The code in SelectionDAG::getMemset for some reason assumes the value passed to
memset is an i32. This breaks the generated code for targets that only have
registers smaller than 32 bits because the value might get split into multiple
registers by the calling convention. See the test for the MSP430 target included
in the patch for an example.
This patch ensures that nothing is assumed about the type of the value. Instead,
the type is taken from the selected overload of the llvm.memset intrinsic.
llvm-svn: 216716
was marked custom. The target independent DAG combine has no way to know if
the shuffles it is introducing are ones that the target could support or not.
llvm-svn: 216678
The included test case would fail, because the MI PHI node would have two
operands from the same predecessor.
This problem occurs when a switch instruction couldn't be selected. This happens
always, because there is no default switch support for FastISel to begin with.
The problem was that FastISel would first add the operand to the PHI nodes and
then fall-back to SelectionDAG, which would then in turn add the same operands
to the PHI nodes again.
This fix removes these duplicate PHI node operands by reseting the
PHINodesToUpdate to its original state before FastISel tried to select the
instruction.
This fixes <rdar://problem/18155224>.
llvm-svn: 216640
Currently instructions are folded very aggressively for AArch64 into the memory
operation, which can lead to the use of killed operands:
%vreg1<def> = ADDXri %vreg0<kill>, 2
%vreg2<def> = LDRBBui %vreg0, 2
... = ... %vreg1 ...
This usually happens when the result is also used by another non-memory
instruction in the same basic block, or any instruction in another basic block.
This fix teaches hasTrivialKill to not only check the LLVM IR that the value has
a single use, but also to check if the register that represents that value has
already been used. This can happen when the instruction with the use was folded
into another instruction (in this particular case a load instruction).
This fixes rdar://problem/18142857.
llvm-svn: 216634
FastEmitInst_ri was constraining the first operand without checking if it is
a virtual register. Use constrainOperandRegClass as all the other
FastEmitInst_* functions.
llvm-svn: 216613
This teaches the AArch64 backend to deal with the operations required
to deal with the operations on v4f16 and v8f16 which are exposed by
NEON intrinsics, plus the add, sub, mul and div operations.
llvm-svn: 216555
This combine is essentially combining target-specific nodes back into target
independent nodes that it "knows" will be combined yet again by a target
independent DAG combine into a different set of target-independent nodes that
are legal (not custom though!) and thus "ok". This seems... deeply flawed. The
crux of the problem is that we don't combine un-legalized shuffles that are
introduced by legalizing other operations, and thus we don't see a very
profitable combine opportunity. So the backend just forces the input to that
combine to re-appear.
However, for this to work, the conditions detected to re-form the unlegalized
nodes must be *exactly* right. Previously, failing this would have caused poor
code (if you're lucky) or a crasher when we failed to select instructions.
After r215611 we would fall back into the legalizer. In some cases, this just
"fixed" the crasher by produces bad code. But in the test case added it caused
the legalizer and the dag combiner to iterate forever.
The fix is to make the alignment checking in the x86 side of things match the
alignment checking in the generic DAG combine exactly. This isn't really a
satisfying or principled fix, but it at least make the code work as intended.
It also highlights that it would be nice to detect the availability of under
aligned loads for a given type rather than bailing on this optimization. I've
left a FIXME to document this.
Original commit message for r215611 which covers the rest of the chang:
[SDAG] Fix a case where we would iteratively legalize a node during
combining by replacing it with something else but not re-process the
node afterward to remove it.
In a truly remarkable stroke of bad luck, this would (in the test case
attached) end up getting some other node combined into it without ever
getting re-processed. By adding it back on to the worklist, in addition
to deleting the dead nodes more quickly we also ensure that if it
*stops* being dead for any reason it makes it back through the
legalizer. Without this, the test case will end up failing during
instruction selection due to an and node with a type we don't have an
instruction pattern for.
It took many million runs of the shuffle fuzz tester to find this.
llvm-svn: 216537
There's no need to do this if the user doesn't call va_start. In the
future, we're going to have thunks that forward these register
parameters with musttail calls, and they won't need these spills for
handling va_start.
Most of the test suite changes are adding va_start calls to existing
tests to keep things working.
llvm-svn: 216294
isPow2DivCheap
That name doesn't specify signed or unsigned.
Lazy as I am, I eventually read the function and variable comments. It turns out that this is strictly about signed div. But I discovered that the comments are wrong:
srl/add/sra
is not the general sequence for signed integer division by power-of-2. We need one more 'sra':
sra/srl/add/sra
That's the sequence produced in DAGCombiner. The first 'sra' may be removed when dividing by exactly '2', but that's a special case.
This patch corrects the comments, changes the name of the flag bit, and changes the name of the accessor methods.
No functional change intended.
Differential Revision: http://reviews.llvm.org/D5010
llvm-svn: 216237
The FPv4-SP floating-point unit is generally referred to as
single-precision only, but it does have double-precision registers and
load, store and GPR<->DPR move instructions which operate on them.
This patch enables the use of these registers, the main advantage of
which is that we now comply with the AAPCS-VFP calling convention.
This partially reverts r209650, which added some AAPCS-VFP support,
but did not handle return values or alignment of double arguments in
registers.
This patch also adds tests for Thumb2 code generation for
floating-point instructions and intrinsics, which previously only
existed for ARM.
llvm-svn: 216172
Store TargetSelectionDAGInfo as a pointer instead of a reference:
getSelectionDAGInfo() may not be implemented for certain backends
(e.g. it's not currently implemented for R600).
This bug is reported by UBSan.
llvm-svn: 216129
legalization stage. With those two optimizations, fewer signed/zero extension
instructions can be inserted, and then we can expose more opportunities to
Machine CSE pass in back-end.
llvm-svn: 216066
Note: This was originally reverted to track down a buildbot error. This commit
exposed a latent bug that was fixed in r215753. Therefore it is reapplied
without any modifications.
I run it through SPEC2k and SPEC2k6 for AArch64 and it didn't introduce any new
regeressions.
Original commit message:
This changes the order in which FastISel tries to materialize a constant.
Originally it would try to use a simple target-independent approach, which
can lead to the generation of inefficient code.
On X86 this would result in the use of movabsq to materialize any 64bit
integer constant - even for simple and small values such as 0 and 1. Also
some very funny floating-point materialization could be observed too.
On AArch64 it would materialize the constant 0 in a register even the
architecture has an actual "zero" register.
On ARM it would generate unnecessary mov instructions or not use mvn.
This change simply changes the order and always asks the target first if it
likes to materialize the constant. This doesn't fix all the issues
mentioned above, but it enables the targets to implement such
optimizations.
Related to <rdar://problem/17420988>.
llvm-svn: 216006
When combining a pair of shuffle nodes, check if the combined shuffle mask is
trivially Undef. In case, immediately fold that pair of shuffles to Undef.
The lack of checks for undef masks was the root-cause of a poor-codegen bug
in the dag combiner.
Example:
%1 = shufflevector <4 x i32> %A, <4 x i32> %B, <4 x i32> <i32 4, i32 1, i32 1, i32 6>
%2 = shufflevector <4 x i32> %1, <4 x i32> undef, <4 x i32> <i32 0, i32 4, i32 1, i32 6>
%3 = shufflevector <4 x i32> %2, <4 x i32> undef, <4 x i32> <i32 1, i32 5, i32 3, i32 3>
Before this patch, on x86 (with -mcpu=corei7) we failed to fold the entire
sequence to Undef value and therefore we generated:
shufps $-123, %xmm1, $xmm0
pshufd $-46, %xmm0, %xmm0
With this patch, the entire shuffle sequence is folded to Undef and no
shuffles are generated in the output assembly.
Added new test cases to test 'combine-vec-shuffle-5.ll'.
llvm-svn: 215797
As Jim pointed out this assert isn't really needed to test for correctness,
because the code right afterwards does the same check and falls-back to
SelectionDAG - as intended.
llvm-svn: 215735
This reverts:
r215595 "[FastISel][X86] Add large code model support for materializing floating-point constants."
r215594 "[FastISel][X86] Use XOR to materialize the "0" value."
r215593 "[FastISel][X86] Emit more efficient instructions for integer constant materialization."
r215591 "[FastISel][AArch64] Make use of the zero register when possible."
r215588 "[FastISel] Let the target decide first if it wants to materialize a constant."
r215582 "[FastISel][AArch64] Cleanup constant materialization code. NFCI."
llvm-svn: 215673
This patch allows a vector fneg of a bitcasted integer value to be optimized in the same way that we already optimize a scalar fneg. If the integer variable is a constant, we can precompute the result and not require any logic ops.
This patch is very similar to a fabs patch committed at r214892.
Differential Revision: http://reviews.llvm.org/D4852
llvm-svn: 215646
input node after manually adding it to the worklist and using CombineTo.
Once we use CombineTo the input node may have been deleted. Despite this
being *completely confusing* and somewhat broken, the only way to
"correctly" return from a DAG combine after potentially deleting the
input node is to return *that exact node*....
But really, this code should just never have used CombineTo. It won't do
what it wants (returning the node as mentioned above just causes the
combine to infloop). The correct way to combine away a casted load to
a load of the correct type is to RAUW the chain directly and then return
the loaded value to replace the actual value node.
I managed to find this with the vector shuffle fuzzer even though it
clearly has nothing at all to do with vector shuffles and rather those
happen to trigger a load of a constant pool that hits this combine *just
right*. I've included the test as it is small and a nice stress test
that the infrastructure isn't asserting.
llvm-svn: 215622
combining by replacing it with something else but not re-process the
node afterward to remove it.
In a truly remarkable stroke of bad luck, this would (in the test case
attached) end up getting some other node combined into it without ever
getting re-processed. By adding it back on to the worklist, in addition
to deleting the dead nodes more quickly we also ensure that if it
*stops* being dead for any reason it makes it back through the
legalizer. Without this, the test case will end up failing during
instruction selection due to an and node with a type we don't have an
instruction pattern for.
It took many million runs of the shuffle fuzz tester to find this.
llvm-svn: 215611
This changes the order in which FastISel tries to materialize a constant.
Originally it would try to use a simple target-independent approach, which
can lead to the generation of inefficient code.
On X86 this would result in the use of movabsq to materialize any 64bit
integer constant - even for simple and small values such as 0 and 1. Also
some very funny floating-point materialization could be observed too.
On AArch64 it would materialize the constant 0 in a register even the
architecture has an actual "zero" register.
On ARM it would generate unnecessary mov instructions or not use mvn.
This change simply changes the order and always asks the target first if it
likes to materialize the constant. This doesn't fix all the issues
mentioned above, but it enables the targets to implement such
optimizations.
Related to <rdar://problem/17420988>.
llvm-svn: 215588
Add header guards to files that were missing guards. Remove #endif comments
as they don't seem common in LLVM (we can easily add them back if we decide
they're useful)
Changes made by clang-tidy with minor tweaks.
llvm-svn: 215558
This patch improves the existing algorithm in DAGCombiner that
attempts to fold shuffles according to rule:
shuffle(shuffle(x, y, M1), undef, M2) -> shuffle(y, undef, M3)
Before this change, there were cases where the DAGCombiner conservatively
avoided folding shuffles even if the resulting mask would have been legal.
That is because the algorithm wrongly assumed that commuting
an illegal shuffle mask would always produce an illegal mask.
With this change, we now correctly compute the commuted shuffle mask before
calling method 'isShuffleMaskLegal' on it.
On X86, this improves for example the codegen for the following function:
define <4 x i32> @test(<4 x i32> %A, <4 x i32> %B) {
%1 = shufflevector <4 x i32> %B, <4 x i32> %A, <4 x i32> <i32 1, i32 2, i32 6, i32 7>
%2 = shufflevector <4 x i32> %1, <4 x i32> undef, <4 x i32> <i32 2, i32 3, i32 2, i32 3>
ret <4 x i32> %2
}
Before this change the X86 backend (-mcpu=corei7) generated
the following assembly code for function @test:
shufps $-23, %xmm0, %xmm1 # xmm1 = xmm1[1,2],xmm0[2,3]
movhlps %xmm1, %xmm1 # xmm1 = xmm1[1,1]
movaps %xmm1, %xmm0
Now we produce:
movhlps %xmm0, %xmm0 # xmm0 = xmm0[1,1]
Added extra test cases in combine-vec-shuffle-2.ll to verify that we correctly
fold according to the above-mentioned rule.
llvm-svn: 215555
This implements PPCTargetLowering::getTgtMemIntrinsic for Altivec load/store
intrinsics. As with the construction of the MachineMemOperands for the
intrinsic calls used for unaligned load/store lowering, the only slight
complication is that we need to represent a larger memory range than the
loaded/stored value-type size (because the address is rounded down to an
aligned address, and we need to conservatively represent the entire possible
range of the actual access). This required adding an extra size field to
TargetLowering::IntrinsicInfo, and this was done in a way that required no
modifications to other targets (the size defaults to the store size of the
provided memory data type).
This fixes test/CodeGen/PowerPC/unal-altivec-wint.ll (so it can be un-XFAILed).
llvm-svn: 215512
Follow up to r214266. Add missing case in ScalarizeVectorResult() for
cttz_zero_undef.
Differential Revision: http://reviews.llvm.org/D4813
llvm-svn: 215330
floating point exceptions, added use of flag to fold potentially exception
raising floating point math in selection DAG. No functionality change, as
targets have to explicitly ask for this behavior and none does today.
llvm-svn: 215222
__stack_chk_guard.
Handle the case where the pointer operand of the load instruction that loads the
stack guard is not a global variable but instead a bitcast.
%StackGuard = load i8** bitcast (i64** @__stack_chk_guard to i8**)
call void @llvm.stackprotector(i8* %StackGuard, i8** %StackGuardSlot)
Original test case provided by Ana Pazos.
This fixes PR20558.
llvm-svn: 215167
be deleted. This will be reapplied as soon as possible and before
the 3.6 branch date at any rate.
Approved by Jim Grosbach, Lang Hames, Rafael Espindola.
This reverts commits r215111, 215115, 215116, 215117, 215136.
llvm-svn: 215154
I am sure we will be finding bits and pieces of dead code for years to
come, but this is a good start.
Thanks to Lang Hames for making MCJIT a good replacement!
llvm-svn: 215111
Allow vector fabs operations on bitcasted constant integer values to be optimized
in the same way that we already optimize scalar fabs.
So for code like this:
%bitcast = bitcast i64 18446744069414584320 to <2 x float> ; 0xFFFF_FFFF_0000_0000
%fabs = call <2 x float> @llvm.fabs.v2f32(<2 x float> %bitcast)
%ret = bitcast <2 x float> %fabs to i64
Instead of generating something like this:
movabsq (constant pool loadi of mask for sign bits)
vmovq (move from integer register to vector/fp register)
vandps (mask off sign bits)
vmovq (move vector/fp register back to integer return register)
We should generate:
mov (put constant value in return register)
I have also removed a redundant clause in the first 'if' statement:
N0.getOperand(0).getValueType().isInteger()
is the same thing as:
IntVT.isInteger()
Testcases for x86 and ARM added to existing files that deal with vector fabs.
One existing testcase for x86 removed because it is no longer ideal.
For more background, please see:
http://reviews.llvm.org/D4770
And:
http://llvm.org/bugs/show_bug.cgi?id=20354
Differential Revision: http://reviews.llvm.org/D4785
llvm-svn: 214892
shorter/easier and have the DAG use that to do the same lookup. This
can be used in the future for TargetMachine based caching lookups from
the MachineFunction easily.
Update the MIPS subtarget switching machinery to update this pointer
at the same time it runs.
llvm-svn: 214838
This code is completely wrong. It is also dead, as if it were to *ever*
run, it would crash. Fortunately, after my work to the combiner, it is
at least *possible* to reach the code, and llvm-stress has found a test
case. Thanks to Patrick for reporting.
It would be really good if anyone who remembers how this code works and
what it was intended to do could add some more obvious test coverage
instead of my completely contrived and reduced test case. My test case
was so brittle I left a bread crumb comment in it to help the next
person to stumble on it and not know what it was actually testing for.
llvm-svn: 214785
combines) until they are legal.
Doing it the old way could, when the stars align *just* right, cause
a node to get into the combine set prior to being legalized. Then, when
the same node showed up as an operand to another node later on (but not
so much later on that it had been deleted as dead) we would fail to add
it back to the worklist thinking it had already been combined. This
would in turn cause it to not be legalized. Fortunately, we can also
walk the operands looking for uncombined (and thus potentially
un-legalized) nodes late. It will still ensure that we walk all operands
of all nodes and send all of them through both the legalizer without
changes and the combiner at least once. (Which was the original goal of
this).
I have a test case for this bug, but it is terribly brittle. For
example, it will stop finding the bug the moment I enable the new
shuffle lowering. I don't yet have any test case that reliably exercises
this bug, and it isn't clear that it will be possible to craft one. It
is entirely possible that with the new shuffle lowering the two forms of
doing this are precisely equivalent. That doesn't mean we shouldn't take
the more conservative approach of insisting on things in the combined
set having survived the legalizer.
llvm-svn: 214673
This is intended to be the minimal change needed to fix PR20354 ( http://llvm.org/bugs/show_bug.cgi?id=20354 ). The check for a vector operation was wrong; we need to check that the fabs itself is not a vector operation.
This patch will not generate the optimal code. A constant pool load and 'and' op will be generated instead of just returning a value that we can calculate in advance (as we do for the scalar case). I've put a 'TODO' comment for that here and expect to have that patch ready soon.
There is a very similar optimization that we can do in visitFNEG, so I've put another 'TODO' there and expect to have another patch for that too.
llvm-svn: 214670
so using a single helper which adds operands back onto the worklist.
Several places didn't rigorously do this but a couple already did.
Factoring them together and doing it rigorously is important to delete
things recursively early on in the combiner and get a chance to see
accurate hasOneUse values. While no existing test cases change, an
upcoming patch to add DAG combining logic for PSHUFB requires this to
work correctly.
llvm-svn: 214623
during DAGCombine in certain circumstances. Unfortunately, the circumstances required
to trigger the issue seem to require a pretty specific interaction of DAGCombines,
and I haven't been able to find a testcase that reproduces on X86, ARM, or AArch64.
The functionality added here is replicated in essentially every other DAG combine,
so it seems pretty obviously correct.
llvm-svn: 214622
fromulation of the node, which isn't really the desired behavior from
within the combiner or legalizer, but is necessary within ISel. I've
added a hopefully helpful comment and fixed the only two places where
this took place.
Yet another step toward the combiner and legalizer not needing to use
update listeners with virtual calls to manage the worklists behind
legalization and combining.
llvm-svn: 214574
This lifts the (very few) places the legalizer would delete dead nodes
into the outer loop around the legalizer. This is significantly simpler
because it doesn't require the legalizer itself to manage the iterator
validity, and it doesn't require the legalizer to be a DAG update
listener in order to remove things from the legalized set. It also makes
the interface much less contrived for the case of the legalizer running
inside the last phase of DAG combining.
I'm working on centralizing the deletion of nodes during both legalizing
and combining as much as possible. My hope is to remove the need for DAG
update listeners from the combiner next, which would remove a costly
virtual dispatch chain on every deletion. This in turn should allow us
to more aggressively delete DAG nodes during combining which will in
turn allow us to combine more aggressively by exposing the actual nodes
which have single users to the combine phases.
llvm-svn: 214546
Altivec vector loads on PowerPC have an interesting property: They always load
from an aligned address (by rounding down the address actually provided if
necessary). In order to generate an actual unaligned load, you can generate two
load instructions, one with the original address, one offset by one vector
length, and use a special permutation to extract the bytes desired.
When this was originally implemented, I generated these two loads using regular
ISD::LOAD nodes, now marked as aligned. Unfortunately, there is a problem with
this:
The alignment of a load does not contribute to its identity, and SDNodes
are uniqued. So, imagine that we have some unaligned load, L1, that is not
aligned. The routine will create two loads, L1(aligned) and (L1+16)(aligned).
Further imagine that there had already existed a load (L1+16)(unaligned) with
the same chain operand as the load L1. When (L1+16)(aligned) is created as part
of the lowering of L1, this load *is* also the (L1+16)(unaligned) node, just
now marked as aligned (because the new alignment overwrites the old). But the
original users of (L1+16)(unaligned) now get the data intended for the
permutation yielding the data for L1, and (L1+16)(unaligned) no longer exists
to get its own permutation-based expansion. This was PR19991.
A second potential problem has to do with the MMOs on these loads, which can be
used by AA during instruction scheduling to break chain-based dependencies. If
the new "aligned" loads get the MMO from the original unaligned load, this does
not represent the fact that it will load data from below the original address.
Normally, this would not matter, but this load might be combined with another
load pair for a previous vector, and then the dependency on the otherwise-
ignored lower bytes can matter.
To fix both problems, instead of generating the necessary loads using regular
ISD::LOAD instructions, ppc_altivec_lvx intrinsics are used instead. These are
provided with MMOs with a conservative address range.
Unfortunately, I no longer have a failing test case (since PR19991 was
reported, other changes in CodeGen have forced this bug back into hiding it
again). Nevertheless, this should fix the underlying problem.
llvm-svn: 214481
Currently when DAGCombine converts loads feeding a switch into a switch of
addresses feeding a load the new load inherits the isInvariant flag of the left
side. This is incorrect since invariant loads can be reordered in cases where it
is illegal to reoarder normal loads.
This patch adds an isInvariant parameter to getExtLoad() and updates all call
sites to pass in the data if they have it or false if they don't. It also
changes the DAGCombine to use that data to make the right decision when
creating the new load.
llvm-svn: 214449
This fixes a mistake where I accidentially dropped the upper 32bit of a
64bit pointer during FastISel lowering of the patchpoint intrinsic.
llvm-svn: 214367
DAGCombine may choose to rewrite graphs where two loads feed a select into
graphs where a select of two addresses feed a load. While it sanity checks the
loads to make sure they are broadly equivalent it currently just uses the
alignment restriction of the left node. In cases where the right node has
stronger alignment requiresment this may lead to bad codegen, such as generating
an aligned load where an unaligned load is required. This patch makes the
combine generate a load with an alignment that is the same as whichever is more
restrictive of the two alignments.
Tests included.
rdar://17762530
llvm-svn: 214322
Fix the missing case in ScalarizeVectorResult() that was exposed with
libclcore.bc in Android.
Differential Revision: http://reviews.llvm.org/D4645
llvm-svn: 214266
We need to make sure we use the softened version of all appropriate operands in
the libcall, or things go horribly wrong. This may entail actually executing a
1-stage softening.
llvm-svn: 214175
inspection in the proccess, and shuffle the logging in the DAG combiner
around a bit.
With this it is much easier to follow what the legalizer is doing. It
should even accurately present most of the strange legalization
operations where a single node is replaced by multiple nodes, etc. There
is still some information lost (we log SDNodes not SDValues so we don't
log which result is used for which thing), but I think this is much
closer to a usable system. Notably, this will make it *much* more
apparant when legalization is actually happening inside the combiner, or
when there is a cycle caused by interactions of the legalizer and the
combiner.
The "bug" I fixed here I'm not sure is remotely possible to trigger. We
were only adding one of the nodes in a replacement to the updated set
rather than all of the nodes in the replacement. Realistically, the
worst result of this are nodes not getting back onto the worklist in the
DAG combiner. I doubt it is possible to trigger this today, and
I certainly don't have any ideas about how, but this at least brings the
code into alignment with the principled operation of the routine.
llvm-svn: 214105
Rename to allowsMisalignedMemoryAccess.
On R600, 8 and 16 byte accesses are mostly OK with 4-byte alignment,
and don't need to be split into multiple accesses. Vector loads with
an alignment of the element type are not uncommon in OpenCL code.
llvm-svn: 214055
over each node in the worklist prior to combining.
This allows the combiner to produce new nodes which need to go back
through legalization. This is particularly useful when generating
operands to target specific nodes in a post-legalize DAG combine where
the operands are significantly easier to express as pre-legalized
operations. My immediate use case will be PSHUFB formation where we need
to build a constant shuffle mask with a build_vector node.
This also refactors the relevant functionality in the legalizer to
support this, and updates relevant tests. I've spoken to the R600 folks
and these changes look like improvements to them. The avx512 change
needs to be investigated, I suspect there is a disagreement between the
legalizer and the DAG combiner there, but it seems a minor issue so
leaving it to be re-evaluated after this patch.
Differential Revision: http://reviews.llvm.org/D4564
llvm-svn: 214020
This is the first commit in a series that add an @llvm.assume intrinsic which
can be used to provide the optimizer with a condition it may assume to be true
(when the control flow would hit the intrinsic call). Some basic properties are added here:
- llvm.invariant(true) is dead.
- llvm.invariant(false) is unreachable (this directly corresponds to the
documented behavior of MSVC's __assume(0)), so is llvm.invariant(undef).
The intrinsic is tagged as writing arbitrarily, in order to maintain control
dependencies. BasicAA has been updated, however, to return NoModRef for any
particular location-based query so that we don't unnecessarily block code
motion.
llvm-svn: 213973
address of the stack guard was being spilled to the stack.
Previously the address of the stack guard would get spilled to the stack if it
was impossible to keep it in a register. This patch introduces a new target
independent node and pseudo instruction which gets expanded post-RA to a
sequence of instructions that load the stack guard value. Register allocator
can now just remat the value when it can't keep it in a register.
<rdar://problem/12475629>
llvm-svn: 213967
with a result number outside the range of results for the node.
I don't know how we managed to not really check this very basic
invariant for so long, but the code is *very* broken at this point.
I have over 270 test failures with the assert enabled. I'm committing it
disabled so that others can join in the cleanup effort and reproduce the
issues. I've also included one of the obvious fixes that I already
found. More fixes to come.
llvm-svn: 213926
which have successfully round-tripped through the combine phase, and use
this to ensure all operands to DAG nodes are visited by the combiner,
even if they are only added during the combine phase.
This is critical to have the combiner reach nodes that are *introduced*
during combining. Previously these would sometimes be visited and
sometimes not be visited based on whether they happened to end up on the
worklist or not. Now we always run them through the combiner.
This fixes quite a few bad codegen test cases lurking in the suite while
also being more principled. Among these, the TLS codegeneration is
particularly exciting for programs that have this in the critical path
like TSan-instrumented binaries (although I think they engineer to use
a different TLS that is faster anyways).
I've tried to check for compile-time regressions here by running llc
over a merged (but not LTO-ed) clang bitcode file and observed at most
a 3% slowdown in llc. Given that this is essentially a worst case (none
of opt or clang are running at this phase) I think this is tolerable.
The actual LTO case should be even less costly, and the cost in normal
compilation should be negligible.
With this combining logic, it is possible to re-legalize as we combine
which is necessary to implement PSHUFB formation on x86 as
a post-legalize DAG combine (my ultimate goal).
Differential Revision: http://reviews.llvm.org/D4638
llvm-svn: 213898
vector operation legalization with support for custom target lowering
and fallback to expand when it fails, and use this to implement sext and
anyext load lowering for x86 in a more principled way.
Previously, the x86 backend relied on a target DAG combine to "combine
away" sextload and extload nodes prior to legalization, or would expand
them during legalization with terrible code. This is particularly
problematic because the DAG combine relies on running over non-canonical
DAG nodes at just the right time to match several common and important
patterns. It used a combine rather than lowering because we didn't have
good lowering support, and to expose some tricks being employed to more
combine phases.
With this change it becomes a proper lowering operation, the backend
marks that it can lower these nodes, and I've added support for handling
the canonical forms that don't have direct legal representations such as
sextload of a v4i8 -> v4i64 on AVX1. With this change, our test cases
for this behavior continue to pass even after the DAG combiner beigns
running more systematically over every node.
There is some noise caused by this in the test suite where we actually
use vector extends instead of subregister extraction. This doesn't
really seem like the right thing to do, but is unlikely to be a critical
regression. We do regress in one case where by lowering to the
target-specific patterns early we were able to combine away extraneous
legal math nodes. However, this regression is completely addressed by
switching to a widening based legalization which is what I'm working
toward anyways, so I've just switched the test to that mode.
Differential Revision: http://reviews.llvm.org/D4654
llvm-svn: 213897
In order to enable the preservation of noalias function parameter information
after inlining, and the representation of block-level __restrict__ pointer
information (etc.), additional kinds of aliasing metadata will be introduced.
This metadata needs to be carried around in AliasAnalysis::Location objects
(and MMOs at the SDAG level), and so we need to generalize the current scheme
(which is hard-coded to just one TBAA MDNode*).
This commit introduces only the necessary refactoring to allow for the
introduction of other aliasing metadata types, but does not actually introduce
any (that will come in a follow-up commit). What it does introduce is a new
AAMDNodes structure to hold all of the aliasing metadata nodes associated with
a particular memory-accessing instruction, and uses that structure instead of
the raw MDNode* in AliasAnalysis::Location, etc.
No functionality change intended.
llvm-svn: 213859
Constant fold the lanes of the input constant build_vector individually
so we correctly handle when the vector elements are not all the same
constant value.
PR20394
llvm-svn: 213798
The target-independent DAGcombiner will generate:
asr w1, X, #31 w1 = splat sign bit.
add X, X, w1, lsr #28 X = X + 0 or pow2-1
asr w0, X, asr #4 w0 = X/pow2
However, the add + shifts is expensive, so generate:
add w0, X, 15 w0 = X + pow2-1
cmp X, wzr X - 0
csel X, w0, X, lt X = (X < 0) ? X + pow2-1 : X;
asr w0, X, asr 4 w0 = X/pow2
llvm-svn: 213758
insertions.
The old behavior could cause arbitrarily bad memory usage in the DAG
combiner if there was heavy traffic of adding nodes already on the
worklist to it. This commit switches the DAG combine worklist to work
the same way as the instcombine worklist where we null-out removed
entries and only add new entries to the worklist. My measurements of
codegen time shows slight improvement. The memory utilization is
unsurprisingly dominated by other factors (the IR and DAG itself
I suspect).
