llvm-svn: 126463

Chris Jefferson spotted a problem with messages_base::catalog while getting libc++ to work on boost.  The standard says this type must be an int.  But this type is the key returned by the OS facility catopen.  On OS X the type returned by catopen is void*, which doesn't fit into an int on 64 bit platforms.  Chris suggested using ptrdiff_t instead of void*.  It still isn't compliant with the standard, but chances are that this change will fix what is ailing boost.  Chris also supplied the algorithm for distinguishing high-order pointers from error conditions.  Thanks Chris.

llvm-svn: 126462

llvm-svn: 126461

llvm-svn: 126460

UnresolvedUsingValueDecl to use NestedNameSpecifierLoc rather than the
extremely-lossy NestedNameSpecifier/SourceRange pair it used to use,
improving source-location information.

Various infrastructure updates to support NestedNameSpecifierLoc:
  - AST/PCH (de-)serialization
  - Recursive AST visitor
  - libclang traversal (including the first tests of this
    functionality)

llvm-svn: 126459

llvm-svn: 126458

D registers since the vpush list may not have gaps. Make sure the stack
adjustment instruction isn't moved between them. Ditto for vpop in
epilogues.

Sorry, can't reduce a small test case.
rdar://9043312

llvm-svn: 126457

llvm-svn: 126456

llvm-svn: 126455

Intoduce '-analyzer-checker-help' flag which outputs a list of all available static analyzer checkers.

This is pretty basic for now, eventually checkers should be grouped according to package, hidden checkers should be indicated etc.

llvm-svn: 126454

I tried to add test cases for these, but I can't because variables
aren't warned on the way functions are and the codegen layer appears to
use different logic for determining that 'a' and 'g' in the test case
should receive C mangling. I've included the test so that if we ever
switch the codegen layer to use these functions, we won't regress due to
latent bugs.

llvm-svn: 126453

llvm-svn: 126452

New live ranges are assigned in long -> short order, but live ranges that have
been evicted at least once are deferred and assigned in short -> long order.

Also disable splitting and spilling for live ranges seen for the first time.

The intention is to create a realistic interference pattern from the heavy live
ranges before starting splitting and spilling around it.

llvm-svn: 126451

llvm-svn: 126450

This is the error I got:

Debugger entered--Lisp error: (void-function split-string-and-unquote)
  (split-string-and-unquote command-line)
  (let* ((words ...) (program ...) (dir default-directory) (file-word ...) (file-subst ...) (args ...) (file ...) (filepart ...) (existing-buffer ...)) (pop-to$
  gud-common-init("/Volumes/data/lldb/svn/trunk/build/Debug/lldb" nil gud-lldb-marker-filter)
  lldb("/Volumes/data/lldb/svn/trunk/build/Debug/lldb")
  call-interactively(lldb)
  execute-extended-command(nil)
  call-interactively(execute-extended-command)

The gud-common-init elisp function references split-string-and-unquote function
which is not defined there.

llvm-svn: 126449

Don't warn about using PredefinedExprs as format string literals.  These never can be a real security issue.

Fixes PR 9314.

llvm-svn: 126447

llvm-svn: 126445

Modifed lldb_private::Process to be able to handle connecting to a remote 
target that isn't running a process. This leaves lldb_private::Process in the
eStateConnected state from which we can then do an attach or launch.

Modified ProcessGDBRemote to be able to set stdin, stdout, stderr, working
dir, disable ASLR and a few other settings down by using new GDB remote 
packets. This allows us to keep all of our current launch flags and settings
intact and still be able to communicate them over to the remote GDB server.
Previously these were being sent as arguments to the debugserver binary that
we were spawning. Also modified ProcessGDBRemote to handle losing connection
to the remote GDB server and always exit immediately. We do this by watching
the lldb_private::Communication event bit for the read thread exiting in the
ProcessGDBRemote async thread.

Added support for many of the new 'Q' packets for setting stdin, stdout,
stderr, working dir and disable ASLR to the GDBRemoteCommunication class for
easy accesss.

Modified debugserver for all of the new 'Q' packets and also made it so that
debugserver always exists if it loses connection with the remote debugger.

llvm-svn: 126444

Introduce a variable in the AsmParserExtension whether [] is valid in an
expression. If it is true, parse them like (). Enable this for ELF only.

llvm-svn: 126443

llvm-svn: 126442

llvm-svn: 126441

[analyzer] Remove '-analyzer-experimental-internal-checks' flag, it doesn't have any checkers associated with it anymore.

llvm-svn: 126440

For example, if 'core.experimental.UnreachableCode' is hidden, it should not be enabled with 'core.experimental'.
Note that this requires llvm commit r126436.

llvm-svn: 126439

llvm-svn: 126438

llvm-svn: 126436

llvm-svn: 126435

When making a DataExtractor from a Value that's got a ClangType, set the AddressByteSize from the AST Context.

llvm-svn: 126433

6m30.

llvm-svn: 126426

Patch by Nathan Jeffords!

llvm-svn: 126425

Limit the folding of any_ext and sext  into the load operation to scalars.
Limit the active-bits trunc optimization to scalars.
Document vector trunc and vector sext in LangRef.

Similar to commit 126080 (for enabling zext).

llvm-svn: 126424

switch to Jazelle state fails, thus treating BXJ as a BX operation.

llvm-svn: 126423

in PR9301.

llvm-svn: 126422

The problem was codegen guessing the wrong values and printing

	.section	.eh_frame,"aMS",@progbits,4

It is not clear at all if Codegen should try to guess, MC is the
one that should know the default flags.

llvm-svn: 126421

llvm-svn: 126406

Also fix a bug where we were not lazily parsing the ELF header and thus
returning an ArchSpec with invalid cpu type components.  Initialize the cpu
subtype as LLDB_INVALID_CPUTYPE for compatibility with the new ArchSpec
implementation.

llvm-svn: 126405

Previously we were using a set of preprocessor defines and returning an ArchSpec
without any OS/Vendor information.  This fixes an issue with plugin resolution
on Linux where a valid OS component is needed.

llvm-svn: 126404

The major issue this patch solves is that ArchSpec::SetTriple no longer depends
on the implementation of Host::GetArchitecture.  On linux, Host::GetArchitecture
calls ArchSpec::SetTriple, thus blowing the stack.

A second smaller point is that SetTriple now defaults to Host defined components
iff all OS, vendor and environment fields are not set.

llvm-svn: 126403

llvm-svn: 126402

llvm-svn: 126401

namespace blanket rule for variables and functions declared 'extern
"C"'.

llvm-svn: 126400