000 (00:49 +0000)]
Do more addrspacecast transforms that happen for bitcast.
Makes addrspacecast (gep) do addrspacecast (gep) instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201376
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 13 Feb 2014 23:34:15 +0000 (23:34 +0000)]
R600/SI: Expand all v8[if]32 operations
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201371
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 13 Feb 2014 23:34:13 +0000 (23:34 +0000)]
R600/SI: Add a pattern for i32 anyext
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201370
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 13 Feb 2014 23:34:12 +0000 (23:34 +0000)]
R600/SI: Completely Disable TypeRewriter on compute
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201369
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 13 Feb 2014 23:34:10 +0000 (23:34 +0000)]
R600/SI: Split global vector loads with more than 4 elements
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201368
91177308-0d34-0410-b5e6-
96231b3b80d8
Tom Stellard [Thu, 13 Feb 2014 23:34:07 +0000 (23:34 +0000)]
R600/SI: Add ShaderType attribute to some tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201367
91177308-0d34-0410-b5e6-
96231b3b80d8