This change results in subtle, frustrating churn in the particular order
in which DAG combines are applied which causes a number of minor
regressions where we fail to match a pattern previously matched by
accident. AFAICT, all of these should be using AddToWorklist to directly
or should be written in a less brittle way. None of the changes seem
drastically bad, and a few of the changes seem distinctly better.
A major change required to make this work is to significantly harden the
way in which the DAG combiner handle nodes which become dead
(zero-uses). Previously, we relied on the ability to "priority-bump"
them on the combine worklist to achieve recursive deletion of these
nodes and ensure that the frontier of remaining live nodes all were
added to the worklist. Instead, I've introduced a routine to just
implement that precise logic with no indirection. It is a significantly
simpler operation than that of the combiner worklist proper. I suspect
this will also fix some other problems with the combiner.
I think the x86 changes are really minor and uninteresting, but the
avx512 change at least is hiding a "regression" (despite the test case
being just noise, not testing some performance invariant) that might be
looked into. Not sure if any of the others impact specific "important"
code paths, but they didn't look terribly interesting to me, or the
changes were really minor. The consensus in review is to fix any
regressions that show up after the fact here.
Thanks to the other reviewers for checking the output on other
architectures. There is a specific regression on ARM that Tim already
has a fix prepped to commit.
Differential Revision: http://reviews.llvm.org/D4616
llvm-svn: 213727
DAG into a helper function.
This adds a trip through the (very minimal) verification logic in
a bunch of places that were missing it, but shouldn't have any other
impact outside of refactoring. I'm hoping to use this to do more clever
things when DAG nodes are inserted into the graph.
llvm-svn: 213612
a bug in 2010 when they were added but are adding no value today.
In fact, they are utter lies. NodeAllocator is used to allocate almost
all of these node types. I don't know what we were trying to assert
here, and the docs don't give any answer. Until we once again stumble
upon a bug needing help, let's clear the path for improvements.
llvm-svn: 213610
We should update the usages to all of the results;
otherwise, we might get assertion failure or SEGV during
the type legalization of ATOMIC_CMP_SWAP_WITH_SUCCESS
with two or more illegal types.
For example, in the following sequence, both i8 and i1
might be illegal in some target, e.g. armv5, mipsel, mips64el,
%0 = cmpxchg i8* %ptr, i8 %desire, i8 %new monotonic monotonic
%1 = extractvalue { i8, i1 } %0, 1
Since both i8 and i1 should be legalized, the corresponding
ATOMIC_CMP_SWAP_WITH_SUCCESS dag will be checked/replaced/updated
twice.
If we don't update the usage to *ALL* of the results in the
first round, the DAG for extractvalue might be processed earlier.
The GetPromotedInteger() will result in assertion failure,
because its operand (i.e. the success bit of cmpxchg) is not
promoted beforehand.
llvm-svn: 213569
This makes the first stage DAG for @llvm.convert.to.fp16 an fptrunc,
and correspondingly @llvm.convert.from.fp16 an fpext. The legalisation
path is now uniform, regardless of the input IR:
fptrunc -> FP_TO_FP16 (if f16 illegal) -> libcall
fpext -> FP16_TO_FP (if f16 illegal) -> libcall
Each target should be able to select the version that best matches its
operations and not be required to duplicate patterns for both fptrunc
and FP_TO_FP16 (for example).
As a result we can remove some redundant AArch64 patterns.
llvm-svn: 213507
'Worklist' consistently rather than a deeply confusing mixture of
'WorkList' and 'Worklist'.
Notably, the very 'WorkList' of the DAG combiner was exposed to target
specific DAG combines under an interface 'AddToWorklist' which was
implemented by in turn calling 'AddToWorkList' in the combiner. This has
sent me circling with the wrong case in grep one too many times.
I chose to normalize on 'Worklist' because that one won the grep-vote
for llvm/lib/... by a hundered hits or so, and it is used in places
relatively "canonical" such as InstCombine's Worklist. Let's all jsut
pick this casing, whether "correct", "good", or "bad" and be
consistent...
llvm-svn: 213506
stack, filter all handle nodes from the DAG combiner worklist.
This will also handle cases where other handle nodes might be
(erroneously) added to the worklist and then cause bugs and explosions
when deleted. For example, when running the legalizer within the DAG
combiner, there are times when other handle nodes are used and can end
up here.
llvm-svn: 213505
Canonicalize shuffles according to rules:
* shuffle(A, shuffle(A, B)) -> shuffle(shuffle(A,B), A)
* shuffle(B, shuffle(A, B)) -> shuffle(shuffle(A,B), B)
* shuffle(B, shuffle(A, Undef)) -> shuffle(shuffle(A, Undef), B)
This patch helps identifying more shuffle pairs that could be combined reusing
the already existing rules in the DAGCombiner.
Added new test 'combine-vec-shuffle-5.ll' to verify that the canonicalized
shuffles are now folded into a single shuffle node by the DAGCombiner.
Added more test cases to 'combine-vec-shuffle-4.ll'.
llvm-svn: 213504
This patch removes function 'CommuteVectorShuffle' from X86ISelLowering.cpp
and moves its logic into SelectionDAG.cpp as method 'getCommutedVectorShuffles'.
This refactoring is in preperation of an upcoming change to the DAGCombiner.
llvm-svn: 213503
Summary: This patch introduces two new iterator ranges and updates existing code to use it. No functional change intended.
Test Plan: All tests (make check-all) still pass.
Reviewers: dblaikie
Reviewed By: dblaikie
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D4481
llvm-svn: 213474
Since the result of a SETCC for AArch64 is 0 or -1 in each lane, we can
move unary operations, in this case [su]int_to_fp through the mask
operation and constant fold the operation away. Generally speaking:
UNARYOP(AND(VECTOR_CMP(x,y), constant))
--> AND(VECTOR_CMP(x,y), constant2)
where constant2 is UNARYOP(constant).
This implements the transform where UNARYOP is [su]int_to_fp.
For example, consider the simple function:
define <4 x float> @foo(<4 x float> %val, <4 x float> %test) nounwind {
%cmp = fcmp oeq <4 x float> %val, %test
%ext = zext <4 x i1> %cmp to <4 x i32>
%result = sitofp <4 x i32> %ext to <4 x float>
ret <4 x float> %result
}
Before this change, the code is generated as:
fcmeq.4s v0, v0, v1
movi.4s v1, #0x1 // Integer splat value.
and.16b v0, v0, v1 // Mask lanes based on the comparison.
scvtf.4s v0, v0 // Convert each lane to f32.
ret
After, the code is improved to:
fcmeq.4s v0, v0, v1
fmov.4s v1, #1.00000000 // f32 splat value.
and.16b v0, v0, v1 // Mask lanes based on the comparison.
ret
The svvtf.4s has been constant folded away and the floating point 1.0f
vector lanes are materialized directly via fmov.4s.
Rather than do the folding manually in the target code, teach getNode()
in the generic SelectionDAG to handle folding constant operands of
vector [su]int_to_fp nodes. It is reasonable (as noted in a FIXME) to do
additional constant folding there as well, but I don't have test cases
for those operations, so leaving them for another time when it becomes
appropriate.
rdar://17693791
llvm-svn: 213341
Previously we asserted on this code. Currently compiler-rt doesn't
actually implement any of these new libcalls, but external help is
pretty much the only viable option for LLVM.
I've followed the much more generic "__truncST2" naming, as opposed to
the odd name for f32 -> f16 truncation. This can obviously be changed
later, or overridden by any targets that need to.
llvm-svn: 213252
This makes the two intrinsics @llvm.convert.from.f16 and
@llvm.convert.to.f16 accept types other than simple "float". This is
only strictly needed for the truncate operation, since otherwise
double rounding occurs and there's no way to represent the strict IEEE
conversion. However, for symmetry we allow larger types in the extend
too.
During legalization, we can expand an "fp16_to_double" operation into
two extends for convenience, but abort when the truncate isn't legal. A new
libcall is probably needed here.
Even after this commit, various target tweaks are needed to actually use the
extended intrinsics. I've put these into separate commits for clarity, so there
are no actual tests of f64 conversion here.
llvm-svn: 213248
This fixes an issue where a local value is defined before and used after an
inline asm call with side effects.
This fix simply flushes the local value map, which updates the insertion point
for the inline asm call to be above any previously defined local values.
This fixes <rdar://problem/17694203>
llvm-svn: 213203
It turns out that in most cases (the main exception being i1-related
types) once these operations are formed we cannot separate them and
the targets end up having to deal with them whether they want to or
not.
This is not a good situation, and a more reasonable default can be
formed by ackowledging this and having targets leave them as Legal.
Only x86 seems to be affected (other targets don't even try marking
the operation Expand).
Mostly there's no visible change here yet, but it will be useful to
have truly expanded EXTLOADS for MVT::f16 softening support.
llvm-svn: 213162
There is no need to pass on TLI separately to the function. As Eric pointed out
the Target Machine already provides everything we need.
llvm-svn: 213108
This patch adds two new rules to the DAGCombiner:
1. shuffle (shuffle A, Undef, M0), B, M1 -> shuffle A, B, M2
2. shuffle (shuffle A, Undef, M0), A, M1 -> shuffle A, Undef, M2
We only do this if the combined shuffle is legal for the target.
Example:
;;
define <4 x float> @test(<4 x float> %a, <4 x float> %b) {
%1 = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32><i32 6, i32 0, i32 1, i32 7>
%2 = shufflevector <4 x float> %1, <4 x float> %b, <4 x i32><i32 1, i32 2, i32 4, i32 5>
ret <4 x i32> %2
}
;;
(using llc -mcpu=corei7 -march=x86-64)
Before, the x86 backend generated:
pshufd $120, %xmm0, %xmm0
shufps $-108, %xmm0, %xmm1
movaps %xmm1, %xmm0
Now the x86 backend generates:
movsd %xmm1, %xmm0
llvm-svn: 213069
The patchpoint instruction should have been inserted before the target
generated call instruction to be inside the ADJSTACKDOWN/ADJSTACKUP call
sequence window.
llvm-svn: 213034
Always update the value map with the result register (if there is one), for the
patchpoint instruction we created to replace the target-specific call
instruction.
llvm-svn: 213033
This patch fixes a crasher in method 'DAGCombiner::visitOR' due to an invalid
call to method 'isShuffleMaskLegal'. On x86, method 'isShuffleMaskLegal'
always expects a legal vector value type in input.
With this patch, we immediately check if the input OR dag node has a legal
vector type; we only try to fold a OR dag node into a single shufflevector
if we know that the resulting shuffle will have a legal type.
This is to avoid calling method 'isShuffleMaskLegal' on a potentially
illegal vector value type.
Added a new test-case to file 'CodeGen/X86/combine-or.ll' to verify that
DAGCombiner doesn't crash in the attempt to check/combine an OR between shuffles
with illegal types.
llvm-svn: 213020
This patch teaches the DAGCombiner how to fold a pair of shuffles
according to rules:
1. shuffle(shuffle A, B, M0), B, M1) -> shuffle(A, B, M2)
2. shuffle(shuffle A, B, M0), A, M1) -> shuffle(A, B, M3)
The new rules would only trigger if the resulting shuffle has legal type and
legal mask.
Added test 'combine-vec-shuffle-3.ll' to verify that DAGCombiner correctly
folds shuffles on x86 when the resulting mask is legal. Also added some negative
cases to verify that we avoid introducing illegal shuffles.
llvm-svn: 213001
Verify that DAGCombiner does not crash when trying to fold a pair of shuffles
according to rule (added at r212539):
(shuffle (shuffle A, Undef, M0), Undef, M1) -> (shuffle A, Undef, M2)
The DAGCombiner avoids folding shuffles if the resulting shuffle dag node
is not legal for the target. That means, the resulting shuffle must have
legal type and legal mask.
Before, the DAGCombiner only called method
'TargetLowering::isShuffleMaskLegal' to check if it was "safe" to fold according
to the above-mentioned rule. However, this caused a crash in the x86 backend
since method 'isShuffleMaskLegal' always expects to be called on a
legal vector type.
llvm-svn: 212915
This implements the target-independent lowering for the patchpoint
intrinsic. Targets have to implement the FastLowerCall
hook to support this intrinsic.
Related to <rdar://problem/17427052>
llvm-svn: 212849
The infrastructure mimics the call lowering we have already in place for
SelectionDAG, but with limitations. For example structure return demotion and
non-simple types are not supported (yet).
Currently every backend has its own implementation and duplicated code for call
lowering. There is also no specified interface that could be called from
target-independent code. The target-hook is opt-in and doesn't affect current
implementations.
llvm-svn: 212848
Create a separate helper function for target-independent intrinsic lowering. Also
add an target-hook that allows to directly call into a target-sepcific intrinsic
lowering method. Currently the implementation is opt-in and doesn't affect
existing target implementations.
llvm-svn: 212843
Move the code to a helper function to allow calls from TypeLegalizer.
No functionality change intended
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Tom Stellard <tom@stellard.net>
Reviewed-by: Owen Anderson <resistor@mac.com>
llvm-svn: 212772
This patch teaches the DAGCombiner how to fold shuffles according to the
following new rules:
1. shuffle(shuffle(x, y), undef) -> x
2. shuffle(shuffle(x, y), undef) -> y
3. shuffle(shuffle(x, y), undef) -> shuffle(x, undef)
4. shuffle(shuffle(x, y), undef) -> shuffle(y, undef)
The backend avoids to combine shuffles according to rules 3. and 4. if
the resulting shuffle does not have a legal mask. This is to avoid introducing
illegal shuffles that are potentially expanded into a sub-optimal sequence of
target specific dag nodes during vector legalization.
Added test case combine-vec-shuffle-2.ll to verify that we correctly triggers
the new rules when combining shuffles.
llvm-svn: 212748
to the zero-extend-vector-inreg node introduced previously for the same
purpose: manage the type legalization of widened extend operations,
especially to support the experimental widening mode for x86.
I'm adding both because sign-extend is expanded in terms of any-extend
with shifts to propagate the sign bit. This removes the last
fundamental scalarization from vec_cast2.ll (a test case that hit many
really bad edge cases for widening legalization), although the trunc
tests in that file still appear scalarized because the the shuffle
legalization is scalarizing. Funny thing, I've been working on that.
Some initial experiments with this and SSE2 scenarios is showing
moderately good behavior already for sign extension. Still some work to
do on the shuffle combining on X86 before we're generating optimal
sequences, but avoiding scalarization is a huge step forward.
llvm-svn: 212714
Summary:
On MIPS32r6/MIPS64r6, floating point comparisons return 0 or -1 but integer
comparisons return 0 or 1.
Updated the various uses of getBooleanContents. Two simplifications had to be
disabled when float and int boolean contents differ:
- ScalarizeVecRes_VSELECT except when the kind of boolean contents is trivially
discoverable (i.e. when the condition of the VSELECT is a SETCC node).
- visitVSELECT (select C, 0, 1) -> (xor C, 1).
Come to think of it, this one could test for the common case of 'C'
being a SETCC too.
Preserved existing behaviour for all other targets and updated the affected
MIPS32r6/MIPS64r6 tests. This also fixes the pi benchmark where the 'low'
variable was counting in the wrong direction because it thought it could simply
add the result of the comparison.
Reviewers: hfinkel
Reviewed By: hfinkel
Subscribers: hfinkel, jholewinski, mcrosier, llvm-commits
Differential Revision: http://reviews.llvm.org/D4389
llvm-svn: 212697
not widening the input type to the node sufficiently to let the ext take
place in a register.
This would in turn result in a mysterious bitcast assertion failure
downstream. First change here is to add back the helpful assert I had in
an earlier version of the code to catch this immediately.
Next change is to add support to the type legalization to detect when we
have widened the operand either too little or too much (for whatever
reason) and find a size-matched legal vector type to convert it to
first. This can also fail so we get a new fallback path, but that seems
OK.
With this, we no longer crash on vec_cast2.ll when using widening. I've
also added the CHECK lines for the zero-extend cases here. We still need
to support sign-extend and trunc (or something) to get plausible code
for the other two thirds of this test which is one of the regression
tests that showed the most scalarization when widening was
force-enabled. Slowly closing in on widening being a viable legalization
strategy without it resorting to scalarization at every turn. =]
llvm-svn: 212614
vector types to be legal and a ZERO_EXTEND node is encountered.
When we use widening to legalize vector types, extend nodes are a real
challenge. Either the input or output is likely to be legal, but in many
cases not both. As a consequence, we don't really have any way to
represent this situation and the prior code in the widening legalization
framework would just scalarize the extend operation completely.
This patch introduces a new DAG node to represent doing a zero extend of
a vector "in register". The core of the idea is to allow legal but
different vector types in the input and output. The output vector must
have fewer lanes but wider elements. The operation is defined to zero
extend the low elements of the input to the size of the output elements,
and drop all of the high elements which don't have a corresponding lane
in the output vector.
It also includes generic expansion of this node in terms of blending
a zero vector into the high elements of the vector and bitcasting
across. This in turn yields extremely nice code for x86 SSE2 when we use
the new widening legalization logic in conjunction with the new shuffle
lowering logic.
There is still more to do here. We need to support sign extension, any
extension, and potentially int-to-float conversions. My current plan is
to continue using similar synthetic nodes to model each of these
transitions with generic lowering code for each one.
However, with this patch LLVM already reaches performance parity with
GCC for the core C loops of the x264 code (assuming you disable the
hand-written assembly versions) when compiling for SSE2 and SSE3
architectures and enabling the new widening and lowering logic for
vectors.
Differential Revision: http://reviews.llvm.org/D4405
llvm-svn: 212610
tracks which elements of the build vector are in fact undef.
This should make actually inpsecting them (likely in my next patch)
reasonably pretty. Also makes the output parameter optional as it is
clear now that *most* users are happy with undefs in their splats.
llvm-svn: 212581
This patch teaches how to fold a shuffle according to rule:
shuffle (shuffle (x, undef, M0), undef, M1) -> shuffle(x, undef, M2)
We do this only if the resulting mask M2 is legal; this is to avoid introducing
illegal shuffles that are potentially expanded into a sub-optimal sequence
of target specific dag nodes.
This patch has the advantage of being target independent, since it works on ISD
nodes. Therefore, all targets (not only x86) can take advantage of this rule.
The idea behind this patch is that most shuffle pairs can be safely combined
before we run the legalizer on vector operations. This allows us to
combine/simplify dag nodes earlier in the process and not only immediately
before instruction selection stage.
That said. This patch is not meant to replace any existing target specific
combine rules; backends might still introduce new shuffles during legalization
stage. Also, this rule is very simple and avoids to aggressively optimize
shuffles.
llvm-svn: 212539
aggressively from the x86 shuffle lowering to the generic SDAG vector
shuffle formation code.
This code already tried to fold away shuffles of splats! It just had
lots of bugs and couldn't handle the case my new x86 shuffle lowering
needed.
First, it failed to correctly compute whether N2 was undef because it
pre-computed this, then did transformations which could *make* N2 undef,
then failed to ever re-consider the precomputed state.
Second, it didn't look through bitcasts at all, even in the safe cases
where they are just element-type bitcasts with no change to the number
of elements.
Third, it didn't handle all-zero bit casts nicely the way my code in the
x86 side of things did, which is essential to getting good zext-shuffle
lowerings.
But all of these are generic. I just ported the code down to this layer
and fixed the surrounding bugs. Tests exercising this in the x86 backend
still pass and some silly code in widen_cast-6.ll gets better. I updated
that test to be a bit more precise but it's still pretty unclear what
the value of the test is in this day and age.
llvm-svn: 212517
around the handling of UNDEF lanes in boolean vector content analysis.
The code before my changes here also failed to check for non-constant
splats in a buildvector. I have no idea how to trigger this, I just
spotted by inspection when trying to understand the code. It seems
extremely unlikely to be worth the trouble to teach the only caller of
this code (DAG combining setcc patterns) how to cleverly handle undef
lanes, so I've just commented more thoroughly that we're giving up
there.
llvm-svn: 212515
nodes about whether they are splats. This is factored out and improved
from r212324 which got reverted as it was far too aggressive. The new
API should help more conservatively handle buildvectors that are
a mixture of splatted and undef values.
No functionality change at this point. The hope is to slowly
re-introduce the undef-tolerant optimization of splats, but each time
being forced to make a concious decision about how to handle the undefs
in a way that doesn't lead to contradicting assumptions about the
collapsed value.
Hal has pointed out in discussions that this may not end up being the
desired API and instead it may be more convenient to get a mask of the
undef elements or something similar. I'm starting simple and will expand
the API as I adapt actual callers and see exactly what they need.
llvm-svn: 212514
lanes in vector splats.
The core problem here is that undef lanes can't *unilaterally* be
considered to contribute to splats. Their handling needs to be more
cautious. There is also a reported failure of the nightly testers
(thanks Tobias!) that may well stem from the same core issue. I'm going
to fix this theoretical issue, factor the APIs a bit better, and then
verify that I don't see anything bad with Tobias's reduction from the
test suite before recommitting.
Original commit message for r212324:
[x86] Generalize BuildVectorSDNode::getConstantSplatValue to work for
any constant, constant FP, or undef splat and to tolerate any undef
lanes in a splat, then replace all uses of isSplatVector in X86's
lowering with it.
This fixes issues where undef lanes in an otherwise splat vector would
prevent the splat logic from firing. It is a touch more awkward to use
this interface, but it is much more accurate. Suggestions for better
interface structuring welcome.
With this fix, the code generated with the widening legalization
strategy for widen_cast-4.ll is *dramatically* improved as the special
lowering strategies for a v16i8 SRA kick in even though the high lanes
are undef.
We also get a slightly different choice for broadcasting an aligned
memory location, and use vpshufd instead of vbroadcastss. This looks
like a minor win for pipelining and domain crossing, but a minor loss
for the number of micro-ops. I suspect its a wash, but folks can
easily tweak the lowering if they want.
llvm-svn: 212475
any constant, constant FP, or undef splat and to tolerate any undef
lanes in a splat, then replace all uses of isSplatVector in X86's
lowering with it.
This fixes issues where undef lanes in an otherwise splat vector would
prevent the splat logic from firing. It is a touch more awkward to use
this interface, but it is much more accurate. Suggestions for better
interface structuring welcome.
With this fix, the code generated with the widening legalization
strategy for widen_cast-4.ll is *dramatically* improved as the special
lowering strategies for a v16i8 SRA kick in even though the high lanes
are undef.
We also get a slightly different choice for broadcasting an aligned
memory location, and use vpshufd instead of vbroadcastss. This looks
like a minor win for pipelining and domain crossing, but a minor loss
for the number of micro-ops. I suspect its a wash, but folks can easily
tweak the lowering if they want.
llvm-svn: 212324
subtarget. This involved having the movt predicate take the current
function - since we care about size in instruction selection for
whether or not to use movw/movt take the function so we can check
the attributes. This required adding the current MachineFunction to
FastISel and propagating through.
llvm-svn: 212309
The PowerPC 128-bit long double data type (ppcf128 in LLVM) is in fact a
pair of two doubles, where one is considered the "high" or
more-significant part, and the other is considered the "low" or
less-significant part. When a ppcf128 value is stored in memory or a
register pair, the high part always comes first, i.e. at the lower
memory address or in the lower-numbered register, and the low part
always comes second. This is true both on big-endian and little-endian
PowerPC systems. (Similar to how with a complex number, the real part
always comes first and the imaginary part second, no matter the byte
order of the system.)
This was implemented incorrectly for little-endian systems in LLVM.
This commit fixes three related issues:
- When printing an immediate ppcf128 constant to assembler output
in emitGlobalConstantFP, emit the high part first on both big-
and little-endian systems.
- When lowering a ppcf128 type to a pair of f64 types in SelectionDAG
(which is used e.g. when generating code to load an argument into a
register pair), use correct low/high part ordering on little-endian
systems.
- In a related issue, because lowering ppcf128 into a pair of f64 must
operate differently from lowering an int128 into a pair of i64,
bitcasts between ppcf128 and int128 must not be optimized away by the
DAG combiner on little-endian systems, but must effect a word-swap.
Reviewed by Hal Finkel.
llvm-svn: 212274
This operation was classified as a binary operation in the widening
logic for some reason (clearly, untested). It is in fact a unary
operation. Add a RUN line to a test to exercise this for x86.
Note that again the vector widening strategy doesn't regress anything
and in one case removes a totally unecessary instruction that we
couldn't avoid when promoting the element type.
llvm-svn: 212257
switches) into a switch, and sink them into a dispatch function that can
return the result rather than awkward variable setting with breaks.
llvm-svn: 212166
The argument list vector is never used after it has been passed to the
CallLoweringInfo and moving it to the CallLoweringInfo is cleaner and
pretty much as cheap as keeping a pointer to it.
llvm-svn: 212135
string_ostream is a safe and efficient string builder that combines opaque
stack storage with a built-in ostream interface.
small_string_ostream<bytes> additionally permits an explicit stack storage size
other than the default 128 bytes to be provided. Beyond that, storage is
transferred to the heap.
This convenient class can be used in most places an
std::string+raw_string_ostream pair or SmallString<>+raw_svector_ostream pair
would previously have been used, in order to guarantee consistent access
without byte truncation.
The patch also converts much of LLVM to use the new facility. These changes
include several probable bug fixes for truncated output, a programming error
that's no longer possible with the new interface.
llvm-svn: 211749
We handle this by spilling the whole thing to the stack and doing the
insertion as a store.
PR19492. This happens in real code because the vectorizer creates v2i128 when AVX is enabled.
llvm-svn: 211435
Summary:
With this patch, range metadata can be added to call/invoke including
IntrinsicInst. Previously, it could only be added to load.
Rename computeKnownBitsLoad to computeKnownBitsFromRangeMetadata because
range metadata is not only used by load.
Update the language reference to reflect this change.
Test Plan:
Add several tests in range-2.ll to confirm the verifier is happy with
having range metadata on call/invoke.
Add two tests in AddOverFlow.ll to confirm annotating range metadata to
call/invoke can benefit InstCombine.
Reviewers: meheff, nlewycky, reames, hfinkel, eliben
Reviewed By: eliben
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D4187
llvm-svn: 211281
It looks like there are two versions of LowerCallTo here: the
SelectionDAGBuilder one is designed to operate on LLVM IR, and the
TargetLowering one in the case where everything is at DAG level.
Previously, only the SelectionDAGBuilder variant could handle demoting
an impossible return to sret semantics (before delegating to the
TargetLowering version), but this functionality is also useful for
certain libcalls (e.g. 128-bit operations on 32-bit x86). So this
commit moves the sret handling down a level.
rdar://problem/17242889
llvm-svn: 211155
It's valid to use FP_TO_SINT when asking for a smaller type (e.g. all
"unsigned int16" values fit into a "signed int32"), but the reverse
isn't true.
Unfortunately, I'm not actually aware of any architecture with
asymmetric FP_TO_SINT and FP_TO_UINT handling and the logic happens to
work in the symmetric case, so I can't actually write a test for this.
llvm-svn: 210986
This commit adds a weak variant of the cmpxchg operation, as described
in C++11. A cmpxchg instruction with this modifier is permitted to
fail to store, even if the comparison indicated it should.
As a result, cmpxchg instructions must return a flag indicating
success in addition to their original iN value loaded. Thus, for
uniformity *all* cmpxchg instructions now return "{ iN, i1 }". The
second flag is 1 when the store succeeded.
At the DAG level, a new ATOMIC_CMP_SWAP_WITH_SUCCESS node has been
added as the natural representation for the new cmpxchg instructions.
It is a strong cmpxchg.
By default this gets Expanded to the existing ATOMIC_CMP_SWAP during
Legalization, so existing backends should see no change in behaviour.
If they wish to deal with the enhanced node instead, they can call
setOperationAction on it. Beware: as a node with 2 results, it cannot
be selected from TableGen.
Currently, no use is made of the extra information provided in this
patch. Test updates are almost entirely adapting the input IR to the
new scheme.
Summary for out of tree users:
------------------------------
+ Legacy Bitcode files are upgraded during read.
+ Legacy assembly IR files will be invalid.
+ Front-ends must adapt to different type for "cmpxchg".
+ Backends should be unaffected by default.
llvm-svn: 210903
This commit adds MachineMemOperands to load and store instructions. This allows
the peephole optimizer to fold load instructions. Unfortunatelly the peephole
optimizer currently doesn't run at -O0.
llvm-svn: 210858
The SelectionDAG bad a special case for ISD::SELECT_CC, where it would
allow targets to specify:
setOperationAction(ISD::SELECT_CC, MVT::Other, Expand);
to indicate that they wanted to expand ISD::SELECT_CC for all types.
This wasn't applied correctly everywhere, and it makes writing new
DAG patterns with ISD::SELECT_CC difficult.
llvm-svn: 210541
This patch adds new target specific combine rules to identify horizontal
add/sub idioms from BUILD_VECTOR dag nodes.
This patch also teaches the DAGCombiner how to canonicalize sequences of
insert_vector_elt dag nodes according to the following rule:
(insert_vector_elt (insert_vector_elt A, I0), I1) ->
(insert_vecto_elt (insert_vector_elt A, I1), I0)
This new canonicalization rule only triggers if the inner insert_vector
dag node has exactly one use; also, both indices must be known constants,
and I1 < I0.
This last rule made it possible to write a simpler algorithm to identify
horizontal add/sub patterns because now we don't have to worry about the
ordering of insert_vector_elt dag nodes.
llvm-svn: 210477
This patch modifies SelectionDAGBuilder to construct SDNodes with associated
NoSignedWrap, NoUnsignedWrap and Exact flags coming from IR BinaryOperator
instructions.
