Commit Graph

161247 Commits

Author SHA1 Message Date
Rafael Espindola 2e2995bf50 Produce direct calls instead of alias to linkonce_odr functions.
This is a small optimization on linux, but should help more on windows
where msvc only outputs one destructor if there would be two identical ones.

llvm-svn: 194095
2013-11-05 21:37:29 +00:00
Tim Northover f02287db27 ARM: permit bare dmb/dsb/isb aliases on Cortex-M0
Cortex-M0 supports these 32-bit instructions despite being Thumb1 only
(mostly). We knew about that but not that the aliases without the default "sy"
operand were also permitted.

llvm-svn: 194094
2013-11-05 21:36:02 +00:00
Dmitri Gribenko 75e12236cc Convert comments to documentation comments (// -> ///)
Patch by MathOnNapkins

llvm-svn: 194093
2013-11-05 21:28:42 +00:00
James Dennett 42c4e50f63 Documentation tweaks: Fix a typo in a reference to the C++ standard,
another typo in a comment, and update formatting of some comments to
use Doxygen \brief annotations instead of repeating the name of the
documented entity.

llvm-svn: 194092
2013-11-05 19:37:38 +00:00
Rafael Espindola 2b11ad4fe9 Use error_code in GVMaterializer.
They just propagate out the bitcode reader error, so we don't need a new enum.

llvm-svn: 194091
2013-11-05 19:36:34 +00:00
Daniel Jasper bbf5c1c965 clang-format: Allow line merging and partial formatting of nested blocks
Before, clang-format would always format entire nested blocks, which
can be unwanted e.g. for long DEBUG({...}) statements. Also
clang-format would not allow to merge lines in nested blocks (e.g. to
put "if (a) return;" on one line in Google style).

This is the first step of several refactorings mostly focussing on the
additional functionality (by reusing the "format many lines" code to
format the children of a nested block). The next steps are:
* Pull out the line merging into its own class.
* Seperate the formatting of many lines from the formatting of a single
  line (and the analysis of the solution space).

llvm-svn: 194090
2013-11-05 19:10:03 +00:00
Rui Ueyama aa1335990f Remove extraneous parentheses.
llvm-svn: 194089
2013-11-05 18:43:10 +00:00
Kaelyn Uhrain 89035721b9 Revert "Try to correct a mistyped "-" or ">" to "->" for some C++ cases."
Revert this patch until cases of rejected valid code (e.g. identifiers
that require ADL to be resolved properly) are fixed.

llvm-svn: 194088
2013-11-05 18:28:21 +00:00
Jim Ingham afbb0af827 Give a better error when the index argument for “frame select” can’t be parsed.
<rdar://problem/15390829>

llvm-svn: 194087
2013-11-05 18:25:23 +00:00
Jiangning Liu 34a7109b47 Implement AArch64 Neon Crypto instruction classes AES, SHA, and 3 SHA.
llvm-svn: 194086
2013-11-05 17:42:24 +00:00
Jiangning Liu d7c52676f6 Implement AArch64 Neon Crypto instruction classes AES, SHA, and 3 SHA.
llvm-svn: 194085
2013-11-05 17:42:05 +00:00
Rafael Espindola 7d712031e1 Convert FindFunctionInStream to return an error_code.
llvm-svn: 194084
2013-11-05 17:16:08 +00:00
Michael Gottesman 24b2f6fdda [objc-arc] Convert the one directional retain/release relation assert to a conditional check + fail.
Due to the previously added overflow checks, we can have a retain/release
relation that is one directional. This occurs specifically when we run into an
additive overflow causing us to drop state in only one direction. If that
occurs, we should bail and not optimize that retain/release instead of
asserting.

