For any @llvm.assume intrinsic, if there is another which dominates it and uses
the same condition, then it is redundant and can be removed. While this does
not alter the semantics of the @llvm.assume intrinsics, it makes subsequent
handling more efficient (and the resulting IR easier to read).
Yaron Keren [Sat, 4 Oct 2014 19:58:30 +0000 (19:58 +0000)]
Solve Visual C++ warning C4805 on getAsInteger<bool>.
Fix http://llvm.org/PR21158 by adding a cast to unsigned long long,
so the comparison would be between two unsigned long longs instead
of bool and unsigned long long.
if (getAsUnsignedInteger(*this, Radix, ULLVal) ||
static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal)
David Blaikie [Sat, 4 Oct 2014 16:24:00 +0000 (16:24 +0000)]
Sink DwarfDebug::updateSubprogramScopeDIE into DwarfCompileUnit
This requires exposing some of the current function state from
DwarfDebug. I hope there's not too much of that to expose as I go
through all the functions, but it still seems nicer to expose singular
data down to multiple consumers, than have consumers expose raw mapping
data structures up to DwarfDebug for building subprograms.
Part of a series of refactoring to allow subprograms in both the
skeleton and dwo CUs under Fission.
David Blaikie [Sat, 4 Oct 2014 15:49:50 +0000 (15:49 +0000)]
Move DwarfCompileUnit from DwarfUnit.h to its own header (DwarfCompileUnit.h)
In preparation for sinking all the subprogram emission code down from
DwarfDebug into DwarfCompileUnit, this will avoid bloating
DwarfUnit.h/cpp greatly and make concerns a bit more clear/isolated.
(sinking this handling down is part of the work to handle emitting
minimal subprograms for -gmlt-like data into the skeleton CU under
fission)
[x86] Slap a triple on this test since it is poking around at the stack
and calling conventions. Otherwise its too hard to craft a usefully
generic set of assertions.
[x86] Enable the new vector shuffle lowering by default.
Update the entire regression test suite for the new shuffles. Remove
most of the old testing which was devoted to the old shuffle lowering
path and is no longer relevant really. Also remove a few other random
tests that only really exercised shuffles and only incidently or without
any interesting aspects to them.
Benchmarking that I have done shows a few small regressions with this on
LNT, zero measurable regressions on real, large applications, and for
several benchmarks where the loop vectorizer fires in the hot path it
shows 5% to 40% improvements for SSE2 and SSE3 code running on Sandy
Bridge machines. Running on AMD machines shows even more dramatic
improvements.
When using newer ISA vector extensions the gains are much more modest,
but the code is still better on the whole. There are a few regressions
being tracked (PR21137, PR21138, PR21139) but by and large this is
expected to be a win for x86 generated code performance.
It is also more correct than the code it replaces. I have fuzz tested
this extensively with ISA extensions up through AVX2 and found no
crashes or miscompiles (yet...). The old lowering had a few miscompiles
and crashers after a somewhat smaller amount of fuzz testing.
There is one significant area where the new code path lags behind and
that is in AVX-512 support. However, there was *extremely little*
support for that already and so this isn't a significant step backwards
and the new framework will probably make it easier to implement lowering
that uses the full power of AVX-512's table-based shuffle+blend (IMO).
Many thanks to Quentin, Andrea, Robert, and others for benchmarking
assistance. Thanks to Adam and others for help with AVX-512. Thanks to
Hal, Eric, and *many* others for answering my incessant questions about
how the backend actually works. =]
I will leave the old code path in the tree until the 3 PRs above are at
least resolved to folks' satisfaction. Then I will rip it (and 1000s of
lines of code) out. =] I don't expect this flag to stay around for very
long. It may not survive next week.
[x86] Fix a bug in the VZEXT DAG combine that I just made more powerful.
