We've rejected these kinds of functions since r28405 in 2006 because
it's impossible to lower the return of a callee cleanup varargs
function. However there are lots of legal ways to leave such a function
without returning, such as aborting. Today we can leave a function with
a musttail call to another function with the correct prototype, and
everything works out.
I'm removing the verifier check declaring that a normal return from such
a function is UB.
Reviewed By: nlewycky
Differential Revision: http://reviews.llvm.org/D5059
llvm-svn: 216779
This patch checks for DAG patterns that are an add or a sub followed by a
compare on 16 and 8 bit inputs. Since AArch64 does not support those types
natively they are legalized into 32 bit values, which means that mask operations
are inserted into the DAG to emulate overflow behaviour. In many cases those
masks do not change the result of the processing and just introduce a dependent
operation, often in the middle of a hot loop.
This patch detects the relevent DAG patterns and then tests to see if the
transforms are equivalent with and without the mask, removing the mask if
possible. The exact mechanism of this patch was discusses in
http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-July/074444.html
There is a reasonably good chance there are missed oppurtunities due to similiar
(but not identical) DAG patterns that could be funneled into this test, adding
them should be simple if we see test cases.
Tests included.
rdar://13754426
llvm-svn: 216776
The new solution is to not use this lowering if there are any dynamic
allocas in the current function. We know up front if there are dynamic
allocas, but we don't know if we'll need to create stack temporaries
with large alignment during lowering. Conservatively assume that we will
need such temporaries.
Reviewed By: hans
Differential Revision: http://reviews.llvm.org/D5128
llvm-svn: 216775
I hope this is the last fix for x64 relocations as I've wasted
a few days on this.
This caused a mysterious issue that some C++ programs crash on
startup. It was because a null pointer is passed as argv to main.
__tmainCRTStartup calls main, but before that it calls all
initialization routines between .text$xc_a and .text$xc_z.
pre_cpp_init is one of such routines, and it is the one who
initializes a heap pointer for argv for later use. That routine
was not called for some reason.
It turned out that __tmainCRTStartup was skipping a block of
code because of the relocation bug. A condition in the function
depends on a memory load, and that memory load was referring
a wrong location. As a result a jump instruction took the
wrong branch, skipping pre_cpp_init and so on.
This patch fixes the issue. Also added more tests to fix them
once and for all.
llvm-svn: 216772
Even loads/stores that have a stronger ordering than monotonic can be safe.
The rule is no release-acquire pair on the path from the QueryInst, assuming that
the QueryInst is not atomic itself.
llvm-svn: 216771
llgs Linux is no longer doing the translation to some gdb fixed signal
numbers. This change modifies the test method to take in the signo
expected for a segfault. The debugserver test uses the fixed gdb number,
and everything else uses signal.SIGSEGV for the platform.
Sidenote: I somehow did not see this reported in multicore tests.
I will need to verify that we aren't missing test failures in that
script. I need to verify why this wasn't more obvious with
test/dosep.py.
llvm-svn: 216770
in the super class, do not issue the warning about property
in current class's protocol will not be auto synthesized.
// rdar://18179833
llvm-svn: 216769
Summary:
Mostly renaming the (not very explicit) variables Tmp0, .. Tmp4, and grouping
related statements together, along with a few lines of comments for the
surprising parts.
No functional change intended.
Test Plan: make check-all
Reviewers: jfb
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D5088
llvm-svn: 216768
People have been incorrectly using "-analyzer-disable-checker" to
silence analyzer warnings on a file, when analyzing a project. Add
the "-analyzer-disable-all-checks" option, which would allow the
suppression and suggest it as part of the error message for
"-analyzer-disable-checker". The idea here is to compose this with
"--analyze" so that users can selectively opt out specific files from
static analysis.
llvm-svn: 216763
I suspect llvm::ilist should take elements by unique_ptr, since it does
take ownership of the element (by stitching it into the linked list) -
one day.
llvm-svn: 216761
Do not warn when property declared in class's protocol will be auto-synthesized
by its uper class implementation because super class has also declared this
property while this class has not. Continue to warn if current class
has declared the property also (because this declaration will not result
in a 2nd synthesis).
rdar://18152478
llvm-svn: 216753
Again, if shared ownership is the right model here (I assume it is,
given graph algorithms & such) this could be tidied up (the 'release'
call removed in favor of something safer) by having
IntrunsiveRefCntPointer constructible from a unique_ptr.
(& honestly I'd probably favor taking a page out of shared_ptr's book,
allowing implicit construction from a unique_ptr rvalue, and only allow
explicit from a raw pointer - currently IntrusiveRefCntPointer can
implicitly own from a raw pointer, which seems unsafe)
llvm-svn: 216752
When we select a trunc instruction we don't emit any code if the type is already
i32 or smaller. This is because the instruction that uses the truncated value
will deal with it.
This behavior can incorrectly transfer a kill flag, which was meant for the
result of the truncate, onto the source register.
%2 = trunc i32 %1 to i16
... = ... %2 -> ... = ... vreg1 <kill>
... = ... %1 ... = ... vreg1
This commit fixes this by emitting a COPY instruction, so that the result and
source register are distinct virtual registers.