Added a new SDNode type called 'BinaryWithFlagsSDNode' to allow accessing
nsw/nuw/exact flags during codegen.
Patch by Marcello Maggioni.
llvm-svn: 210467
DAG cycle detection is only enabled with ENABLE_EXPENSIVE_CHECKS. However we
can run it just before we would crash in order to provide more informative
diagnostics.
Now in addition to the "Overran sorted position" message we also get the Node
printed if a cycle was detected.
Tested by building several configs: Debug+Assert, Debug+Assert+Check (this is
ENABLE_EXPENSIVE_CHECKS), Release+Assert and Release. Also tried that the
AssignTopologicalOrder assert produces the expected results.
llvm-svn: 209977
Pass the DAG down to checkForCycles from all callers where we have it. This
allows target-specific nodes to be printed properly.
Also print some missing newlines.
llvm-svn: 209976
This patch teaches the backend how to simplify/canonicalize dag node
sequences normally introduced by the backend when promoting certain dag nodes
with illegal vector type.
This patch adds two new combine rules:
1) fold (shuffle (bitcast (BINOP A, B)), Undef, <Mask>) ->
(shuffle (BINOP (bitcast A), (bitcast B)), Undef, <Mask>)
2) fold (BINOP (shuffle (A, Undef, <Mask>)), (shuffle (B, Undef, <Mask>))) ->
(shuffle (BINOP A, B), Undef, <Mask>).
Both rules are only triggered on the type-legalized DAG.
In particular, rule 1. is a target specific combine rule that attempts
to sink a bitconvert into the operands of a binary operation.
Rule 2. is a target independet rule that attempts to move a shuffle
immediately after a binary operation.
llvm-svn: 209930
Summary:
If both vector args to vselect are concat_vectors and the condition is
constant and picks half a vector from each argument, convert the vselect
into a concat_vectors.
Added a test.
The ConvertSelectToConcatVector is assuming it doesn't get vselects with
arguments of, for example, <undef, undef, true, true>. Those get taken
care of in the checks above its call.
Reviewers: nadav, delena, grosbach, hfinkel
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3916
llvm-svn: 209929
Unordered is strictly weaker than monotonic, so if the latter doesn't have any
barriers then the former certainly shouldn't.
rdar://problem/16548260
llvm-svn: 209901
An address only use of an extract element of a load can be simplified to a
load. Without this the result of the extract element is spilled to the
stack so that an address is available.
llvm-svn: 209788
No test because no in-tree targets change the bitwidth of the
setcc type depending on the bitwidth of the compared type.
Patch by Ke Bai
llvm-svn: 209771
This matches gcc's behavior. It also seems natural given that aliases
contain other properties that govern how it is accessed (linkage,
visibility, dll storage).
Clang still has to be updated to expose this feature to C.
llvm-svn: 209759
This reverts r208640 (I've just XFAILed the test) because it broke ppc64/Linux
self-hosting. Because nearly every regression test triggers a segfault, I hope
this will be easy to fix.
llvm-svn: 209747
Cortex-M4 only has single-precision floating point support, so any LLVM
"double" type will have been split into 2 i32s by now. Fortunately, the
consecutive-register framework turns out to be precisely what's needed to
reconstruct the double and follow AAPCS-VFP correctly!
rdar://problem/17012966
llvm-svn: 209650
- On ARM/ARM64 we get a vrev because the shuffle matching code is really smart. We still unroll anything that's not v4i32 though.
- On X86 we get a pshufb with SSSE3. Required more cleverness in isShuffleMaskLegal.
- On PPC we get a vperm for v8i16 and v4i32. v2i64 is unrolled.
llvm-svn: 209123
This is mostly a mechanical change changing all the call sites to the newer
chained-function construction pattern. This removes the horrible 15-parameter
constructor for the CallLoweringInfo in favour of setting properties of the call
via chained functions. No functional change beyond the removal of the old
constructors are intended.
llvm-svn: 209082
This is a preliminary step to help ease the construction of CallLoweringInfo.
Changing the construction to a chained function pattern requires that the
parameter be nullable. However, rather than copying the vector, save a pointer
rather than the reference to permit a late binding of the arguments.
llvm-svn: 209080
The problem occurs when a non-i1 setcc is inverted. For example 'i8 = setcc' will get 'xor 0xff' to invert this. This is clearly wrong when the boolean contents are ZeroOrOne.
This patch introduces getLogicalNOT and updates SetCC legalisation to use it.
Reviewed by Hal Finkel.
llvm-svn: 208641
Right now the load may not get DCE'd because of the side-effect of updating
the base pointer.
This can happen if we lower a read-modify-write of an illegal larger type
(e.g. i48) such that the modification only affects one of the subparts (the
lower i32 part but not the higher i16 part). See the testcase.
In order to spot the dead load we need to revisit it when SimplifyDemandedBits
decided that the value of the load is masked off. This is the
CommitTargetLoweringOpt piece.
I checked compile time with ARM64 by sending SPEC bitcode files through llc.
No measurable change.
Fixes <rdar://problem/16031651>
llvm-svn: 208640
We must validate the value type in TLI::getRegisterByName, because if we
don't and the wrong type was used with the IR intrinsic, then we'll assert
(because we won't be able to find a valid register class with which to
construct the requested copy operation). For PPC64, additionally, the type
information is necessary to decide between the 64-bit register and the 32-bit
subregister.
No functionality change.
llvm-svn: 208508
When using the ARM AAPCS, HFAs (Homogeneous Floating-point Aggregates) must
be passed in a block of consecutive floating-point registers, or on the stack.
This means that unused floating-point registers cannot be back-filled with
part of an HFA, however this can currently happen. This patch, along with the
corresponding clang patch (http://reviews.llvm.org/D3083) prevents this.
llvm-svn: 208413
When reducing the bitwidth of a comparison against a constant, the
original setcc's result type was used, which was incorrect.
No test since I don't think any other in tree targets change the
bitwidth of the setcc type depending on the bitwidth of the compared
type.
llvm-svn: 208236
This patch implements the infrastructure to use named register constructs in
programs that need access to specific registers (bare metal, kernels, etc).
So far, only the stack pointer is supported as a technology preview, but as it
is, the intrinsic can already support all non-allocatable registers from any
architecture.
llvm-svn: 208104
For pattern like ((x >> C1) & Mask) << C2, DAG combiner may convert it
into (x >> (C1-C2)) & (Mask << C2), which makes pattern matching of ubfx
more difficult.
For example:
Given
%shr = lshr i64 %x, 4
%and = and i64 %shr, 15
%arrayidx = getelementptr inbounds [8 x [64 x i64]]* @arr, i64 0, %i64 2, i64 %and
%0 = load i64* %arrayidx
With current shift folding, it takes 3 instrs to compute base address:
lsr x8, x0, #1
and x8, x8, #0x78
add x8, x9, x8
If using ubfx, it only needs 2 instrs:
ubfx x8, x0, #4, #4
add x8, x9, x8, lsl #3
This fixes bug 19589
llvm-svn: 207702
Otherwise the legalizer would just scalarize everything. Support for
mulhi in the targets isn't that great yet so on most targets we get
exactly the same scalarized output. Add a test for x86 vector udiv.
I had to disable the mulhi nodes on ARM because there aren't any patterns
for it. As far as I know ARM has instructions for getting the high part of
a multiply so this should be fixed.
llvm-svn: 207315
The included test case would return the incorrect results, because the expansion
of an shift with a constant shift amount of 0 would generate undefined behavior.
This is because ExpandShiftByConstant assumes that all shifts by constants with
a value of 0 have already been optimized away. This doesn't happen for opaque
constants and usually this isn't a problem, because opaque constants won't take
this code path - they are not supposed to. In the case that the opaque constant
has to be expanded by the legalizer, the legalizer would drop the opaque flag.
In this case we hit the limitations of ExpandShiftByConstant and create incorrect
code.
This commit fixes the legalizer by not dropping the opaque flag when expanding
opaque constants and adding an assertion to ExpandShiftByConstant to catch this
not supported case in the future.
This fixes <rdar://problem/16718472>
llvm-svn: 207304
buildbot - do not insert debug intrinsics before phi nodes.
Debug info for optimized code: Support variables that are on the stack and
described by DBG_VALUEs during their lifetime.
Previously, when a variable was at a FrameIndex for any part of its
lifetime, this would shadow all other DBG_VALUEs and only a single
fbreg location would be emitted, which in fact is only valid for a small
range and not the entire lexical scope of the variable. The included
dbg-value-const-byref testcase demonstrates this.
This patch fixes this by
Local
- emitting dbg.value intrinsics for allocas that are passed by reference
- dropping all dbg.declares (they are now fully lowered to dbg.values)
SelectionDAG
- renamed constructors for SDDbgValue for better readability.
- fix UserValue::match() to handle indirect values correctly
- not inserting an MMI table entries for dbg.values that describe allocas.
- lowering dbg.values that describe allocas into *indirect* DBG_VALUEs.
CodeGenPrepare
- leaving dbg.values for an alloca were they are (see comment)
Other
- regenerated/updated instcombine.ll testcase and included source
rdar://problem/16679879
http://reviews.llvm.org/D3374
llvm-svn: 207269
AllocaInst that was missing in one location.
Debug info for optimized code: Support variables that are on the stack and
described by DBG_VALUEs during their lifetime.
Previously, when a variable was at a FrameIndex for any part of its
lifetime, this would shadow all other DBG_VALUEs and only a single
fbreg location would be emitted, which in fact is only valid for a small
range and not the entire lexical scope of the variable. The included
dbg-value-const-byref testcase demonstrates this.
This patch fixes this by
Local
- emitting dbg.value intrinsics for allocas that are passed by reference
- dropping all dbg.declares (they are now fully lowered to dbg.values)
SelectionDAG
- renamed constructors for SDDbgValue for better readability.
- fix UserValue::match() to handle indirect values correctly
- not inserting an MMI table entries for dbg.values that describe allocas.
- lowering dbg.values that describe allocas into *indirect* DBG_VALUEs.
CodeGenPrepare
- leaving dbg.values for an alloca were they are (see comment)
Other
- regenerated/updated instcombine.ll testcase and included source
rdar://problem/16679879
http://reviews.llvm.org/D3374
llvm-svn: 207235
AllocaInst that was missing in one location.
Debug info for optimized code: Support variables that are on the stack and
described by DBG_VALUEs during their lifetime.
Previously, when a variable was at a FrameIndex for any part of its
lifetime, this would shadow all other DBG_VALUEs and only a single
fbreg location would be emitted, which in fact is only valid for a small
range and not the entire lexical scope of the variable. The included
dbg-value-const-byref testcase demonstrates this.
This patch fixes this by
Local
- emitting dbg.value intrinsics for allocas that are passed by reference
- dropping all dbg.declares (they are now fully lowered to dbg.values)
SelectionDAG
- renamed constructors for SDDbgValue for better readability.
- fix UserValue::match() to handle indirect values correctly
- not inserting an MMI table entries for dbg.values that describe allocas.
- lowering dbg.values that describe allocas into *indirect* DBG_VALUEs.
CodeGenPrepare
- leaving dbg.values for an alloca were they are (see comment)
Other
- regenerated/updated instcombine.ll testcase and included source
rdar://problem/16679879
http://reviews.llvm.org/D3374
llvm-svn: 207165
described by DBG_VALUEs during their lifetime.
Previously, when a variable was at a FrameIndex for any part of its
lifetime, this would shadow all other DBG_VALUEs and only a single
fbreg location would be emitted, which in fact is only valid for a small
range and not the entire lexical scope of the variable. The included
dbg-value-const-byref testcase demonstrates this.
This patch fixes this by
Local
- emitting dbg.value intrinsics for allocas that are passed by reference
- dropping all dbg.declares (they are now fully lowered to dbg.values)
SelectionDAG
- renamed constructors for SDDbgValue for better readability.
- fix UserValue::match() to handle indirect values correctly
- not inserting an MMI table entries for dbg.values that describe allocas.
- lowering dbg.values that describe allocas into *indirect* DBG_VALUEs.
CodeGenPrepare
- leaving dbg.values for an alloca were they are (see comment)
Other
- regenerated/updated instcombine-intrinsics testcase and included source
rdar://problem/16679879
http://reviews.llvm.org/D3374
llvm-svn: 207130
define below all header includes in the lib/CodeGen/... tree. While the
current modules implementation doesn't check for this kind of ODR
violation yet, it is likely to grow support for it in the future. It
also removes one layer of macro pollution across all the included
headers.
Other sub-trees will follow.
llvm-svn: 206837
behavior based on other files defining DEBUG_TYPE, which means it cannot
define DEBUG_TYPE at all. This is actually better IMO as it forces folks
to define relevant DEBUG_TYPEs for their files. However, it requires all
files that currently use DEBUG(...) to define a DEBUG_TYPE if they don't
already. I've updated all such files in LLVM and will do the same for
other upstream projects.
This still leaves one important change in how LLVM uses the DEBUG_TYPE
macro going forward: we need to only define the macro *after* header
files have been #include-ed. Previously, this wasn't possible because
Debug.h required the macro to be pre-defined. This commit removes that.
By defining DEBUG_TYPE after the includes two things are fixed:
- Header files that need to provide a DEBUG_TYPE for some inline code
can do so by defining the macro before their inline code and undef-ing
it afterward so the macro does not escape.
- We no longer have rampant ODR violations due to including headers with
different DEBUG_TYPE definitions. This may be mostly an academic
violation today, but with modules these types of violations are easy
to check for and potentially very relevant.
Where necessary to suppor headers with DEBUG_TYPE, I have moved the
definitions below the includes in this commit. I plan to move the rest
of the DEBUG_TYPE macros in LLVM in subsequent commits; this one is big
enough.
The comments in Debug.h, which were hilariously out of date already,
have been updated to reflect the recommended practice going forward.
llvm-svn: 206822
Win64 stack unwinder gets confused when execution flow "falls through" after
a call to 'noreturn' function. This fixes the "missing epilogue" problem by
emitting a trap instruction for IR 'unreachable' on x86_x64-pc-windows.
A secondary use for it would be for anyone wanting to make double-sure that
'noreturn' functions, indeed, do not return.
llvm-svn: 206684
This particular DAG combine is designed to kick in when both ConstantFPs will
end up being loaded via a litpool, however those nodes have a semi-legal
status, dictated by isFPImmLegal so in some cases there wouldn't have been a
litpool in the first place. Don't try to be clever in those circumstances.
Picked up while merging some AArch64 tests.
llvm-svn: 206365
handles Intrinsic::trap if TargetOptions::TrapFuncName is set.
This fixes a bug in which the trap function was not taken into consideration
when a program was compiled without optimization (at -O0).
<rdar://problem/16291933>
llvm-svn: 206323
ARM64 suffered multiple -verify-machineinstr failures (principally over the
xsp/xzr issue) because FastISel was completely ignoring which subset of the
general-purpose registers each instruction required.
More fixes are coming in ARM64 specific FastISel, but this should cover the
generic problems.
llvm-svn: 206283
We had disabled use of TBAA during CodeGen (even when otherwise using AA)
because the ptrtoint/inttoptr used by CGP for address sinking caused BasicAA to
miss basic type punning that it should catch (and, thus, we'd fail to override
TBAA when we should).
However, when AA is in use during CodeGen, CGP now uses normal GEPs and
bitcasts, instead of ptrtoint/inttoptr, when doing address sinking. As a
result, BasicAA should be able to make us do the right thing in the face of
type-punning, and it seems safe to enable use of TBAA again. self-hosting seems
fine on PPC64/Linux on the P7, with TBAA enabled and -misched=shuffle.
Note: We still don't update TBAA when merging stack slots, although because
BasicAA should now catch all such cases, this is no longer a blocking issue.
Nevertheless, I plan to commit code to deal with this properly in the near
future.
llvm-svn: 206093
The TargetLowering::expandMUL() helper contains lowering code extracted
from the DAGTypeLegalizer and allows the SelectionDAGLegalizer to expand more
ISD::MUL patterns without having to use a library call.
llvm-svn: 206037
This code has been moved to a new function in the TargetLowering
class called expandMUL(). The purpose of this is to be able
to share lowering code between the SelectionDAGLegalize and
DAGTypeLegalizer classes.
No functionality changed intended.
llvm-svn: 206036
FoldConstantArithmetic() only knows how to deal with a few target independent
ISD opcodes. Bail early if it sees a target-specific ISD node. These node do
funny things with operand types which may break the assumptions of the code
that follows, and there's no actual folding that can be done anyway. For example,
non-constant 256 bit vector shifts on X86 have a shift-amount operand that's a
128-bit v4i32 vector regardless of what the first operand type is and that breaks
the assumption that the operand types must match.
rdar://16530923
llvm-svn: 205937
sign/zero/any extensions. However a few places were not checking properly the
property of the load and were turning an indexed load into a regular extended
load. Therefore the indexed value was lost during the process and this was
triggering an assertion.
<rdar://problem/16389332>
llvm-svn: 205923
Fixes PR16365 - Extremely slow compilation in -O1 and -O2.
The SD scheduler has a quadratic implementation of load clustering
which absolutely blows up compile time for large blocks with constant
pool loads. The MI scheduler has a better implementation of load
clustering. However, we have not done the work yet to completely
eliminate the SD scheduler. Some benchmarks still seem to benefit from
early load clustering, although maybe by chance.
As an intermediate term fix, I just put a nice limit on the number of
DAG users to search before finding a match. With this limit there are no
binary differences in the LLVM test suite, and the PR16365 test case
does not suffer any compile time impact from this routine.
llvm-svn: 205738
When LLVM sees something like (v1iN (vselect v1i1, v1iN, v1iN)) it can
decide that the result is OK (v1i64 is legal on AArch64, for example)
but it still need scalarising because of that v1i1. There was no code
to do this though.
AArch64 and ARM64 have DAG combines to produce efficient code and
prevent that occuring in *most* such situations, but there are edge
cases that they miss. This adds a legalization to cope with that.
llvm-svn: 205626
There were several overlapping problems here, and this solution is
closely inspired by the one adopted in AArch64 in r201381.
Firstly, scalarisation of v1i1 setcc operations simply fails if the
input types are legal. This is fixed in LegalizeVectorTypes.cpp this
time, and allows AArch64 code to be simplified slightly.
Second, vselect with such a setcc feeding into it ends up in
ScalarizeVectorOperand, where it's not handled. I experimented with an
implementation, but found that whatever DAG came out was rather
horrific. I think Hao's DAG combine approach is a good one for
quality, though there are edge cases it won't catch (to be fixed
separately).
Should fix PR19335.
llvm-svn: 205625
llc doesn't generate nodes for unconditional fall-through branches for targets
without FastISel implementation (X86 has it, but can be disabled by
"-fast-isel=false") in SelectionDAGBuilder::visitBr().
So for line 4 in the following testcase
1: void foo(int i){
2: switch(i){
3: default:
4: break;
5: }
6: return;
7: }
there is no corresponding line in .debug_line section, and a debugger
cannot set a breakpoint at line 4.
Fix this by always emitting a branch when we're not optimizing and add a
testcase to ensure that there's code on every line we'd want to break.
Patch by Daniil Fukalov.
llvm-svn: 205529
This adds the ability to expand large (meaning with more than two unique
defined values) BUILD_VECTOR nodes in terms of SCALAR_TO_VECTOR and (legal)
vector shuffles. There is now no limit of the size we are capable of expanding
this way, although we don't currently do this for vectors with many unique
values because of the default implementation of TLI's
shouldExpandBuildVectorWithShuffles function.
There is currently no functional change to any existing targets because the new
capabilities are not used unless some target overrides the TLI
shouldExpandBuildVectorWithShuffles function. As a result, I've not included a
test case for the new functionality in this commit, but regression tests will
(at least) be added soon when I commit support for the PPC QPX vector
instruction set.
The benefit of committing this now is that it makes the
shouldExpandBuildVectorWithShuffles callback, which had to be added for other
reasons regardless, fully functional. I suspect that other targets will
also benefit from tuning the heuristic.
llvm-svn: 205243
There are two general methods for expanding a BUILD_VECTOR node:
1. Use SCALAR_TO_VECTOR on the defined scalar values and then shuffle
them together.
2. Build the vector on the stack and then load it.
Currently, we use a fixed heuristic: If there are only one or two unique
defined values, then we attempt an expansion in terms of SCALAR_TO_VECTOR and
vector shuffles (provided that the required shuffle mask is legal). Otherwise,
always expand via the stack. Even when SCALAR_TO_VECTOR is not legal, this
can still be a good idea depending on what tricks the target can play when
lowering the resulting shuffle. If the target can't do anything special,
however, and if SCALAR_TO_VECTOR is expanded via the stack, this heuristic
leads to sub-optimal code (two stack loads instead of one).
Because only the target knows whether the SCALAR_TO_VECTORs and shuffles for a
build vector of a particular type are likely to be optimial, this adds a new
TLI function: shouldExpandBuildVectorWithShuffles which takes the vector type
and the count of unique defined values. If this function returns true, then
method (1) will be used, subject to the constraint that all of the necessary
shuffles are legal (as determined by isShuffleMaskLegal). If this function
returns false, then method (2) is always used.
This commit does not enhance the current code to support expanding a
build_vector with more than two unique values using shuffles, but I'll commit
an implementation of the more-general case shortly.
llvm-svn: 205230
When the loop vectorizer vectorizes code that uses the loop induction variable,
we often end up with IR like this:
%b1 = insertelement <2 x i32> undef, i32 %v, i32 0
%b2 = shufflevector <2 x i32> %b1, <2 x i32> undef, <2 x i32> zeroinitializer
%i = add <2 x i32> %b2, <i32 2, i32 3>
If the add in this example is not legal (as is the case on PPC with VSX), it
will be scalarized, and we'll end up with a number of extract_vector_elt nodes
with the vector shuffle as the input operand, and that vector shuffle is fed by
one or more build_vector nodes. By the time that vector operations are
expanded, visitEXTRACT_VECTOR_ELT will not create new extract_vector_elt by
looking through the vector shuffle (to make sure that no illegal operations are
created), and so the extract_vector_elt -> vector shuffle -> build_vector is
never simplified to an operand of the build vector.
By looking at build_vectors through a shuffle we fix this particular situation,
preventing a vector from being built, only to be deconstructed again (for the
scalarized add) -- an expensive proposition when this all needs to be done via
the stack. We probably want a more comprehensive fix here where we look back
recursively through any shuffles to any build_vectors or scalar_to_vectors,
etc. but that can come later.
llvm-svn: 205179
When expanding EXTRACT_VECTOR_ELT and EXTRACT_SUBVECTOR using
SelectionDAGLegalize::ExpandExtractFromVectorThroughStack, we store the entire
vector and then load the piece we want. This is fine in isolation, but
generating a new store (and corresponding stack slot) for each extraction ends
up producing code of poor quality. When we scalarize a vector operation (using
SelectionDAG::UnrollVectorOp for example) we generate one EXTRACT_VECTOR_ELT
for each element in the vector. This used to generate one stored copy of the
vector for each element in the vector. Now we search the uses of the vector for
a suitable store before generating a new one, which results in much more
efficient scalarization code.
llvm-svn: 205153
This adds back r204781.
Original message:
Aliases are just another name for a position in a file. As such, the
regular symbol resolutions are not applied. For example, given
define void @my_func() {
ret void
}
@my_alias = alias weak void ()* @my_func
@my_alias2 = alias void ()* @my_alias
We produce without this patch:
.weak my_alias
my_alias = my_func
.globl my_alias2
my_alias2 = my_alias
That is, in the resulting ELF file my_alias, my_func and my_alias are
just 3 names pointing to offset 0 of .text. That is *not* the
semantics of IR linking. For example, linking in a
@my_alias = alias void ()* @other_func
would require the strong my_alias to override the weak one and
my_alias2 would end up pointing to other_func.
There is no way to represent that with aliases being just another
name, so the best solution seems to be to just disallow it, converting
a miscompile into an error.
llvm-svn: 204934
Implementing the LLVM part of the call to __builtin___clear_cache
which translates into an intrinsic @llvm.clear_cache and is lowered
by each target, either to a call to __clear_cache or nothing at all
incase the caches are unified.
Updating LangRef and adding some tests for the implemented architectures.
Other archs will have to implement the method in case this builtin
has to be compiled for it, since the default behaviour is to bail
unimplemented.
A Clang patch is required for the builtin to be lowered into the
llvm intrinsic. This will be done next.
llvm-svn: 204802
This reverts commit r204781.
I will follow up to with msan folks to see what is what they
were trying to do with aliases to weak aliases.
llvm-svn: 204784
Aliases are just another name for a position in a file. As such, the
regular symbol resolutions are not applied. For example, given
define void @my_func() {
ret void
}
@my_alias = alias weak void ()* @my_func
@my_alias2 = alias void ()* @my_alias
We produce without this patch:
.weak my_alias
my_alias = my_func
.globl my_alias2
my_alias2 = my_alias
That is, in the resulting ELF file my_alias, my_func and my_alias are
just 3 names pointing to offset 0 of .text. That is *not* the
semantics of IR linking. For example, linking in a
@my_alias = alias void ()* @other_func
would require the strong my_alias to override the weak one and
my_alias2 would end up pointing to other_func.
There is no way to represent that with aliases being just another
name, so the best solution seems to be to just disallow it, converting
a miscompile into an error.
llvm-svn: 204781
Usually opaque constants shouldn't be folded, unless they are simple unary
operations that don't create new constants. Although this shouldn't drop the
opaque constant flag. This commit fixes this.
Related to <rdar://problem/14774662>
llvm-svn: 204737
If GT/UGT or LT/ULT were set to expand, a comparison
with a constant would replace it with the illegal
cond code.
There are several more places later in this function that
will have the same basic problem.
Theoretically R600 should hit this problem for a test,
but for some reason it doesn't.
llvm-svn: 204727
This patch renames method 'isConstantSplat' as 'getConstantSplatValue'
(mainly for consistency reasons), and rewrites its logic to ensure
that we always perform a legal 'cast<ConstantSDNode>'.
Added test shift-combine-crash.ll to verify that DAGCombiner no longer crashes with an assertion failure in the attempt to simplify a vector shift by a vector of all undef counts.
llvm-svn: 204536
Summary:
SLP Vectorization of intrinsics (r203707) has exposed cases where the
expansion of vector bswap is failing (PR19151).
Reviewers: hfinkel
CC: chandlerc
Differential Revision: http://llvm-reviews.chandlerc.com/D3104
llvm-svn: 204163
Rather than LegalizeAction::Expand, this needs LegalizeAction::Promote to get
promoted to fp_to_sint v8f32->v8i32. This is a legal operation on AVX.
For that to work properly, we also need to teach the legalizer about the
specific promotion required here. The default vector promotion uses
bitcasting to a vector type of the same total size. We want to promote the
vector element type, effectively widening the operation and then truncating
the result. This is analogous to the current logic of how int_to_fp is
promoted.
The change also factors out some code from the int_to_fp promotion code to
ValueType::widenIntegerVectorElementType. This is now shared between
int_to_fp and fp_to_int.
There is no longer need for the custom lowering of fp_to_sint f32->v8i16 in
X86. It can now go through the new target-independent fp_to_*int promotion
logic.
I also checked that no other target uses Promote for these ops yet, so there
shouldn't be any unexpected change in behavior.
Fixes <rdar://problem/16202247>
llvm-svn: 204058
operator* on the by-operand iterators to return a MachineOperand& rather than
a MachineInstr&. At this point they almost behave like normal iterators!
Again, this requires making some existing loops more verbose, but should pave
the way for the big range-based for-loop cleanups in the future.
llvm-svn: 203865
for use with C++11 range-based for-loops.
The gist of phase 1 is to remove the skipInstruction() and skipBundle()
methods from these iterators, instead splitting each iterator into a version
that walks operands, a version that walks instructions, and a version that
walks bundles. This has the result of making some "clever" loops in lib/CodeGen
more verbose, but also makes their iterator invalidation characteristics much
more obvious to the casual reader. (Making them concise again in the future is a
good motivating case for a pre-incrementing range adapter!)
Phase 2 of this undertaking with consist of removing the getOperand() method,
and changing operator*() of the operand-walker to return a MachineOperand&. At
that point, it should be possible to add range views for them that work as one
might expect.
llvm-svn: 203757
The syntax for "cmpxchg" should now look something like:
cmpxchg i32* %addr, i32 42, i32 3 acquire monotonic
where the second ordering argument gives the required semantics in the case
that no exchange takes place. It should be no stronger than the first ordering
constraint and cannot be either "release" or "acq_rel" (since no store will
have taken place).
rdar://problem/15996804
llvm-svn: 203559
This requires a number of steps.
1) Move value_use_iterator into the Value class as an implementation
detail
2) Change it to actually be a *Use* iterator rather than a *User*
iterator.
3) Add an adaptor which is a User iterator that always looks through the
Use to the User.
4) Wrap these in Value::use_iterator and Value::user_iterator typedefs.
5) Add the range adaptors as Value::uses() and Value::users().
6) Update *all* of the callers to correctly distinguish between whether
they wanted a use_iterator (and to explicitly dig out the User when
needed), or a user_iterator which makes the Use itself totally
opaque.
Because #6 requires churning essentially everything that walked the
Use-Def chains, I went ahead and added all of the range adaptors and
switched them to range-based loops where appropriate. Also because the
renaming requires at least churning every line of code, it didn't make
any sense to split these up into multiple commits -- all of which would
touch all of the same lies of code.
The result is still not quite optimal. The Value::use_iterator is a nice
regular iterator, but Value::user_iterator is an iterator over User*s
rather than over the User objects themselves. As a consequence, it fits
a bit awkwardly into the range-based world and it has the weird
extra-dereferencing 'operator->' that so many of our iterators have.
I think this could be fixed by providing something which transforms
a range of T&s into a range of T*s, but that *can* be separated into
another patch, and it isn't yet 100% clear whether this is the right
move.