It turns out this combine was always somewhat flawed -- there are cases
where nested VZEXT nodes *can't* be combined: if their types have
a mismatch that can be observed in the result. While none of these show
up in currently, once I switch to the new vector shuffle lowering a few
test cases actually form such nested VZEXT nodes. I've not come up with
any IR pattern that I can sensible write to exercise this, but it will
be covered by tests once I flip the switch.
[x86] Sink a generic combine of VZEXT nodes from the lowering to VZEXT
nodes to the DAG combining of them.
This will allow the combine to fire on both old vector shuffle lowering
and the new vector shuffle lowering and generally seems like a cleaner
design. I've trimmed down the code a bit and tried to make it and the
surrounding combine fairly clean while moving it around.
[x86] Add a really preposterous number of patterns for matching all of
the various ways in which blends can be used to do vector element
insertion for lowering with the scalar math instruction forms that
effectively re-blend with the high elements after performing the
operation.
This then allows me to bail on the element insertion lowering path when
we have SSE4.1 and are going to be doing a normal blend, which in turn
restores the last of the blends lost from the new vector shuffle
lowering when I got it to prioritize insertion in other cases (for
example when we don't *have* a blend instruction).
Without the patterns, using blends here would have regressed
sse-scalar-fp-arith.ll *completely* with the new vector shuffle
lowering. For completeness, I've added RUN-lines with the new lowering
here. This is somewhat superfluous as I'm about to flip the default, but
hey, it shows that this actually significantly changed behavior.
The patterns I've added are just ridiculously repetative. Suggestions on
making them better very much welcome. In particular, handling the
commuted form of the v2f64 patterns is somewhat obnoxious.
[x86] Adjust the patterns for lowering X86vzmovl nodes which don't
perform a load to use blendps rather than movss when it is available.
For non-loads, blendps is *much* faster. It can execute on two ports in
Sandy Bridge and Ivy Bridge, and *three* ports on Haswell. This fixes
one of the "regressions" from aggressively taking the "insertion" path
in the new vector shuffle lowering.
This does highlight one problem with blendps -- it isn't commuted as
heavily as it should be. That's future work though.
Richard Smith [Fri, 3 Oct 2014 20:17:06 +0000 (20:17 +0000)]
PR21145: Teach LLVM about C++14 sized deallocation functions.
C++14 adds new builtin signatures for 'operator delete'. This change allows
new/delete pairs to be removed in C++14 onwards, as they were in C++11 and
before.
Revert "Revert "DI: Fold constant arguments into a single MDString""
This reverts commit r218918, effectively reapplying r218914 after fixing
an Ocaml bindings test and an Asan crash. The root cause of the latter
was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a
PR to investigate who requires the loose check (and why).
Original commit message follows.
--
This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString. Integers are stringified and
a `\0` character is used as a separator.
Part of PR17891.
Note: I've attached my testcases upgrade scripts to the PR. If I've
just broken your out-of-tree testcases, they might help.
Adam Nemet [Fri, 3 Oct 2014 20:00:34 +0000 (20:00 +0000)]
[ISel] Keep matching state consistent when folding during X86 address match
In the X86 backend, matching an address is initiated by the 'addr' complex
pattern and its friends. During this process we may reassociate and-of-shift
into shift-of-and (FoldMaskedShiftToScaledMask) to allow folding of the
shift into the scale of the address.
However as demonstrated by the testcase, this can trigger CSE of not only the
shift and the AND which the code is prepared for but also the underlying load
node. In the testcase this node is sitting in the RecordedNode and MatchScope
data structures of the matcher and becomes a deleted node upon CSE. Returning
from the complex pattern function, we try to access it again hitting an assert
because the node is no longer a load even though this was checked before.
Now obviously changing the DAG this late is bending the rules but I think it
makes sense somewhat. Outside of addresses we prefer and-of-shift because it
may lead to smaller immediates (FoldMaskAndShiftToScale is an even better
example because it create a non-canonical node). We currently don't recognize
addresses during DAGCombiner where arguably this canonicalization should be
performed. On the other hand, having this in the matcher allows us to cover
all the cases where an address can be used in an instruction.
