AVX512 compare instructions return v*i1 types.
In cases where the number of elements in the returned value are less than 8, clang adds zeroes to get a mask of v8i1 type.
Later on it's replaced with CONCAT_VECTORS, which then is lowered to many DAG nodes including insert/extract element and shift right/left nodes.
The fact that AVX512 compare instructions put the result in a k register and zeroes all its upper bits allows us to remove the extra nodes simply by copying the result to the required register class.
When lowering, identify these cases and transform them into an INSERT_SUBVECTOR node (marked legal), then catch this pattern in instructions selection phase and transform it into one avx512 cmp instruction.
Differential Revision: https://reviews.llvm.org/D33188
llvm-svn: 305465
Summary:
This patch is part of 3 patches that together form a single patch, but must be introduced in stages in order not to break things.
The way that LLVM interprets DW_OP_plus in DIExpression nodes is basically that of the DW_OP_plus_uconst operator since LLVM expects an unsigned constant operand. This unnecessarily restricts the DW_OP_plus operator, preventing it from being used to describe the evaluation of runtime values on the expression stack. These patches try to align the semantics of DW_OP_plus and DW_OP_minus with that of the DWARF definition, which pops two elements off the expression stack, performs the operation and pushes the result back on the stack.
This is done in three stages:
• The first patch (LLVM) adds support for DW_OP_plus_uconst.
• The second patch (Clang) contains changes all its uses from DW_OP_plus to DW_OP_plus_uconst.
• The third patch (LLVM) changes the semantics of DW_OP_plus and DW_OP_minus to be in line with its DWARF meaning. This patch includes the bitcode upgrade from legacy DIExpressions.
Patch by Sander de Smalen.
Reviewers: echristo, pcc, aprantl
Reviewed By: aprantl
Subscribers: fhahn, javed.absar, aprantl, llvm-commits
Differential Revision: https://reviews.llvm.org/D33894
llvm-svn: 305386
The dream of a unified check-line auto-generator for all phases of compilation is dead.
The llc script has already diverged to be better at its goal, so having 2 scripts that
do almost the same thing is just causing confusion.
We can rip out the llc ability in update_test_checks.py next and rename it, so it will
be clear that we have one script for llc check auto-generation and another for opt.
llvm-svn: 305206
Summary:
This change enables the sin(x) cos(x) -> sincos(x) optimization on GNU
target triples. This optimization was being inhibited when -ffast-math
wasn't set because sincos in GLibC does not set errno, while sin and cos
do. However, this optimization will only run if the attributes on the
sin/cos calls include readnone, which is how clang represents the fact
that it doesn't care about the errno values set by these functions (via
the -fno-math-errno flag).
Reviewers: hfinkel, bogner
Subscribers: mcrosier, javed.absar, llvm-commits, paul.redmond
Differential Revision: https://reviews.llvm.org/D32921
llvm-svn: 305204
The dream of a unified check-line auto-generator for all phases of compilation is dead.
The llc script has already diverged to be better at its goal, so having 2 scripts that
do almost the same thing is just causing confusion for newcomers. I plan to fix up more
x86 tests in a next commit. We can rip out the llc ability in update_test_checks.py after
that.
llvm-svn: 305202
Summary:
The old check for slot overlap treated 2 slots `S` and `T` as
overlapping if there existed a CFG node in which both of the slots could
possibly be active. That is overly conservative and caused stack blowups
in Rust programs. Instead, check whether there is a single CFG node in
which both of the slots are possibly active *together*.
Fixes PR32488.
Patch by Ariel Ben-Yehuda <ariel.byd@gmail.com>
Reviewers: thanm, nagisa, llvm-commits, efriedma, rnk
Reviewed By: thanm
Subscribers: dotdash
Differential Revision: https://reviews.llvm.org/D31583
llvm-svn: 305193
I was looking closer at the x86 test diffs in D33866, and the first change seems like it
shouldn't happen in the first place. So this patch will resolve that.
Using Agner's tables and AMD docs, vperm2f128 and vinsertf128 have identical timing for
any given CPU model, so we should be able to interchange those without affecting perf.
But as we can see in some of the diffs here, using vperm2f128 allows load folding, so
we should take that opportunity to reduce code size and register pressure.
A secondary advantage is making AVX1 and AVX2 codegen more similar. Given that vperm2f128
was introduced with AVX1, we should be selecting it in all of the same situations that we
would with AVX2. If there's some reason that an AVX1 CPU would not want to use this
instruction, that should be fixed up in a later pass.
Differential Revision: https://reviews.llvm.org/D33938
llvm-svn: 305171
Summary: UADDO has 2 result, and one must check the result no before doing any kind of combine. Without it, the transform is invalid.
Reviewers: joerg
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34088
llvm-svn: 305162
This prevents against assertion errors like PR32659 which occur from a
replacement deleting a node after it's been added to the list argument
of RemoveDeadNodes. The specific failure from PR32659 does not
currently happen, but it is still potentially possible. The underlying
cause is that the callers of the change dfunction builds up a list of
nodes to delete after having moved their uses and it possible that a
move of a later node will cause a previously deleted nodes to be
deleted.
Reviewers: bkramer, spatel, davide
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D33731
llvm-svn: 305070
If we know that both operands of an unsigned integer vector comparison are non-negative,
then it's safe to directly use a signed-compare-greater-than instruction (the only non-equality
integer vector compare predicate provided by SSE/AVX).
