The DetectDeadLaneMask already ensures that we have no dead subregister
definitions making the special handling in LiveIntervalAnalysis
unnecessary. This reverts most of r248335.
llvm-svn: 267937
Dead phi nodes are needed for code motion (such as copy propagation),
where a new use would be placed in a location that would be dominated
by a dead phi. Such a transformation is not legal for copy propagation,
and the existence of the phi would prevent it, but if the phi is not
there, it may appear to be valid.
llvm-svn: 267932
Summary: Using libtool instead of ar and ranlib on Darwin shaves a minute off my clang build. This is because on Darwin libtool is optimized to give hints to the kernel about filesystem interactions that allow it to be faster.
Reviewers: bogner, pete
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D19611
llvm-svn: 267930
We now read out the rest of the substreams from the DBI streams. One of
these substreams, the FileInfo substream, contains information about which
source files contribute to each module (aka compiland). This patch
additionally parses out the file information from that substream, and
dumps it in llvm-pdbdump.
Differential Revision: http://reviews.llvm.org/D19634
Reviewed by: ruiu
llvm-svn: 267928
Revert "[Power9] Implement add-pc, multiply-add, modulo, extend-sign-shift, random number, set bool, and dfp test significance".
This patch has caused a functional regression in SPEC2k6 namd, and a performance regression in mesa-pipe.
llvm-svn: 267927
Using multiple context used to be a really big memory saving because we
could free memory from each file while the linker proceeded with the
symbol resolution. We are getting lazier about reading data from the
bitcode, so I was curious if this was still a good tradeoff.
One thing that is a bit annoying is that we still have to copy the
symbol names. The problem is that the names are stored in the Module and
get freed when we move the module bits during linking.
Long term I think the solution is to add a symbol table to the bitcode.
That way IRObject file will not need to use a Module or a Context and we
can drop it while still keeping a StringRef to the names.
This patch is still be an interesting medium term improvement.
When linking llvm-as without debug info this patch is a small speedup:
master: 29.861877513 seconds
patch: 29.814533787 seconds
With debug info the numbers are
master: 34.765181469 seconds
patch: 34.563351584 seconds
The peak memory usage when linking llvm-as with debug info was
master: 599.10MB
patch: 600.13MB
llvm-svn: 267921
ScheduleDAGMI::initQueues changes the RegionBegin to the first non-debug
instruction. Since it does not track register pressure, it does not affect
any RP trackers. ScheduleDAGMILive inherits initQueues from ScheduleDAGMI,
and it does reset the TopTPTracker in its schedule method. Any derived,
target-specific scheduler will need to do it as well, but the TopRPTracker
is only exposed as a "const" object to derived classes. Without the ability
to modify the tracker directly, this leaves a derived scheduler with a
potential of having the TopRPTracker out-of-sync with the CurrentTop.
The symptom of the problem:
void llvm::ScheduleDAGMILive::scheduleMI(llvm::SUnit *, bool):
Assertion `TopRPTracker.getPos() == CurrentTop && "out of sync"' failed.
Differential Revision: http://reviews.llvm.org/D19438
llvm-svn: 267918
Relocations against sections with no SHF_ALLOC bit are R_ABS relocations.
Currently we are creating Relocations vector for them, but that is wasteful.
This patch is to skip vector construction and to directly apply relocations
in place.
This patch seems to be pretty effective for large executables with debug info.
r266158 (Rafael's patch to change the way how we apply relocations) caused a
temporary performance degradation for such executables, but this patch makes
it even faster than before.
Time to link clang with debug info (output size is 1070 MB):
before r266158: 15.312 seconds (0%)
r266158: 17.301 seconds (+13.0%)
Head: 16.484 seconds (+7.7%)
w/patch: 13.166 seconds (-14.0%)
Differential Revision: http://reviews.llvm.org/D19645
llvm-svn: 267917
The canonical form for allocas is a single allocation of the array type.
In case we see a non-canonical array alloca, make sure we aren't
replacing this with an array N times smaller.
llvm-svn: 267916
This happens to be working now because the includes exist in another CMake file that is included before this one. That will change with upcoming refactoring.
llvm-svn: 267912
On Darwin, MAP_ANONYMOUS is a synonym for MAP_ANON. However, some SDK's
don't define MAP_ANONYMOUS. Use MAP_ANON to work around this.
(As a point of interest, the situation is exactly reversed on Linux.)
llvm-svn: 267907
When comparing unqualified types, canonical types should be used, otherwise equivalent types may be treated as different type.
Differential Revision: http://reviews.llvm.org/D19662
llvm-svn: 267906
Currently Mips::emitAtomicBinaryPartword() does not properly respect the
width of pointers. For MIPS64 this causes the memory address that the ll/sc
sequence uses to be truncated. At runtime this causes a segmentation fault.
This can be fixed by applying similar changes as r266204, so that a full 64bit
pointer is loaded.
Reviewers: dsanders
Differential Review: http://reviews.llvm.org/D19651
llvm-svn: 267900
The DWARF2 specification of DW_AT_bit_offset is ambiguous for
little-endian machines, but by restoring to the old behavior
we match what debuggers expect and what other popular compilers
generate.
llvm-svn: 267896
The DWARF2 specification of DW_AT_bit_offset was written from the perspective of
a big-endian machine with unclear semantics for other systems. DWARF4
deprecated DW_AT_bit_offset and introduced a new attribute DW_AT_data_bit_offset
that simply counts the number of bits from the beginning of the containing
entity regardless of endianness.
After this patch LLVM emits DW_AT_bit_offset for DWARF 2 or 3 and
DW_AT_data_bit_offset when DWARF 4 or later is requested.
llvm-svn: 267895
Matthias Braun