This fixes rdar://problem/18178188.
llvm-svn: 216750
See http://reviews.llvm.org/D5108 for details.
This change does the following:
* eliminates the Process::GetUnixSignals() virtual method and replaces with a fixed getter.
* replaces the Process UnixSignals storage with a shared pointer.
* adds a Process constructor variant that can be passed the UnixSignalsSP. When the constructor without the UnixSignalsSP is specified, the Host's default UnixSignals is used.
* adds a host-specific version of GetUnixSignals() that is used when we need the host's appropriate UnixSignals variant.
* replaces GetUnixSignals() overrides in PlatformElfCore, ProcessGDBRemote, ProcessFreeBSD and ProcessLinux with code that appropriately sets the Process::UnixSignals for the process.
This change also enables some future patches that will enable llgs to be used for local Linux debugging.
llvm-svn: 216748
This patch accepts environment variables of the form:
LLDB_DEBUGSERVER_EXTRA_ARG_n
where n starts with 1, and may continue nearly indefinitely (up through std::numeric_limits<uint32_t>::max()).
The code loops around, starting with 1, until it doesn't find one of the environment variables. For each one it does find defined, it appends the environment variable's contents to the end of the debugserver/llgs startup command line issued when the stub is started for local debugging.
I am using this to add arbitrary startup commands to the llgs command line for turning on additional logging. For example:
export LLDB_DEBUGSERVER_EXTRA_ARG_1="-c"
export LLDB_DEBUGSERVER_EXTRA_ARG_2="log enable -f /tmp/llgs_packets.log gdb-remote packets"
export LLDB_DEBUGSERVER_EXTRA_ARG_3="-c"
export LLDB_DEBUGSERVER_EXTRA_ARG_4="log enable -f /tmp/llgs_process.log lldb process"
llvm-svn: 216745
In theory, it'd be nice if we could move to a case where all buried
pointers were buried via unique_ptr to demonstrate that the program had
finished with the value (that we could really have cleanly deallocated
it) but instead chose to bury it.
I think the main reason that's not possible right now is the various
IntrusiveRefCntPtrs in the Frontend, sharing ownership for a variety of
compiler bits (see the various similar
"CompilerInstance::releaseAndLeak*" functions). I have yet to figure out
their correct ownership semantics - but perhaps, even if the
intrusiveness can be removed, the shared ownership may yet remain and
that would lead to a non-unique burying as is there today. (though we
could model that a little better - by passing in a shared_ptr, etc -
rather than needing the two step that's currently used in those other
releaseAndLeak* functions)
This might be a bit more robust if BuryPointer took the boolean:
BuryPointer(bool, unique_ptr<T>)
and the choice to bury was made internally - that way, even when
DisableFree was not set, the unique_ptr would still be null in the
caller and there'd be no chance of accidentally having a different
codepath where the value is used after burial in !DisableFree, but it
becomes null only in DisableFree, etc...
llvm-svn: 216742
Yet another step toward ARM64 support. With this commit, lldb-gdbserver started on ARM64 target can be accessed by lldb running on desktop PC and it can process simple commands (like 'continue'). Still ARM64 support lacks NativeRegisterContextLinux_arm64.* code which waits to be implemented.
Based on similar files for Linux x86_64 and Darwin ARM64. Due to common code extraction from Darwin related files, lldb should be tested for any unexpected regression on Darwin ARM64 machines too.
See the following for more details:
http://reviews.llvm.org/D4580http://lists.cs.uiuc.edu/pipermail/lldb-commits/Week-of-Mon-20140825/012670.html
Change by Paul Osmialowski.
llvm-svn: 216737
Summary:
Instead of specifying the alignment as metadata which may be destroyed by
transformation passes, make the alignment the second argument to ldu/ldg
intrinsic calls.
Test Plan:
ldu-ldg.ll
ldu-i8.ll
ldu-reg-plus-offset.ll
Reviewers: eliben, meheff, jholewinski
Reviewed By: meheff, jholewinski
Subscribers: jholewinski, llvm-commits
Differential Revision: http://reviews.llvm.org/D5093
llvm-svn: 216731
Summary: Since the personality functions do the actual unwinding on ARM,
and will also stop unwinding when they encounter a handler, we invoke
_Unwind_VRS_Interpret() directly form _Unwind_Backtrace().
To simplify, the logic for decoding an EHT is moved out of
unwindOneFrame() and into its own function, decode_eht_entry(). Unlike
unwindOneFrame(), which could only handle ARM's compact personality
function entries (section 6.3) decode_eht_entry() can handle the generic
entries (section 6.2).
Reviewers: jroelofs
Reviewed By: jroelofs
Subscribers: piman, aemerson, cfe-commits
Differential Revision: http://reviews.llvm.org/D5112
llvm-svn: 216730
While working on a Thumb-2 code size optimization I just realized that we don't have any regression tests for it.
So here's a first test case, I plan to increase the coverage over time.
llvm-svn: 216728
Reid Kleckner