On modern Intel processors hardware SQRT in many cases is faster than RSQRT
followed by Newton-Raphson refinement. The patch introduces a simple heuristic
to choose between hardware SQRT instruction and Newton-Raphson software
estimation.
The patch treats scalars and vectors differently. The heuristic is that for
scalars the compiler should optimize for latency while for vectors it should
optimize for throughput. It is based on the assumption that throughput bound
code is likely to be vectorized.
Basically, the patch disables scalar NR for big cores and disables NR completely
for Skylake. Firstly, scalar SQRT has shorter latency than NR code in big cores.
Secondly, vector SQRT has been greatly improved in Skylake and has better
throughput compared to NR.
Differential Revision: https://reviews.llvm.org/D21379
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277725 91177308-0d34-0410-b5e6-96231b3b80d8
When using orbis-llvm-cov.exe to generate the HTML report, the HTML report
can look quite different to the source file if it includes tabs.The default
tab size is 2 spaces instead of 8 spaces. A command line switch is
be added to set the tab size.
Differential Revision: https://reviews.llvm.org/D23087
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277715 91177308-0d34-0410-b5e6-96231b3b80d8
Enable tail calls by default for (micro)MIPS(64).
microMIPS is slightly more tricky than doing it for MIPS(R6) or microMIPSR6.
microMIPS has two instruction encodings: 16bit and 32bit along with some
restrictions on the size of the instruction that can fill the delay slot.
For safe tail calls for microMIPS, the delay slot filler attempts to find
a correct size instruction for the delay slot of TAILCALL pseudos.
Reviewers: dsanders, vkalintris
Subscribers: jfb, dsanders, sdardis, llvm-commits
Differential Revision: https://reviews.llvm.org/D21138
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277708 91177308-0d34-0410-b5e6-96231b3b80d8
This should ensure that we can atomically write two bytes (on top of the
retq and the one past it) and have those two bytes not straddle cache
lines.
We also move the label past the alignment instruction so that we can refer
to the actual first instruction, as opposed to potential padding before the
aligned instruction.
Update the tests to allow us to reflect the new order of assembly.
Reviewers: rSerge, echristo, majnemer
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D23101
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277701 91177308-0d34-0410-b5e6-96231b3b80d8
This just tests that the register limit isn't exceeded,
so the regisetr allocation doesn't need to be great.'
The critically slow part is all in greedy RA, so
switch to basic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277700 91177308-0d34-0410-b5e6-96231b3b80d8
This reinstates r277611 + r277614 and reverts r277642. A cast_or_null
should have been a dyn_cast_or_null.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277691 91177308-0d34-0410-b5e6-96231b3b80d8
overloaded (and simpler).
Sean rightly pointed out in code review that we've started using
"wrapper pass" as a specific part of the old pass manager, and in fact
it is more applicable there. Here, we really have a pass *template* to
build a repeated pass, so call it that.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277689 91177308-0d34-0410-b5e6-96231b3b80d8
pdbdump calls DbiStreamBuilder::commit through PDBFileBuilder::commit
without calling DbiStreamBuilder::finalize. Because `finalize` initializes
`Header` member, `Header` remained nullptr which caused a crash bug.
Differential Revision: https://reviews.llvm.org/D23143
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277681 91177308-0d34-0410-b5e6-96231b3b80d8
rewriteOperands() always performed liveness queries at the base index
rather than the RegSlot/Base as apropriate for the machine operand. This
could lead to illegal rewriting in some cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277661 91177308-0d34-0410-b5e6-96231b3b80d8
changing them to Expected<> to allow them to pass through llvm Errors.
No functional change.
This commit by itself will break the next lld builds. I’ll be committing the
matching change for lld immediately next.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277656 91177308-0d34-0410-b5e6-96231b3b80d8
This patch fixes pr25548.