We're intentionally not changing the condition code to signed in order to preserve the
existing transforms that use min/max/psubus below here.
This should solve PR33276:
https://bugs.llvm.org/show_bug.cgi?id=33276
Differential Revision: https://reviews.llvm.org/D33862
llvm-svn: 304909
Summary:
The patch makes instruction count the highest priority for
LSR solution for X86 (previously registers had highest priority).
Reviewers: qcolombet
Differential Revision: http://reviews.llvm.org/D30562
From: Evgeny Stupachenko <evstupac@gmail.com>
llvm-svn: 304824
If -simplify-mir option is passed then MIRPrinter will not print such fields.
This change also required some lit test cases in CodeGen directory to be changed.
Reviewed By: MatzeB
Differential Revision: https://reviews.llvm.org/D32304
llvm-svn: 304779
This is a negative test as pextrw doesn't write to all 32-bits of the
spilled GPR. This fold ended up happening when D32684 was landed and
covers the regression that motivated reverting it in r304762.
llvm-svn: 304763
In testing, we've found yet another miscompile caused by the new tables.
And this one is even less clear how to fix (we could teach it to fold
a 16-bit load instead of the 32-bit load it wants, or block folding
entirely).
Also, the approach to excluding instructions seems increasingly to not
scale well.
I have left a more detailed analysis on the review log for the original
patch (https://reviews.llvm.org/D32684) along with suggested path
forward. I will land an additional test case that I wrote which covers
the code that was miscompiling (folding into the output of `pextrw`) in
a subsequent commit to keep this a pure revert.
For each commit reverted here, I've restricted the revert to the
non-test code touching the x86 fold table emission until the last commit
where I did revert the test updates. This means the *new* test cases
added for `insertps` and `xchg` remain untouched (and continue to pass).
Reverted commits:
r304540: [X86] Don't fold into memory operands into insertps in the ...
r304347: [TableGen] Adapt more places to getValueAsString now ...
r304163: [X86] Don't fold away the memory operand of an xchg.
r304123: Don't capture a temporary std::string in a StringRef.
r304122: Resubmit "[X86] Adding new LLVM TableGen backend that ..."
Original commit was in r304088, and after a string of fixes was reverted
previously in r304121 to fix build bots, and then re-landed in r304122.
llvm-svn: 304762
- Move ISel (and pre-isel) pass construction into TargetPassConfig
- Extract AsmPrinter construction into a helper function
Putting the ISel code into TargetPassConfig seems a lot more natural and
both changes together make make it easier to build custom pipelines
involving .mir in an upcoming commit. This moves MachineModuleInfo to an
earlier place in the pass pipeline which shouldn't have any effect.
llvm-svn: 304754
There's nothing darwin-specific in these tests, and using that
setting causes extra phantom diffs when the auto-generated check
lines are regenerated today.
llvm-svn: 304753
Running `llc -verify-dom-info` on the attached testcase results in a
crash in the verifier, due to a stale dominator tree.
i.e.
DominatorTree is not up to date!
Computed:
=============================--------------------------------
Inorder Dominator Tree:
[1] %safe_mod_func_uint8_t_u_u.exit.i.i.i {0,7}
[2] %lor.lhs.false.i61.i.i.i {1,2}
[2] %safe_mod_func_int8_t_s_s.exit.i.i.i {3,6}
[3] %safe_div_func_int64_t_s_s.exit66.i.i.i {4,5}
Actual:
=============================--------------------------------
Inorder Dominator Tree:
[1] %safe_mod_func_uint8_t_u_u.exit.i.i.i {0,9}
[2] %lor.lhs.false.i61.i.i.i {1,2}
[2] %safe_mod_func_int8_t_s_s.exit.i.i.i {3,8}
[3] %safe_div_func_int64_t_s_s.exit66.i.i.i {4,5}
[3] %safe_mod_func_int8_t_s_s.exit.i.i.i.lor.lhs.false.i61.i.i.i_crit_edge {6,7}
This is because in `SelectionDAGIsel` we split critical edges without
updating the corresponding dominator for the function (and we claim
in `MachineFunctionPass::getAnalysisUsage()` that the domtree is preserved).
We could either stop preserving the domtree in `getAnalysisUsage`
or tell `splitCriticalEdge()` to update it.
As the second option is easy to implement, that's the one I chose.
Differential Revision: https://reviews.llvm.org/D33800
llvm-svn: 304742
We currently generate BUILD_VECTOR as a tree of UNPCKL shuffles of the same type:
e.g. for v4f32:
Step 1: unpcklps 0, 2 ==> X: <?, ?, 2, 0>
: unpcklps 1, 3 ==> Y: <?, ?, 3, 1>
Step 2: unpcklps X, Y ==> <3, 2, 1, 0>
The issue is because we are not placing sequential vector elements together early enough, we fail to recognise many combinable patterns - consecutive scalar loads, extractions etc.
Instead, this patch unpacks progressively larger sequential vector elements together:
e.g. for v4f32:
Step 1: unpcklps 0, 2 ==> X: <?, ?, 1, 0>
: unpcklps 1, 3 ==> Y: <?, ?, 3, 2>
Step 2: unpcklpd X, Y ==> <3, 2, 1, 0>
This does mean that we are creating UNPCKL shuffle of different value types, but the relevant combines that benefit from this are quite capable of handling the additional BITCASTs that are now included in the shuffle tree.
