keyword. We now handle this keyword in HandleIdentifier, making a note
for ourselves when we've seen the __import_module__ keyword so that
the next lexed token can trigger a module import (if needed). This
greatly simplifies Preprocessor::Lex(), and completely erases the 5.5%
-Eonly slowdown Argiris noted when I originally implemented
__import_module__. Big thanks to Argiris for noting that horrible
regression!
llvm-svn: 139265
Need branch relocation support to distinguish this encoding from the
16-bit Thumb1 encoding w/o the explicit .w suffix. That comes later, though.
llvm-svn: 139257
The immediate offset of the non-writeback i8 form (encoding T4) allows
negative offsets only. The positive offset form of the encoding is the
LDRT instruction. Immediate offsets in the range [0,255] use encoding T3
instead.
llvm-svn: 139254
ctor-initializer, remember to call the Sema action to generate default
ctor-initializers. What a delightful little miscompile. Fixes PR10578
/ <rdar://problem/9877267>.
llvm-svn: 139253
In some cases such as interpreters using indirectbr, the CFG can be very
complicated, and live range splitting may be forced to insert a large
number of phi-defs. When that happens, traceSiblingValue can spend a
lot of time zipping around in the CFG looking for defs and reloads.
This patch causes more information to be cached in SibValues, and the
cached values are used to terminate searches early. This speeds up
spilling by 20x in one interpreter test case. For more typical code,
this is just a 10% speedup of spilling.
llvm-svn: 139247
(MachThreadList::EnableHardwareWatchpoint()) where the watchpoint is not associated
with a thread and the current thread, if set, is returned, otherwise we return the
first thread.
Plus minor change to RNBRemote::HandlePacket_z() to use the existing macros to check
the validity of break_id/watch_id.
llvm-svn: 139246
There is no 16-bit wide encoding, so the .w suffix isn't needed (indeed, isn't
documented as allowed). Also add the missing '!' token on the _UPD
variant.
llvm-svn: 139243
Choose 32-bit vs. 16-bit encoding when there's no .w suffix in post-processing
as match classes are insufficient to handle the context-sensitiveness of
the writeback operand's legality for the 16-bit encodings.
llvm-svn: 139242
duplicate tests are eliminated (for example if the two functions both have
a catch clause catching the same type, ensure the redundant one is removed).
Note that it would probably be safe to say that eh.typeid.for is 'const',
but since two calls to it with the same argument can give different results
(but only if the calls are in different functions), it seems more correct to
mark it only 'pure'; this doesn't get in the way of the optimization.
llvm-svn: 139236
function. This is really the beginning of the second phase of
refactorings here. The end goal is to have (roughly) three interfaces:
1) Base class to format a single diagnostic suitable for display on the
console.
2) Extension of the base class which also displays a caret diagnostic
suitable for display on the console.
3) An adaptor that implements the DiagnosticClient by delegating to #1
and/or #2 as appropriate.
Once we have these, things like libclang's formatDiagnostic can use #1
and #2 to provide really well formatted (and consistently formatted!)
textual formatting of diagnostics.
Getting there is going to be quite a bit of shuffling. I'm basically
sketching out where the interface boundaries can be drawn for #1 and #2
within the existing classes. That lets me shuffle with a minimum of fuss
and delta. Once that's done, and any of the related interfaces that need
to change are updated, I'll hoist these into separate headers and
re-implement libclang in terms of their interfaces. Long WIP, but
comments at each step welcome. =D
llvm-svn: 139228
(The fix for the related failures on x86 is going to be nastier because we actually need Acquire memoperands attached to the atomic load instrs, etc.)
llvm-svn: 139221
a stack array of a magical size with an assert() that we never
overflowed it. That seems incredibly risky. We also have a very nice API
for bundling up a vector we expect to usually have a small size without
loss of functionality or security if the size is excessive.
The fallout is to remove the last pointer+size parameter pair that are
traced through the recursive caret diagnostic emission.
llvm-svn: 139217
There are still a few issues which need to be resolved with code generation for atomic load and store, so I'm not converting the places which need those for now.
I'm not entirely sure what to do about __builtin_llvm_memory_barrier: the fence instruction doesn't expose all the possibilities which can be expressed by __builtin_llvm_memory_barrier. I would appreciate hearing from anyone who is using this intrinsic.
llvm-svn: 139216
Jim Grosbach