Current implementation of PPCBoolRetToInt doesn't handle CallInst correctly, so it failed to do the intended optimization when there is a CallInst with parameters. This patch fixed that.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277655 91177308-0d34-0410-b5e6-96231b3b80d8
Not a correctness issue, but it would be nice if we didn't have to
recompute our block numbering (worst-case) every time we move MSSA.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277652 91177308-0d34-0410-b5e6-96231b3b80d8
Some of these tests need to be cleaned up further to make it obvious
what they're testing, but as a first step remove all instances of
"grep".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277648 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r277611 and the followup r277614.
Bootstrap builds and chromium builds are crashing during inlining after
this change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277642 91177308-0d34-0410-b5e6-96231b3b80d8
Didn't want to fold this in with r277640, since it touches bits that
aren't entirely related to r277640.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277641 91177308-0d34-0410-b5e6-96231b3b80d8
This is a follow-up to r277637. It teaches MemorySSA that invariant
loads (and loads of provably constant memory) are always liveOnEntry.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277640 91177308-0d34-0410-b5e6-96231b3b80d8
This patch makes MemorySSA recognize atomic/volatile loads, and makes
MSSA treat said loads specially. This allows us to be a bit more
aggressive in some cases.
Administrative note: Revision was LGTM'ed by reames in person.
Additionally, this doesn't include the `invariant.load` recognition in
the differential revision, because I feel it's better to commit that
separately. Will commit soon.
Differential Revision: https://reviews.llvm.org/D16875
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277637 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
InstCombine unfolds expressions of the form `zext(or(icmp, icmp))` to `or(zext(icmp), zext(icmp))` such that in a later iteration of InstCombine the exposed `zext(icmp)` instructions can be optimized. We now combine this unfolding and the subsequent `zext(icmp)` optimization to be performed together. Since the unfolding doesn't happen separately anymore, we also again enable the folding of `logic(cast(icmp), cast(icmp))` expressions to `cast(logic(icmp, icmp))` which had been disabled due to its interference with the unfolding transformation.
Tested via `make check` and `lnt`.
Background
==========
For a better understanding on how it came to this change we subsequently summarize its history. In commit r275989 we've already tried to enable the folding of `logic(cast(icmp), cast(icmp))` to `cast(logic(icmp, icmp))` which had to be reverted in r276106 because it could lead to an endless loop in InstCombine (also see http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20160718/374347.html). The root of this problem is that in `visitZExt()` in InstCombineCasts.cpp there also exists a reverse of the above folding transformation, that unfolds `zext(or(icmp, icmp))` to `or(zext(icmp), zext(icmp))` in order to expose `zext(icmp)` operations which would then possibly be eliminated by subsequent iterations of InstCombine. However, before these `zext(icmp)` would be eliminated the folding from r275989 could kick in and cause InstCombine to endlessly switch back and forth between the folding and the unfolding transformation. This is the reason why we now combine the `zext`-unfolding and the elimination of the exposed `zext(icmp)` to happen at one go because this enables us to still allow the cast-folding in `logic(cast(icmp), cast(icmp))` without entering an endless loop again.
Details on the submitted changes
================================
- In `visitZExt()` we combine the unfolding and optimization of `zext` instructions.
- In `transformZExtICmp()` we have to use `Builder->CreateIntCast()` instead of `CastInst::CreateIntegerCast()` to make sure that the new `CastInst` is inserted in a `BasicBlock`. The new calls to `transformZExtICmp()` that we introduce in `visitZExt()` would otherwise cause according assertions to be triggered (in our case this happend, for example, with lnt for the MultiSource/Applications/sqlite3 and SingleSource/Regression/C++/EH/recursive-throw tests). The subsequent usage of `replaceInstUsesWith()` is necessary to ensure that the new `CastInst` replaces the `ZExtInst` accordingly.
- In InstCombineAndOrXor.cpp we again allow the folding of casts on `icmp` instructions.
- The instruction order in the optimized IR for the zext-or-icmp.ll test case is different with the introduced changes.