However, this change gets us most of the benefit and cleans up
a substantial amount of code around Use and User. =]
llvm-svn: 203364
This is already done for shifts. Allow it for rotations as well. E.g.:
(rotl:i32 x, (trunc (and y, 31))) -> (rotl:i32 x, (and (trunc y), 31))
Use the newly factored-out distributeTruncateThroughAnd.
With this patch and some X86.td tweaks we should be able to remove redundant
masking of the rotation amount like in the example above. HW implicitly
performs this masking.
The testcase will be added as part of the X86 patch.
llvm-svn: 203316
This is the new idiom:
x<<(y&31) | x>>((0-y)&31)
which is recognized as:
x ROTL (y&31)
The change refines matchRotateSub. In
Neg & (OpSize - 1) == (OpSize - Pos) & (OpSize - 1), if Pos is
Pos' & (OpSize - 1) we can just use Pos' instead of Pos.
llvm-svn: 203315
Slightly change the wording in the function comment. Originally, it can be
misunderstood as we turned the input into two subsequent rotates.
Better connect the comment which talks about Mask and the code which used
LoBits. Renamed variable to MaskLoBits.
llvm-svn: 203314
be split and the result type widened.
When the condition of a vselect has to be split it makes no sense widening the
vselect and thereby widening the condition. We end up in an endless loop of
widening (vselect result type) and splitting (condition mask type) doing this.
Instead, split both the condition and the vselect and widen the result.
I ran this over the test suite with i686 and mattr=+sse and saw no regressions.
Fixes PR18036.
llvm-svn: 203311
This patch teaches the DAGCombiner how to fold a binary OR between two
shufflevector into a single shuffle vector when possible.
The rules are:
1. fold (or (shuf A, V_0, MA), (shuf B, V_0, MB)) -> (shuf A, B, Mask1)
2. fold (or (shuf A, V_0, MA), (shuf B, V_0, MB)) -> (shuf B, A, Mask2)
The DAGCombiner can take advantage of the fact that OR is commutative and
compute two possible shuffle masks (Mask1 and Mask2) for the resulting
shuffle node.
Before folding a dag according to either rule 1 or 2, DAGCombiner verifies
that the resulting shuffle mask is legal for the target.
DAGCombiner would firstly try to fold according to 1.; If not possible
then it will try to fold according to 2.
If both Mask1 and Mask2 are illegal then we conservatively don't fold
the OR instruction.
llvm-svn: 203156
already lived there and it is where it belongs -- this is the in-memory
debug location representation.
This is just cleanup -- Modules can actually cope with this, but that
doesn't make it right. After chatting with folks that have out-of-tree
stuff, going ahead and moving the rest of the headers seems preferable.
llvm-svn: 202960
Patchpoints already did this. Doing it for stackmaps is a convenience
for the runtime in the event that it needs to scratch register to
patch or perform a runtime call thunk.
Unlike patchpoints, we just assume the AnyRegCC calling
convention. This is the only language and target independent calling
convention specific to stackmaps so makes sense. Although the calling
convention is not currently used to select the scratch registers.
llvm-svn: 202943
selection dag (PR19012)
In X86SelectionDagInfo::EmitTargetCodeForMemcpy we check with MachineFrameInfo
to make sure that ESI isn't used as a base pointer register before we choose to
emit rep movs (which clobbers esi).
The problem is that MachineFrameInfo wouldn't know about dynamic allocas or
inline asm that clobbers the stack pointer until SelectionDAGBuilder has
encountered them.
This patch fixes the problem by checking for such things when building the
FunctionLoweringInfo.
Differential Revision: http://llvm-reviews.chandlerc.com/D2954
llvm-svn: 202930
Currently this code is duplicated across visitSHL, visitSRA and visitSRL. The
plan is to add rotates as clients to this new function.
There is no functional change intended here.
llvm-svn: 202908
This extract-and-trunc vector optimization cannot work for i1 values as
currently implemented, and so I'm disabling this for now for i1 values. In the
future, this can be fixed properly.
Soon I'll commit support for i1 CR bit tracking in the PowerPC backend, and
this will be covered by one of the existing regression tests.
llvm-svn: 202449
shifted mask rather than masking and shifting separately.
The patch adds this transformation to the DAGCombiner:
(shl (and (setcc:i8v16 ...) N01C) N1C) -> (and (setcc:i8v16 ...) N01C<<N1C)
<rdar://problem/16054492>
Patch by Adam Nemet <anemet@apple.com>
llvm-svn: 201906
This fix checks the original LLVM IR node to identify opaque constants by
looking for the bitcast-constant pattern. Originally we looked at the generated
SDNode, but this might lead to incorrect results. The SDNode could have been
generated by an constant expression that was folded to a constant.
This fixes <rdar://problem/16050719>
llvm-svn: 201291
We are now no longer relying on the target-specific call lowering implementation
to lower a stackmap intrinsic call. Instead we perform the call lowering in a
target-independent way directly in the stackmap lowering code. This simplifies
the code and removes the need to fixup the code after the target-specific call
lowering.
llvm-svn: 201263
The ID type for the stackmap and patchpoint intrinsics are in both cases i64.
This fixes an zero extend in the SelectionDAGBuilder that still used i32. This
also updates the target independent instructions STACKMAP and PATCHPOINT to use
the correct type.
llvm-svn: 201262
BUILD_VECTOR nodes, e.g.:
(concat_vectors (BUILD_VECTOR a1, a2, a3, a4), (BUILD_VECTOR b1, b2, b3, b4))
->
(BUILD_VECTOR a1, a2, a3, a4, b1, b2, b3, b4)
This fixes an issue with AVX, where a sequence was not recognized as a 256-bit
vbroadcast due to the concat_vectors.
llvm-svn: 201158
During DAGCombine visitShiftByConstant assumes that certain binary operations
with only constant operands can always be folded successfully. This is no longer
true when the constant is opaque. This commit fixes visitShiftByConstant by not
performing the optimization for opaque constants. Otherwise we would end up in
an infinite DAGCombine loop.
llvm-svn: 200900
Calls with inalloca are lowered by skipping all stores for arguments
passed in memory and the initial stack adjustment to allocate argument
memory.
Now the frontend is responsible for the memory layout, and the backend
doesn't have to do any work. As a result these changes are pretty
minimal.
Reviewers: echristo
Differential Revision: http://llvm-reviews.chandlerc.com/D2637
llvm-svn: 200596
Allocas marked inalloca are never static, but we were trying to put them
into the static alloca map if they were in the entry block. Also add an
assertion in x86 fastisel.
llvm-svn: 200593
when the input is a concat_vectors and the insert replaces one of the
concat halves:
Lower half: fold (insert_subvector (concat_vectors X, Y), Z) ->
(concat_vectors Z, Y)
Upper half: fold (insert_subvector (concat_vectors X, Y), Z) ->
(concat_vectors X, Z)
This can be seen with the following IR:
define <8 x float> @lower_half(<4 x float> %v1, <4 x float> %v2, <4 x
float> %v3) {
%1 = shufflevector <4 x float> %v1, <4 x float> %v2, <8 x i32> <i32
0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%2 = tail call <8 x float> @llvm.x86.avx.vinsertf128.ps.256(<8 x
float> %1, <4 x float> %v3, i8 0)
The vinsertf128 intrinsic is converted into an insert_subvector node
in SelectionDAGBuilder.cpp.
Using AVX, without the patch this generates two vinsertf128 instructions:
vinsertf128 $1, %xmm1, %ymm0, %ymm0
vinsertf128 $0, %xmm2, %ymm0, %ymm0
With the patch this is optimized into:
vinsertf128 $1, %xmm1, %ymm2, %ymm0
Patch by Robert Lougher.
llvm-svn: 200506
When converting from "or + br" to two branches, or converting from
"and + br" to two branches, we correctly update the edge weights of
the two branches.
The previous attempt at r200431 was reverted at r200434 because of
two testing case failures. I modified my patch a little, but forgot
to re-run "make check-all".
Testing case CodeGen/ARM/lsr-unfolded-offset.ll is updated because of
the patch's impact on branch probability which causes changes in
spill placement.
llvm-svn: 200502
When converting from "or + br" to two branches, or converting from
"and + br" to two branches, we correctly update the edge weights of
the two branches.
llvm-svn: 200431
Make sure that we don't introduce illegal build_vector dag nodes
when trying to fold a sign_extend of a build_vector.
This fixes a regression introduced by r200234.
Added test CodeGen/X86/fold-vector-sext-crash.ll
to verify that llc no longer crashes with an assertion failure
due to an illegal build_vector of type MVT::v4i64.
Thanks to Ilia Filippov for spotting this regression and for
providing a reproducible test case.
llvm-svn: 200313
Before this patch we used getIntImmCost from TargetTransformInfo to determine if
a load of a constant should be converted to just a constant, but the threshold
for this was set to an arbitrary value. This value works well for the two
targets (X86 and ARM) that implement this target-hook, but it isn't
target-independent at all.
Now targets have the possibility to decide directly if this optimization should
be performed. The default value is set to false to preserve the current
behavior. The target hook has been moved to TargetLowering, which removed the
last use and need of TargetTransformInfo in SelectionDAG.
llvm-svn: 200271
Also update the comment, since it actually produces a
select (setcc) instead of select_cc.
It was checking and using the setcc result type for the
type of the sext, instead of the type of the compared items.
In my problem case, the sext was to i32 and was used as the setcc type,
but the expected type was i64.
No test since I haven't been able to hit the problem with
this on any in-tree targets.
llvm-svn: 200249
This patch teaches the DAGCombiner how to fold a sext/aext/zext dag node when
the operand in input is a build vector of constants (or UNDEFs).
The inability to fold a sext/zext of a constant build_vector was the root
cause of some pcg bugs affecting vselect expansion on x86-64 with AVX support.
Before this change, the DAGCombiner only knew how to fold a sext/zext/aext of a
ConstantSDNode.
llvm-svn: 200234
Issue outcomes from DAGCombiner::MergeConsequtiveStores, more precisely from
mem-ops sequence sorting.
Consider, how MergeConsequtiveStores works for next example:
store i8 1, a[0]
store i8 2, a[1]
store i8 3, a[1] ; a[1] again.
return ; DAG starts here
1. Method will collect all the 3 stores.
2. It sorts them by distance from the base pointer (farthest with highest
index).
3. It takes first consecutive non-overlapping stores and (if possible) replaces
them with a single store instruction.
The point is, we can't determine here which 'store' instruction
would be the second after sorting ('store 2' or 'store 3').
It happens that 'store 3' would be the second, and 'store 2' would be the third.
So after merging we have the next result:
store i16 (1 | 3 << 8), base ; is a[0] but bit-casted to i16
store i8 2, a[1]
So actually we swapped 'store 3' and 'store 2' and got wrong contents in a[1].
Fix: In sort routine just also take into account mem-op sequence number.
llvm-svn: 200201
There are currently two issues, of which I currently know, that prevent TBAA
from being correctly usable in CodeGen:
1. Stack coloring does not update TBAA when merging allocas. This is easy
enough to fix, but is not the largest problem.
2. CGP inserts ptrtoint/inttoptr pairs when sinking address computations.
Because BasicAA does not handle inttoptr, we'll often miss basic type punning
idioms that we need to catch so we don't miscompile real-world code (like LLVM).
I don't yet have a small test case for this, but this fixes self hosting a
non-asserts build of LLVM on PPC64 when using -enable-aa-sched-mi and -misched=shuffle.
llvm-svn: 200093
This option (which is !NDEBUG only) allows restricting the use of alias
analysis in DAGCombiner to a specific function. This has proved extremely
valuable to isolating bugs related to this feature, and mirrors the
misched-only-func option provided by the new instruction scheduler.
llvm-svn: 200088
This commit caused -Woverloaded-virtual warnings. The two new
TargetTransformInfo::getIntImmCost functions were only added to the superclass,
and to the X86 subclass. The other targets were not updated, and the
warning highlighted this by pointing out that e.g. ARMTTI::getIntImmCost was
hiding the two new getIntImmCost variants.
We could pacify the warning by adding "using TargetTransformInfo::getIntImmCost"
to the various subclasses, or turning it off, but I suspect that it's wrong to
leave the functions unimplemnted in those targets. The default implementations
return TCC_Free, which I don't think is right e.g. for ARM.
llvm-svn: 200058
Retry commit r200022 with a fix for the build bot errors. Constant expressions
have (unlike instructions) module scope use lists and therefore may have users
in different functions. The fix is to simply ignore these out-of-function uses.
llvm-svn: 200034
DAGCombiner::GatherAllAliases, which is only used when AA used is enabled
during DAGCombine, had a fundamentally incorrect assumption for which this
change compensates. GatherAllAliases, which is used to find aliasing
predecessor chain nodes (so that a better chain can be selected for a load or
store to enable subsequent optimizations) assumed that walking up the chain
would always catch all possibly-aliasing loads and stores. This is not true: To
really find all aliases, we also need to search for aliases through the value
operand of a store, etc. Consider the following situation:
Token1 = ...
L1 = load Token1, %52
S1 = store Token1, L1, %51
L2 = load Token1, %52+8
S2 = store Token1, L2, %51+8
Token2 = Token(S1, S2)
L3 = load Token2, %53
S3 = store Token2, L3, %52
L4 = load Token2, %53+8
S4 = store Token2, L4, %52+8
If we search for aliases of S3 (which loads address %52), and we look only
through the chain, then we'll miss the trivial dependence on L1 (which loads
from %52). We then might change all loads and stores to use Token1 as their
chain operand, which could result in copying %53 into %52 before copying
%52 into %51 (which should happen first).
The problem is, however, that searching for such data dependencies can become
expensive, and the cost is not directly related to the chain depth. Instead,
we'll rule out such configurations by insisting that we've visited all chain
users (except for users of the original chain, which is not necessary). When
doing this, we need to look through nodes we don't care about (otherwise,
things like register copies will interfere with trivial use cases).
Unfortunately, I don't have a small test case for this problem. Creating the
underlying situation is not hard (a pair of memcpys will do it), but arranging
for the default instruction schedule to be incorrect is very fragile.
This unbreaks self hosting on PPC64 when using
-mllvm -combiner-global-alias-analysis -mllvm -combiner-alias-analysis.
llvm-svn: 200033
These transformations obviously won't work for indexed (pre/post-inc) loads and
stores. In practice, I'm not sure there is any benefit to enabling them for
indexed nodes because other transformations that these might enable likely also
won't handle indexed nodes.
I don't have an in-tree test case that hits this problem, but an upcoming bug
fix will make it much more likely.
llvm-svn: 200023
This pass identifies expensive constants to hoist and coalesces them to
better prepare it for SelectionDAG-based code generation. This works around the
limitations of the basic-block-at-a-time approach.
First it scans all instructions for integer constants and calculates its
cost. If the constant can be folded into the instruction (the cost is
TCC_Free) or the cost is just a simple operation (TCC_BASIC), then we don't
consider it expensive and leave it alone. This is the default behavior and
the default implementation of getIntImmCost will always return TCC_Free.
If the cost is more than TCC_BASIC, then the integer constant can't be folded
into the instruction and it might be beneficial to hoist the constant.
Similar constants are coalesced to reduce register pressure and
materialization code.
When a constant is hoisted, it is also hidden behind a bitcast to force it to
be live-out of the basic block. Otherwise the constant would be just
duplicated and each basic block would have its own copy in the SelectionDAG.
The SelectionDAG recognizes such constants as opaque and doesn't perform
certain transformations on them, which would create a new expensive constant.
This optimization is only applied to integer constants in instructions and
simple (this means not nested) constant cast experessions. For example:
%0 = load i64* inttoptr (i64 big_constant to i64*)
Reviewed by Eric
llvm-svn: 200022
This is a horrible bit of code. We're calling a simplification routine *in the middle* of type legalization. We tell the
simplification routine that it's running after legalization, but some of the types it will encounter will be illegal! The
fix is only to invoke the simplification if the types in question were legal, so that none of its invariants will be violated.
llvm-svn: 199847
This fixes a regression intruced by r199135.
Revision 199135 tried to simplify part of the logic in method
DAGCombiner::SimplifyVBinOp introducing calls to method BuildVectorSDNode::isConstant().
However, that revision wrongly changed the check performed by method
SimplifyVBinOp to identify dag nodes that can be folded.
Before revision 199135, that method only tried to simplify vector binary operations
if both operands were build_vector of Constant/ConstantFP/Undef only.
After revision 199135, method SimplifyVBinop tried to
simplify also vector binary operations with only one constant operand.
This fixes the problem restoring the old behavior of SimplifyVBinOp.
llvm-svn: 199328
When creating a virtual register for a def, the value type should be
used to pick the register class. If we only use the register class
constraint on the instruction, we might pick a too large register class.
Some registers can store values of different sizes. For example, the x86
xmm registers can hold f32, f64, and 128-bit vectors. The three
different value sizes are represented by register classes with identical
register sets: FR32, FR64, and VR128. These register classes have
different spill slot sizes, so it is important to use the right one.
The register class constraint on an instruction doesn't necessarily care
about the size of the value its defining. The value type determines
that.
This fixes a problem where InstrEmitter was picking 32-bit register
classes for 64-bit values on SPARC.
llvm-svn: 199187
This commit teaches DAG to reassociate vector ops, which in turn enables
constant folding of vector op chains that appear later on during custom lowering
and DAG combine.
Reviewed by Andrea Di Biagio
llvm-svn: 199135
This is a very confusing option for a feature that will go away.
-enable-misched is exposed instead to help triage issues with the new
scheduler.
llvm-svn: 199133
At the moment we expect rotates to have the form:
(or (shl X, Y), (shr X, Z))
where Y == bitsize(X) - Z or Z == bitsize(X) - Y. This form means that
the (or ...) is undefined for Y == 0 or Z == 0. This undefinedness can
be avoided by using Y == (C * bitsize(X) - Z) & (bitsize(X) - 1) or
Z == (C * bitsize(X) - Y) & (bitsize(X) - 1) for any integer C
(including 0, the most natural choice).
llvm-svn: 198861
InstCombine converts (sub 32, (add X, C)) into (sub 32-C, X),
so a rotate left of a 32-bit Y by X+C could appear as either:
(or (shl Y, (add X, C)), (shr Y, (sub 32, (add X, C))))
without InstCombine or:
(or (shl Y, (add X, C)), (shr Y, (sub 32-C, X)))
with it.
We already matched the first form. This patch handles the second too.
llvm-svn: 198860
operand into the Value interface just like the core print method is.
That gives a more conistent organization to the IR printing interfaces
-- they are all attached to the IR objects themselves. Also, update all
the users.
This removes the 'Writer.h' header which contained only a single function
declaration.
llvm-svn: 198836
are part of the core IR library in order to support dumping and other
basic functionality.
Rename the 'Assembly' include directory to 'AsmParser' to match the
library name and the only functionality left their -- printing has been
in the core IR library for quite some time.
Update all of the #includes to match.
All of this started because I wanted to have the layering in good shape
before I started adding support for printing LLVM IR using the new pass
infrastructure, and commandline support for the new pass infrastructure.
llvm-svn: 198688
There is a wrong assumption that the vector element type and the
type of each ConstantSDNode in the build_vector were the same.
However, when promoting the integer operand of a legally typed
build_vector, the operand type and the vector element type do not
need to be the same
(See method 'DAGTypeLegalizer::PromoteIntOp_BUILD_VECTOR' in
LegalizeIntegerTypes.cpp).
in AArch64 backend, the following dag sequence:
C0: i1 = Constant<0>
C1: i1 = Constant<-1>
V: v8i1 = BUILD_VECTOR C1, C1, C0, C0, C0, C0, C0, C0
is type-legalized into:
NewC0: i32 = Constant<0>
NewC1: i32 = Constant<1>
V: v8i8 = BUILD_VECTOR NewC1, NewC1, NewC0, NewC0, NewC0, NewC0, NewC0, NewC0
Forcing a getZeroExtend to VTBits to ensure that the new constant
is correctly.
llvm-svn: 198582
This moves the check up into the parent class so that all targets can use it
without having to copy (and keep in sync) the same error message.
llvm-svn: 198579
For AArch64 backend, if DAGCombiner see "sext(setcc)", it will
combine them together to a single setcc with extended value type.
Then if it see "zext(setcc)", it assumes setcc is Vxi1, and try to
create "(and (vsetcc), (1, 1, ...)". While setcc isn't Vxi1,
DAGcombiner will create wrong node and get wrong code emitted.
llvm-svn: 198190
ConstantSDNodes (or UNDEFs) into a simple BUILD_VECTOR.
For example, given the following sequence of dag nodes:
i32 C = Constant<1>
v4i32 V = BUILD_VECTOR C, C, C, C
v4i32 Result = SIGN_EXTEND_INREG V, ValueType:v4i1
The SIGN_EXTEND_INREG node can be folded into a build_vector since
the vector in input is a BUILD_VECTOR of constants.
The optimized sequence is:
i32 C = Constant<-1>
v4i32 Result = BUILD_VECTOR C, C, C, C
llvm-svn: 198084
This changes the MachineFrameInfo API to use the new SSPLayoutKind information
produced by the StackProtector pass (instead of a boolean flag) and updates a
few pass dependencies (to preserve the SSP analysis).
The stack layout follows the same approach used prior to this change - i.e.,
only LargeArray stack objects will be placed near the canary and everything
else will be laid out normally. After this change, structures containing large
arrays will also be placed near the canary - a case previously missed by the
old implementation.
Out of tree targets will need to update their usage of
MachineFrameInfo::CreateStackObject to remove the MayNeedSP argument.
The next patch will implement the rules for sspstrong and sspreq. The end goal
is to support ssp-strong stack layout rules.
WIP.
Differential Revision: http://llvm-reviews.chandlerc.com/D2158
llvm-svn: 197653
This optional register liveness analysis pass can be enabled with either
-enable-stackmap-liveness, -enable-patchpoint-liveness, or both. The pass
traverses each basic block in a machine function. For each basic block the
instructions are processed in reversed order and if a patchpoint or stackmap
instruction is encountered the current live-out register set is encoded as a
register mask and attached to the instruction.
Later on during stackmap generation the live-out register mask is processed and
also emitted as part of the stackmap.
This information is optional and intended for optimization purposes only. This
will enable a client of the stackmap to reason about the registers it can use
and which registers need to be preserved.
Reviewed by Andy
llvm-svn: 197317
This reverts commit r197254.
This was an accidental merge of Juergen's patch. It will be checked in
shortly, but wasn't meant to go in quite yet.
Conflicts:
include/llvm/CodeGen/StackMaps.h
lib/CodeGen/StackMaps.cpp
test/CodeGen/X86/stackmap-liveness.ll
llvm-svn: 197260
DAGCombiner could fold (truncate (load)) -> smaller load if the original
load was the width of the truncation result or wider. This patch extends
it to handle cases where the original load was narrower (and so the
extension type stays the same).
llvm-svn: 197030
This re-lands commit r196876, which was reverted in r196879.
The tests have been fixed to pass on platforms with a stack alignment
larger than 4.
Update to clang side tests will land shortly.
llvm-svn: 196939
One unusual feature of the z architecture is that the result of a
previous load can be reused indefinitely for subsequent loads, even if
a cache-coherent store to that location is performed by another CPU.
A special serializing instruction must be used if you want to force
a load to be reattempted.
Since volatile loads are not supposed to be omitted in this way,
we should insert a serializing instruction before each such load.
The same goes for atomic loads.
The patch implements this at the IR->DAG boundary, in a similar way
to atomic fences. It is a no-op for targets other than SystemZ.
llvm-svn: 196905
For stack frames requiring realignment, three pointers may be needed:
- ebp to address incoming arguments
- esi (could be any callee-saved register) to address locals
- esp to address outgoing arguments
We would use esi unconditionally without verifying that it did not
conflict with inline assembly.
This change doesn't do the verification, it simply emits a fatal error
on functions that use stack realignment, dynamic SP adjustments, and
inline assembly.
Because stack realignment is common on Windows, we also no longer assume
that MS inline assembly clobbers esp. Instead, we analyze the inline
instructions for implicit definitions and check if esp is there. If so,
we require the use of a base pointer and consider it in the condition
above.
Mostly fixes PR16830, but we could try harder to find a non-conflicting
base pointer.
Reviewers: sunfish
Differential Revision: http://llvm-reviews.chandlerc.com/D1317
llvm-svn: 196876
This just extends the existing hack. It should be enough to get a reproducible bootstrap
on 32 bits.
I will open a bug to track getting a real fix for this.
llvm-svn: 196462
A Direct stack map location records the address of frame index. This
address is itself the value that the runtime requested. This differs
from IndirectMemRefOp locations, which refer to a stack locations from
which the requested values must be loaded. Direct locations can
directly communicate the address if an alloca, while IndirectMemRefOp
handle register spills.
For example:
entry:
%a = alloca i64...
llvm.experimental.stackmap(i32 <ID>, i32 <shadowBytes>, i64* %a)
Since both the alloca and stackmap intrinsic are in the entry block,
and the intrinsic takes the address of the alloca, the runtime can
assume that LLVM will not substitute alloca with any intervening
value. This must be verified by the runtime by checking that the stack
map's location is a Direct location type. The runtime can then
determine the alloca's relative location on the stack immediately after
compilation, or at any time thereafter. This differs from Register and
Indirect locations, because the runtime can only read the values in
those locations when execution reaches the instruction address of the
stack map.
llvm-svn: 195712
Summary:
Moved the requirement for SelectionDAG::getConstant() to return legally
typed nodes slightly earlier. There were two optional DAGCombine passes
that were missed out and were required to produce type-legal DAGs.
Simplified a code-path in tryFoldToZero() to use SelectionDAG::getConstant().
This provides support for both promoted and expanded vector types whereas the
previous code only supported promoted vector types.
Fixes a "Type for zero vector elements is not legal" assertion detected by
an llvm-stress generated test.
Reviewers: resistor
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2251
llvm-svn: 195635
Improvements over r195317:
- Set/restore EnableFastISel flag instead of just running FastISel within
SelectAllBasicBlocks; the flag is checked in various places, and
FastISel won't run properly if those places don't do the right thing.
- Test looks for normal ISel versus FastISel behavior, and not
something more subtle that doesn't work everywhere.
Based on work by Andrea Di Biagio.
llvm-svn: 195491
The legalizer can now do this type of expansion for more
type combinations without loading and storing to and
from the stack.
NOTE: This is a candidate for the 3.4 branch.
llvm-svn: 195398
This patch is a rewrite of the original patch commited in r194542. Instead of
relying on the type legalizer to do the splitting for us, we now peform the
splitting ourselves in the DAG combiner. This is necessary for the case where
the vector mask is a legal type after promotion and still wouldn't require
splitting.
Patch by: Juergen Ributzka
NOTE: This is a candidate for the 3.4 branch.
llvm-svn: 195397
Summary:
LegalizeSetCCCondCode can now legalize SETEQ and SETNE by returning the inverse
condition and requesting that the caller invert the result of the condition.
The caller of LegalizeSetCCCondCode must handle the inverted CC, and they do
so as follows:
SETCC, BR_CC:
Invert the result of the SETCC with SelectionDAG::getNOT()
SELECT_CC:
Swap the true/false operands.
This is necessary for MSA which lacks an integer SETNE instruction.
Reviewers: resistor
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2229
llvm-svn: 195355
It broke, at least, i686 target. It is reproducible with "llc -mtriple=i686-unknown".
FYI, it didn't appear to add either "-O0" or "-fast-isel".
llvm-svn: 195339
Stop folding constant adds into GEP when the type size doesn't match.
Otherwise, the adds' operands are effectively being promoted, changing the
conditions of an overflow. Results are different when:
sext(a) + sext(b) != sext(a + b)
Problem originally found on x86-64, but also fixed issues with ARM and PPC,
which used similar code.
<rdar://problem/15292280>
Patch by Duncan Exon Smith!
llvm-svn: 194840
Summary:
When getConstant() is called for an expanded vector type, it is split into
multiple scalar constants which are then combined using appropriate build_vector
and bitcast operations.
In addition to the usual big/little endian differences, the case where the
element-order of the vector does not have the same endianness as the elements
themselves is also accounted for. For example, for v4i32 on big-endian MIPS,
the byte-order of the vector is <3210,7654,BA98,FEDC>. For little-endian, it is
<0123,4567,89AB,CDEF>.
Handling this case turns out to be a nop since getConstant() returns a splatted
vector (so reversing the element order doesn't change the value)
This fixes a number of cases in MIPS MSA where calling getConstant() during
operation legalization introduces illegal types (e.g. to legalize v2i64 UNDEF
into a v2i64 BUILD_VECTOR of illegal i64 zeros). It should also handle bigger
differences between illegal and legal types such as legalizing v2i64 into v8i16.
lowerMSASplatImm() in the MIPS backend no longer needs to avoid calling
getConstant() so this function has been updated in the same patch.
For the sake of transparency, the steps I've taken since the review are:
* Added 'virtual' to isVectorEltOrderLittleEndian() as requested. This revealed
that the MIPS tests were falsely passing because a polymorphic function was
not actually polymorphic in the reviewed patch.
* Fixed the tests that were now failing. This involved deleting the code to
handle the MIPS MSA element-order (which was previously doing an byte-order
swap instead of an element-order swap). This left
isVectorEltOrderLittleEndian() unused and it was deleted.
* Fixed build failures caused by rebasing beyond r194467-r194472. These build
failures involved the bset, bneg, and bclr instructions added in these commits
using lowerMSASplatImm() in a way that was no longer valid after this patch.
Some of these were fixed by calling SelectionDAG::getConstant() instead,
others were fixed by a new function getBuildVectorSplat() that provided the
removed functionality of lowerMSASplatImm() in a more sensible way.