Differential Revision: https://reviews.llvm.org/D33864
llvm-svn: 304688
There's nothing darwin-specific in these tests, and using
that setting causes extra phantom diffs when the auto-generated
check lines are regenerated today.
llvm-svn: 304614
This pass allows to run the register scavenging independently of
PrologEpilogInserter to allow targeted testing.
Also adds some basic register scavenging tests.
llvm-svn: 304606
Since r288804, we try to lower build_vectors on AVX using broadcasts of
float/double. However, when we broadcast integer values that happen to
have a NaN float bitpattern, we lose the NaN payload, thereby changing
the integer value being broadcast.
This is caused by ConstantFP::get, to which we pass the splat i32 as
a float (by bitcasting it using bitsToFloat). ConstantFP::get takes
a double parameter, so we end up lossily converting a single-precision
NaN to double-precision.
Instead, avoid any kinds of conversions by directly building an APFloat
from the splatted APInt.
Note that this also fixes another piece of code (broadcast of
subvectors), that currently isn't susceptible to the same problem.
Also note that we could really just use APInt and ConstantInt
throughout: the constant pool type doesn't matter much. Still, for
consistency, use the appropriate type.
llvm-svn: 304590
This initial patch doesn't actually do much useful. It's just to show where the new code goes. Once this is in, I'll extend the verification logic to check more useful properties.
For those curious, the more complicated version of this patch already found one very suspicious thing.
Differential Revision: https://reviews.llvm.org/D33819
llvm-svn: 304564
insertps behaves differently, the register form selects from an input
register based on the immediate operand while the memory form just loads
the given address. We have custom code to change the immediate in cases
where that's legal, so completely remove insertps from the generated
tables.
llvm-svn: 304540
Summary:
Add an early combine to match patterns such as:
(i16 bitcast (v16i1 x))
->
(i16 movmsk (v16i8 sext (v16i1 x)))
This combine needs to happen early enough before
type-legalization scalarizes the result of the setcc.
Reviewers: igorb, craig.topper, RKSimon
Subscribers: delena, llvm-commits
Differential Revision: https://reviews.llvm.org/D33311
llvm-svn: 304406
Summary: This pattern is no very useful per se, but it exposes optimization for toehr patterns that wouldn't kick in otherwize. It's very common and worth optimizing for.
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32756
llvm-svn: 304402
Summary: LiveRangeShrink pass moves instruction right after the definition with the same BB if the instruction and its operands all have more than one use. This pass is inexpensive and guarantees optimal live-range within BB.
Reviewers: davidxl, wmi, hfinkel, MatzeB, andreadb
Reviewed By: MatzeB, andreadb
Subscribers: hiraditya, jyknight, sanjoy, skatkov, gberry, jholewinski, qcolombet, javed.absar, krytarowski, atrick, spatel, RKSimon, andreadb, MatzeB, mehdi_amini, mgorny, efriedma, davide, dberlin, llvm-commits
Differential Revision: https://reviews.llvm.org/D32563
llvm-svn: 304371
We should have a single call site entry with no landing pad. This
indicates that no EH action should be taken and the unwinder should
unwind to the next frame.
We currently don't recognize __gxx_personality_seh0 as a known
personality, so we forcibly emit a table, and that table was wrong. This
was filed as PR33220. Now we emit a correct table for that personality.
The next step is to recognize that we can completely skip the table for
this personality.
llvm-svn: 304363
Summary:
If we attempt to unfold an SUnit in ScheduleDAG that results in
finding an already scheduled load, we must should abort the
unfold as it will not improve scheduling.
This fixes PR32610.
Reviewers: jmolloy, sunfish, bogner, spatel
Subscribers: llvm-commits, MatzeB
Differential Revision: https://reviews.llvm.org/D32911
llvm-svn: 304321
xchg with a mem operand has different locking semantics. If we unfold it
into a xchg r,r we will loose the implicit lock. Likewise we never want
to fold a register xchg into a memory one as it would be a lot slower.
This triggers during LLVM selfhost.
llvm-svn: 304163
The extending load possibility was missed in:
https://reviews.llvm.org/rL304072
We might want to handle this cases as a follow-up, but bailing out for now
to avoid miscompiling.
llvm-svn: 304153
This was reverted due to buildbot breakages and I was not familiar
with this code to investigate it. But while trying to get a
useful backtrace for the author, it turns out the fix was very
obvious. Resubmitting this patch as is, and will submit the
fix in a followup so that the fix is not hidden in the larger
CL.
llvm-svn: 304122
This reverts commit 28cb1003507f287726f43c771024a1dc102c45fe as well
as all subsequent followups. llvm-tblgen currently segfaults with
this change, and it seems it has been broken on the bots all
day with no fixes in preparation. See, for example:
http://lab.llvm.org:8011/builders/clang-x86-windows-msvc2015/
llvm-svn: 304121
X86 backend holds huge tables in order to map between the register and memory forms of each instruction.
This TableGen Backend automatically generated all these tables with the appropriate flags for each entry.
Differential Revision: https://reviews.llvm.org/D32684
llvm-svn: 304088
If we have (extract_subvector(load wide vector)) with no other users,
that can just be (load narrow vector). This is intentionally conservative.
Follow-ups may loosen the one-use constraint to account for the extract cost
or just remove the one-use check.
The memop chain updating is based on code that already exists multiple times
in x86 lowering, so that should be pulled into a helper function as a follow-up.
Background: this is a potential improvement noticed via regressions caused by
making x86's peekThroughBitcasts() not loop on consecutive bitcasts (see
comments in D33137).