I've also talked a little bit to Dan Gohman on llvm-dev who added the RAUW for
the new shift node in FoldMaskedShiftToScaledMask. This RAUW is responsible
for initiating the recursive CSE on users
(http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-September/076903.html) but it
is not strictly necessary since the shift is hooked into the visited user. Of
course it's safer to keep the DAG consistent at all times (e.g. for accurate
number of uses, etc.).
So rather than changing the fundamentals, I've decided to continue along the
previous patches and detect the CSE. This patch installs a very targeted
DAGUpdateListener for the duration of a complex-pattern match and updates the
matching state accordingly. (Previous patches used HandleSDNode to detect the
CSE but that's not practical here). The listener is only installed on X86.
I tested that there is no measurable overhead due to this while running
through the spec2k BC files with llc. The only thing we pay for is the
creation of the listener. The callback never ever triggers in spec2k since
this is a corner case.
Robin Morisset [Fri, 3 Oct 2014 18:10:07 +0000 (18:10 +0000)]
[Power] Delete redundant test Atomics-32.ll
The test Atomics-32.ll was both redundant (all operations are also checked by
atomics.ll at least) and not actually checking correctness (it was not using
FileCheck, just verifying that the compiler does not crash).
Daniel Sanders [Fri, 3 Oct 2014 15:37:37 +0000 (15:37 +0000)]
[mips] Print warning when using register names not available in N32/64
Summary:
The register names t4-t7 are not available in the N32 and N64 ABIs.
This patch prints a warning, when those names are used in N32/64,
along with a fix-it with the correct register names.
Sid Manning [Fri, 3 Oct 2014 13:18:11 +0000 (13:18 +0000)]
Adding skeleton for unit testing Hexagon Code Emission
Adding and modifying CMakeLists.txt files to run unit tests under
unittests/Target/* if the directory exists. Adding basic unit test to check
that code emitter object can be retrieved.
Differential Revision: http://reviews.llvm.org/D5523
Change by: Colin LeMahieu
[x86] Teach the new vector shuffle lowering to aggressively form MOVSS
and MOVSD nodes for single element vector inserts.
This is particularly important because a number of patterns in the
backend detect these patterns and leverage them to simplify things. It
also fixes quite a few of the insertion bad code examples. However, it
regresses a specific area: when available, blendps and blendpd are
*dramatically* faster than movss and movsd respectively. But it doesn't
really work to form the blend logic first because the blends *aren't* as
crazy efficient when the data is coming from memory anyways, and thus
will have a movss or movsd regardless. Also, doing that would block
a bunch of the patterns that this is designed to hit.
So my plan is to go into the patterns for lowering MOVSS and MOVSD and
lower them via blends when available. However that's a pretty invasive
restructuring so it will need to be a follow-up patch.
I have already gone into the patterns to lower MOVSS and MOVSD from
memory using MOVLPD, etc. Without that, several of the test cases
I already have regress.
Renato Golin [Fri, 3 Oct 2014 12:20:53 +0000 (12:20 +0000)]
Revert 202433 - Provide a target override for the latest regalloc heuristic
That commit was introduced in order to help investigate a problem in ARM
codegen breaking from commit 202304 (Add a limit to the heuristic that register
allocates instructions in local order). Recent analisys indicated that the
problem no longer exists, so I'm reverting this change.
[x86] Refactor the element insertion logic in the new vector shuffle
lowering to handle the potential mirroring of 2-element vectors (because
we can't reliably sort them one way) in the caller rather than in the
insertion logic.
This will simplify things considerably as more ways to fail to match the
insertion are added because now we have a nice try and retry point.
[x86] Fix the RUN-lines of this test to make sense.
I got them quite wrong when updating it and had the SSE4.1 run checked
for SSE2 and the SSE2 run checked for SSE4.1. I think everything was
actually generic SSE, but this still seems good to fix. While here,
hoist the triple into the IR and make the flag set a bit more direct in
what it is trying to test.