Reviewers: bkramer
Reviewed By: bkramer
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D1973
llvm-svn: 194811
This is to avoid this transformation in some cases:
fold (conv (load x)) -> (load (conv*)x)
On architectures that don't natively support some vector
loads efficiently casting the load to a smaller vector of
larger types and loading is more efficient.
Patch by Micah Villmow.
llvm-svn: 194783
If a null call target is provided, don't emit a dummy call. This
allows the runtime to reserve as little nop space as it needs without
the requirement of emitting a call.
llvm-svn: 194676
This patch reapplies r193676 with an additional fix for the Hexagon backend. The
SystemZ backend has already been fixed by r194148.
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask type for the given target. Now the type
legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
Reviewed by Nadav
llvm-svn: 194542
This patch moves the jump address materialization inside the noop slide. This
enables patching of the materialization itself or its complete removal. This
patch also adds the ability to define scratch registers that can be used safely
by the code called from the patchpoint intrinsic. At least one scratch register
is required, because that one is used for the materialization of the jump
address. This patch depends on D2009.
Differential Revision: http://llvm-reviews.chandlerc.com/D2074
Reviewed by Andy
llvm-svn: 194306
The idea of the AnyReg Calling Convention is to provide the call arguments in
registers, but not to force them to be placed in a paticular order into a
specified set of registers. Instead it is up tp the register allocator to assign
any register as it sees fit. The same applies to the return value (if
applicable).
Differential Revision: http://llvm-reviews.chandlerc.com/D2009
Reviewed by Andy
llvm-svn: 194293
MorphNodeTo is not safe to call during DAG building. It eagerly
deletes dependent DAG nodes which invalidates the NodeMap. We could
expose a safe interface for morphing nodes, but I don't think it's
worth it. Just create a new MachineNode and replaceAllUsesWith.
My understaning of the SD design has been that we want to support
early target opcode selection. That isn't very well supported, but
generally works. It seems reasonable to rely on this feature even if
it isn't widely used.
llvm-svn: 194102
When an extend more than doubles the size of the elements (e.g., a zext
from v16i8 to v16i32), the normal legalization method of splitting the
vectors will run into problems as by the time the destination vector is
legal, the source vector is illegal. The end result is the operation
often becoming scalarized, with the typical horrible performance. For
example, on x86_64, the simple input of:
define void @bar(<16 x i8> %a, <16 x i32>* %p) nounwind {
%tmp = zext <16 x i8> %a to <16 x i32>
store <16 x i32> %tmp, <16 x i32>*%p
ret void
}
Generates:
.section __TEXT,__text,regular,pure_instructions
.section __TEXT,__const
.align 5
LCPI0_0:
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.long 255 ## 0xff
.section __TEXT,__text,regular,pure_instructions
.globl _bar
.align 4, 0x90
_bar:
vpunpckhbw %xmm0, %xmm0, %xmm1
vpunpckhwd %xmm0, %xmm1, %xmm2
vpmovzxwd %xmm1, %xmm1
vinsertf128 $1, %xmm2, %ymm1, %ymm1
vmovaps LCPI0_0(%rip), %ymm2
vandps %ymm2, %ymm1, %ymm1
vpmovzxbw %xmm0, %xmm3
vpunpckhwd %xmm0, %xmm3, %xmm3
vpmovzxbd %xmm0, %xmm0
vinsertf128 $1, %xmm3, %ymm0, %ymm0
vandps %ymm2, %ymm0, %ymm0
vmovaps %ymm0, (%rdi)
vmovaps %ymm1, 32(%rdi)
vzeroupper
ret
So instead we can check if there are legal types that enable us to split
more cleverly when the input vector is already legal such that we don't
turn it into an illegal type. If the extend is such that it's more than
doubling the size of the input we check if
- the number of vector elements is even,
- the source type is legal,
- the type of a split source is illegal,
- the type of an extended (by doubling element size) source is legal, and
- the type of that extended source when split is legal.
If the conditions are met, instead of just splitting both the
destination and the source types, we create an extend that only goes up
one "step" (doubling the element width), and the continue legalizing the
rest of the operation normally. The result is that this operates as a
new, more effecient, termination condition for the loop of "split the
operation until the destination type is legal."
With this change, the above example now compiles to:
_bar:
vpxor %xmm1, %xmm1, %xmm1
vpunpcklbw %xmm1, %xmm0, %xmm2
vpunpckhwd %xmm1, %xmm2, %xmm3
vpunpcklwd %xmm1, %xmm2, %xmm2
vinsertf128 $1, %xmm3, %ymm2, %ymm2
vpunpckhbw %xmm1, %xmm0, %xmm0
vpunpckhwd %xmm1, %xmm0, %xmm3
vpunpcklwd %xmm1, %xmm0, %xmm0
vinsertf128 $1, %xmm3, %ymm0, %ymm0
vmovaps %ymm0, 32(%rdi)
vmovaps %ymm2, (%rdi)
vzeroupper
ret
This generalizes a custom lowering that was added a while back to the
ARM backend. That lowering is no longer necessary, and is removed. The
testcases for it, however, provide excellent ARM tests for this change
and so remain.
rdar://14735100
llvm-svn: 193727
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask type for the given target. This mask has
usually the same size as the VSELECT return type (except for Intel KNL). Now the
type legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
Reviewed by Nadav
llvm-svn: 193676
Making useAA() default to true for SystemZ showed that the combiner alias
analysis wasn't handling volatile accesses. This hit many of the SystemZ
tests, but I arbitrarily picked one for the purpose of this patch.
llvm-svn: 193518
Most SelectionDAG code drops the TBAA info when creating a new form of a
load and store (e.g. during legalization, or when converting a plain
load to an extending one). This patch tries to catch all cases where
the TBAA information can legitimately be carried over.
The patch adds alternative forms of getLoad() and getExtLoad() that take
a MachineMemOperand instead of individual fields. (The corresponding
getTruncStore() already exists.) The idea is to use the MachineMemOperand
forms when all fields are carried over (size, pointer info, isVolatile,
isNonTemporal, alignment and TBAA info). If some adjustment is being
made, e.g. to narrow the load, then we still pass the individual fields
but also pass the TBAA info.
llvm-svn: 193517
ARM processors without ldrex/strex need to be able to make libcalls for all
atomic operations, including the newer min/max versions.
The alternative would probably be expanding these operations in terms of
cmpxchg (as x86 does always), but in the configurations where this matters
code-size tends to be paramount so the libcall is more desirable.
llvm-svn: 193398
This optimization is not SSE specific so I am moving it to DAGco.
The new scalar_to_vector dag node exposed a missing pattern in the AArch64 target that I needed to add.
llvm-svn: 193393
For some targets, it is useful to be able to look at the original
type of an argument without having to dig through the original IR.
This also fixes a bug in SelectionDAGBuilder where InputArg.PartOffset
was not taking into account the offset of structure elements.
Patch by: Justin Holewinski
Tom Stellard:
- Changed the type of ArgVT to EVT, so it can store non-simple types
like v3i32.
llvm-svn: 193214
VTList has a long life cycle through the module and getVTList is frequently called. In current getVTList, sequential search over a std::vector is used, this is inefficient in big module.
This patch use FoldingSet to implement hashing mechanism when searching.
Reviewer: Nadav Rotem
Test : Pass unit tests & LNT test suite
llvm-svn: 193150
PR17168 describes a test case that fails when compiling for debug with
fast-isel. Investigation showed that the test was failing because a DBG_VALUE
machine instruction was placed prior to a PHI.
For this problem to occur requires the following:
* Compile for debug
* Compile with fast-isel
* In a block B, fast-isel must partially succeed before punting to DAG-isel
* B must start with a PHI
* The first unhandled node in the DAG must not generate a machine instruction
* A debug value with an order less than that of that first node exists
When all of these circumstances apply, the existing test that an instruction
was not inserted won't fire. Currently it tests whether the block is empty,
or whether the last instruction generated is a phi. When fast-isel has
partially succeeded, the last instruction generated will not be a phi.
Instead, we need to check whether the current insert position is immediately
following a phi. This patch adds that check, and adds the test case from the
PR as a regression test.
llvm-svn: 192976
There are targets that support i128 sized scalars but cannot emit
instructions that modify them directly. The proper thing to do is to
emit a libcall.
This fixes PR17481.
llvm-svn: 192957
When canonicalizing dags according to the rule
(shl (zext (shr X, c1) ), c1) ==> (zext (shl (shr X, c1), c1))
remember to add the new shl dag to the DAGCombiner worklist of nodes.
If we don't explicitly add it to the worklist of nodes to visit, we
may not trigger later on the rule that folds the shift left + logical
shift right into a AND instruction with bitmask.
llvm-svn: 192883
This happens e.g. with <2 x i64> -1 on x86_32. It cannot be generated directly
because i64 is illegal. It would be nice if getNOT would handle this
transparently, but I don't see a way to generate a legal constant there right
now. Fixes PR17487.
llvm-svn: 192795
This is really an extension of the current (shl (shr ...)) -> shl optimization.
The main difference is that certain upper bits must also not be demanded.
The motivating examples are the first two in the testcase, which occur
in llvmpipe output.
llvm-svn: 192783
This should fix the buildbots.
Original commit message:
[DAGCombiner] Slice a big load in two loads when the element are next to each
other in memory and the target has paired load and performs post-isel loads
combining.
E.g., this optimization will transform something like this:
a = load i64* addr
b = trunc i64 a to i32
c = lshr i64 a, 32
d = trunc i64 c to i32
into:
b = load i32* addr1
d = load i32* addr2
Where addr1 = addr2 +/- sizeof(i32), if the target supports paired load and
performs post-isel loads combining.
One should overload TargetLowering::hasPairedLoad to provide this information.
The default is false.
<rdar://problem/14477220>
llvm-svn: 192476
other in memory and the target has paired load and performs post-isel loads
combining.
E.g., this optimization will transform something like this:
a = load i64* addr
b = trunc i64 a to i32
c = lshr i64 a, 32
d = trunc i64 c to i32
into:
b = load i32* addr1
d = load i32* addr2
Where addr1 = addr2 +/- sizeof(i32), if the target supports paired load and
performs post-isel loads combining.
One should overload TargetLowering::hasPairedLoad to provide this information.
The default is false.
<rdar://problem/14477220>
llvm-svn: 192471
DAGCombiner::visitFP_EXTEND will apply the following transformation:
fold (fpext (load x)) -> (fpext (fptrunc (extload x)))
but the implementation does not handle indexed loads (pre/post inc.), but did
not specifically ignore them either (unlike for extending loads, which it
already ignored), causing an assert when the transformation was applied to an
indexed load. This is the minimal fix for correctness (causing the
transformation to be skipped for indexed loads).
Unfortunately, I don't have an in-tree test case.
llvm-svn: 191989
SelectionDAG will now attempt to inverse an illegal conditon in order to
find a legal one and if that doesn't work, it will attempt to swap the
operands using the inverted condition.
There are no new test cases for this, but a nubmer of the existing R600
tests hit this path.
llvm-svn: 191602
This is useful for targets like R600, which only support GT, GE, NE, and EQ
condition codes as it removes the need to handle unsupported condition
codes in target specific code.
There are no tests with this commit, but R600 has been updated to take
advantage of this new feature, so its existing selectcc tests are now
testing the swapped operands path.
llvm-svn: 191601
Interpreting the results of this function is not very intuitive, so I
cleaned it up to make it more clear whether or not a SETCC op was
legalized and how it was legalized (either by swapping LHS and RHS or
replacing with AND/OR).
This patch does change functionality in the LHS and RHS swapping case,
but unfortunately there are no in-tree tests for this. However, this
patch is a prerequisite for R600 to take advantage of the LHS and RHS
swapping, so tests will be added in subsequent commits.
llvm-svn: 191600
This change fixes the problem reported in pr17380 and re-add the dagcombine
transformation ensuring that the value types are always legal if the
transformation is triggered after Legalization took place.
Added the test case from pr17380.
llvm-svn: 191509
(shl (zext (shr A, X)), X) => (zext (shl (shr A, X), X)).
The rule only triggers when there are no other uses of the
zext to avoid materializing more instructions.
This helps the DAGCombiner understand that the shl/shr
sequence can then be converted into an and instruction.
llvm-svn: 191393
Patch by Ana Pazos.
1.Added support for v1ix and v1fx types.
2.Added Scalar Pairwise Reduce instructions.
3.Added initial implementation of Scalar Arithmetic instructions.
llvm-svn: 191263
Sometimes a copy from a vreg -> vreg sneaks into the middle of a terminator
sequence. It is safe to slice this into the stack protector success bb.
This fixes PR16979.
llvm-svn: 191260
Previously, the DAGISel function WalkChainUsers was spotting that it
had entered already-selected territory by whether a node was a
MachineNode (amongst other things). Since it's fairly common practice
to insert MachineNodes during ISelLowering, this was not the correct
check.
Looking around, it seems that other nodes get their NodeId set to -1
upon selection, so this makes sure the same thing happens to all
MachineNodes and uses that characteristic to determine whether we
should stop looking for a loop during selection.
This should fix PR15840.
llvm-svn: 191165
The Type Legalizer recognizes that VSELECT needs to be split, because the type
is to wide for the given target. The same does not always apply to SETCC,
because less space is required to encode the result of a comparison. As a result
VSELECT is split and SETCC is unrolled into scalar comparisons.
This commit fixes the issue by checking for VSELECT-SETCC patterns in the DAG
Combiner. If a matching pattern is found, then the result mask of SETCC is
promoted to the expected vector mask for the given target. This mask has usually
te same size as the VSELECT return type (except for Intel KNL). Now the type
legalizer will split both VSELECT and SETCC.
This allows the following X86 DAG Combine code to sucessfully detect the MIN/MAX
pattern. This fixes PR16695, PR17002, and <rdar://problem/14594431>.
llvm-svn: 191130
C-like languages promote types like unsigned short to unsigned int before
performing an arithmetic operation. Currently the rotate matcher in the
DAGCombiner does not consider this situation.
This commit extends the DAGCombiner in the way that the pattern
(or (shl ([az]ext x), (*ext y)), (srl ([az]ext x), (*ext (sub 32, y))))
is folded into
([az]ext (rotl x, y))
The matching is restricted to aext and zext because in this cases the upper
bits are either undefined or known. Test case is included.
This fixes PR16726.
llvm-svn: 191049
C-like languages promote types like unsigned short to unsigned int before
performing an arithmetic operation. Currently the rotate matcher in the
DAGCombiner does not consider this situation.
This commit extends the DAGCombiner in the way that the pattern
(or (shl ([az]ext x), (*ext y)), (srl ([az]ext x), (*ext (sub 32, y))))
is folded into
([az]ext (rotl x, y))
The matching is restricted to aext and zext because in this cases the upper
bits are either undefined or known. Test case is included.
This fixes PR16726.
llvm-svn: 191045
Use the DIVariable::isIndirect() flag set by the frontend instead of
guessing whether to set the machine location's indirection bit.
Paired commit with CFE.
llvm-svn: 190961
When a truncate node defines a legal vector type but uses an illegal
vector type, the legalization process was splitting the vector until
<1 x vector> type, but then it was failing to scalarize the node because
it did not know how to handle TRUNCATE.
<rdar://problem/14989896>
llvm-svn: 190830
DAGCombiner::isAlias can be called with SrcValue1 or SrcValue2 null, and we
can't use AA in this case (if we try, then the casting code in AA will assert).
llvm-svn: 190763
The vselect mask isn't a setcc.
This breaks in the case when the result of getSetCCResultType
is larger than the vector operands
e.g. %tmp = select i1 %cmp <2 x i8> %a, <2 x i8> %b
when getSetCCResultType returns <2 x i32>, the assertion
that the (MaskTy.getSizeInBits() == Op1.getValueType().getSizeInBits())
is hit.
No test since I don't think I can hit this with any of the current
targets. The R600/SI implementation would break, since it returns a
vector of i1 for this, but it doesn't reach ExpandSELECT for other
reasons.
llvm-svn: 190376
The work on this project was left in an unfinished and inconsistent state.
Hopefully someone will eventually get a chance to implement this feature, but
in the meantime, it is better to put things back the way the were. I have
left support in the bitcode reader to handle the case-range bitcode format,
so that we do not lose bitcode compatibility with the llvm 3.3 release.
This reverts the following commits: 155464, 156374, 156377, 156613, 156704,
156757, 156804 156808, 156985, 157046, 157112, 157183, 157315, 157384, 157575,
157576, 157586, 157612, 157810, 157814, 157815, 157880, 157881, 157882, 157884,
157887, 157901, 158979, 157987, 157989, 158986, 158997, 159076, 159101, 159100,
159200, 159201, 159207, 159527, 159532, 159540, 159583, 159618, 159658, 159659,
159660, 159661, 159703, 159704, 160076, 167356, 172025, 186736
llvm-svn: 190328
Occasionally DAGCombiner can spot that a SETCC operation is completely
redundant and reduce it to "all true" or "all false". If this happens to a
vector, the value produced has to take account of what a normal comparison
would have produced, which may be an all-1s bitmask.
The fix in SelectionDAG.cpp is tested, however, as far as I can see the code in
TargetLowering.cpp is possibly unreachable and almost certainly irrelevant when
triggered so there are no tests. However, I believe it's still clearly the
right change and may save someone else some hassle if it suddenly becomes
reachable. So I'm doing it anyway.
llvm-svn: 190147
This uses the TargetSubtargetInfo::useAA() function to control the defaults of
the -combiner-alias-analysis and -combiner-global-alias-analysis options.
llvm-svn: 189564
We want to convert code like (or (srl N, 8), (shl N, 8)) into (srl (bswap N),
const), but this is only valid if the bits above 16 on the source pattern are
0, the checks we were doing on this were slightly wrong before.
llvm-svn: 189348
If we have a binary operation like ISD:ADD, we can set the result type
equal to the result type of one of its operands rather than using
TargetLowering::getPointerTy().
Also, any use of DAG.getIntPtrConstant(C) as an operand for a binary
operation can be replaced with:
DAG.getConstant(C, OtherOperand.getValueType());
llvm-svn: 189227
This adds minimal support to the SelectionDAG for handling address spaces
with different pointer sizes. The SelectionDAG should now correctly
lower pointer function arguments to the correct size as well as generate
the correct code when lowering getelementptr.
This patch also updates the R600 DataLayout to use 32-bit pointers for
the local address space.
v2:
- Add more helper functions to TargetLoweringBase
- Use CHECK-LABEL for tests
llvm-svn: 189221
When truncated vector stores were being custom lowered in
VectorLegalizer::LegalizeOp(), the old (illegal) and new (legal) node pair
was not being added to LegalizedNodes list. Instead of the legalized
result being passed to VectorLegalizer::TranslateLegalizeResult(),
the result was being passed back into VectorLegalizer::LegalizeOp(),
which ended up adding a (new, new) pair to the list instead.
This was causing an assertion failure when a custom lowered truncated
vector store was the last instruction a basic block and the VectorLegalizer
was unable to find it in the LegalizedNodes list when updating the
DAG root.
llvm-svn: 188953
The small utility function that pattern matches Base + Index +
Offset patterns for loads and stores fails to recognize the base
pointer for loads/stores from/into an array at offset 0 inside a
loop. As a result DAGCombiner::MergeConsecutiveStores was not able
to merge all stores.
This commit fixes the issue by adding an additional pattern match
and also a test case.
Reviewer: Nadav
llvm-svn: 188936
SystemZTargetLowering::emitStringWrapper() previously loaded the character
into R0 before the loop and made R0 live on entry. I'd forgotten that
allocatable registers weren't allowed to be live across blocks at this stage,
and it confused LiveVariables enough to cause a miscompilation of f3 in
memchr-02.ll.
This patch instead loads R0 in the loop and leaves LICM to hoist it
after RA. This is actually what I'd tried originally, but I went for
the manual optimisation after noticing that R0 often wasn't being hoisted.
This bug forced me to go back and look at why, now fixed as r188774.
We should also try to optimize null checks so that they test the CC result
of the SRST directly. The select between null and the SRST GPR result could
then usually be deleted as dead.
llvm-svn: 188779
Previously, generation of stack protectors was done exclusively in the
pre-SelectionDAG Codegen LLVM IR Pass "Stack Protector". This necessitated
splitting basic blocks at the IR level to create the success/failure basic
blocks in the tail of the basic block in question. As a result of this,
calls that would have qualified for the sibling call optimization were no
longer eligible for optimization since said calls were no longer right in
the "tail position" (i.e. the immediate predecessor of a ReturnInst
instruction).
Then it was noticed that since the sibling call optimization causes the
callee to reuse the caller's stack, if we could delay the generation of
the stack protector check until later in CodeGen after the sibling call
decision was made, we get both the tail call optimization and the stack
protector check!
A few goals in solving this problem were:
1. Preserve the architecture independence of stack protector generation.
2. Preserve the normal IR level stack protector check for platforms like
OpenBSD for which we support platform specific stack protector
generation.
The main problem that guided the present solution is that one can not
solve this problem in an architecture independent manner at the IR level
only. This is because:
1. The decision on whether or not to perform a sibling call on certain
platforms (for instance i386) requires lower level information
related to available registers that can not be known at the IR level.
2. Even if the previous point were not true, the decision on whether to
perform a tail call is done in LowerCallTo in SelectionDAG which
occurs after the Stack Protector Pass. As a result, one would need to
put the relevant callinst into the stack protector check success
basic block (where the return inst is placed) and then move it back
later at SelectionDAG/MI time before the stack protector check if the
tail call optimization failed. The MI level option was nixed
immediately since it would require platform specific pattern
matching. The SelectionDAG level option was nixed because
SelectionDAG only processes one IR level basic block at a time
implying one could not create a DAG Combine to move the callinst.
To get around this problem a few things were realized:
1. While one can not handle multiple IR level basic blocks at the
SelectionDAG Level, one can generate multiple machine basic blocks
for one IR level basic block. This is how we handle bit tests and
switches.
2. At the MI level, tail calls are represented via a special return
MIInst called "tcreturn". Thus if we know the basic block in which we
wish to insert the stack protector check, we get the correct behavior
by always inserting the stack protector check right before the return
statement. This is a "magical transformation" since no matter where
the stack protector check intrinsic is, we always insert the stack
protector check code at the end of the BB.
Given the aforementioned constraints, the following solution was devised:
1. On platforms that do not support SelectionDAG stack protector check
generation, allow for the normal IR level stack protector check
generation to continue.
2. On platforms that do support SelectionDAG stack protector check
generation:
a. Use the IR level stack protector pass to decide if a stack
protector is required/which BB we insert the stack protector check
in by reusing the logic already therein. If we wish to generate a
stack protector check in a basic block, we place a special IR
intrinsic called llvm.stackprotectorcheck right before the BB's
returninst or if there is a callinst that could potentially be
sibling call optimized, before the call inst.
b. Then when a BB with said intrinsic is processed, we codegen the BB
normally via SelectBasicBlock. In said process, when we visit the
stack protector check, we do not actually emit anything into the
BB. Instead, we just initialize the stack protector descriptor
class (which involves stashing information/creating the success
mbbb and the failure mbb if we have not created one for this
function yet) and export the guard variable that we are going to
compare.
c. After we finish selecting the basic block, in FinishBasicBlock if
the StackProtectorDescriptor attached to the SelectionDAGBuilder is
initialized, we first find a splice point in the parent basic block
before the terminator and then splice the terminator of said basic
block into the success basic block. Then we code-gen a new tail for
the parent basic block consisting of the two loads, the comparison,
and finally two branches to the success/failure basic blocks. We
conclude by code-gening the failure basic block if we have not
code-gened it already (all stack protector checks we generate in
the same function, use the same failure basic block).
llvm-svn: 188755
This adds a llvm.copysign intrinsic; We already have Libfunc recognition for
copysign (which is turned into the FCOPYSIGN SDAG node). In order to
autovectorize calls to copysign in the loop vectorizer, we need a corresponding
intrinsic as well.
In addition to the expected changes to the language reference, the loop
vectorizer, BasicTTI, and the SDAG builder (the intrinsic is transformed into
an FCOPYSIGN node, just like the function call), this also adds FCOPYSIGN to a
few lists in LegalizeVector{Ops,Types} so that vector copysigns can be
expanded.
In TargetLoweringBase::initActions, I've made the default action for FCOPYSIGN
be Expand for vector types. This seems correct for all in-tree targets, and I
think is the right thing to do because, previously, there was no way to generate
vector-values FCOPYSIGN nodes (and most targets don't specify an action for
vector-typed FCOPYSIGN).
llvm-svn: 188728
- split WidenVecRes_Binary into WidenVecRes_Binary and WidenVecRes_BinaryCanTrap
- WidenVecRes_BinaryCanTrap preserves the original behaviour for operations
that can trap
- WidenVecRes_Binary simply widens the operation and improves codegen for
3-element vectors by allowing widening and promotion on x86 (matches the
behaviour of unary and ternary operation widening)
- use WidenVecRes_Binary for operations on integers.
Reviewed by: nrotem
llvm-svn: 188699
We had previously been asserting when faced with a FCOPYSIGN f64, ppcf128 node
because there was no way to expand the FCOPYSIGN node. Because ppcf128 is the
sum of two doubles, and the first double must have the larger magnitude, we
can take the sign from the first double. As a result, in addition to fixing the
crash, this is also an optimization.
llvm-svn: 188655
Teach the generic instruction selection helper functions to constrain
the register classes of their input operands. For non-physical register
references, the generic code needs to be careful not to mess that up
when replacing references to result registers. As the comment indicates
for MachineRegisterInfo::replaceRegWith(), it's important to call
constrainRegClass() first.
rdar://12594152
llvm-svn: 188593
Generalize r188163 to cope with return types other than MVT::i32, just
as the existing visitMemCmpCall code did. I've split this out into a
subroutine so that it can be used for other upcoming patches.
I also noticed that I'd used the wrong API to record the out chain.
It's a load that uses DAG.getRoot() rather than getRoot(), so the out
chain should go on PendingLoads. I don't have a testcase for that because
we don't do any interesting scheduling on z yet.
llvm-svn: 188540
A common idiom is to use zero and all-ones as sentinal values and to
check for both in a single conditional ("x != 0 && x != (unsigned)-1").
That generates code, for i32, like:
testl %edi, %edi
setne %al
cmpl $-1, %edi
setne %cl
andb %al, %cl
With this transform, we generate the simpler:
incl %edi
cmpl $1, %edi
seta %al
Similar improvements for other integer sizes and on other platforms. In
general, combining the two setcc instructions into one is better.
rdar://14689217
llvm-svn: 188315
LowerCallTo returns a pair with the return value of the call as the first
element and the chain associated with the return value as the second element. If
we lower a call that has a void return value, LowerCallTo returns an SDValue
with a NULL SDNode and the chain for the call. Thus makeLibCall by just
returning the first value makes it impossible for you to set up the chain so
that the call is not eliminated as dead code.
I also updated all references to makeLibCall to reflect the new return type.
llvm-svn: 188300
Previously the asserts were only checking that RHS and LHS were the same type and had the same element type as the result. All downstream code for ISD::VECTOR_SHUFFLE requires the types to be the same.
Also removed one unnecessary check of matched element counts that was present in the code.
llvm-svn: 188051
All libm floating-point rounding functions, except for round(), had their own
ISD nodes. Recent PowerPC cores have an instruction for round(), and so here I'm
adding ISD::FROUND so that round() can be custom lowered as well.
For the most part, this is straightforward. I've added an intrinsic
and a matching ISD node just like those for nearbyint() and friends. The
SelectionDAG pattern I've named frnd (because ISD::FP_ROUND has already claimed
fround).
This will be used by the PowerPC backend in a follow-up commit.
llvm-svn: 187926
This virtual function can be implemented by targets to specify the type
to use for the index operand of INSERT_VECTOR_ELT, EXTRACT_VECTOR_ELT,
INSERT_SUBVECTOR, EXTRACT_SUBVECTOR. The default implementation returns
the result from TargetLowering::getPointerTy()
The previous code was using TargetLowering::getPointerTy() for vector
indices, because this is guaranteed to be legal on all targets. However,
using TargetLowering::getPointerTy() can be a problem for targets with
pointer sizes that differ across address spaces. On such targets,
when vectors need to be loaded or stored to an address space other than the
default 'zero' address space (which is the address space assumed by
TargetLowering::getPointerTy()), having an index that
is a different size than the pointer can lead to inefficient
pointer calculations, (e.g. 64-bit adds for a 32-bit address space).
There is no intended functionality change with this patch.
llvm-svn: 187748
For a testcase like the following:
typedef unsigned long uint64_t;
typedef struct {
uint64_t lo;
uint64_t hi;
} blob128_t;
void add_128_to_128(const blob128_t *in, blob128_t *res) {
asm ("PAND %1, %0" : "+Q"(*res) : "Q"(*in));
}
where we'll fail to allocate the register for the output constraint,
our matching input constraint will not find a register to match,
and could try to search past the end of the current operands array.
On the idea that we'd like to attempt to keep compilation going
to find more errors in the module, change the error cases when
we're visiting inline asm IR to return immediately and avoid
trying to create a node in the DAG. This leaves us with only
a single error message per inline asm instruction, but allows us
to safely keep going in the general case.
llvm-svn: 187470
This patch prevents the following combine when the input vector is used more
than once.
insert_vector_elt (build_vector elt0, ..., eltN), NewEltIdx, idx
=>
build_vector elt0, ..., NewEltIdx, ..., eltN
The reasons are:
- Building a vector may be expensive, so try to reuse the existing part of a
vector instead of creating a new one (think big vectors).
- elt0 to eltN now have two users instead of one. This may prevent some other
optimizations.
llvm-svn: 187396
Adds unit tests for it too.