Differential Revision: https://reviews.llvm.org/D33578
llvm-svn: 304072
Rewrite fixupKills() to use the LivePhysRegs class. Simplifies the code
and fixes a bug where the CSR registers in return blocks where missed
leading to invalid kill flags. Also remove the unnecessary rule that we
wouldn't set kill flags on tied operands.
No tests as I have an upcoming commit improving MachineVerifier checks
to catch these cases in multiple existing lit tests.
llvm-svn: 304055
In the best case:
extract (binop (concat X1, X2), (concat Y1, Y2)), N --> binop XN, YN
...we kill all of the extract/concat and just have narrow binops remaining.
If only one of the binop operands is amenable, this transform is still
worthwhile because we kill some of the extract/concat.
Optional bitcasting makes the code more complicated, but there doesn't
seem to be a way to avoid that.
The TODO about extending to more than bitwise logic is there because we really
will regress several x86 tests including madd, psad, and even a plain
integer-multiply-by-2 or shift-left-by-1. I don't think there's anything
fundamentally wrong with this patch that would cause those regressions; those
folds are just missing or brittle.
If we extend to more binops, I found that this patch will fire on at least one
non-x86 regression test. There's an ARM NEON test in
test/CodeGen/ARM/coalesce-subregs.ll with a pattern like:
t5: v2f32 = vector_shuffle<0,3> t2, t4
t6: v1i64 = bitcast t5
t8: v1i64 = BUILD_VECTOR Constant:i64<0>
t9: v2i64 = concat_vectors t6, t8
t10: v4f32 = bitcast t9
t12: v4f32 = fmul t11, t10
t13: v2i64 = bitcast t12
t16: v1i64 = extract_subvector t13, Constant:i32<0>
There was no functional change in the codegen from this transform from what I
could see though.
For the x86 test changes:
1. PR32790() is the closest call. We don't reduce the AVX1 instruction count in that case,
but we improve throughput. Also, on a core like Jaguar that double-pumps 256-bit ops,
there's an unseen win because two 128-bit ops have the same cost as the wider 256-bit op.
SSE/AVX2/AXV512 are not affected which is expected because only AVX1 has the extract/concat
ops to match the pattern.
2. do_not_use_256bit_op() is the best case. Everyone wins by avoiding the concat/extract.
Related bug for IR filed as: https://bugs.llvm.org/show_bug.cgi?id=33026
3. The SSE diffs in vector-trunc-math.ll are just scheduling/RA, so nothing real AFAICT.
4. The AVX1 diffs in vector-tzcnt-256.ll are all the same pattern: we reduced the instruction
count by one in each case by eliminating two insert/extract while adding one narrower logic op.
https://bugs.llvm.org/show_bug.cgi?id=32790
Differential Revision: https://reviews.llvm.org/D33137
llvm-svn: 303997
Rename the DEBUG_TYPE to match the names of corresponding passes where
it makes sense. Also establish the pattern of simply referencing
DEBUG_TYPE instead of repeating the passname where possible.
llvm-svn: 303921
AVX512_VPOPCNTDQ is a new feature set that was published by Intel.
The patch represents the LLVM side of the addition of two new intrinsic based instructions (vpopcntd and vpopcntq).
Differential Revision: https://reviews.llvm.org/D33169
llvm-svn: 303858
I suspect this buildbot has slow-incdec set by default, most likely due to
the default CPU having this set. This feature bit can prevent optsize from
having an effect on this IR.
llvm-svn: 303720
Summary:
It's rare but a small number of patterns use IntInit's at the root of the match.
On X86, one such rule is enabled by the OptForSize predicate and causes the
compiler to use the smaller:
%0 = MOV32r1
instead of the usual:
%0 = MOV32ri 1
This patch adds support for matching IntInit's at the root and uses this as a
test case for the optsize attribute that was implemented in r301750
Reviewers: qcolombet, ab, t.p.northover, rovka, kristof.beyls, aditya_nandakumar
Reviewed By: qcolombet
Subscribers: igorb, llvm-commits
Differential Revision: https://reviews.llvm.org/D32791
llvm-svn: 303678
Summary:
While this makes some case better and some case worse - so it's unclear if it is a worthy combine just by itself - this is a useful canonicalisation.
As per discussion in D32756 .
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32916
llvm-svn: 303441
This patch defines the i1 type as illegal in the X86 backend for AVX512.
For DAG operations on <N x i1> types (build vector, extract vector element, ...) i8 is used, and should be truncated/extended.
This should produce better scalar code for i1 types since GPRs will be used instead of mask registers.
Differential Revision: https://reviews.llvm.org/D32273
llvm-svn: 303421
This also reverts follow-ups r303292 and r303298.
It broke some Chromium tests under MSan, and apparently also internal
tests at Google.
llvm-svn: 303369
This provides a new way to access the TargetMachine through
TargetPassConfig, as a dependency.
The patterns replaced here are:
* Passes handling a null TargetMachine call
`getAnalysisIfAvailable<TargetPassConfig>`.
* Passes not handling a null TargetMachine
`addRequired<TargetPassConfig>` and call
`getAnalysis<TargetPassConfig>`.
* MachineFunctionPasses now use MF.getTarget().
* Remove all the TargetMachine constructors.
* Remove INITIALIZE_TM_PASS.
This fixes a crash when running `llc -start-before prologepilog`.
PEI needs StackProtector, which gets constructed without a TargetMachine
by the pass manager. The StackProtector pass doesn't handle the case
where there is no TargetMachine, so it segfaults.