Apologies for the size of the testcase. It is necessary to cause the additive
cfg overflow to trigger.

rdar://15377890

llvm-svn: 194083
2013-11-05 16:02:40 +00:00
Timur Iskhodzhanov 5877663622 Fix vbtable indices when a class shares the vbptr with a non-virtual base
llvm-svn: 194082
2013-11-05 15:54:58 +00:00
John Thompson 41778fd28f pp-trace user documentation - beginnings
llvm-svn: 194081
2013-11-05 14:41:29 +00:00
Marshall Clow 9d66b72aa9 Refactor floating point code for num_get::do_get into a template. No functionality change
llvm-svn: 194080
2013-11-05 14:28:52 +00:00
John Thompson 5252325990 [extra] pp-trace - Fixed problems found while writing docs.
llvm-svn: 194079
2013-11-05 14:16:11 +00:00
Alp Toker a2f1b8d238 Provide a test input for opt
This was only working previously due to a quirk in the way lit
concatenates script commands.

llvm-svn: 194078
2013-11-05 13:57:34 +00:00
Benjamin Kramer 77db163645 Silence GCC warning about dropping off a fully covered switch.
llvm-svn: 194077
2013-11-05 13:45:09 +00:00
Alp Toker 3364fe149c Disable test fully with REQUIRES
Third time's the charm, sorry for the noise.

llvm-svn: 194076
2013-11-05 13:27:19 +00:00
Peter Zotov a4b2ac4c30 [OCaml] (PR16190) Add ValueKinds for ConstantDataSequential and subclasses
Original patch by David Monniaux

llvm-svn: 194075
2013-11-05 12:55:43 +00:00
Peter Zotov ae0344b07f [llvm-c] (PR16190) Add LLVMIsA* functions for ConstantDataSequential and subclasses
Original patch by David Monniaux

llvm-svn: 194074
2013-11-05 12:55:37 +00:00
Alp Toker 53b74accc7 Disable the test that crashes
XFAIL wasn't enough for some of the build servers.

llvm-svn: 194073
2013-11-05 12:55:00 +00:00
Alp Toker 6731f38aca Fix a test that was never being run, mark it XFAIL
Turns out this has always been crashing since the test landed in
r186647:

  Assertion failed: (Access != AS_none && "Access specifier is AS_none inside a record decl"), function CheckAccessDeclContext, file clang/lib/AST/DeclBase.cpp, line 690.

Needs attention.

llvm-svn: 194072
2013-11-05 12:45:46 +00:00
Alp Toker f80de76ca9 Make test run reliably
llc waits for input on stdin, which was not provided in this test. It
was running only thanks to a quirk in the way lit concatenates
commands.

llvm-svn: 194071
2013-11-05 12:45:43 +00:00
Alp Toker f9cabd9ddb Fix test that was not being run
llvm-svn: 194070
2013-11-05 12:45:40 +00:00
Alp Toker 52ca4a4e47 Fix test that wasn't testing anything
llvm-svn: 194069
2013-11-05 12:45:37 +00:00
Joey Gouly a7310a8cfa Do not allow functions or kernels called 'main' in OpenCL.
llvm-svn: 194068
2013-11-05 12:30:39 +00:00
Peter Zotov c2a9fe7a77 [OCaml] (PR10016) Add a few missing line in OCamlLangImpl2.rst
Original patch by Damien Schoof

llvm-svn: 194067
2013-11-05 12:14:04 +00:00
Reed Kotler 4d0313d89f Remove the word "thumb" from comments. Remove also an incorrect
command regarding the porting from the ARM version (was an old comment).

llvm-svn: 194066
2013-11-05 12:04:37 +00:00
Peter Zotov 28f6876ecc [OCaml] (PR16318) Add missing argument to Llvm.const_intcast
llvm-svn: 194065
2013-11-05 11:56:20 +00:00
Peter Zotov ce7a91b277 [OCaml] (PR11717) Make declare_qualified_global respect address argument
Original patch by Jonathan Ragan-Kelley

llvm-svn: 194064
2013-11-05 11:56:13 +00:00
Jason Molenda 8c71337abc Add the GetNumThreadOriginExtendedBacktraceTypes and
GetThreadOriginExtendedBacktraceTypeAtIndex methods to
SBProcess.

Add documentation for the GetQueueName and GetQueueID methods
to SBThread.
<rdar://problem/15314369> 

llvm-svn: 194063
2013-11-05 11:00:35 +00:00
David Majnemer 64582671af X86 Disassembler: remove unused bool typedef-name
llvm-svn: 194062
2013-11-05 10:34:42 +00:00
David Majnemer 21db51ab91 Simplify test case added in r194059
llvm-svn: 194061
2013-11-05 10:05:58 +00:00
Alp Toker 9f67932ee7 Suppress OS crash dialog in llvm-rtdyld
All other tools have this -- it's needed to avoid hanging lit on Windows in
case of a crash.

llvm-svn: 194060
2013-11-05 09:33:43 +00:00
David Majnemer 76faf1f525 Lex: Require that '#' be followed by a macro parameter name when preceded by '##'
After lexing a '##', we would look ahead and check to see if it was
followed by '__VA_ARGS__'.  After doing so, we would then go ahead and
lex the token.