Split BasicBlockUtils into an analysis-half and a transforms-half, and put the
analysis bits into a new Analysis/CFG.{h,cpp}. Promote isPotentiallyReachable
into llvm::isPotentiallyReachable and move it into Analysis/CFG.
llvm-svn: 187283
CustomLowerNode was not being called during SplitVectorOperand,
meaning custom legalization could not be used by targets.
This also adds a test case for NVPTX that depends on this custom
legalization.
Differential Revision: http://llvm-reviews.chandlerc.com/D1195
Attempt to fix the buildbots by making the X86 test I just added platform independent
llvm-svn: 187202
This reverts commit 187198. It broke the bots.
The soft float test probably needs a -triple because of name differences.
On the hard float test I am getting a "roundss $1, %xmm0, %xmm0", instead of
"vroundss $1, %xmm0, %xmm0, %xmm0".
llvm-svn: 187201
CustomLowerNode was not being called during SplitVectorOperand,
meaning custom legalization could not be used by targets.
This also adds a test case for NVPTX that depends on this custom
legalization.
Differential Revision: http://llvm-reviews.chandlerc.com/D1195
llvm-svn: 187198
There is a comment at the top of DAGTypeLegalizer::PerformExpensiveChecks
which, in part, says:
// Note that these invariants may not hold momentarily when processing a node:
// the node being processed may be put in a map before being marked Processed.
Unfortunately, this assert would be valid only if the above-mentioned invariant
held unconditionally. This was causing llc to assert when, in fact,
everything was fine.
Thanks to Richard Sandiford for investigating this issue!
Fixes PR16562.
llvm-svn: 186338
Change the informal convention of DBG_VALUE machine instructions so that
we can express a register-indirect address with an offset of 0.
The old convention was that a DBG_VALUE is a register-indirect value if
the offset (operand 1) is nonzero. The new convention is that a DBG_VALUE
is register-indirect if the first operand is a register and the second
operand is an immediate. For plain register values the combination reg,
reg is used. MachineInstrBuilder::BuildMI knows how to build the new
DBG_VALUES.
rdar://problem/13658587
llvm-svn: 185966
Because integer BUILD_VECTOR operands may have a larger type than the result's
vector element type, and all operands must have the same type, when widening a
BUILD_VECTOR node by adding UNDEFs, we cannot use the vector element type, but
rather must use the type of the existing operands.
Another bug found by llvm-stress.
llvm-svn: 185960
in-tree implementations of TargetLoweringBase::isFMAFasterThanMulAndAdd in
order to resolve the following issues with fmuladd (i.e. optional FMA)
intrinsics:
1. On X86(-64) targets, ISD::FMA nodes are formed when lowering fmuladd
intrinsics even if the subtarget does not support FMA instructions, leading
to laughably bad code generation in some situations.
2. On AArch64 targets, ISD::FMA nodes are formed for operations on fp128,
resulting in a call to a software fp128 FMA implementation.
3. On PowerPC targets, FMAs are not generated from fmuladd intrinsics on types
like v2f32, v8f32, v4f64, etc., even though they promote, split, scalarize,
etc. to types that support hardware FMAs.
The function has also been slightly renamed for consistency and to force a
merge/build conflict for any out-of-tree target implementing it. To resolve,
see comments and fixed in-tree examples.
llvm-svn: 185956
When folding sub x, x (and other similar constructs), where x is a vector, the
result is a vector of zeros. After type legalization, make sure that the input
zero elements have a legal type. This type may be larger than the result's
vector element type.
This was another bug found by llvm-stress.
llvm-svn: 185949
This fixes a bug (found by llvm-stress) in
DAGTypeLegalizer::PromoteIntRes_BUILD_VECTOR where it assumed that the result
type would always be larger than the original operands. This is not always
true, however, with boolean vectors. For example, promoting a node of type v8i1
(where the operands will be of type i32, the type to which i1 is promoted) will
yield a node with a result vector element type of i16 (and operands of type
i32). As a result, we cannot blindly assume that we can ANY_EXTEND the operands
to the result type.
llvm-svn: 185794
ReduceLoadWidth unconditionally drops extensions from loads. Limit it to the
case when all of the bits the extension would otherwise produce are dropped by
the shrink. It would be possible to shrink the load in more cases by merging
the extensions, but this isn't trivial and a very rare case. I left a TODO for
that case.
Fixes PR16551.
llvm-svn: 185755
This prevents the emission of DAG-generated vreg definitions after a
tail call be dropping them entirely (on the grounds that nothing could
use them anyway, and they interfere with O0 CodeGen).
llvm-svn: 185754
Stop using the ISD::EXCEPTIONADDR and ISD::EHSELECTION when lowering
landing pad arguments. These nodes were previously legalized into
CopyFromReg nodes, but that never worked properly because the
CopyFromReg node weren't guaranteed to be scheduled at the top of the
basic block.
This meant the exception pointer and selector registers could be
clobbered before being copied to a virtual register.
This patch copies the two physical registers to virtual registers at
the beginning of the basic block, and lowers the landingpad instruction
directly to two CopyFromReg nodes reading the *virtual* registers. This
is safe because virtual registers don't get clobbered.
A future patch will remove the ISD::EXCEPTIONADDR and ISD::EHSELECTION
nodes.
llvm-svn: 185617
Compute the insertion point from the end of the basic block instead of
skipping labels from the front.
This caused failures in landing pads when live-in copies where inserted
before instruction selection.
llvm-svn: 185616
Stop using the ISD::EXCEPTIONADDR and ISD::EHSELECTION when lowering
landing pad arguments. These nodes were previously legalized into
CopyFromReg nodes, but that never worked properly because the
CopyFromReg node weren't guaranteed to be scheduled at the top of the
basic block.
This meant the exception pointer and selector registers could be
clobbered before being copied to a virtual register.
This patch copies the two physical registers to virtual registers at
the beginning of the basic block, and lowers the landingpad instruction
directly to two CopyFromReg nodes reading the *virtual* registers. This
is safe because virtual registers don't get clobbered.
A future patch will remove the ISD::EXCEPTIONADDR and ISD::EHSELECTION
nodes.
llvm-svn: 185595
DAGCombiner was counting all uses of a load node when considering whether it's
worth combining into a zextload. Really, it wants to ignore the chain and just
count real uses.
rdar://problem/13896307
llvm-svn: 185419
should expand ATOMIC_CMP_SWAP nodes the same way that it does for ATOMIC_SWAP.
Since ATOMIC_LOADs on some targets (e.g. older ARM variants) get legalized to
ATOMIC_CMP_SWAPs, the missing case had been causing i64 atomic loads to crash
during isel.
<rdar://problem/14074644>
llvm-svn: 185186
No functionality change.
It should suffice to check the type of a debug info metadata, instead of
calling Verify. For cases where we know the type of a DI metadata, use
assert.
Also update testing cases to make them conform to the format of DI classes.
llvm-svn: 185135
A FastISel optimization was causing us to emit no information for such
parameters & when they go missing we end up emitting a different
function type. By avoiding that shortcut we not only get types correct
(very important) but also location information (handy) - even if it's
only live at the start of a function & may be clobbered later.
Reviewed/discussion by Evan Cheng & Dan Gohman.
llvm-svn: 184604
value is zero.
This allows optmizations to kick in more easily.
Fix some test cases so that they remain meaningful (i.e., not completely dead
coded) when optimizations apply.
<rdar://problem/14096009> superfluous multiply by high part of zero-extended
value.
llvm-svn: 184222
Rather than using the full power of target-specific addressing modes in
DBG_VALUEs with Frame Indicies, simply use Frame Index + Offset. This
reduces the complexity of debug info handling down to two
representations of values (reg+offset and frame index+offset) rather
than three or four.
Ideally we could ensure that frame indicies had been eliminated by the
time we reached an assembly or dwarf generation, but I haven't spent the
time to figure out where the FIs are leaking through into that & whether
there's a good place to convert them. Some FI+offset=>reg+offset
conversion is done (see PrologEpilogInserter, for example) which is
necessary for some SelectionDAG assumptions about registers, I believe,
but it might be possible to make this a more thorough conversion &
ensure there are no remaining FIs no matter how instruction selection
is performed.
llvm-svn: 184066
The TargetLoweringInfo object is owned by the TargetMachine. In the future, the
TargetMachine object may change, which may also change the TargetLoweringInfo
object.
llvm-svn: 183356
Fixes PR16146: gdb.base__call-ar-st.exp fails after
pre-RA-sched=source fixes.
Patch by Xiaoyi Guo!
This also fixes an unsupported dbg.value test case. Codegen was
previously incorrect but the test was passing by luck.
llvm-svn: 182885
When -ffast-math is in effect (on Linux, at least), clang defines
__FINITE_MATH_ONLY__ > 0 when including <math.h>. This causes the
preprocessor to include <bits/math-finite.h>, which renames the sqrt functions.
For instance, "sqrt" is renamed as "__sqrt_finite".
This patch adds the 3 new names in such a way that they will be treated
as equivalent to their respective original names.
llvm-svn: 182739
Use a field in the SelectionDAGNode object to track its IR ordering.
This adds fields and utility classes without changing existing
interfaces or functionality.
llvm-svn: 182701
If the input operands to SETCC are promoted, we need to make sure that we
either use the promoted form of both operands (or neither); a mixture is not
allowed. This can happen, for example, if a target has a custom promoted
i1-returning intrinsic (where i1 is not a legal type). In this case, we need to
use the promoted form of both operands.
This change only augments the behavior of the existing logic in the case where
the input types (which may or may not have already been legalized) disagree,
and should not affect existing target code because this case would otherwise
cause an assert in the SETCC operand promotion code.
This will be covered by (essentially all of the) tests for the new PPCCTRLoops
infrastructure.
llvm-svn: 181926
report a fatal error. This allows us to continue processing the translation
unit. Test case to come on the clang side because we need an inline asm
diagnostics handler in place.
rdar://13446483
llvm-svn: 180873
register-indirect address with an offset of 0.
It used to be that a DBG_VALUE is a register-indirect value if the offset
(operand 1) is nonzero. The new convention is that a DBG_VALUE is
register-indirect if the first operand is a register and the second
operand is an immediate. For plain registers use the combination reg, reg.
rdar://problem/13658587
llvm-svn: 180816
This already helps SSE2 x86 a lot because it lacks an efficient way to
represent a vector select. The long term goal is to enable the backend to match
a canonicalized pattern into a single instruction (e.g. vabs or pabs).
llvm-svn: 180597
This exposed an issue with PowerPC AltiVec where it appears it was setting the wrong vector boolean contents. The included change
fixes the PowerPC tests, and was OK'd by Hal.
llvm-svn: 180129
Rather than just splitting the input type and hoping for the best, apply
a bit more cleverness. Just splitting the types until the source is
legal often leads to an illegal result time, which is then widened and a
scalarization step is introduced which leads to truly horrible code
generation. With the loop vectorizer, these sorts of operations are much
more common, and so it's worth extra effort to do them well.
Add a legalization hook for the operands of a TRUNCATE node, which will
be encountered after the result type has been legalized, but if the
operand type is still illegal. If simple splitting of both types
ends up with the result type of each half still being legal, just
do that (v16i16 -> v16i8 on ARM, for example). If, however, that would
result in an illegal result type (v8i32 -> v8i8 on ARM, for example),
we can get more clever with power-two vectors. Specifically,
split the input type, but also widen the result element size, then
concatenate the halves and truncate again. For example on ARM,
To perform a "%res = v8i8 trunc v8i32 %in" we transform to:
%inlo = v4i32 extract_subvector %in, 0
%inhi = v4i32 extract_subvector %in, 4
%lo16 = v4i16 trunc v4i32 %inlo
%hi16 = v4i16 trunc v4i32 %inhi
%in16 = v8i16 concat_vectors v4i16 %lo16, v4i16 %hi16
%res = v8i8 trunc v8i16 %in16
This allows instruction selection to generate three VMOVN instructions
instead of a sequences of moves, stores and loads.
Update the ARMTargetTransformInfo to take this improved legalization
into account.
Consider the simplified IR:
define <16 x i8> @test1(<16 x i32>* %ap) {
%a = load <16 x i32>* %ap
%tmp = trunc <16 x i32> %a to <16 x i8>
ret <16 x i8> %tmp
}
define <8 x i8> @test2(<8 x i32>* %ap) {
%a = load <8 x i32>* %ap
%tmp = trunc <8 x i32> %a to <8 x i8>
ret <8 x i8> %tmp
}
Previously, we would generate the truly hideous:
.syntax unified
.section __TEXT,__text,regular,pure_instructions
.globl _test1
.align 2
_test1: @ @test1
@ BB#0:
push {r7}
mov r7, sp
sub sp, sp, #20
bic sp, sp, #7
add r1, r0, #48
add r2, r0, #32
vld1.64 {d24, d25}, [r0:128]
vld1.64 {d16, d17}, [r1:128]
vld1.64 {d18, d19}, [r2:128]
add r1, r0, #16
vmovn.i32 d22, q8
vld1.64 {d16, d17}, [r1:128]
vmovn.i32 d20, q9
vmovn.i32 d18, q12
vmov.u16 r0, d22[3]
strb r0, [sp, #15]
vmov.u16 r0, d22[2]
strb r0, [sp, #14]
vmov.u16 r0, d22[1]
strb r0, [sp, #13]
vmov.u16 r0, d22[0]
vmovn.i32 d16, q8
strb r0, [sp, #12]
vmov.u16 r0, d20[3]
strb r0, [sp, #11]
vmov.u16 r0, d20[2]
strb r0, [sp, #10]
vmov.u16 r0, d20[1]
strb r0, [sp, #9]
vmov.u16 r0, d20[0]
strb r0, [sp, #8]
vmov.u16 r0, d18[3]
strb r0, [sp, #3]
vmov.u16 r0, d18[2]
strb r0, [sp, #2]
vmov.u16 r0, d18[1]
strb r0, [sp, #1]
vmov.u16 r0, d18[0]
strb r0, [sp]
vmov.u16 r0, d16[3]
strb r0, [sp, #7]
vmov.u16 r0, d16[2]
strb r0, [sp, #6]
vmov.u16 r0, d16[1]
strb r0, [sp, #5]
vmov.u16 r0, d16[0]
strb r0, [sp, #4]
vldmia sp, {d16, d17}
vmov r0, r1, d16
vmov r2, r3, d17
mov sp, r7
pop {r7}
bx lr
.globl _test2
.align 2
_test2: @ @test2
@ BB#0:
push {r7}
mov r7, sp
sub sp, sp, #12
bic sp, sp, #7
vld1.64 {d16, d17}, [r0:128]
add r0, r0, #16
vld1.64 {d20, d21}, [r0:128]
vmovn.i32 d18, q8
vmov.u16 r0, d18[3]
vmovn.i32 d16, q10
strb r0, [sp, #3]
vmov.u16 r0, d18[2]
strb r0, [sp, #2]
vmov.u16 r0, d18[1]
strb r0, [sp, #1]
vmov.u16 r0, d18[0]
strb r0, [sp]
vmov.u16 r0, d16[3]
strb r0, [sp, #7]
vmov.u16 r0, d16[2]
strb r0, [sp, #6]
vmov.u16 r0, d16[1]
strb r0, [sp, #5]
vmov.u16 r0, d16[0]
strb r0, [sp, #4]
ldm sp, {r0, r1}
mov sp, r7
pop {r7}
bx lr
Now, however, we generate the much more straightforward:
.syntax unified
.section __TEXT,__text,regular,pure_instructions
.globl _test1
.align 2
_test1: @ @test1
@ BB#0:
add r1, r0, #48
add r2, r0, #32
vld1.64 {d20, d21}, [r0:128]
vld1.64 {d16, d17}, [r1:128]
add r1, r0, #16
vld1.64 {d18, d19}, [r2:128]
vld1.64 {d22, d23}, [r1:128]
vmovn.i32 d17, q8
vmovn.i32 d16, q9
vmovn.i32 d18, q10
vmovn.i32 d19, q11
vmovn.i16 d17, q8
vmovn.i16 d16, q9
vmov r0, r1, d16
vmov r2, r3, d17
bx lr
.globl _test2
.align 2
_test2: @ @test2
@ BB#0:
vld1.64 {d16, d17}, [r0:128]
add r0, r0, #16
vld1.64 {d18, d19}, [r0:128]
vmovn.i32 d16, q8
vmovn.i32 d17, q9
vmovn.i16 d16, q8
vmov r0, r1, d16
bx lr
llvm-svn: 179989
This pattern occurs in SROA output due to the way vector arguments are lowered
on ARM.
The testcase from PR15525 now compiles into this, which is better than the code
we got with the old scalarrepl:
_Store:
ldr.w r9, [sp]
vmov d17, r3, r9
vmov d16, r1, r2
vst1.8 {d16, d17}, [r0]
bx lr
Differential Revision: http://llvm-reviews.chandlerc.com/D647
llvm-svn: 179106
For this we need to use a libcall. Previously LLVM didn't implement
libcall support for frem, so I've added it in the usual
straightforward manner. A test case from the bug report is included.
llvm-svn: 178639
This is helps on architectures where i8,i16 are not legal but we have byte, and
short loads/stores. Allowing us to merge copies like the one below on ARM.
copy(char *a, char *b, int n) {
do {
int t0 = a[0];
int t1 = a[1];
b[0] = t0;
b[1] = t1;
radar://13536387
llvm-svn: 178546
We would also like to merge sequences that involve a variable index like in the
example below.
int index = *idx++
int i0 = c[index+0];
int i1 = c[index+1];
b[0] = i0;
b[1] = i1;
By extending the parsing of the base pointer to handle dags that contain a
base, index, and offset we can handle examples like the one above.
The dag for the code above will look something like:
(load (i64 add (i64 copyfromreg %c)
(i64 signextend (i8 load %index))))
(load (i64 add (i64 copyfromreg %c)
(i64 signextend (i32 add (i32 signextend (i8 load %index))
(i32 1)))))
The code that parses the tree ignores the intermediate sign extensions. However,
if there is a sign extension it needs to be on all indexes.
(load (i64 add (i64 copyfromreg %c)
(i64 signextend (add (i8 load %index)
(i8 1))))
vs
(load (i64 add (i64 copyfromreg %c)
(i64 signextend (i32 add (i32 signextend (i8 load %index))
(i32 1)))))
radar://13536387
llvm-svn: 178483
immediate in a register. I don't believe this should ever fail, but I see no
harm in trying to make this code bullet proof.
I've added an assert to ensure my assumtion is correct. If the assertion fires
something is wrong and we should fix it, rather then just silently fall back to
SelectionDAG isel.
llvm-svn: 178305
- Handle the case where the result of 'insert_subvect' is bitcasted
before 'extract_subvec'. This removes the redundant insertf128/extractf128
pair on unaligned 256-bit vector load/store on vectors of non 64-bit integer.
llvm-svn: 177945
For instance, following transformation will be disabled:
x + x + x => 3.0f * x;
The problem of these transformations is that it introduces a FP constant, which
following Instruction-Selection pass cannot handle.
Reviewed by Nadav, thanks a lot!
rdar://13445387
llvm-svn: 177933
Performing this check unilaterally prevented us from generating FMAs when the incoming IR contained illegal vector types which would eventually be legalized to underlying types that *did* support FMA.
For example, an @llvm.fmuladd on an OpenCL float16 should become a sequence of float4 FMAs, not float4 fmul+fadd's.
NOTE: Because we still call the target-specific profitability hook, individual targets can reinstate the old behavior, if desired, by simply performing the legality check inside their callback hook. They can also perform more sophisticated legality checks, if, for example, some illegal vector types can be productively implemented as FMAs, but not others.
llvm-svn: 177820
This reverts commit 06091513c283c863296f01cc7c2e86b56bb50d02.
The code is obviously wrong, but the trivial fix causes
inefficient code generation on X86. Somebody with more
knowledge of the code needs to take a look here.
Signed-off-by: Christian König <christian.koenig@amd.com>
llvm-svn: 177529
A node's ordering is only propagated during legalization if (a) the new node does
not have an ordering (is not a CSE'd node), or (b) the new node has an ordering
that is higher than the node being legalized.
llvm-svn: 177465
This doesn't reset all of the target options within the TargetOptions
object. This is because some of those are ABI-specific and must be determined if
it's okay to change those on the fly.
llvm-svn: 176986
Summary:
Statistics are still available in Release+Asserts (any +Asserts builds),
and stats can also be turned on with LLVM_ENABLE_STATS.
Move some of the FastISel stats that were moved under DEBUG()
back out of DEBUG(), since stats are disabled across the board now.
Many tests depend on grepping "-stats" output. Move those into
a orig_dir/Stats/. so that they can be marked as unsupported
when building without statistics.
Differential Revision: http://llvm-reviews.chandlerc.com/D486
llvm-svn: 176733
LegalizeDAG.cpp uses the value of the comparison operands when checking
the legality of BR_CC, so DAGCombiner should do the same.
v2:
- Expand more BR_CC value types for NVPTX
v3:
- Expand correct BR_CC value types for Hexagon, Mips, and XCore.
llvm-svn: 176694
Code generation makes some basic assumptions about the IR it's been given. In
particular, if there is only one 'invoke' in the function, then that invoke
won't be going away. However, with the advent of the `llvm.donothing' intrinsic,
those invokes may go away. If all of them go away, the landing pad no longer has
any users. This confuses the back-end, which asserts.
This happens with SjLj exceptions, because that's the model that modifies the IR
based on there being invokes, etc. in the function.
Remove any invokes of `llvm.donothing' during SjLj EH preparation. This will
give us a CFG that the back-end won't be confused about. If all of the invokes
in a function are removed, then the SjLj EH prepare pass won't insert the bogus
code the relies upon the invokes being there.
<rdar://problem/13228754&13316637>
llvm-svn: 176677
rdar:13370002 [pre-RA-sched] assertion: released too many times
I tracked this down to an earlier hack that is no longer applicable
and interfered with normal scheduler logic. With the changes in
r176037, it was causing an instruction to be scheduled multiple times.
I have an external test case that I tried hard to reduce and
failed. I can't even reproduce with llc.
llvm-svn: 176636
- ISD::SHL/SRL/SRA must have either both scalar or both vector operands
but TLI.getShiftAmountTy() so far only return scalar type. As a
result, backend logic assuming that breaks.
- Rename the original TLI.getShiftAmountTy() to
TLI.getScalarShiftAmountTy() and re-define TLI.getShiftAmountTy() to
return target-specificed scalar type or the same vector type as the
1st operand.
- Fix most TICG logic assuming TLI.getShiftAmountTy() a simple scalar
type.
llvm-svn: 176364
SelectionDAGIsel::LowerArguments needs a function, not a basic block. So it
makes sense to pass it the function instead of extracting a basic-block from
the function and then tossing it. This is also more self-documenting (functions
have arguments, BBs don't).
In addition, added comments to a couple of Select* methods.
llvm-svn: 176305
fewer scalar integer (i32 or i64) arguments. It completely eliminates the need
for SDISel for trivial functions.
Also, add the new llc -fast-isel-abort-args option, which is similar to
-fast-isel-abort option, but for formal argument lowering.
llvm-svn: 176052
memory intrinsics in the SDAG builder.
When alignment is zero, the lang ref says that *no* alignment
assumptions can be made. This is the exact opposite of the internal API
contracts of the DAG where alignment 0 indicates that the alignment can
be made to be anything desired.
There is another, more explicit alignment that is better suited for the
role of "no alignment at all": an alignment of 1. Map the intrinsic
alignment to this early so that we don't end up generating aligned DAGs.
It is really terrifying that we've never seen this before, but we
suddenly started generating a large number of alignment 0 memcpys due to
the new code to do memcpy-based copying of POD class members. That patch
contains a bug that rounds bitfield alignments down when they are the
first field. This can in turn produce zero alignments.
This fixes weird crashes I've seen in library users of LLVM on 32-bit
hosts, etc.
llvm-svn: 176022
One of the phases of SelectionDAG is LegalizeVectors. We don't need to sort the DAG and copy nodes around if there are no vector ops.
Speeds up the compilation time of SelectionDAG on a big scalar workload by ~8%.
llvm-svn: 175929
It was incorrectly checking a Function* being an IntrinsicInst* which
isn't possible. It should always have been checking the CallInst* instead.
Added test case for x86 which ensures we only get one constant load.
It was 2 before this change.
rdar://problem/13267920
llvm-svn: 175853
A legal BUILD_VECTOR goes in and gets constant folded into another legal
BUILD_VECTOR so we don't lose any legality here. The problematic PPC
optimization that made this check necessary was fixed recently.
llvm-svn: 175759
(2xi32) (truncate ((2xi64) bitcast (buildvector i32 a, i32 x, i32 b, i32 y)))
can be folded into a (2xi32) (buildvector i32 a, i32 b).
Such a DAG would cause uneccessary vdup instructions followed by vmovn
instructions.
We generate this code on ARM NEON for a setcc olt, 2xf64, 2xf64. For example, in
the vectorized version of the code below.
double A[N];
double B[N];
void test_double_compare_to_double() {
int i;
for(i=0;i<N;i++)
A[i] = (double)(A[i] < B[i]);
}
radar://13191881
Fixes bug 15283.
llvm-svn: 175670
- When extloading from a vector with non-byte-addressable element, e.g.
<4 x i1>, the current logic breaks. Extend the current logic to
fix the case where the element type is not byte-addressable by loading
all bytes, bit-extracting/packing each element.
llvm-svn: 175642
If the frame pointer is omitted, and any stack changes occur in the inline
assembly, e.g.: "pusha", then any C local variable or C argument references
will be incorrect.
I pass no judgement on anyone who would do such a thing. ;)
rdar://13218191
llvm-svn: 175334
If two functions require different features (e.g., `-mno-sse' vs. `-msse') then
we want to honor that, especially during LTO. We can do that by resetting the
subtarget's features depending upon the 'target-feature' attribute.
llvm-svn: 175314
- add sincos to runtime library if target triple environment is GNU
- added canCombineSinCosLibcall() which checks that sincos is in the RTL and
if the environment is GNU then unsafe fpmath is enabled (required to
preserve errno)
- extended sincos-opt lit test
Reviewed by: Hal Finkel
llvm-svn: 175283
DAGCombiner::ReduceLoadWidth was converting (trunc i32 (shl i64 v, 32))
into (shl i32 v, 32) into undef. To prevent this, check the shift count
against the final result size.
Patch by: Kevin Schoedel
Reviewed by: Nadav Rotem
llvm-svn: 174972
Sorry for the lack of a test case. I tried writing one for i386 as i know selects are illegal on this target, but they are actually considered legal by isel and expanded later.
I can't see any targets to trigger this, but checking for the legality of a node before forming it is general goodness.
llvm-svn: 174934
function is successfully handled by fast-isel. That's because function
arguments are *always* handled by SDISel. Introduce FastLowerArguments to
allow each target to provide hook to handle formal argument lowering.
As a proof-of-concept, add ARMFastIsel::FastLowerArguments to handle
functions with 4 or fewer scalar integer (i8, i16, or i32) arguments. It
completely eliminates the need for SDISel for trivial functions.
rdar://13163905
llvm-svn: 174855
Previously, even when a pre-increment load or store was generated,
we often needed to keep a copy of the original base register for use
with other offsets. If all of these offsets are constants (including
the offset which was combined into the addressing mode), then this is
clearly unnecessary. This change adjusts these other offsets to use the
new incremented address.
llvm-svn: 174746
Aside from the question of whether we report a warning or an error when we
can't satisfy a requested stack object alignment, the current implementation
of this is not good. We're not providing any source location in the diagnostics
and the current warning is not connected to any warning group so you can't
control it. We could improve the source location somewhat, but we can do a
much better job if this check is implemented in the front-end, so let's do that
instead. <rdar://problem/13127907>
llvm-svn: 174741
base point of a load, and the overall alignment of the load. This caused infinite loops in DAG combine with the
original application of this patch.
ORIGINAL COMMIT LOG:
When the target-independent DAGCombiner inferred a higher alignment for a load,
it would replace the load with one with the higher alignment. However, it did
not place the new load in the worklist, which prevented later DAG combines in
the same phase (for example, target-specific combines) from ever seeing it.
This patch corrects that oversight, and updates some tests whose output changed
due to slightly different DAGCombine outputs.
llvm-svn: 174431
it would replace the load with one with the higher alignment. However, it did
not place the new load in the worklist, which prevented later DAG combines in
the same phase (for example, target-specific combines) from ever seeing it.
This patch corrects that oversight, and updates some tests whose output changed
due to slightly different DAGCombine outputs.
llvm-svn: 174343
This required disabling a PowerPC optimization that did the following:
input:
x = BUILD_VECTOR <i32 16, i32 16, i32 16, i32 16>
lowered to:
tmp = BUILD_VECTOR <i32 8, i32 8, i32 8, i32 8>
x = ADD tmp, tmp
The add now gets folded immediately and we're back at the BUILD_VECTOR we
started from. I don't see a way to fix this currently so I left it disabled
for now.
Fix some trivially foldable X86 tests too.
llvm-svn: 174325
conditions are met:
1. They share the same operand and are in the same BB.
2. Both outputs are used.
3. The target has a native instruction that maps to ISD::FSINCOS node or
the target provides a sincos library call.
Implemented the generic optimization in sdisel and enabled it for
Mac OSX. Also added an additional optimization for x86_64 Mac OSX by
using an alternative entry point __sincos_stret which returns the two
results in xmm0 / xmm1.
rdar://13087969
PR13204
llvm-svn: 173755
Fix that by adding a cast to the shift expander. This came up with vector shifts
on sse-less X86 CPUs.
<2 x i64> = shl <2 x i64> <2 x i64>
-> i64,i64 = shl i64 i64; shl i64 i64
-> i32,i32,i32,i32 = shl_parts i32 i32 i64; shl_parts i32 i32 i64
Now we cast the last two i64s to the right type. Fixes the crash in PR14668.
llvm-svn: 173615
with an initial number of elements, instead of DenseMap, which has
zero initial elements, in order to avoid the copying of elements
when the size changes and to avoid allocating space every time
LegalizeTypes is run. This patch will not affect the memory footprint,
because DenseMap will increase the element size to 64
when the first element is added.