Related to PR30324.
Differential Revision: https://reviews.llvm.org/D33222
llvm-svn: 303360
According to Intel's Optimization Reference Manual for SNB+:
" For LEA instructions with three source operands and some specific situations, instruction latency has increased to 3 cycles, and must
dispatch via port 1:
- LEA that has all three source operands: base, index, and offset
- LEA that uses base and index registers where the base is EBP, RBP,or R13
- LEA that uses RIP relative addressing mode
- LEA that uses 16-bit addressing mode "
This patch currently handles the first 2 cases only.
Differential Revision: https://reviews.llvm.org/D32277
llvm-svn: 303333
- '-verify-mahcineinstrs' starts to complain allocatable live-in physical
registers on non-entry or non-landing-pad basic blocks.
- Refactor the XBEGIN translation to define EAX on a dedicated fallback code
path due to XABORT. Add a pseudo instruction to define EAX explicitly to
avoid add physical register live-in.
Differential Revision: https://reviews.llvm.org/D33168
llvm-svn: 303306
Summary:
In SelectionDAG, when a store is immediately chained to another store
to the same address, elide the first store as it has no observable
effects. This is causes small improvements dealing with intrinsics
lowered to stores.
Test notes:
* Many testcases overwrite store addresses multiple times and needed
minor changes, mainly making stores volatile to prevent the
optimization from optimizing the test away.
* Many X86 test cases optimized out instructions associated with
associated with va_start.
* Note that test_splat in CodeGen/AArch64/misched-stp.ll no longer has
dependencies to check and can probably be removed and potentially
replaced with another test.
Reviewers: rnk, john.brawn
Subscribers: aemerson, rengolin, qcolombet, jyknight, nemanjai, nhaehnle, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D33206
llvm-svn: 303198
According to Intel's Optimization Reference Manual for SNB+:
" For LEA instructions with three source operands and some specific situations, instruction latency has increased to 3 cycles, and must
dispatch via port 1:
- LEA that has all three source operands: base, index, and offset
- LEA that uses base and index registers where the base is EBP, RBP,or R13
- LEA that uses RIP relative addressing mode
- LEA that uses 16-bit addressing mode "
This patch currently handles the first 2 cases only.
Differential Revision: https://reviews.llvm.org/D32277
llvm-svn: 303183
Shrink-wrapping uses post-dominators to find a restore point that
post-dominates all the uses of CSR / stack.
The way dominator trees are modeled in LLVM today is that unreachable
blocks are not present in a generic dominator tree, so, an unreachable node is
dominated by anything: include/llvm/Support/GenericDomTree.h:467.
Since for post-dominators, a no-return block is considered
"unreachable", calling findNearestCommonDominator on an unreachable node
A and a non-unreachable node B, will return B, which can be false. If we
find such node, we bail out since there is no good restore point
available.
rdar://problem/30186931
llvm-svn: 303130
At O3 we are more willing to increase size if we believe it will improve
performance. The current threshold for tail-duplication of 2 instructions is
conservative, and can be relaxed at O3.
Benchmark results:
llvm test-suite:
6% improvement in aha, due to duplication of loop latch
3% improvement in hexxagon
2% slowdown in lpbench. Seems related, but couldn't completely diagnose.
Internal google benchmark:
Produces 4% improvement on internal google protocol buffer serialization
benchmarks.
Differential-Revision: https://reviews.llvm.org/D32324
llvm-svn: 303084
Currently, when masked load, store, gather or scatter intrinsics are used, we check in CodeGenPrepare pass if the subtarget support these intrinsics, if not we replace them with scalar code - this is a functional transformation not an optimization (not optional).
CodeGenPrepare pass does not run when the optimization level is set to CodeGenOpt::None (-O0).
Functional transformation should run with all optimization levels, so here I created a new pass which runs on all optimization levels and does no more than this transformation.
Differential Revision: https://reviews.llvm.org/D32487
llvm-svn: 303050
I noticed the 512-bit lzcnts don't use the X86 specific lookup table code and instead use the EXPAND case in LegalizeDAG. I was toying around with fixing this and noticed it would require compare instructions that generate i1 masks and then converting from mask to vector. Then I noticed that we don't test which compares are used with avx512vl and no avx512cd.
llvm-svn: 303020
Remove an unneeded prefix from the 32-bit command line. Make all the 64-bit triples match. Replace ALL with X64 and remove it from the 32-bit test.
llvm-svn: 303019
Summary: LiveRangeShrink pass moves instruction right after the definition with the same BB if the instruction and its operands all have more than one use. This pass is inexpensive and guarantees optimal live-range within BB.
Reviewers: davidxl, wmi, hfinkel, MatzeB, andreadb
Reviewed By: MatzeB, andreadb
Subscribers: hiraditya, jyknight, sanjoy, skatkov, gberry, jholewinski, qcolombet, javed.absar, krytarowski, atrick, spatel, RKSimon, andreadb, MatzeB, mehdi_amini, mgorny, efriedma, davide, dberlin, llvm-commits
Differential Revision: https://reviews.llvm.org/D32563
llvm-svn: 302938
manages to form a VSELECT with a non-i1 element type condition. Those
are technically allowed in SDAG (at least, the generic type legalization
logic will form them and I wouldn't want to try to audit everything te
preclude forming them) so we need to be able to lower them.