However we would fail in the case where the '##' was followed by a '#'
followed by an identifier because we would have lexed the '#' separately
from the identifier, bypassing our parameter validation logic.

Instead, lex the tokens coming after the '##' later.

This fixes PR17804.

llvm-svn: 194059
2013-11-05 09:30:17 +00:00
Peter Zotov 04a0e5050f [OCaml] Properly tag the custom operations of Llvm.llbuilder
All other custom operations tags have LLVM prefix.

llvm-svn: 194058
2013-11-05 09:13:46 +00:00
Peter Zotov a4c2a3e6b6 [OCaml] Llvm_linker: do not use external in module interface
Workaround for an OCaml bug:
http://caml.inria.fr/mantis/view.php?id=4166

llvm-svn: 194057
2013-11-05 09:13:39 +00:00
Richard Smith 96c07b165e Remove stray comma from documentation.
llvm-svn: 194056
2013-11-05 09:12:55 +00:00
Richard Smith b47c36f8e1 C++1y sized deallocation: if we have a use, but not a definition, of a sized
deallocation function (and the corresponding unsized deallocation function has
been declared), emit a weak discardable definition of the function that
forwards to the corresponding unsized deallocation.

This allows a C++ standard library implementation to provide both a sized and
an unsized deallocation function, where the unsized one does not just call the
sized one, for instance by putting both in the same object file within an
archive.

llvm-svn: 194055
2013-11-05 09:12:18 +00:00
Richard Smith 2d1d3ca1cc Enable links now that they work.
llvm-svn: 194054
2013-11-05 08:27:00 +00:00
Reed Kotler 0f007fc4ce Fix r194019 as requested by Eric Christopher.
Submit the basic port of the rest of ARM constant islands code to Mips. 
Two test cases are added which reflect the next level of functionality:
constants getting moved to water areas that are out of range from the
initial placement at the end of the function and basic blocks being split to
create water when none exists that can be used. There is a bunch of this
code that is not complete and has been marked with IN_PROGRESS. I will
finish cleaning this all up during the next week or two and submit the
rest of the test cases. I have elminated some code for dealing with
inline assembly because to me it unecessarily complicates things and
some of the newer features of llvm like function attributies and builtin
assembler give me better tools to solve the alignment issues created
there. Also, for Mips16 I even have the option of not doing constant
islands in the present of inline assembler if I chose. When everything
has been completed I will summarize the port and notify people that
are knowledgable regarding the ARM Constant Islands code so they can
review it in it's entirety if they wish.

llvm-svn: 194053
2013-11-05 08:14:14 +00:00
David Majnemer 2748da9be4 Sema: Improve comment introduced in r193397
llvm-svn: 194052
2013-11-05 08:01:18 +00:00
Alp Toker e5e3bc0c04 Fix symbol defines in config.h.cmake
These were incorrectly pointing to HAVE_LOG despite being checked for
correctly in config-ix.cmake.

Patch by James Lyon!

llvm-svn: 194051
2013-11-05 07:27:18 +00:00
Rafael Espindola 26cc9bdfc0 Expand the test a bit.
llvm-svn: 194050
2013-11-05 07:11:42 +00:00
Rafael Espindola 1cc563d558 Add a testcase from the recent bootstrap failure.
llvm-svn: 194049
2013-11-05 06:44:57 +00:00
Craig Topper be79768b6a Lift alignment restrictions on load folding for a significant portion of AVX instructions.
llvm-svn: 194048
2013-11-05 06:31:43 +00:00
Rafael Espindola b5ac86865a Revert "Produce direct calls instead of alias to linkonce_odr functions."
This reverts commit r194046.
Debugging a bootstrap issue.

llvm-svn: 194047
2013-11-05 05:29:54 +00:00
Rafael Espindola 5d79a51a45 Produce direct calls instead of alias to linkonce_odr functions.
This is a small optimization on linux, but should help more on windows
where msvc only outputs one destructor if there would be two identical ones.

llvm-svn: 194046
2013-11-05 05:22:36 +00:00