Patch by Wan Xiaofei.
llvm-svn: 173448
Previously we tried to infer it from the bit width size, with an added
IsIEEE argument for the PPC/IEEE 128-bit case, which had a default
value. This default value allowed bugs to creep in, where it was
inappropriate.
llvm-svn: 173138
The optimization handles esoteric cases but adds a lot of complexity both to the X86 backend and to other backends.
This optimization disables an important canonicalization of chains of SEXT nodes and makes SEXT and ZEXT asymmetrical.
Disabling the canonicalization of consecutive SEXT nodes into a single node disables other DAG optimizations that assume
that there is only one SEXT node. The AVX mask optimizations is one example. Additionally this optimization does not update the cost model.
llvm-svn: 172968
The included test case is derived from one of the GCC compatibility tests.
The problem arises after the selection DAG has been converted to type-legalized
form. The combiner first sees a 64-bit load that can be converted into a
pre-increment form. The original load feeds into a SRL that isolates the
upper 32 bits of the loaded doubleword. This looks like an opportunity for
DAGCombiner::ReduceLoadWidth() to replace the 64-bit load with a 32-bit load.
However, this transformation is not valid, as the replacement load is not
a pre-increment load. The pre-increment load produces an extra result,
which feeds a subsequent add instruction. The replacement load only has
one result value, and this value is propagated to all uses of the pre-
increment load, including the add. Because the add is looking for the
second result value as its operand, it ends up attempting to add a constant
to a token chain, resulting in a crash.
So the patch simply disables this transformation for any load with more than
two result values.
llvm-svn: 172480
This fixes some of the cycles between libCodeGen and libSelectionDAG. It's still
a complete mess but as long as the edges consist of virtual call it doesn't
cause breakage. BasicTTI did static calls and thus broke some build
configurations.
llvm-svn: 172246
- recognize string "{memory}" in the MI generation
- mark as mayload/maystore when there's a memory clobber constraint.
PR14859.
Patch by Krzysztof Parzyszek
llvm-svn: 172228
requirement when creating stack objects in MachineFrameInfo.
Add CreateStackObjectWithMinAlign to throw error when the minimal alignment
can't be achieved and to clamp the alignment when the preferred alignment
can't be achieved. Same is true for CreateVariableSizedObject.
Will not emit error in CreateSpillStackObject or CreateStackObject.
As long as callers of CreateStackObject do not assume the object will be
aligned at the requested alignment, we should not have miscompile since
later optimizations which look at the object's alignment will have the correct
information.
rdar://12713765
llvm-svn: 172027
It cahced XOR's operands before calling visitXOR() but failed to update the
operands when visitXOR changed the XOR node.
rdar://12968664
llvm-svn: 171999
fp128 is almost but not quite completely illegal as a type on AArch64. As a
result it needs to have a register class (for argument passing mainly), but all
operations need to be lowered to runtime calls. Currently there's no way for
targets to do this (without duplicating code), as the relevant functions are
hidden in SelectionDAG. This patch changes that.
llvm-svn: 171971
one file where it is called as a static function. Nuke the declaration
and the definition in lib/CodeGen, along with the include of
SelectionDAG.h from this file.
There is no dependency edge from lib/CodeGen to
lib/CodeGen/SelectionDAG, so it isn't valid for a routine in lib/CodeGen
to reference the DAG. There is a dependency from
lib/CodeGen/SelectionDAG on lib/CodeGen. This breaks one violation of
this layering.
llvm-svn: 171842
peculiar headers under include/llvm.
This struct still doesn't make a lot of sense, but it makes more sense
down in TargetLowering than it did before.
llvm-svn: 171739
a TargetMachine to construct (and thus isn't always available), to an
analysis group that supports layered implementations much like
AliasAnalysis does. This is a pretty massive change, with a few parts
that I was unable to easily separate (sorry), so I'll walk through it.
The first step of this conversion was to make TargetTransformInfo an
analysis group, and to sink the nonce implementations in
ScalarTargetTransformInfo and VectorTargetTranformInfo into
a NoTargetTransformInfo pass. This allows other passes to add a hard
requirement on TTI, and assume they will always get at least on
implementation.
The TargetTransformInfo analysis group leverages the delegation chaining
trick that AliasAnalysis uses, where the base class for the analysis
group delegates to the previous analysis *pass*, allowing all but tho
NoFoo analysis passes to only implement the parts of the interfaces they
support. It also introduces a new trick where each pass in the group
retains a pointer to the top-most pass that has been initialized. This
allows passes to implement one API in terms of another API and benefit
when some other pass above them in the stack has more precise results
for the second API.
The second step of this conversion is to create a pass that implements
the TargetTransformInfo analysis using the target-independent
abstractions in the code generator. This replaces the
ScalarTargetTransformImpl and VectorTargetTransformImpl classes in
lib/Target with a single pass in lib/CodeGen called
BasicTargetTransformInfo. This class actually provides most of the TTI
functionality, basing it upon the TargetLowering abstraction and other
information in the target independent code generator.
The third step of the conversion adds support to all TargetMachines to
register custom analysis passes. This allows building those passes with
access to TargetLowering or other target-specific classes, and it also
allows each target to customize the set of analysis passes desired in
the pass manager. The baseline LLVMTargetMachine implements this
interface to add the BasicTTI pass to the pass manager, and all of the
tools that want to support target-aware TTI passes call this routine on
whatever target machine they end up with to add the appropriate passes.
The fourth step of the conversion created target-specific TTI analysis
passes for the X86 and ARM backends. These passes contain the custom
logic that was previously in their extensions of the
ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces.
I separated them into their own file, as now all of the interface bits
are private and they just expose a function to create the pass itself.
Then I extended these target machines to set up a custom set of analysis
passes, first adding BasicTTI as a fallback, and then adding their
customized TTI implementations.
The fourth step required logic that was shared between the target
independent layer and the specific targets to move to a different
interface, as they no longer derive from each other. As a consequence,
a helper functions were added to TargetLowering representing the common
logic needed both in the target implementation and the codegen
implementation of the TTI pass. While technically this is the only
change that could have been committed separately, it would have been
a nightmare to extract.
The final step of the conversion was just to delete all the old
boilerplate. This got rid of the ScalarTargetTransformInfo and
VectorTargetTransformInfo classes, all of the support in all of the
targets for producing instances of them, and all of the support in the
tools for manually constructing a pass based around them.
Now that TTI is a relatively normal analysis group, two things become
straightforward. First, we can sink it into lib/Analysis which is a more
natural layer for it to live. Second, clients of this interface can
depend on it *always* being available which will simplify their code and
behavior. These (and other) simplifications will follow in subsequent
commits, this one is clearly big enough.
Finally, I'm very aware that much of the comments and documentation
needs to be updated. As soon as I had this working, and plausibly well
commented, I wanted to get it committed and in front of the build bots.
I'll be doing a few passes over documentation later if it sticks.
Commits to update DragonEgg and Clang will be made presently.
llvm-svn: 171681
pass into the SelectionDAG itself rather than snooping on the
implementation of that pass as exposed by the TargetMachine. This
removes the last direct client of the ScalarTargetTransformInfo class
outside of the TTI pass implementation.
llvm-svn: 171625
DAGCombiner::reduceBuildVecConvertToConvertBuildVec() was making two
mistakes:
1. It was checking the legality of scalar INT_TO_FP nodes and then generating
vector nodes.
2. It was passing the result value type to
TargetLoweringInfo::getOperationAction() when it should have been
passing the value type of the first operand.
llvm-svn: 171420
into their new header subdirectory: include/llvm/IR. This matches the
directory structure of lib, and begins to correct a long standing point
of file layout clutter in LLVM.
There are still more header files to move here, but I wanted to handle
them in separate commits to make tracking what files make sense at each
layer easier.
The only really questionable files here are the target intrinsic
tablegen files. But that's a battle I'd rather not fight today.
I've updated both CMake and Makefile build systems (I think, and my
tests think, but I may have missed something).
I've also re-sorted the includes throughout the project. I'll be
committing updates to Clang, DragonEgg, and Polly momentarily.
llvm-svn: 171366
utils/sort_includes.py script.
Most of these are updating the new R600 target and fixing up a few
regressions that have creeped in since the last time I sorted the
includes.
llvm-svn: 171362
directly.
This is in preparation for removing the use of the 'Attribute' class as a
collection of attributes. That will shift to the AttributeSet class instead.
llvm-svn: 171253
This is supposed to be a mechanical change with no functional effects.
InstrEmitter can generate all types of MachineOperands which revealed
that MachineInstrBuilder was missing a few methods, added by this patch.
Besides providing a context pointer to MI::addOperand(),
MachineInstrBuilder seems like a better fit for this code.
llvm-svn: 170712
bitwidth op back to the original size. If we reduce ANDs then this can cause
an endless loop. This patch changes the ZEXT to ANY_EXTEND if the demanded bits
are equal or smaller than the size of the reduced operation.
llvm-svn: 170505
A register can be associated with several distinct register classes.
For example, on PPC, the floating point registers are each associated with
both F4RC (which holds f32) and F8RC (which holds f64). As a result, this code
would fail when provided with a floating point register and an f64 operand
because it would happen to find the register in the F4RC class first and
return that. From the F4RC class, SDAG would extract f32 as the register
type and then assert because of the invalid implied conversion between
the f64 value and the f32 register.
Instead, search all register classes. If a register class containing the
the requested register has the requested type, then return that register
class. Otherwise, as before, return the first register class found that
contains the requested register.
llvm-svn: 170436
TargetLowering::getRegClassFor).
Some isSimple() guards were missing, or getSimpleVT() were hoisted too
far, resulting in asserts on valid LLVM assembly input.
llvm-svn: 170336
Accordingly, add helper funtions getSimpleValueType (in parallel to
getValueType) in SDValue, SDNode, and TargetLowering.
This is the first, in a series of patches.
This is the second attempt. In the first attempt (r169837), a few
getSimpleVT() were hoisted too far, detected by bootstrap failures.
llvm-svn: 170104
mention the inline memcpy / memset expansion code is a mess?
This patch split the ZeroOrLdSrc argument into two: IsMemset and ZeroMemset.
The first indicates whether it is expanding a memset or a memcpy / memmove.
The later is whether the memset is a memset of zero. It's totally possible
(likely even) that targets may want to do different things for memcpy and
memset of zero.
llvm-svn: 169959
Also added more comments to explain why it is generally ok to return true.
- Rename getOptimalMemOpType argument IsZeroVal to ZeroOrLdSrc. It's meant to
be true for loaded source (memcpy) or zero constants (memset). The poor name
choice is probably some kind of legacy issue.
llvm-svn: 169954
ScalarTargetTransformInfo::getIntImmCost() instead. "Legal" is a poorly defined
term for something like integer immediate materialization. It is always possible
to materialize an integer immediate. Whether to use it for memcpy expansion is
more a "cost" conceern.
llvm-svn: 169929
Accordingly, add helper funtions getSimpleValueType (in parallel to
getValueType) in SDValue, SDNode, and TargetLowering.
This is the first, in a series of patches.
llvm-svn: 169837
try to reduce the width of this load, and would end up transforming:
(truncate (lshr (sextload i48 <ptr> as i64), 32) to i32)
to
(truncate (zextload i32 <ptr+4> as i64) to i32)
We lost the sext attached to the load while building the narrower i32
load, and replaced it with a zext because lshr always zext's the
results. Instead, bail out of this combine when there is a conflict
between a sextload and a zext narrowing. The rest of the DAG combiner
still optimize the code down to the proper single instruction:
movswl 6(...),%eax
Which is exactly what we wanted. Previously we read past the end *and*
missed the sign extension:
movl 6(...), %eax
llvm-svn: 169802
This shouldn't affect codegen for -O0 compiles as tail call markers are not
emitted in unoptimized compiles. Testing with the external/internal nightly
test suite reveals no change in compile time performance. Testing with -O1,
-O2 and -O3 with fast-isel enabled did not cause any compile-time or
execution-time failures. All tests were performed on my x86 machine.
I'll monitor our arm testers to ensure no regressions occur there.
In an upcoming clang patch I will be marking the objc_autoreleaseReturnValue
and objc_retainAutoreleaseReturnValue as tail calls unconditionally. While
it's theoretically true that this is just an optimization, it's an
optimization that we very much want to happen even at -O0, or else ARC
applications become substantially harder to debug.
Part of rdar://12553082
llvm-svn: 169796
1. Teach it to use overlapping unaligned load / store to copy / set the trailing
bytes. e.g. On 86, use two pairs of movups / movaps for 17 - 31 byte copies.
2. Use f64 for memcpy / memset on targets where i64 is not legal but f64 is. e.g.
x86 and ARM.
3. When memcpy from a constant string, do *not* replace the load with a constant
if it's not possible to materialize an integer immediate with a single
instruction (required a new target hook: TLI.isIntImmLegal()).
4. Use unaligned load / stores more aggressively if target hooks indicates they
are "fast".
5. Update ARM target hooks to use unaligned load / stores. e.g. vld1.8 / vst1.8.
Also increase the threshold to something reasonable (8 for memset, 4 pairs
for memcpy).
This significantly improves Dhrystone, up to 50% on ARM iOS devices.
rdar://12760078
llvm-svn: 169791
understand target implementation of any_extend / extload, just generate
zero_extend in place of any_extend for liveouts when the target knows the
zero_extend will be implicit (e.g. ARM ldrb / ldrh) or folded (e.g. x86 movz).
rdar://12771555
llvm-svn: 169536
missed in the first pass because the script didn't yet handle include
guards.
Note that the script is now able to handle all of these headers without
manual edits. =]
llvm-svn: 169224
Sooooo many of these had incorrect or strange main module includes.
I have manually inspected all of these, and fixed the main module
include to be the nearest plausible thing I could find. If you own or
care about any of these source files, I encourage you to take some time
and check that these edits were sensible. I can't have broken anything
(I strictly added headers, and reordered them, never removed), but they
may not be the headers you'd really like to identify as containing the
API being implemented.
Many forward declarations and missing includes were added to a header
files to allow them to parse cleanly when included first. The main
module rule does in fact have its merits. =]
llvm-svn: 169131
If we need to split the operand of a VSELECT, it must be the mask operand. We
split the entire VSELECT operand with EXTRACT_SUBVECTOR.
llvm-svn: 168883
This allows me to begin enabling (or backing out) misched by default
for one subtarget at a time. To run misched we typically want to:
- Disable SelectionDAG scheduling (use the source order scheduler)
- Enable more aggressive coalescing (until we decide to always run the coalescer this way)
- Enable MachineScheduler pass itself.
Disabling PostRA sched may follow for some subtargets.
llvm-svn: 167826
This adds support for weak DAG edges to the general scheduling
infrastructure in preparation for MachineScheduler support for
heuristics based on weak edges.
llvm-svn: 167738
r165941: Resubmit the changes to llvm core to update the functions to
support different pointer sizes on a per address space basis.
Despite this commit log, this change primarily changed stuff outside of
VMCore, and those changes do not carry any tests for correctness (or
even plausibility), and we have consistently found questionable or flat
out incorrect cases in these changes. Most of them are probably correct,
but we need to devise a system that makes it more clear when we have
handled the address space concerns correctly, and ideally each pass that
gets updated would receive an accompanying test case that exercises that
pass specificaly w.r.t. alternate address spaces.
However, from this commit, I have retained the new C API entry points.
Those were an orthogonal change that probably should have been split
apart, but they seem entirely good.
In several places the changes were very obvious cleanups with no actual
multiple address space code added; these I have not reverted when
I spotted them.
In a few other places there were merge conflicts due to a cleaner
solution being implemented later, often not using address spaces at all.
In those cases, I've preserved the new code which isn't address space
dependent.
This is part of my ongoing effort to clean out the partial address space
code which carries high risk and low test coverage, and not likely to be
finished before the 3.2 release looms closer. Duncan and I would both
like to see the above issues addressed before we return to these
changes.
llvm-svn: 167222
getIntPtrType support for multiple address spaces via a pointer type,
and also introduced a crasher bug in the constant folder reported in
PR14233.
These commits also contained several problems that should really be
addressed before they are re-committed. I have avoided reverting various
cleanups to the DataLayout APIs that are reasonable to have moving
forward in order to reduce the amount of churn, and minimize the number
of commits that were reverted. I've also manually updated merge
conflicts and manually arranged for the getIntPtrType function to stay
in DataLayout and to be defined in a plausible way after this revert.
Thanks to Duncan for working through this exact strategy with me, and
Nick Lewycky for tracking down the really annoying crasher this
triggered. (Test case to follow in its own commit.)
After discussing with Duncan extensively, and based on a note from
Micah, I'm going to continue to back out some more of the more
problematic patches in this series in order to ensure we go into the
LLVM 3.2 branch with a reasonable story here. I'll send a note to
llvmdev explaining what's going on and why.
Summary of reverted revisions:
r166634: Fix a compiler warning with an unused variable.
r166607: Add some cleanup to the DataLayout changes requested by
Chandler.
r166596: Revert "Back out r166591, not sure why this made it through
since I cancelled the command. Bleh, sorry about this!
r166591: Delete a directory that wasn't supposed to be checked in yet.
r166578: Add in support for getIntPtrType to get the pointer type based
on the address space.
llvm-svn: 167221
the MachineInstr MayLoad/MayLoad flags are based on the tablegen implementation.
For inline assembly, however, we need to compute these based on the constraints.
Revert r166929 as this is no longer needed, but leave the test case in place.
rdar://12033048 and PR13504
llvm-svn: 167040
checks to avoid performing compile-time arithmetic on PPCDoubleDouble.
Now that APFloat supports arithmetic on PPCDoubleDouble, those checks
are no longer needed, and we can treat the type like any other.
llvm-svn: 166958
- If more than 1 elemennts are defined and target supports the vectorized
conversion, use the vectorized one instead to reduce the strength on
conversion operation.
llvm-svn: 166546
which is supposed to consistently raise SIGTRAP across all systems. In contrast,
__builtin_trap() behave differently on different systems. e.g. it raises SIGTRAP on ARM, and
SIGILL on X86. The purpose of __builtin_debugtrap() is to consistently provide "trap"
functionality, in the mean time preserve the compatibility with on gcc on __builtin_trap().
The X86 backend is already able to handle debugtrap(). This patch is to:
1) make front-end recognize "__builtin_debugtrap()" (emboddied in the one-line change to Clang).
2) In DAG legalization phase, by default, "debugtrap" will be replaced with "trap", which
make the __builtin_debugtrap() "available" to all existing ports without the hassle of
changing their code.
3) If trap-function is specified (via -trap-func=xyz to llc), both __builtin_debugtrap() and
__builtin_trap() will be expanded into the function call of the specified trap function.
This behavior may need change in the future.
The provided testing-case is to make sure 2) and 3) are working for ARM port, and we
already have a testing case for x86.
llvm-svn: 166300
- Folding (trunc (concat ... X )) to (concat ... (trunc X) ...) is valid
when '...' are all 'undef's.
- r166125 relies on this transformation.
llvm-svn: 166155
- If the extracted vector has the same type of all vectored being concatenated
together, it should be simplified directly into v_i, where i is the index of
the element being extracted.
llvm-svn: 166125
any scheduling heuristics nor does it build up any scheduling data structure
that other heuristics use. It essentially linearize by doing a DFA walk but
it does handle glues correctly.
IMPORTANT: it probably can't handle all the physical register dependencies so
it's not suitable for x86. It also doesn't deal with dbg_value nodes right now
so it's definitely is still WIP.
rdar://12474515
llvm-svn: 166122
Also provide an MRI::getReservedRegs() function to access the frozen
register set, and isReserved() and isAllocatable() methods to test
individual registers.
The various implementations of TRI::getReservedRegs() are quite
complicated, and many passes need to look at the reserved register set.
This patch makes it possible for these passes to use the cached copy in
MRI, avoiding a lot of malloc traffic and repeated calculations.
llvm-svn: 165982
On PowerPC, a bitcast of <16 x i8> to i128 may run through a code
path in ExpandRes_BITCAST that attempts to do an intermediate
bitcast to a <4 x i32> vector, and then construct the Hi and Lo parts
of the resulting i128 by pairing up two of those i32 vector elements
each. The code already recognizes that on a big-endian system, the
first two vector elements form the Hi part, and the final two vector
elements form the Lo part (vice-versa from the little-endian situation).
However, we also need to take endianness into account when forming each
of those separate pairs: on a big-endian system, vector element 0 is
the *high* part of the pair making up the Hi part of the result, and
vector element 1 is the low part of the pair. The code currently always
uses vector element 0 as the low part and vector element 1 as the high
part, as is appropriate for little-endian platforms only.
This patch fixes this by swapping the vector elements as they are
paired up as appropriate.
llvm-svn: 165802
not legal. However, it should use a div instruction + mul + sub if divide is
legal. The rem legalization code was missing a check and incorrectly uses a
divrem libcall even when div is legal.
rdar://12481395
llvm-svn: 165778
The minimum set of required instructions is ISD::AND, ISD::OR, ISD::SETO(or ISD::SETOEQ) and ISD::SETUO(or ISD::SETUNE). Everything is expanded into one of two patterns:
Pattern 1: (LHS CC1 RHS) Opc (LHS CC2 RHS)
Pattern 2: (LHS CC1 LHS) Opc (RHS CC2 RHS)
llvm-svn: 165655
- Due to the current matching vector elements constraints in ISD::FP_EXTEND,
rounding from v2f32 to v2f64 is scalarized. Add a customized v2f32 widening
to convert it into a target-specific X86ISD::VFPEXT to work around this
constraints. This patch also reverts a previous attempt to fix this issue by
recovering the scalarized ISD::FP_EXTEND pattern and thus significantly
reduces the overhead of supporting non-power-2 vector FP extend.
llvm-svn: 165625
SchedulerDAGInstrs::buildSchedGraph ignores dependencies between FixedStack
objects and byval parameters. So loading byval parameters from stack may be
inserted *before* it will be stored, since these operations are treated as
independent.
Fix:
Currently ARMTargetLowering::LowerFormalArguments saves byval registers with
FixedStack MachinePointerInfo. To fix the problem we need to store byval
registers with MachinePointerInfo referenced to first the "byval" parameter.
Also commit adds two new fields to the InputArg structure: Function's argument
index and InputArg's part offset in bytes relative to the start position of
Function's argument. E.g.: If function's argument is 128 bit width and it was
splitted onto 32 bit regs, then we got 4 InputArg structs with same arg index,
but different offset values.
llvm-svn: 165616
We use the enums to query whether an Attributes object has that attribute. The
opaque layer is responsible for knowing where that specific attribute is stored.
llvm-svn: 165488
This class is used by LSR and a number of places in the codegen.
This is the first step in de-coupling LSR from TLI, and creating
a new interface in between them.
llvm-svn: 165455
multiple stores with a single load. We create the wide loads and stores (and their chains)
before we remove the scalar loads and stores and fix the DAG chain. We attempted to merge
loads with a different chain. When that happened, the assumption that it is safe to RAUW
broke and a cycle was introduced.
llvm-svn: 165148
is not profitable in many cases because modern processors perform multiple stores
in parallel and merging stores prior to merging requires extra work. We handle two main cases:
1. Store of multiple consecutive constants:
q->a = 3;
q->4 = 5;
In this case we store a single legal wide integer.
2. Store of multiple consecutive loads:
int a = p->a;
int b = p->b;
q->a = a;
q->b = b;
In this case we load/store either ilegal vector registers or legal wide integer registers.
llvm-svn: 165125
because moden processos can store multiple values in parallel, and preparing the consecutive store requires
some work. We only handle these cases:
1. Consecutive stores where the values and consecutive loads. For example:
int a = p->a;
int b = p->b;
q->a = a;
q->b = b;
2. Consecutive stores where the values are constants. Foe example:
q->a = 4;
q->b = 5;
llvm-svn: 164910
buildbots. Original commit message:
A DAGCombine optimization for merging consecutive stores. This optimization is not profitable in many cases
because moden processos can store multiple values in parallel, and preparing the consecutive store requires
some work. We only handle these cases:
1. Consecutive stores where the values and consecutive loads. For example:
int a = p->a;
int b = p->b;
q->a = a;
q->b = b;
2. Consecutive stores where the values are constants. Foe example:
q->a = 4;
q->b = 5;
llvm-svn: 164890
because moden processos can store multiple values in parallel, and preparing the consecutive store requires
some work. We only handle these cases:
1. Consecutive stores where the values and consecutive loads. For example:
int a = p->a;
int b = p->b;
q->a = a;
q->b = b;
2. Consecutive stores where the values are constants. Foe example:
q->a = 4;
q->b = 5;
llvm-svn: 164885
scalar-to-vector conversion that we cannot handle. For instance, when an invalid
constraint is used in an inline asm statement.
<rdar://problem/12284092>
llvm-svn: 164662
scalar-to-vector conversion that we cannot handle. For instance, when an invalid
constraint is used in an inline asm statement.
<rdar://problem/12284092>
llvm-svn: 164657
Provide interface in TargetLowering to set or get the minimum number of basic
blocks whereby jump tables are generated for switch statements rather than an
if sequence.
getMinimumJumpTableEntries() defaults to 4.
setMinimumJumpTableEntries() allows target configuration.
This patch changes the default for the Hexagon architecture to 5
as it improves performance on some benchmarks.
llvm-svn: 164628
- Find a legal vector type before casting and extracting element from it.
- As the new vector type may have more than 2 elements, build the final
hi/lo pair by BFS pairing them from bottom to top.
llvm-svn: 163830
by xoring the high-bit. This fails if the source operand is a vector because we need to negate
each of the elements in the vector.
Fix rdar://12281066 PR13813.
llvm-svn: 163802
- BlockAddress has no support of BA + offset form and there is no way to
propagate that offset into machine operand;
- Add BA + offset support and a new interface 'getTargetBlockAddress' to
simplify target block address forming;
- All targets are modified to use new interface and X86 backend is enhanced to
support BA + offset addressing.
llvm-svn: 163743
SelectionDAG::getConstantFP(double Val, EVT VT, bool isTarget);
should not be used when Val is not a simple constant (as the comment in
SelectionDAG.h indicates). This patch avoids using this function
when folding an unknown constant through a bitcast, where it cannot be
guaranteed that Val will be a simple constant.
llvm-svn: 163703
This folding happens as early as possible for performance reasons, and to make sure it isn't foiled by other transforms (e.g. forming FMAs).
llvm-svn: 163519
- CodeGenPrepare pass for identifying div/rem ops
- Backend specifies the type mapping using addBypassSlowDivType
- Enabled only for Intel Atom with O2 32-bit -> 8-bit
- Replace IDIV with instructions which test its value and use DIVB if the value
is positive and less than 256.
- In the case when the quotient and remainder of a divide are used a DIV
and a REM instruction will be present in the IR. In the non-Atom case
they are both lowered to IDIVs and CSE removes the redundant IDIV instruction,
using the quotient and remainder from the first IDIV. However,
due to this optimization CSE is not able to eliminate redundant
IDIV instructions because they are located in different basic blocks.
This is overcome by calculating both the quotient (DIV) and remainder (REM)
in each basic block that is inserted by the optimization and reusing the result
values when a subsequent DIV or REM instruction uses the same operands.
- Test cases check for the presents of the optimization when calculating
either the quotient, remainder, or both.
Patch by Tyler Nowicki!
llvm-svn: 163150
This reverts commit 5dd9e214fb92847e947f9edab170f9b4e52b908f.
Thanks to Duncan for explaining how this should have been done.
Conflicts:
test/CodeGen/X86/vec_select.ll
llvm-svn: 163064
Manage tied operands entirely internally to MachineInstr. This makes it
possible to change the representation of tied operands, as I will do
shortly.
The constraint that tied uses and defs must be in the same order was too
restrictive.
llvm-svn: 163021
because it does not support CMOV of vectors. To implement this efficientlyi, we broadcast the condition bit and use a sequence of NAND-OR
to select between the two operands. This is the same sequence we use for targets that don't have vector BLENDs (like SSE2).
rdar://12201387
llvm-svn: 162926
When there are multiple tied use-def pairs on an inline asm instruction,
the tied uses must appear in the same order as the defs.
It is possible to write an LLVM IR inline asm instruction that breaks
this constraint, but there is no reason for a front end to emit the
operands out of order.
The gnu inline asm syntax specifies tied operands as a single read/write
constraint "+r", so ouf of order operands are not possible.
llvm-svn: 162878
For normal instructions, isTied() is set automatically by addOperand(),
based on MCInstrDesc, but inline asm has tied operands outside the
descriptor.
llvm-svn: 162869
These extra flags are not required to properly order the atomic
load/store instructions. SelectionDAGBuilder chains atomics as if they
were volatile, and SelectionDAG::getAtomic() sets the isVolatile bit on
the memory operands of all atomic operations.
The volatile bit is enough to order atomic loads and stores during and
after SelectionDAG.
This means we set mayLoad on atomic_load, mayStore on atomic_store, and
mayLoad+mayStore on the remaining atomic read-modify-write operations.
llvm-svn: 162733
In SelectionDAGLegalize::ExpandLegalINT_TO_FP, expand INT_TO_FP nodes without
using any f64 operations if f64 is not a legal type.
Patch by Stefan Kristiansson.
llvm-svn: 162728
the case of multiple edges from one block to another.
A simple example is a switch statement with multiple values to the same
destination. The definition of an edge is modified from a pair of blocks to
a pair of PredBlock and an index into the successors.
Also set the weight correctly when building SelectionDAG from LLVM IR,
especially when converting a Switch.