This isn't too hard to implement. We mark VSELECT as custom so we get
a chance in C++, add a fast path for i1 conditions to get directly
handled by the patterns, and a fallback when we need to manually force
the condition to be an i1 that uses the vptestm instruction to turn
a non-mask into a mask.
This, unsurprisingly, generates awful code. But it at least doesn't
crash. This was actually impacting open source packages built with LLVM
for AVX-512 in the wild, so quickly landing a patch that at least stops
the immediate bleeding.
I think I've found where to fix the codegen quality issue, but less
confident of that change so separating it out from the thing that
doesn't change the result of any existing test case but causes mine to
not crash.
llvm-svn: 302785
This reverts r302712.
The change fails with ASAN enabled:
ERROR: AddressSanitizer: use-after-poison on address ... at ...
READ of size 2 at ... thread T0
#0 ... in llvm::SDNode::getNumValues() const <snip>/include/llvm/CodeGen/SelectionDAGNodes.h:855:42
#1 ... in llvm::SDNode::hasAnyUseOfValue(unsigned int) const <snip>/lib/CodeGen/SelectionDAG/SelectionDAG.cpp:7270:3
#2 ... in llvm::SDValue::use_empty() const <snip> include/llvm/CodeGen/SelectionDAGNodes.h:1042:17
#3 ... in (anonymous namespace)::DAGCombiner::MergeConsecutiveStores(llvm::StoreSDNode*) <snip>/lib/CodeGen/SelectionDAG/DAGCombiner.cpp:12944:7
Reviewers: niravd
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D33081
llvm-svn: 302746
Summary:
Allow consecutive stores whose values come from consecutive loads to
merged in the presense of other uses of the loads. Previously this was
disallowed as in general the merged load cannot be shared with the
other uses. Merging N stores into 1 may cause as many as N redundant
loads. However in the context of caching this should have neglible
affect on memory pressure and reduce instruction count making it
almost always a win.
Fixes PR32086.
Reviewers: spatel, jyknight, andreadb, hfinkel, efriedma
Reviewed By: efriedma
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D30471
llvm-svn: 302712
This pass uses a new target hook to decide whether or not to expand a particular
intrinsic to the shuffevector sequence.
Differential Revision: https://reviews.llvm.org/D32245
llvm-svn: 302631
This is a follow-up to r302611, which moved an -O0 computation of DT
from SDAGISel to TwoAddress.
Don't use it here either, and avoid computing it completely. The only
use was forwarding the analysis as an optional argument to utility
functions.
Differential Revision: https://reviews.llvm.org/D32766
llvm-svn: 302612
Before r247167, the pass manager builder controlled which AA
implementations were used, exporting them all in the AliasAnalysis
analysis group.
Now, AAResultsWrapperPass always uses BasicAA, but still uses other AA
implementations if made available in the pass pipeline.
But regardless, SDAGISel is required at O0, and really doesn't need to
be doing fancy optimizations based on useful AA results.
Don't require AA at CodeGenOpt::None, and only use it otherwise.
This does have a functional impact (and one testcase is pessimized
because we can't reuse a load). But I think that's desirable no matter
what.
Note that this alone doesn't result in less DT computations: TwoAddress
was previously able to reuse the DT we computed for SDAG. That will be
fixed separately.
Differential Revision: https://reviews.llvm.org/D32766
llvm-svn: 302611
We currently require SCEV, which requires DT/LI. Those are expensive to
compute, but the pass only runs for functions that have the safestack
attribute.
Compute DT/LI to build SCEV lazily, only when the pass is actually going
to transform the function.
Differential Revision: https://reviews.llvm.org/D31302
llvm-svn: 302610
This should hopefully makes changes to the O0 pipeline obvious; it's
easy to require expensive passes, and this helps make informed
decisions.
Case in point: in the few weeks separating the time when I initially
wrote this patch to the time when I committed, the test regressed as
r302103 added another use of DT!
llvm-svn: 302608
This patch adds more patterns that a reasonable person might write that can be compiled to BZHI.
This adds support for
(~0U >> (32 - b)) & a;
and
a << (32 - b) >> (32 - b);
This was inspired by the code in APInt::clearUnusedBits.
This can pass an index of 32 to the bzhi instruction which a quick test of Haswell hardware shows will not mask any bits. Though the description text in the Intel manual says the "index is saturated to OperandSize-1". The pseudocode in the same manual indicates no bits will be zeroed for this case.
I think this is still missing cases where the subtract portion is an 8-bit operation.
Differential Revision: https://reviews.llvm.org/D32616
llvm-svn: 302549
for scalar masked instructions only the lower bit of the mask is relevant. so for constant masks we should either do an unmasked operation or no operation, depending on the value of the lower bit.
This patch handles cases where the lower bit is '1'.
Differential Revision: https://reviews.llvm.org/D32805
llvm-svn: 302546
The modified tests should test the masked intrinsics.
Currently the mask is constant, which with a future patch (https://reviews.llvm.org/D32805) will cause the intrinsics to be replaced with an unmasked version.
This patch changes the constant mask to be a variable one.
llvm-svn: 302529
Using arguments with attribute inalloca creates problems for verification
of machine representation. This attribute instructs the backend that the
argument is prepared in stack prior to CALLSEQ_START..CALLSEQ_END
sequence (see http://llvm.org/docs/InAlloca.htm for details). Frame size
stored in CALLSEQ_START in this case does not count the size of this
argument. However CALLSEQ_END still keeps total frame size, as caller can
be responsible for cleanup of entire frame. So CALLSEQ_START and
CALLSEQ_END keep different frame size and the difference is treated by
MachineVerifier as stack error. Currently there is no way to distinguish
this case from actual errors.
