This reverts commit r238795, as it broke the Thumb2 self-hosting buildbot.
Since self-hosting issues with Clang are hard to investigate, I'm taking the
liberty to revert now, so we can investigate it offline.
llvm-svn: 238821
Summary: These directives are used to set the current value of the SoftFloat feature.
Reviewers: dsanders
Reviewed By: dsanders
Subscribers: llvm-commits, mpf
Differential Revision: http://reviews.llvm.org/D9074
llvm-svn: 238813
This create a MCSymbolELF class and moves SymbolSize since only ELF
needs a size expression.
This reduces the size of MCSymbol from 56 to 48 bytes.
llvm-svn: 238801
The existing code would unnecessarily break LDRD/STRD apart with
non-adjacent registers, on thumb2 this is not necessary.
Ideally on thumb2 we shouldn't match for ldrd/strd pre-regalloc anymore
as there is not reason to set register hints anymore, changing that is
something for a future patch however.
Differential Revision: http://reviews.llvm.org/D9694
llvm-svn: 238795
Previously CCMP/FCCMP instructions were only used by the
AArch64ConditionalCompares pass for control flow. This patch uses them
for SELECT like instructions as well by matching patterns in ISelLowering.
PR20927, rdar://18326194
Differential Revision: http://reviews.llvm.org/D8232
llvm-svn: 238793
Summary: Implement bswap intrinsic for MIPS FastISel. It's very different for misp32 r1/r2 .
Based on a patch by Reed Kotler.
Test Plan:
bswap1.ll
test-suite
Reviewers: dsanders, rkotler
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D7219
llvm-svn: 238760
Summary:
Implement the intrinsics memset, memcopy and memmove in MIPS FastISel.
Make some needed infrastructure fixes so that this can work.
Based on a patch by Reed Kotler.
Test Plan:
memtest1.ll
The patch passes test-suite for mips32 r1/r2 and at O0/O2
Reviewers: rkotler, dsanders
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D7158
llvm-svn: 238759
Summary: Implement the LLVM assembly urem/srem and sdiv/udiv instructions in MIPS FastISel.
Based on a patch by Reed Kotler.
Test Plan:
srem1.ll
div1.ll
test-suite at O0/O2 for mips32 r1/r2
Reviewers: dsanders, rkotler
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D7028
llvm-svn: 238757
Summary: Implement the LLVM IR select statement for MIPS FastISelsel.
Based on a patch by Reed Kotler.
Test Plan:
"Make check" test included now.
Passes test-suite at O2/O0 mips32 r1/r2.
Reviewers: dsanders, rkotler
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D6774
llvm-svn: 238756
Summary:
The contents of the HI/LO registers are unpredictable after the execution of
the MUL instruction. In addition to implicitly defining these registers in the
MUL instruction definition, we have to mark those registers as dead too.
Without this the fast register allocator is running out of registers when the
MUL instruction is followed by another one that tries to allocate the AC0
register.
Based on a patch by Reed Kotler.
Reviewers: dsanders, rkotler
Subscribers: llvm-commits, rfuhler
Differential Revision: http://reviews.llvm.org/D9825
llvm-svn: 238755
This is important because of different addressing modes
depending on the address space for GPU targets.
This only adds the argument, and does not update
any of the uses to provide the correct address space.
llvm-svn: 238723
The original version didn't properly account for the base register
being modified before the final jump, so caused miscompilations in
Chromium and LLVM. I've fixed this and tested with an LLVM self-host
(I don't have the means to build & test Chromium).
The general idea remains the same: in pathological cases jump tables
can be too far away from the instructions referencing them (like other
constants) so they need to be movable.
Should fix PR23627.
llvm-svn: 238680
best approach of each.
For vNi16, we use SHL + ADD + SRL pattern that seem easily the best.
For vNi32, we use the PUNPCK + PSADBW + PACKUSWB pattern. In some cases
there is a huge improvement with this in IACA's estimated throughput --
over 2x higher throughput!!!! -- but the measurements are too good to be
true. In one narrow case, the SHL + ADD + SHL + ADD + SRL pattern looks
slightly faster, but I'm not sure I believe any of the measurements at
this point. Both are the exact same uops though. Hard to be confident of
anything past that.
If anyone wants to collect very detailed (Agner-level) timings with the
result of this patch, or with the i32 case replaced with SHL + ADD + SHl
+ ADD + SRL, I'd be very interested. Note that you'll need to test it on
both Ivybridge and Haswell, with both SSE3, SSSE3, and AVX selected as
I saw unique behavior in each of these buckets with IACA all of which
should be checked against measured performance.
But this patch is still a useful improvement by dropping duplicate work
and getting the much nicer PSADBW lowering for v2i64.