IntegersSubsetMapping is updated to calculate the weight for each cluster.
llvm-svn: 162572
SelectionDAG's 'init' has not been called when the SelectionDAGBuilder is
constructed (in SelectionDAGISel's constructor), so this was previously always
initialized with 0.
llvm-svn: 162333
IR that hasn't been through SimplifyCFG can look like this:
br i1 %b, label %r, label %r
Make sure we don't create duplicate Machine CFG edges in this case.
Fix the machine code verifier to accept conditional branches with a
single CFG edge.
llvm-svn: 162230
The DAGCombiner tries to optimise a BUILD_VECTOR by checking if it
consists purely of get_vector_elts from one or two source vectors. If
so, it either makes a concat_vectors node or a shufflevector node.
However, it doesn't check the element type width of the underlying
vector, so if you have this sequence:
Node0: v4i16 = ...
Node1: i32 = extract_vector_elt Node0
Node2: i32 = extract_vector_elt Node0
Node3: v16i8 = BUILD_VECTOR Node1, Node2, ...
It will attempt to:
Node0: v4i16 = ...
NewNode1: v16i8 = concat_vectors Node0, ...
Where this is actually invalid because the element width is completely
different. This causes an assertion failure on DAG legalization stage.
Fix:
If output item type of BUILD_VECTOR differs from input item type.
Make concat_vectors based on input element type and then bitcast it to the output vector type. So the case described above will transformed to:
Node0: v4i16 = ...
NewNode1: v8i16 = concat_vectors Node0, ...
NewNode2: v16i8 = bitcast NewNode1
llvm-svn: 162195
This adds support for TargetIndex operands during isel. The meaning of
these (index, offset, flags) operands is entirely defined by the target.
llvm-svn: 161453
This patch is mostly just refactoring a bunch of copy-and-pasted code, but
it also adds a check that the call instructions are readnone or readonly.
That check was already present for sin, cos, sqrt, log2, and exp2 calls, but
it was missing for the rest of the builtins being handled in this code.
llvm-svn: 161282
The previous change caused fast isel to not attempt handling any calls to
builtin functions. That included things like "printf" and caused some
noticable regressions in compile time. I wanted to avoid having fast isel
keep a separate list of functions that had to be kept in sync with what the
code in SelectionDAGBuilder.cpp was handling. I've resolved that here by
moving the list into TargetLibraryInfo. This is somewhat redundant in
SelectionDAGBuilder but it will ensure that we keep things consistent.
llvm-svn: 161263
I noticed that SelectionDAGBuilder::visitCall was missing a check for memcmp
in TargetLibraryInfo, so that it would use custom code for memcmp calls even
with -fno-builtin. I also had to add a new -disable-simplify-libcalls option
to llc so that I could write a test for this.
llvm-svn: 161262
Fast isel doesn't currently have support for translating builtin function
calls to target instructions. For embedded environments where the library
functions are not available, this is a matter of correctness and not
just optimization. Most of this patch is just arranging to make the
TargetLibraryInfo available in fast isel. <rdar://problem/12008746>
llvm-svn: 161232
large immediates. Add dag combine logic to recover in case the large
immediates doesn't fit in cmp immediate operand field.
int foo(unsigned long l) {
return (l>> 47) == 1;
}
we produce
%shr.mask = and i64 %l, -140737488355328
%cmp = icmp eq i64 %shr.mask, 140737488355328
%conv = zext i1 %cmp to i32
ret i32 %conv
which codegens to
movq $0xffff800000000000,%rax
andq %rdi,%rax
movq $0x0000800000000000,%rcx
cmpq %rcx,%rax
sete %al
movzbl %al,%eax
ret
TargetLowering::SimplifySetCC would transform
(X & -256) == 256 -> (X >> 8) == 1
if the immediate fails the isLegalICmpImmediate() test. For x86,
that's immediates which are not a signed 32-bit immediate.
Based on a patch by Eli Friedman.
PR10328
rdar://9758774
llvm-svn: 160346
In the added testcase the constant 55 was behind an AssertZext of type i1, and ComputeDemandedBits
reported that some of the bits were both known to be one and known to be zero.
Together with Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 160305
Add a micro-optimization to getNode of CONCAT_VECTORS when both operands are undefs.
Can't find a testcase for this because VECTOR_SHUFFLE already handles undef operands, but Duncan suggested that we add this.
Together with Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 160229
When dumping the DAG for a fatal 'Cannot select' back-end error, also
provide the name of the function the construct is in. Useful when dealing
with large testcases, as the next step is to llvm-extract the function
in question to get a small(er) testcase.
llvm-svn: 160152
the input vector, it can be bigger (this is helpful for powerpc where <2 x i16>
is a legal vector type but i16 isn't a legal type, IIRC). However this wasn't
being taken into account by ExpandRes_EXTRACT_VECTOR_ELT, causing PR13220.
Lightly tweaked version of a patch by Michael Liao.
llvm-svn: 160116
r1025 = s/zext r1024, 4
r1026 = extract_subreg r1025, 4
to a copy:
r1026 = copy r1024
This is correct. However it uses TII->isCoalescableExtInstr() which can return
true for instructions which essentially does a sext_in_reg so this can end up
with an illegal copy where the source and destination register classes do not
match. Add a check to avoid it. Sorry, no test case possible at this time.
rdar://11849816
llvm-svn: 160059
multiple scalars and insert them into a vector. Next, we shuffle the elements
into the correct places, as before.
Also fix a small dagcombine bug in SimplifyBinOpWithSameOpcodeHands, when the
migration of bitcasts happened too late in the SelectionDAG process.
llvm-svn: 159991
subtarget CPU descriptions and support new features of
MachineScheduler.
MachineModel has three categories of data:
1) Basic properties for coarse grained instruction cost model.
2) Scheduler Read/Write resources for simple per-opcode and operand cost model (TBD).
3) Instruction itineraties for detailed per-cycle reservation tables.
These will all live side-by-side. Any subtarget can use any
combination of them. Instruction itineraries will not change in the
near term. In the long run, I expect them to only be relevant for
in-order VLIW machines that have complex contraints and require a
precise scheduling/bundling model. Once itineraries are only actively
used by VLIW-ish targets, they could be replaced by something more
appropriate for those targets.
This tablegen backend rewrite sets things up for introducing
MachineModel type #2: per opcode/operand cost model.
llvm-svn: 159891
Also allow trailing register mask operands on non-variadic both
MachineSDNodes and MachineInstrs.
The extra physreg RegisterSDNode operands are added to the MI as
<imp-use> operands. This makes it possible to have non-variadic call
instructions.
Call and return instructions really are non-variadic, the argument
registers should only be used implicitly - they are not part of the
encoding.
llvm-svn: 159727
IntegersSubsetMapping
- Replaced type of Items field from std::list with std::map. In neares future I'll test it with DenseMap and do the correspond replacement
if possible.
llvm-svn: 159703
IntegersSubsetMapping
- Replaced type of Items field from std::list with std::map. In neares future I'll test it with DenseMap and do the correspond replacement
if possible.
llvm-svn: 159659
Teach vector legalization how to honor Promote for int to float
conversions. The code checking whether to promote the operation knew
to look at the operand, but the actual promotion code didn't. This
fixes that. The operand is promoted up via [zs]ext.
rdar://11762659
llvm-svn: 159378
include/llvm/Analysis/DebugInfo.h to include/llvm/DebugInfo.h.
The reasoning is because the DebugInfo module is simply an interface to the
debug info MDNodes and has nothing to do with analysis.
llvm-svn: 159312
boolean flag to an enum: { Fast, Standard, Strict } (default = Standard).
This option controls the creation by optimizations of fused FP ops that store
intermediate results in higher precision than IEEE allows (E.g. FMAs). The
behavior of this option is intended to match the behaviour specified by a
soon-to-be-introduced frontend flag: '-ffuse-fp-ops'.
Fast mode - allows formation of fused FP ops whenever they're profitable.
Standard mode - allow fusion only for 'blessed' FP ops. At present the only
blessed op is the fmuladd intrinsic. In the future more blessed ops may be
added.
Strict mode - allow fusion only if/when it can be proven that the excess
precision won't effect the result.
Note: This option only controls formation of fused ops by the optimizers. Fused
operations that are explicitly requested (e.g. FMA via the llvm.fma.* intrinsic)
will always be honored, regardless of the value of this option.
Internally TargetOptions::AllowExcessFPPrecision has been replaced by
TargetOptions::AllowFPOpFusion.
llvm-svn: 158956
_umodsi3 libcalls if they have the same arguments. This optimization
was apparently broken if one of the node was replaced in place.
rdar://11714607
llvm-svn: 158900
This patch adds DAG combines to form FMAs from pairs of FADD + FMUL or
FSUB + FMUL. The combines are performed when:
(a) Either
AllowExcessFPPrecision option (-enable-excess-fp-precision for llc)
OR
UnsafeFPMath option (-enable-unsafe-fp-math)
are set, and
(b) TargetLoweringInfo::isFMAFasterThanMulAndAdd(VT) is true for the type of
the FADD/FSUB, and
(c) The FMUL only has one user (the FADD/FSUB).
If your target has fast FMA instructions you can make use of these combines by
overriding TargetLoweringInfo::isFMAFasterThanMulAndAdd(VT) to return true for
types supported by your FMA instruction, and adding patterns to match ISD::FMA
to your FMA instructions.
llvm-svn: 158757
expression (a * b + c) that can be implemented as a fused multiply-add (fma)
if the target determines that this will be more efficient. This intrinsic
will be used to implement FP_CONTRACT support and an aggressive FMA formation
mode.
If your target has a fast FMA instruction you should override the
isFMAFasterThanMulAndAdd method in TargetLowering to return true.
llvm-svn: 158014
This allows a subtarget to explicitly specify the issue width and
other properties without providing pipeline stage details for every
instruction.
llvm-svn: 157979
IntegersSubsetGeneric, IntegersSubsetMapping: added IntTy template parameter, that allows use either APInt or IntItem. This change allows to write unittest for these classes.
llvm-svn: 157880
No functional change intended.
Sorry for the churn. The iterator classes are supposed to help avoid
giant commits like this one in the future. The TableGen-produced
register lists are getting quite large, and it may be necessary to
change the table representation.
This makes it possible to do so without changing all clients (again).
llvm-svn: 157854
IntegersSubset devided into IntegersSubsetGeneric and into IntegersSubset itself. The first has no references to ConstantInt and works with IntItem only.
IntegersSubsetMapping also made generic. Here added second template parameter "IntegersSubsetTy" that allows to use on of two IntegersSubset types described below.
llvm-svn: 157815
Implemented IntItem - the wrapper around APInt. Why not to use APInt item directly right now?
1. It will very difficult to implement case ranges as series of small patches. We got several large and heavy patches. Each patch will about 90-120 kb. If you replace ConstantInt with APInt in SwitchInst you will need to changes at the same time all Readers,Writers and absolutely all passes that uses SwitchInst.
2. We can implement APInt pool inside and save memory space. E.g. we use several switches that works with 256 bit items (switch on signatures, or strings). We can avoid value duplicates in this case.
3. IntItem can be easyly easily replaced with APInt.
4. Currenly we can interpret IntItem both as ConstantInt and as APInt. It allows to provide SwitchInst methods that works with ConstantInt for non-updated passes.
Why I need it right now? Currently I need to update SimplifyCFG pass (EqualityComparisons). I need to work with APInts directly a lot, so peaces of code
ConstantInt *V = ...;
if (V->getValue().ugt(AnotherV->getValue()) {
...
}
will look awful. Much more better this way:
IntItem V = ConstantIntVal->getValue();
if (AnotherV < V) {
}
Of course any reviews are welcome.
P.S.: I'm also going to rename ConstantRangesSet to IntegersSubset, and CRSBuilder to IntegersSubsetMapping (allows to map individual subsets of integers to the BasicBlocks).
Since in future these classes will founded on APInt, it will possible to use them in more generic ways.
llvm-svn: 157576
SimplifyCFG tends to form a lot of 2-3 case switches when merging branches. Move
the most likely condition to the front so it is checked first and the others can
be skipped. This is currently not as effective as it could be because SimplifyCFG
destroys profiling metadata when merging branches and switches. Merging branch
weight metadata is tricky though.
This code touches at most 3 cases so I didn't use a proper sorting algorithm.
llvm-svn: 157521
to pass around a struct instead of a large set of individual values. This
cleans up the interface and allows more information to be added to the struct
for future targets without requiring changes to each and every target.
NV_CONTRIB
llvm-svn: 157479
may be RAUW'd by the recursive call to LegalizeOps; instead, retrieve
the other operands when calling UpdateNodeOperands. Fixes PR12889.
llvm-svn: 157162
SelectionDAGBuilder::Clusterify : main functinality was replaced with CRSBuilder::optimize, so big part of Clusterify's code was reduced.
llvm-svn: 157046
When a combine twiddles an extract_vector, care should be take to preserve
the type of the index operand. No luck extracting a reasonable testcase,
unfortunately.
rdar://11391009
llvm-svn: 156419
The getPointerRegClass() hook can return register classes that depend on
the calling convention of the current function (ptr_rc_tailcall).
So far, we have been able to infer the calling convention from the
subtarget alone, but as we add support for multiple calling conventions
per target, that no longer works.
Patch by Yiannis Tsiouris!
llvm-svn: 156328
This will be used to determine whether it's profitable to turn a select into a
branch when the branch is likely to be predicted.
Currently enabled for everything but Atom on X86 and Cortex-A9 devices on ARM.
I'm not entirely happy with the name of this flag, suggestions welcome ;)
llvm-svn: 156233
We want the representative register class to contain the largest
super-registers available. This makes the function less sensitive to the
register class numbering.
llvm-svn: 156220
The masks returned by SuperRegClassIterator are computed automatically
by TableGen. This is better than depending on the manually specified
SuperRegClasses.
llvm-svn: 156147
The ensures that virtual registers always belong to an allocatable class.
If your target attempts to create a vreg for an operand that has no
allocatable register subclass, you will crash quickly.
This ensures that targets define register classes as intended.
llvm-svn: 156046
This time, also fix the caller of AddGlue to properly handle
incomplete chains. AddGlue had failure modes, but shamefully hid them
from its caller. It's luck ran out.
Fixes rdar://11314175: BuildSchedUnits assert.
llvm-svn: 155749
DAGCombine strangeness may result in multiple loads from the same
offset. They both may try to glue themselves to another load. We could
insist that the redundant loads glue themselves to each other, but the
beter fix is to bail out from bad gluing at the time we detect it.
Fixes rdar://11314175: BuildSchedUnits assert.
llvm-svn: 155668
The X86 target is editing the selection DAG while isel is selecting
nodes following a topological ordering. When the DAG hacking triggers
CSE, nodes can be deleted and bad things happen.
llvm-svn: 155257
Now that multiple DAGUpdateListeners can be active at the same time,
ISelPosition can become a local variable in DoInstructionSelection.
We simply register an ISelUpdater with CurDAG while ISelPosition exists.
llvm-svn: 155249
Instead of passing listener pointers to RAUW, let SelectionDAG itself
keep a linked list of interested listeners.
This makes it possible to have multiple listeners active at once, like
RAUWUpdateListener was already doing. It also makes it possible to
register listeners up the call stack without controlling all RAUW calls
below.
DAGUpdateListener uses an RAII pattern to add itself to the SelectionDAG
list of active listeners.
llvm-svn: 155248
transformation:
(X op C1) ^ C2 --> (X op C1) & ~C2 iff (C1&C2) == C2
should be done.
This change has been tested:
Using a debug+asserts build:
on the specific test case that brought this bug to light
make check-all
lnt nt
using this clang to build a release version of clang
Using the release+asserts clang-with-clang build:
on the specific test case that brought this bug to light
make check-all
lnt nt
Checking in because Evan wants it checked in. Test case forthcoming after
scrubbing.
llvm-svn: 154955
Fix a dagcombine optimization which assumes that the vsetcc result type is always
of the same size as the compared values. This is ture for SSE/AVX/NEON but not
for all targets.
llvm-svn: 154490
legalizer always use the DAG entry node. This is wrong when the libcall is
emitted as a tail call since it effectively folds the return node. If
the return node's input chain is not the entry (i.e. call, load, or store)
use that as the tail call input chain.
PR12419
rdar://9770785
rdar://11195178
llvm-svn: 154370
when -ffast-math, i.e. don't just always do it if the reciprocal can
be formed exactly. There is already an IR level transform that does
that, and it does it more carefully.
llvm-svn: 154296
in TargetLowering. There was already a FIXME about this location being
odd. The interface is simplified as a consequence. This will also make
it easier to change TLS models when compiling with PIE.
llvm-svn: 154292
shuffle node because it could introduce new shuffle nodes that were not
supported efficiently by the target.
2. Add a more restrictive shuffle-of-shuffle optimization for cases where the
second shuffle reverses the transformation of the first shuffle.
llvm-svn: 154266
reciprocal if converting to the reciprocal is exact. Do it even if inexact
if -ffast-math. This substantially speeds up ac.f90 from the polyhedron
benchmarks.
llvm-svn: 154265
LSR always tries to make the ICmp in the loop latch use the incremented
induction variable. This allows the induction variable to be kept in a
single register.
When the induction variable limit is equal to the stride,
SimplifySetCC() would break LSR's hard work by transforming:
(icmp (add iv, stride), stride) --> (cmp iv, 0)
This forced us to use lea for the IC update, preventing the simpler
incl+cmp.
<rdar://problem/7643606>
<rdar://problem/11184260>
llvm-svn: 154119
This allows us to keep passing reduced masks to SimplifyDemandedBits, but
know about all the bits if SimplifyDemandedBits fails. This allows instcombine
to simplify cases like the one in the included testcase.
llvm-svn: 154011
When folding X == X we need to check getBooleanContents() to determine if the
result is a vector of ones or a vector of negative ones.
I tried creating a test case, but the problem seems to only be exposed on a
much older version of clang (around r144500).
rdar://10923049
llvm-svn: 153966
Do not try to optimize swizzles of shuffles if the source shuffle has more than
a single user, except when the source shuffle is also a swizzle.
llvm-svn: 153864
This is the CodeGen equivalent of r153747. I tested that there is not noticeable
performance difference with any combination of -O0/-O2 /-g when compiling
gcc as a single compilation unit.
llvm-svn: 153817
here but it has no other uses, then we have a problem. E.g.,
int foo (const int *x) {
char a[*x];
return 0;
}
If we assign 'a' a vreg and fast isel later on has to use the selection
DAG isel, it will want to copy the value to the vreg. However, there are
no uses, which goes counter to what selection DAG isel expects.
<rdar://problem/11134152>
llvm-svn: 153705
execution-time regression for nsieve-bits on the ARMv7 -O0 -g nightly tester.
This may also improve compile-time on architectures that would otherwise
generate a libcall for urem (e.g., ARM) or fall back to the DAG selector.
rdar://10810716
llvm-svn: 153230
Type legalization can zero-extend the elements of the build_vector node, so,
for example, we may have an <8 x i8> with i32 elements of value 255. That
should return 'true' for the vector being all ones.
llvm-svn: 153203
a variable. The previous code would break the debug info changing
code invariant. This will regress debug info for arguments where
we elide the alloca created.
Fixes rdar://11066468
llvm-svn: 153074
It caused MSP430DAGToDAGISel::SelectIndexedBinOp() to be miscompiled.
When two ReplaceUses()'s are expanded as inline, vtable in base class is stored to latter (ISelUpdater)ISU.
llvm-svn: 152877
(i16 load $addr+c*sizeof(i16)) and replace uses of (i32 vextract) with the
i16 load. It should issue an extload instead: (i32 extload $addr+c*sizeof(i16)).
rdar://11035895
llvm-svn: 152675
Renamed methods caseBegin, caseEnd and caseDefault with case_begin, case_end, and case_default.
Added some notes relative to case iterators.
llvm-svn: 152532
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120130/136146.html
Implemented CaseIterator and it solves almost all described issues: we don't need to mix operand/case/successor indexing anymore. Base iterator class is implemented as a template since it may be initialized either from "const SwitchInst*" or from "SwitchInst*".
ConstCaseIt is just a read-only iterator.
CaseIt is read-write iterator; it allows to change case successor and case value.
Usage of iterator allows totally remove resolveXXXX methods. All indexing convertions done automatically inside the iterator's getters.
Main way of iterator usage looks like this:
SwitchInst *SI = ... // intialize it somehow
for (SwitchInst::CaseIt i = SI->caseBegin(), e = SI->caseEnd(); i != e; ++i) {
BasicBlock *BB = i.getCaseSuccessor();
ConstantInt *V = i.getCaseValue();
// Do something.
}
If you want to convert case number to TerminatorInst successor index, just use getSuccessorIndex iterator's method.
If you want initialize iterator from TerminatorInst successor index, use CaseIt::fromSuccessorIndex(...) method.
There are also related changes in llvm-clients: klee and clang.
llvm-svn: 152297
ScheduleDAG is responsible for the DAG: SUnits and SDeps. It provides target hooks for latency computation.
ScheduleDAGInstrs extends ScheduleDAG and defines the current scheduling region in terms of MachineInstr iterators. It has access to the target's scheduling itinerary data. ScheduleDAGInstrs provides the logic for building the ScheduleDAG for the sequence of MachineInstrs in the current region. Target's can implement highly custom schedulers by extending this class.
ScheduleDAGPostRATDList provides the driver and diagnostics for current postRA scheduling. It maintains a current Sequence of scheduled machine instructions and logic for splicing them into the block. During scheduling, it uses the ScheduleHazardRecognizer provided by the target.
Specific changes:
- Removed driver code from ScheduleDAG. clearDAG is the only interface needed.
- Added enterRegion/exitRegion hooks to ScheduleDAGInstrs to delimit the scope of each scheduling region and associated DAG. They should be used to setup and cleanup any region-specific state in addition to the DAG itself. This is necessary because we reuse the same ScheduleDAG object for the entire function. The target may extend these hooks to do things at regions boundaries, like bundle terminators. The hooks are called even if we decide not to schedule the region. So all instructions in a block are "covered" by these calls.
- Added ScheduleDAGInstrs::begin()/end() public API.
- Moved Sequence into the driver layer, which is specific to the scheduling algorithm.
llvm-svn: 152208
To avoid problems with zero shifts when getting the bits that move between words
we use a trick: first shift the by amount-1, then do another shift by one. When
amount is 0 (and size 32) we first shift by 31, then by one, instead of by 32.
Also fix a latent bug that emitted the low and high words in the wrong order
when shifting right.
Fixes PR12113.
llvm-svn: 151637
When the GEP index is a vector of pointers, the code that calculated the size
of the element started from the vector type, and not the contained pointer type.
As a result, instead of looking at the data element pointed by the vector, this
code used the size of the vector. This works for 32bit members (on 32bit
systems), but not for other types. Added code to peel the vector type and
added a test.
llvm-svn: 151626
the processor keeps a return addresses stack (RAS) which stores the address
and the instruction execution state of the instruction after a function-call
type branch instruction.
Calling a "noreturn" function with normal call instructions (e.g. bl) can
corrupt RAS and causes 100% return misprediction so LLVM should use a
unconditional branch instead. i.e.
mov lr, pc
b _foo
The "mov lr, pc" is issued in order to get proper backtrace.
rdar://8979299
llvm-svn: 151623
variable declaration as an argument because we want that address
anyhow for our debug information.
This seems to fix rdar://9965111, at least we have more debug
information than before and from reading the assembly it appears
to be the correct location.
llvm-svn: 151335
that are greater than the vector element type. For example BUILD_VECTOR
of type <1 x i1> with a constant i8 operand.
This patch fixes the assertion.
llvm-svn: 150477
The scheduler will sometimes check the implicit-def list on instructions
to properly handle pre-colored DAG edges.
Also check any register mask operands for physreg clobbers.
llvm-svn: 150428
v8i8 -> v8i32 on AVX machines. The codegen often scalarizes ANY_EXTEND nodes.
The DAGCombiner has two optimizations that can mitigate the problem. First,
if all of the operands of a BUILD_VECTOR node are extracted from an ZEXT/ANYEXT
nodes, then it is possible to create a new simplified BUILD_VECTOR which uses
UNDEFS/ZERO values to eliminate the scalar ZEXT/ANYEXT nodes.
Second, another dag combine optimization lowers BUILD_VECTOR into a shuffle
vector instruction.
In the case of zext v8i8->v8i32 on AVX, a value in an XMM register is to be
shuffled into a wide YMM register.
This patch modifes the second optimization and allows the creation of
shuffle vectors even when the newly generated vector and the original vector
from which we extract the values are of different types.
llvm-svn: 150340
Make them accessible through MCInstrInfo. They are only used for debugging purposes so this doesn't
have an impact on performance. X86MCTargetDesc.o goes from 630K to 461K on x86_64.
llvm-svn: 150245
but with a critical fix to the SelectionDAG code that optimizes copies
from strings into immediate stores: the previous code was stopping reading
string data at the first nul. Address this by adding a new argument to
llvm::getConstantStringInfo, preserving the behavior before the patch.
llvm-svn: 149800
SelectionDAG has 4 different ways of passing physreg defs to users.
Collect all of the uses at the same time, and pass all of them to
MI->setPhysRegsDeadExcept() to mark the remaining defs dead.
The setPhysRegsDeadExcept() function will soon add the required
implicit-defs to instructions with register mask operands.
llvm-svn: 149708
In this patch we optimize this pattern and convert the sequence into extract op of a narrow type.
This allows the BUILD_VECTOR dag optimizations to construct efficient shuffle operations in many cases.
llvm-svn: 149692
This new scheduler plugs into the existing selection DAG scheduling framework. It is a top-down critical path scheduler that tracks register pressure and uses a DFA for pipeline modeling.
Patch by Sergei Larin!
llvm-svn: 149547
The purpose of refactoring is to hide operand roles from SwitchInst user (programmer). If you want to play with operands directly, probably you will need lower level methods than SwitchInst ones (TerminatorInst or may be User). After this patch we can reorganize SwitchInst operands and successors as we want.
What was done:
1. Changed semantics of index inside the getCaseValue method:
getCaseValue(0) means "get first case", not a condition. Use getCondition() if you want to resolve the condition. I propose don't mix SwitchInst case indexing with low level indexing (TI successors indexing, User's operands indexing), since it may be dangerous.
2. By the same reason findCaseValue(ConstantInt*) returns actual number of case value. 0 means first case, not default. If there is no case with given value, ErrorIndex will returned.
3. Added getCaseSuccessor method. I propose to avoid usage of TerminatorInst::getSuccessor if you want to resolve case successor BB. Use getCaseSuccessor instead, since internal SwitchInst organization of operands/successors is hidden and may be changed in any moment.
4. Added resolveSuccessorIndex and resolveCaseIndex. The main purpose of these methods is to see how case successors are really mapped in TerminatorInst.
4.1 "resolveSuccessorIndex" was created if you need to level down from SwitchInst to TerminatorInst. It returns TerminatorInst's successor index for given case successor.
4.2 "resolveCaseIndex" converts low level successors index to case index that curresponds to the given successor.
Note: There are also related compatability fix patches for dragonegg, klee, llvm-gcc-4.0, llvm-gcc-4.2, safecode, clang.
llvm-svn: 149481
This SelectionDAG node will be attached to call nodes by LowerCall(),
and eventually becomes a MO_RegisterMask MachineOperand on the
MachineInstr representing the call instruction.
LowerCall() will attach a register mask that depends on the calling
convention.
llvm-svn: 148436
We know that the blend instructions only use the MSB, so if the mask is
sign-extended then we can convert it into a SHL instruction. This is a
common pattern because the type-legalizer sign-extends the i1 type which
is used by the LLVM-IR for the condition.
Added a new optimization in SimplifyDemandedBits for SIGN_EXTEND_INREG -> SHL.
llvm-svn: 148225
overly conservative. It was concerned about cases where it would prohibit
folding simple [r, c] addressing modes. e.g.
ldr r0, [r2]
ldr r1, [r2, #4]
=>
ldr r0, [r2], #4
ldr r1, [r2]
Change the logic to look for such cases which allows it to form indexed memory
ops more aggressively.
rdar://10674430
llvm-svn: 148086
When we load the v12i32 type, the GenWidenVectorLoads method generates two loads: v8i32 and v4i32
and attempts to use CONCAT_VECTORS to join them. In this fix I concat undef values to widen
the smaller value. The test "widen_load-2.ll" also exposes this bug on AVX.
llvm-svn: 147964
detect a pattern which can be implemented with a small 'shl' embedded in
the addressing mode scale. This happens in real code as follows:
unsigned x = my_accelerator_table[input >> 11];
Here we have some lookup table that we look into using the high bits of
'input'. Each entity in the table is 4-bytes, which means this
implicitly gets turned into (once lowered out of a GEP):
*(unsigned*)((char*)my_accelerator_table + ((input >> 11) << 2));
The shift right followed by a shift left is canonicalized to a smaller
shift right and masking off the low bits. That hides the shift right
which x86 has an addressing mode designed to support. We now detect
masks of this form, and produce the longer shift right followed by the
proper addressing mode. In addition to saving a (rather large)
instruction, this also reduces stalls in Intel chips on benchmarks I've
measured.
In order for all of this to work, one part of the DAG needs to be
canonicalized *still further* than it currently is. This involves
removing pointless 'trunc' nodes between a zextload and a zext. Without
that, we end up generating spurious masks and hiding the pattern.
llvm-svn: 147936
of several newly un-defaulted switches. This also helps optimizers
(including LLVM's) recognize that every case is covered, and we should
assume as much.
llvm-svn: 147861
a combined-away node and the result of the combine isn't substantially
smaller than the input, it's just canonicalized. This is the first part
of a significant (7%) performance gain for Snappy's hot decompression
loop.
llvm-svn: 147604
Job Noorman