This patch adds additional argument to CALLSEQ_START and its
target-specific counterparts to keep size of stack that is set up prior to
the call frame sequence. This argument allows MachineVerifier to calculate
actual frame size associated with frame setup instruction and correctly
process the case of inalloca arguments.
The changes made by the patch are:
- Frame setup instructions get the second mandatory argument. It
affects all targets that use frame pseudo instructions and touched many
files although the changes are uniform.
- Access to frame properties are implemented using special instructions
rather than calls getOperand(N).getImm(). For X86 and ARM such
replacement was made previously.
- Changes that reflect appearance of additional argument of frame setup
instruction. These involve proper instruction initialization and
methods that access instruction arguments.
- MachineVerifier retrieves frame size using method, which reports sum of
frame parts initialized inside frame instruction pair and outside it.
The patch implements approach proposed by Quentin Colombet in
https://bugs.llvm.org/show_bug.cgi?id=27481#c1.
It fixes 9 tests failed with machine verifier enabled and listed
in PR27481.
Differential Revision: https://reviews.llvm.org/D32394
llvm-svn: 302527
Similar to what we do for vXi8 ASHR(X, 7), use SSE42's PCMPGTQ to splat the sign instead of using the PSRAD+PSHUFD.
Avoiding bitcasts this improves combines that utilize computeNumSignBits, permits memory folding and reduces pipe pressure. Although it does require a second register, given that this is a (cheap) zero register the impact is minimal.
Differential Revision: https://reviews.llvm.org/D32973
llvm-svn: 302525
This reverts commit r302461.
It appears to be causing failures compiling gtest with debug info on the
Linux sanitizer bot. I was unable to reproduce the failure locally,
however.
llvm-svn: 302504
Summary:
For inalloca functions, this is a very common code pattern:
%argpack = type <{ i32, i32, i32 }>
define void @f(%argpack* inalloca %args) {
entry:
%a = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 0
%b = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 1
%c = getelementptr inbounds %argpack, %argpack* %args, i32 0, i32 2
tail call void @llvm.dbg.declare(metadata i32* %a, ... "a")
tail call void @llvm.dbg.declare(metadata i32* %c, ... "b")
tail call void @llvm.dbg.declare(metadata i32* %b, ... "c")
Even though these GEPs can be simplified to a constant offset from EBP
or RSP, we don't do that at -O0, and each GEP is computed into a
register. Registers used to compute argument addresses are typically
spilled and clobbered very quickly after the initial computation, so
live debug variable tracking loses information very quickly if we use
DBG_VALUE instructions.
This change moves processing of dbg.declare between argument lowering
and basic block isel, so that we can ask if an argument has a frame
index or not. If the argument lives in a register as is the case for
byval arguments on some targets, then we don't put it in the side table
and during ISel we emit DBG_VALUE instructions.
Reviewers: aprantl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32980
llvm-svn: 302483
Summary:
An llvm.dbg.declare of a static alloca is always added to the
MachineFunction dbg variable map, so these values are entirely
redundant. They survive all the way through codegen to be ignored by
DWARF emission.
Effectively revert r113967
Two bugpoint-reduced test cases from 2012 broke as a result of this
change. Despite my best efforts, I haven't been able to rewrite the test
case using dbg.value. I'm not too concerned about the lost coverage
because these were reduced from the test-suite, which we still run.
Reviewers: aprantl, dblaikie
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32920
llvm-svn: 302461
Currently combineLogicBlendIntoPBLENDV can only match ASHR to detect sign splatting of a bit mask, this patch generalises this to use computeNumSignBits instead.
This is a first step in several things we can do to improve PBLENDV support:
* Better matching of X86ISD::ANDNP patterns.
* Handle floating point cases.
* Better vector and bitcast support in computeNumSignBits.
* Recognise that PBLENDV only uses the sign bit of the mask, we should be able strip away sign splats (ASHR, PCMPGT isNeg tests etc.).
Differential Revision: https://reviews.llvm.org/D32953
llvm-svn: 302424
This patch introduces an LLVM intrinsic and a target opcode for custom event
logging in XRay. Initially, its use case will be to allow users of XRay to log
some type of string ("poor man's printf"). The target opcode compiles to a noop
sled large enough to enable calling through to a runtime-determined relative
function call. At runtime, when X-Ray is enabled, the sled is replaced by
compiler-rt with a trampoline to the logic for creating the custom log entries.
Future patches will implement the compiler-rt parts and clang-side support for
emitting the IR corresponding to this intrinsic.
Reviewers: timshen, dberris
Subscribers: igorb, pelikan, rSerge, timshen, echristo, dberris, llvm-commits
Differential Revision: https://reviews.llvm.org/D27503
llvm-svn: 302405
rL294581 broke unnecessary register dependencies on partial v16i8/v8i16 BUILD_VECTORs, but on SSE41 we (currently) use insertion for full BUILD_VECTORs as well. By allowing full insertion to occur on SSE41 targets we can break register dependencies here as well.
llvm-svn: 302355
Remove an extra canonicalization step if ISD::ABS is going to be used anyway.
Updated x86 abs combine to check that we are lowering from both canonicalizations.
llvm-svn: 302337
- MIParser: If the successor list is not specified successors will be
added based on basic block operands in the block and possible
fallthrough.