I'd still like to rephrase this in terms of generic horizontal sum. It's
a bit lame to have a special case of that just for popcount.
llvm-svn: 238652
The plan was to move the whole table into the already existing ArchExtNames
but some fields depend on a table-generated file, and we don't yet have this
feature in the generic lib/Support side.
Once the minimum target-specific table-generated files are available in a
generic fashion to these libraries, we'll have to keep it in the ASM parser.
llvm-svn: 238651
shorter one. NFC.
In addition to being much shorter to type and requiring fewer arguments,
this change saves over 30 lines from this one file, all wasted on total
boilerplate...
llvm-svn: 238640
shifting vectors of bytes as x86 doesn't have direct support for that.
This removes a bunch of redundant masking in the generated code for SSE2
and SSE3.
In order to avoid the really significant code size growth this would
have triggered, I also factored the completely repeatative logic for
shifting and masking into two lambdas which in turn makes all of this
much easier to read IMO.
llvm-svn: 238637
in-register LUT technique.
Summary:
A description of this technique can be found here:
http://wm.ite.pl/articles/sse-popcount.html
The core of the idea is to use an in-register lookup table and the
PSHUFB instruction to compute the population count for the low and high
nibbles of each byte, and then to use horizontal sums to aggregate these
into vector population counts with wider element types.
On x86 there is an instruction that will directly compute the horizontal
sum for the low 8 and high 8 bytes, giving vNi64 popcount very easily.
Various tricks are used to get vNi32 and vNi16 from the vNi8 that the
LUT computes.
The base implemantion of this, and most of the work, was done by Bruno
in a follow up to D6531. See Bruno's detailed post there for lots of
timing information about these changes.
I have extended Bruno's patch in the following ways:
0) I committed the new tests with baseline sequences so this shows
a diff, and regenerated the tests using the update scripts.
1) Bruno had noticed and mentioned in IRC a redundant mask that
I removed.
2) I introduced a particular optimization for the i32 vector cases where
we use PSHL + PSADBW to compute the the low i32 popcounts, and PSHUFD
+ PSADBW to compute doubled high i32 popcounts. This takes advantage
of the fact that to line up the high i32 popcounts we have to shift
them anyways, and we can shift them by one fewer bit to effectively
divide the count by two. While the PSHUFD based horizontal add is no
faster, it doesn't require registers or load traffic the way a mask
would, and provides more ILP as it happens on different ports with
high throughput.
3) I did some code cleanups throughout to simplify the implementation
logic.
4) I refactored it to continue to use the parallel bitmath lowering when
SSSE3 is not available to preserve the performance of that version on
SSE2 targets where it is still much better than scalarizing as we'll
still do a bitmath implementation of popcount even in scalar code
there.
With #1 and #2 above, I analyzed the result in IACA for sandybridge,
ivybridge, and haswell. In every case I measured, the throughput is the
same or better using the LUT lowering, even v2i64 and v4i64, and even
compared with using the native popcnt instruction! The latency of the
LUT lowering is often higher than the latency of the scalarized popcnt
instruction sequence, but I think those latency measurements are deeply
misleading. Keeping the operation fully in the vector unit and having
many chances for increased throughput seems much more likely to win.
With this, we can lower every integer vector popcount implementation
using the LUT strategy if we have SSSE3 or better (and thus have
PSHUFB). I've updated the operation lowering to reflect this. This also
fixes an issue where we were scalarizing horribly some AVX lowerings.
Finally, there are some remaining cleanups. There is duplication between
the two techniques in how they perform the horizontal sum once the byte
population count is computed. I'm going to factor and merge those two in
a separate follow-up commit.
Differential Revision: http://reviews.llvm.org/D10084
llvm-svn: 238636
a separate routine, generalize it to work for all the integer vector
sizes, and do general code cleanups.
This dramatically improves lowerings of byte and short element vector
popcount, but more importantly it will make the introduction of the
LUT-approach much cleaner.
The biggest cleanup I've done is to just force the legalizer to do the
bitcasting we need. We run these iteratively now and it makes the code
much simpler IMO. Other changes were minor, and mostly naming and
splitting things up in a way that makes it more clear what is going on.
The other significant change is to use a different final horizontal sum
approach. This is the same number of instructions as the old method, but
shifts left instead of right so that we can clear everything but the
final sum with a single shift right. This seems likely better than
a mask which will usually have to read the mask from memory. It is
certaily fewer u-ops. Also, this will be temporary. This and the LUT
approach share the need of horizontal adds to finish the computation,
and we have more clever approaches than this one that I'll switch over
to.
llvm-svn: 238635
It turns out that _except_handler3 and _except_handler4 really use the
same stack allocation layout, at least today. They just make different
choices about encoding the LSDA.
This is in preparation for lowering the llvm.eh.exceptioninfo().
llvm-svn: 238627
Renato Golin