[x86] Significantly improve the ability of the new vector shuffle
lowering to match VZEXT_MOVL patterns.
I hadn't realized that these had sufficient pattern smarts in the
backend to lower zext-ing from the low element of a vector without it
being a scalar_to_vector node. They do, and this is how to match a bunch
of patterns for movq, movss, etc.
There is a weird propensity to end up using pshufd to place the element
afterward even though it means domain crossing (or rather, to use
xorps+movss to zext the element rather than movq) but that's an
orthogonal problem with VZEXT_MOVL that someone should probably look at.
[x86] Add some important, missing test coverage for blending from one
vector to a zero vector for the v2 cases and fix the v4 integer cases to
actually blend from a vector.
There are already seprate tests for the case of inserting from a scalar.
These cases cover a lot of the regressions I've seen in the regression
test suite for the new vector shuffle lowering and specifically cover
the reported lack of using various zext-ing instruction patterns. My
next patch should fix a big chunk of this, but wanted to get a nice
baseline for these patterns in the test cases first.
[x86] Remove a couple of fairly pointless tests. These were merely
testing that we generated divps and divss but not in a very systematic
way. There are other tests for widening binary operations already that
make these unnecessary.
The second one seems mostly about testing Atom as well as normal X86,
but despite the comment claiming it is testing a different instruction
sequence, it then tests for exactly the same div instruction sequence!
(The sequence of instructions is actually quite different on Atom, but
not the sequence of div instructions....)
And then it has an "execution" test that simply isn't run? Very strange.
Anyways, none of this is really needed so clean this up.
[x86] Add triples to these tests so that we see fewer calling convention
differences and they're a bit easier to maintain. This should fix the
tests on cygwin bots, etc.
[x86] Regenerate precise checks for a couple of test cases and remove
a test case that was just grepping the debug stats output rather than
actually checking the generated code for anything useful.
[x86] Remove an over-reduced test case. This would need to be
intergrated much more fully into some logical part of the backend to
really understand what it is trying to accomplish and how to update it.
I suspect it no longer holds enough value to be worth having.
[x86] This is a terrible SSE1 test, but we should keep it. I've deleted
two functions that really didn't have any interesting assertions, and
generated more precise tests for one of the others.
[x86] Regenerate a number of FileCheck assertions with my script for
test cases that will change with the new vector shuffle lowering. This
gives us a nice baseline for deltas against. I've checked and removed
the cases where there were weird register usage being pinned down, and
all of these are extremely pin-pointed tests so fully checking them
seems very appropriate.
[x86] Remove a couple of other overly isolated tests that are low-value
at this point. We have lots of tests of peephole optimizations with
insert and extract on vectors.
[x86] Regenerate a bunch more avx512 test cases using my script to have
tighter, more strict FileCheck assertions. Some of these I really like
as they show case exactly what instruction sequences come out of these
microscopic functionality tests.
[x86] Regenerate an avx512 test with my script to provide a nice
baseline for updates from the new vector shuffle lowering.
I've inspected the results here, and I couldn't find any register
allocation decisions where there should be any realistic way to register
allocate things differently. The closest was the imul test case. If you
see something here you'd like register number variables on, just shout
and I'll add them.
[x86] Remove some of the --show-mc-encoding flags from avx512 tests that
need to be updated for the new vector shuffle lowering.
After talking to Adam Nemet, Tim Northover, etc., it seems that testing
MC encodings in the same suite as the basic codegen isn't the right
approach. Instead, we're going to want dedicated MC tests for the
encodings. These encodings are starting to get in my way so I wanted to
cut them out early. The total set of instructions that should have
encoding tests added is:
Hal Finkel [Thu, 2 Oct 2014 22:34:22 +0000 (22:34 +0000)]
[PowerPC] Modern Book-E cores support sync
Older Book-E cores, such as the PPC 440, support only msync (which has the same
encoding as sync 0), but not any of the other sync forms. Newer Book-E cores,
however, do support sync, and for performance reasons we should allow the use
of the more-general form.