- MIRPrinter: Adds a new `simplify-mir` option, with that option set:
Skip printing of block successor lists in cases where the
parser is guaranteed to reconstruct it. This means we still print the
list if some successor cannot be determined (happens for example for
jump tables), if the successor order changes or branch probabilities
being unequal.
Differential Revision: https://reviews.llvm.org/D31262
llvm-svn: 302289
This avoids problems on code like this:
char buf[16];
__asm {
movups xmm0, [buf]
mov [buf], eax
}
The frontend size in this case (1) is wrong, and the register makes the
instruction matching unambiguous. There are also enough bytes available
that we shouldn't complain to the user that they are potentially using
an incorrectly sized instruction to access the variable.
Supersedes D32636 and D26586 and fixes PR28266
llvm-svn: 302179
According to psABI, PLT stub clobbers XMM8-XMM15.
In Regcall calling convention those registers are used for passing parameters.
Thus we need to prevent lazy binding in Regcall.
Differential Revision: https://reviews.llvm.org/D32430
llvm-svn: 302124
Summary:
This change adds a new section to the xray-instrumented binary that
stores an index into ranges of the instrumentation map, where sleds
associated with the same function can be accessed as an array. At
runtime, we can get access to this index by function ID offset allowing
for selective patching and unpatching by function ID.
Each entry in this new section (xray_fn_idx) will include two pointers
indicating the start and one past the end of the sleds associated with
the same function. These entries will be 16 bytes long on x86 and
aarch64. On arm, we align to 16 bytes anyway so the runtime has to take
that into consideration.
__{start,stop}_xray_fn_idx will be the symbols that the runtime will
look for when we implement the selective patching/unpatching by function
id APIs. Because XRay synthesizes the function id's in a monotonically
increasing manner at runtime now, implementations (and users) can use
this table to look up the sleds associated with a specific function.
This is useful in implementations that want to do things like:
- Implement coverage mode for functions by patching everything
pre-main, then as functions are encountered, the installed handler
can unpatch the function that's been encountered after recording
that it's been called.
- Do "learning mode", so that the implementation can figure out some
statistical information about function calls by function id for a
time being, and then determine which functions are worth
uninstrumenting at runtime.
- Do "selective instrumentation" where an implementation can
specifically instrument only certain function id's at runtime
(either based on some external data, or through some other
heuristics) instead of patching all the instrumented functions at
runtime.
Reviewers: dblaikie, echristo, chandlerc, javed.absar
Subscribers: pelikan, aemerson, kpw, llvm-commits, rengolin
Differential Revision: https://reviews.llvm.org/D32693
llvm-svn: 302109
Summary:
This is an implementation of the loop detection logic that XRay needs to
determine whether a function might take time at runtime. Without this
heuristic, XRay will tend to not instrument short functions that have
loops that might have runtime dependent on inputs or external values.
While this implementation doesn't do any further analysis than just
figuring out whether there is a loop in the MachineFunction being
code-gen'ed, we're paving the way for being able to perform more
sophisticated analysis of the function in the future (for example to
determine whether the trip count for the loop might be constant, and
make a decision on that instead). This enables us to cover more
functions with the default heuristics, and potentially identify ones
that have variable runtime latency just by looking for the presence of
loops.
Reviewers: chandlerc, rnk, pelikan
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32274
llvm-svn: 302103
Summary:
The existing implementation creates a symbolic SCEV expression every
time we analyze a phi node and then has to remove it, when the analysis
is finished. This is very expensive, and in most of the cases it's also
unnecessary. According to the data I collected, ~60-70% of analyzed phi
nodes (measured on SPEC) have the following form:
PN = phi(Start, OP(Self, Constant))
Handling such cases separately significantly speeds this up.
Reviewers: sanjoy, pete
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32663
llvm-svn: 302096
Summary: Do the transform when the carry isn't used. It's a pattern exposed when legalizing large integers.
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32755
llvm-svn: 302047
This patch adds support for the the LightWeight Profiling (LWP) instructions which are available on all AMD Bulldozer class CPUs (bdver1 to bdver4).
Reapplied - this time without changing line endings of existing files.
Differential Revision: https://reviews.llvm.org/D32769
llvm-svn: 302041
This patch adds support for the the LightWeight Profiling (LWP) instructions which are available on all AMD Bulldozer class CPUs (bdver1 to bdver4).
Differential Revision: https://reviews.llvm.org/D32769
llvm-svn: 302028
Fixes PR31789 - When loop-vectorize tries to use these intrinsics for a
non-default address space pointer we fail with a "Calling a function with a
bad singature!" assertion. This patch solves this by adding the 'vector of
pointers' argument as an overloaded type which will determine the address
space.
Differential revision: https://reviews.llvm.org/D31490
llvm-svn: 302018
The "macosx" OS type is still the canonical type. In the future "macos" will
become the canonical OS type (but we will still support "macosx").
rdar://27043820
Differential Revision: https://reviews.llvm.org/D32748
llvm-svn: 302011
Summary: This is a common pattern that arise when legalizing large integers operations. Only do it when Y + 1 cannot overflow as this would change the carry behavior of uaddo .
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D32687
llvm-svn: 301922
Summary: Common pattern when legalizing large integers operations. Similar to D32687, when the carry isn't used.
Reviewers: jyknight, nemanjai, mkuper, spatel, RKSimon, zvi, bkramer
Differential Revision: https://reviews.llvm.org/D32738
llvm-svn: 301919
Ayman Musa