| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
| |
llvm-svn: 144003
|
| |
|
|
|
|
|
|
|
| |
The xorps instruction is smaller than pxor, so prefer that encoding.
The ExecutionDepsFix pass will switch the encoding to pxor and xorpd
when appropriate.
llvm-svn: 143996
|
| |
|
|
| |
llvm-svn: 143915
|
| |
|
|
| |
llvm-svn: 143904
|
| |
|
|
|
|
| |
qualified with HasAVX2 instead of HasAVX. Mark VINSERTF128 and VEXTRACTF128 as never having side effects.
llvm-svn: 143902
|
| |
|
|
| |
llvm-svn: 143895
|
| |
|
|
| |
llvm-svn: 143861
|
| |
|
|
|
|
|
| |
zero-extend the constant integer encoding. Test case provides testing for
both call parameters and materialization of i1, i8, and i16 types.
llvm-svn: 143821
|
| |
|
|
|
|
|
|
| |
encoding from the DWARF asm printer.
As a side effect we now print dwarf ulebs with .ascii directives.
llvm-svn: 143809
|
| |
|
|
|
|
|
|
| |
is unnecessary through local analysis. Patch from Bruno Cardoso Lopes, with some additional changes.
I'm going to wait for any review comments and perform some additional testing before turning this on by default.
llvm-svn: 143750
|
| |
|
|
| |
llvm-svn: 143682
|
| |
|
|
| |
llvm-svn: 143669
|
| |
|
|
|
|
|
| |
across calls, and only check for nested dependences on the special
call-sequence-resource register.
llvm-svn: 143660
|
| |
|
|
| |
llvm-svn: 143646
|
| |
|
|
| |
llvm-svn: 143640
|
| |
|
|
| |
llvm-svn: 143612
|
| |
|
|
| |
llvm-svn: 143603
|
| |
|
|
|
|
| |
hoist them out of loops. Fixes <rdar://problem/6027699>
llvm-svn: 143600
|
| |
|
|
|
|
| |
alignment permits.
llvm-svn: 143582
|
| |
|
|
|
|
| |
implied by directory index zero.
llvm-svn: 143570
|
| |
|
|
| |
llvm-svn: 143559
|
| |
|
|
| |
llvm-svn: 143536
|
| |
|
|
| |
llvm-svn: 143529
|
| |
|
|
|
|
| |
constant splat value. Fixes PR11289.
llvm-svn: 143498
|
| |
|
|
|
|
| |
This can happen if the address + addend is less than the start of the cp / dp.
llvm-svn: 143459
|
| |
|
|
|
|
| |
single test.
llvm-svn: 143458
|
| |
|
|
| |
llvm-svn: 143457
|
| |
|
|
| |
llvm-svn: 143455
|
| |
|
|
| |
llvm-svn: 143336
|
| |
|
|
| |
llvm-svn: 143332
|
| |
|
|
| |
llvm-svn: 143331
|
| |
|
|
|
|
| |
-enable-dwarf-directory.
llvm-svn: 143326
|
| |
|
|
|
|
| |
and zero out the bits where zeros should've been shifted in.
llvm-svn: 143315
|
| |
|
|
|
|
| |
vector of 8-bit integers, but produces a vector of 16-bit integers.
llvm-svn: 143313
|
| |
|
|
|
|
|
|
|
| |
On x86: (shl V, 1) -> add V,V
Hardware support for vector-shift is sparse and in many cases we scalarize the
result. Additionally, on sandybridge padd is faster than shl.
llvm-svn: 143311
|
| |
|
|
| |
llvm-svn: 143307
|
| |
|
|
|
|
|
| |
If all of the inputs are zero/any_extended, create a new simple BV
which can be further optimized by other BV optimizations.
llvm-svn: 143297
|
| |
|
|
| |
llvm-svn: 143291
|
| |
|
|
| |
llvm-svn: 143262
|
| |
|
|
|
|
| |
"REQUIRES: asserts" for now.
llvm-svn: 143247
|
| |
|
|
|
|
|
|
|
| |
fixes: Use a separate register, instead of SP, as the
calling-convention resource, to avoid spurious conflicts with
actual uses of SP. Also, fix unscheduling of calling sequences,
which can be triggered by pseudo-two-address dependencies.
llvm-svn: 143206
|
| |
|
|
|
|
|
|
|
|
|
| |
host-endian-neutral.
Don't assume APInt::getRawData() would hold target-aware endianness nor host-compliant endianness. rawdata[0] holds most lower i64, even on big endian host.
FIXME: Add a testcase for big endian target.
FIXME: Ditto on CompileUnit::addConstantFPValue() ?
llvm-svn: 143194
|
| |
|
|
|
|
| |
-mtriple=i686-linux. It must be for elf!
llvm-svn: 143189
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
it fixes the dragonegg self-host (it looks like gcc is miscompiled).
Original commit messages:
Eliminate LegalizeOps' LegalizedNodes map and have it just call RAUW
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.
Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.
Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.
Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.
This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.
Delete #if 0 code accidentally left in.
llvm-svn: 143188
|
| |
|
|
|
|
|
| |
tools that read the debug info in the .o files by making the DIE sizes more
consistent.
llvm-svn: 143186
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.
Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.
Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.
Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.
This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.
llvm-svn: 143177
|
| |
|
|
| |
llvm-svn: 143164
|
| |
|
|
|
|
|
| |
Previously, we were only setting the alignment bits on over-aligned
loads and stores.
llvm-svn: 143160
|
| |
|
|
| |
llvm-svn: 143145
|
| |
|
|
|
|
| |
might change if optimised
llvm-svn: 143116
|