This refactors msync use into its own feature group so that it applies by
default only to older Book-E cores (of the relevant cores, we only have
definitions for the PPC440/450 currently).
Robin Morisset [Thu, 2 Oct 2014 22:27:07 +0000 (22:27 +0000)]
[Power] Improve the expansion of atomic loads/stores
Summary:
Atomic loads and store of up to the native size (32 bits, or 64 for PPC64)
can be lowered to a simple load or store instruction (as the synchronization
is already handled by AtomicExpand, and the atomicity is guaranteed thanks to
the alignment requirements of atomic accesses). This is exactly what this patch
does. Previously, these were implemented by complex
load-linked/store-conditional loops.. an obvious performance problem.
For example, this patch turns
```
define void @store_i8_unordered(i8* %mem) {
store atomic i8 42, i8* %mem unordered, align 1
ret void
}
```
from
```
_store_i8_unordered: ; @store_i8_unordered
; BB#0:
rlwinm r2, r3, 3, 27, 28
li r4, 42
xori r5, r2, 24
rlwinm r2, r3, 0, 0, 29
li r3, 255
slw r4, r4, r5
slw r3, r3, r5
and r4, r4, r3
LBB4_1: ; =>This Inner Loop Header: Depth=1
lwarx r5, 0, r2
andc r5, r5, r3
or r5, r4, r5
stwcx. r5, 0, r2
bne cr0, LBB4_1
; BB#2:
blr
```
into
```
_store_i8_unordered: ; @store_i8_unordered
; BB#0:
li r2, 42
stb r2, 0(r3)
blr
```
which looks like a pretty clear win to me.
Test Plan:
fixed the tests + new test for indexed accesses + make check-all
Fix the threshold added in r186434 (a re-apply of r185393) and updaated
to be a ManagedStatic in r218163 to not be a global variable written and
read to from within the innards of SpillPlacement.
This will fix a really scary race condition for anyone that has two
copies of LLVM running spill placement concurrently. Yikes!
This will also fix a really significant compile time hit that r218163
caused because the spill placement threshold read is actually in the
*very* hot path of this code. The memory fence on each read was showing
up as huge compile time regressions when spilling is responsible for
most of the compile time. For example, optimizing sanitized code showed
over 50% compile time regressions here. =/
Rui Ueyama [Thu, 2 Oct 2014 22:05:29 +0000 (22:05 +0000)]
llvm-readobj: print COFF imported symbols
This patch defines a new iterator for the imported symbols.
Make a change to COFFDumper to use that iterator to print
out imported symbols and its ordinals.
DI: Fold constant arguments into a single MDString
This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString. Integers are stringified and
a `\0` character is used as a separator.
Part of PR17891.
Note: I've attached my testcases upgrade scripts to the PR. If I've
just broken your out-of-tree testcases, they might help.
[x86] Teach the new vector shuffle lowering to widen floating point
elements as well as integer elements in order to form simpler shuffle
patterns.
This is the primary reason why we were failing to match some of the
2-and-2 floating point shuffles such as PR21140. Even after fixing this
we need to support some extra patterns in the backend in order to match
the resulting X86ISD::UNPCKL nodes into the correct instructions. This
commit should fix PR21140 and includes more comprehensive testing of
insertion patterns in v4 shuffles.
Not all of the added tests are beautiful. For example, we don't have
clever instructions to insert-via-load in the integer domain. There are
also some places where we aren't sufficiently cunning with our use of
movq and movd, but that's future work.
Justin Bogner [Thu, 2 Oct 2014 17:14:18 +0000 (17:14 +0000)]
InstrProf: Avoid linear search in a hot loop
Every time we were adding or removing an expression when generating a
coverage mapping we were doing a linear search to try and deduplicate
the list. The indices in the list are important, so we can't just
replace it by a DenseMap entirely, but an auxilliary DenseMap for fast
lookup massively improves the performance issues I was seeing here.
projects / llvm / log