- The test cases in zext.ll have been adopted from the reverted commits r275989 and r276105.
Reviewers: grosser, majnemer, spatel
Subscribers: eli.friedman, majnemer, llvm-commits
Differential Revision: https://reviews.llvm.org/D22864
Contributed-by: Matthias Reisinger <d412vv1n@gmail.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277635 91177308-0d34-0410-b5e6-96231b3b80d8
We currently only support combining target shuffles that consist of a single source input (plus elements known to be undef/zero).
This patch generalizes the recursive combining of the target shuffle to collect all the inputs, merging any duplicates along the way, into a full set of src ops and its shuffle mask.
We uncover a number of cases where we have failed to combine a unary shuffle because the input has been duplicated and separated during lowering.
This will allow us to combine to 2-input shuffles in a future patch.
Differential Revision: https://reviews.llvm.org/D22859
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277631 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit the revert commit r277627. The build errors
mentioned in r277627 were likely caused by an unclean build directory.
Sorry for the noise.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277630 91177308-0d34-0410-b5e6-96231b3b80d8
This removes the restriction for the icmp constant, but as noted by the FIXME comments,
we still need to change individual checks for binop operand constants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277629 91177308-0d34-0410-b5e6-96231b3b80d8
This reverts commit r277540. It breaks the build with:
../lib/Object/Archive.cpp:264:41: error: return type of out-of-line definition of 'llvm::object::ArchiveMemberHeader::getUID' differs from that in the declaration
Expected<unsigned> ArchiveMemberHeader::getUID() const {
~~~~~~~~~~~~~~~~~~ ^
include/llvm/Object/Archive.h:53:12: note: previous declaration is here
unsigned getUID() const;
~~~~~~~~ ^
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277627 91177308-0d34-0410-b5e6-96231b3b80d8
This is a fix for PR28697.
An MDNode can indirectly refer to a GlobalValue, through a
ConstantAsMetadata. When the GlobalValue is deleted, the MDNode operand
is reset to `nullptr`. If the node is uniqued, this can lead to a
hard-to-detect cache invalidation in a Metadata map that's shared across
an LLVMContext.
Consider:
1. A map from Metadata* to `T` called RemappedMDs.
2. A node that references a global variable, `!{i1* @GV}`.
3. Insert `!{i1* @GV} -> SomeT` in the map.
4. Delete `@GV`, leaving behind `!{null} -> SomeT`.
Looking up the generic and uninteresting `!{null}` gives you `SomeT`,
which is likely related to `@GV`. Worse, `SomeT`'s lifetime may be tied
to the deleted `@GV`.
This occurs in practice in the shared ValueMap used since r266579 in the
IRMover. Other code that handles more than one Module (with different
lifetimes) in the same LLVMContext could hit it too.
The fix here is a partial revert of r225223: in the rare case that an
MDNode operand is a ConstantAsMetadata (i.e., wrapping a node from the
Value hierarchy), drop uniquing if it gets replaced with `nullptr`.
This changes step #4 above to leave behind `distinct !{null} -> SomeT`,
which can't be confused with the generic `!{null}`.
In theory, this can cause some churn in the LLVMContext's MDNode
uniquing map when Values are being deleted. However:
- The number of GlobalValues referenced from uniqued MDNodes is
expected to be quite small. E.g., the debug info metadata schema
only references GlobalValues from distinct nodes.
- Other Constants have the lifetime of the LLVMContext, whose teardown
is careful to drop references before deleting the constants.
As a result, I don't expect a compile time regression from this change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277625 91177308-0d34-0410-b5e6-96231b3b80d8
Currently we have a number of tests that fail with -verify-machineinstrs.
To detect this cases earlier we add the option to the testcases with the
exception of tests that will currently fail with this option. PR 27456 keeps
track of this failures.
No code review, as discussed with Hal Finkel.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@277624 91177308-0d34-0410-b5e6-96231b3b80d8
