getSCEV for an ashr instruction creates an intermediate zext
expression when it truncates its operand.
The operand is initially inside the loop, so the narrow zext
expression has a non-loop-invariant loop disposition.
LoopSimplify then runs on an outer loop, hoists the ashr operand, and
properly invalidate the SCEVs that are mapped to value.
The SCEV expression for the ashr is now an AddRec with the hoisted
value as the now loop-invariant start value.
The LoopDisposition of this wide value was properly invalidated during
LoopSimplify.
However, if we later get the ashr SCEV again, we again try to create
the intermediate zext expression. We get the same SCEV that we did
earlier, and it is still cached because it was never mapped to a
Value. When we try to create a new AddRec we abort because we're using
the old non-loop-invariant LoopDisposition.
I don't have a solution for this other than to clear LoopDisposition
when LoopSimplify hoists things.
I think the long-term strategy should be to perform LoopSimplify on
all loops before computing SCEV and before running any loop opts on
individual loops. It's possible we may want to rerun LoopSimplify on
individual loops, but it should rarely do anything, so rarely require
invalidating SCEV.
llvm-svn: 198478
The motivation is to mark dump methods as used in debug builds so that they can
be called from lldb, but to not do so in release builds so that they can be
dead-stripped.
There's lots of potential follow-up work suggested in the thread
"Should dump methods be LLVM_ATTRIBUTE_USED only in debug builds?" on cfe-dev,
but everyone seems to agreen on this subset.
Macro name chosen by fair coin toss.
llvm-svn: 198456
This reverts commit r198441.
This change doesn't build on Windows, and doesn't do the right thing on
Linux and other platforms that don't use a _Z prefix instead of __Z for
C++ names.
It also had no tests, so it wasn't clear how to fix it forward.
llvm-svn: 198445
symbol name, also put the human readable name in a comment.
Also fix a bug in LLVMDisasmInstruction() that was not flushing
the raw_svector_ostream for the disassembled instruction string
before copying it to the output buffer that was causing truncation
of the output.
rdar://10173828
llvm-svn: 198441
Before this patch any program that wanted to know the final symbol name of a
GlobalValue had to link with Target.
This patch implements a compromise solution where the mangler uses DataLayout.
This way, any tool that already links with Target (llc, clang) gets the exact
behavior as before and new IR files can be mangled without linking with Target.
With this patch the mangler is constructed with just a DataLayout and DataLayout
is extended to include the information the Mangler needs.
llvm-svn: 198438
This reverts commit r198398, thus reapplying r198397.
I had accidentally introduced an endianness issue when applying the hash
to the type unit. Using support::ulittle64_t in the reinterpret_cast in
addDwarfTypeUnitType fixes this issue.
Original commit message:
Debug Info: Type Units: Simplify type hashing using IR-provided unique
names.
What's good for LTO metadata size problems ought to be good for non-LTO
debug info size too, so let's rely on the same uniqueness in both cases.
If it's insufficient for non-LTO for whatever reason (since we now won't
be uniquing CU-local types or any C types - but these are likely to not
be the most significant contributors to type bloat) we should consider a
frontend solution that'll help both LTO and non-LTO alike, rather than
using DWARF-level DIE-hashing that only helps non-LTO debug info size.
It's also much simpler this way and benefits C++ even more since we can
deduplicate lexically separate definitions of the same C++ type since
they have the same mangled name.
llvm-svn: 198436
The loop rerolling pass was failing with an assertion failure from a
failed cast on loops like this:
void foo(int *A, int *B, int m, int n) {
for (int i = m; i < n; i+=4) {
A[i+0] = B[i+0] * 4;
A[i+1] = B[i+1] * 4;
A[i+2] = B[i+2] * 4;
A[i+3] = B[i+3] * 4;
}
}
The code was casting the SCEV-expanded code for the new
induction variable to a phi-node. When the loop had a non-constant
lower bound, the SCEV expander would end the code expansion with an
add insted of a phi node and the cast would fail.
It looks like the cast to a phi node was only needed to get the
induction variable value coming from the backedge to compute the end
of loop condition. This patch changes the loop reroller to compare
the induction variable to the number of times the backedge is taken
instead of the iteration count of the loop. In other words, we stop
the loop when the current value of the induction variable ==
IterationCount-1. Previously, the comparison was comparing the
induction variable value from the next iteration == IterationCount.
This problem only seems to occur on 32-bit targets. For some reason,
the loop is not rerolled on 64-bit targets.
PR18290
llvm-svn: 198425
cycles
This allows the value equality check to work even if we don't have a dominator
tree. Also add some more comments.
I was worried about compile time impacts and did not implement reachability but
used the dominance check in the initial patch. The trade-off was that the
dominator tree was required.
The llvm utility function isPotentiallyReachable cuts off the recursive search
after 32 visits. Testing did not show any compile time regressions showing my
worries unjustfied.
No compile time or performance regressions at O3 -flto -mavx on test-suite +
externals.
Addresses review comments from r198290.
llvm-svn: 198400
What's good for LTO metadata size problems ought to be good for non-LTO
debug info size too, so let's rely on the same uniqueness in both cases.
If it's insufficient for non-LTO for whatever reason (since we now won't
be uniquing CU-local types or any C types - but these are likely to not
be the most significant contributors to type bloat) we should consider a
frontend solution that'll help both LTO and non-LTO alike, rather than
using DWARF-level DIE-hashing that only helps non-LTO debug info size.
It's also much simpler this way and benefits C++ even more since we can
deduplicate lexically separate definitions of the same C++ type since
they have the same mangled name.
llvm-svn: 198397
The cgo problem was that it wants dwarf2 which doesn't support direct
constant encoding of the location. So let's add support for dwarf2
encoding (using a location expression) of data member locations.
This reverts commit r198385.
llvm-svn: 198389
The greedy register allocator tries to split a live-range around each
instruction where it is used or defined to relax the constraints on the entire
live-range (this is a last chance split before falling back to spill).
The goal is to have a big live-range that is unconstrained (i.e., that can use
the largest legal register class) and several small local live-range that carry
the constraints implied by each instruction.
E.g.,
Let csti be the constraints on operation i.
V1=
op1 V1(cst1)
op2 V1(cst2)
V1 live-range is constrained on the intersection of cst1 and cst2.
tryInstructionSplit relaxes those constraints by aggressively splitting each
def/use point:
V1=
V2 = V1
V3 = V2
op1 V3(cst1)
V4 = V2
op2 V4(cst2)
Because of how the coalescer infrastructure works, each new variable (V3, V4)
that is alive at the same time as V1 (or its copy, here V2) interfere with V1.
Thus, we end up with an uncoalescable copy for each split point.
To make tryInstructionSplit less aggressive, we check if the split point
actually relaxes the constraints on the whole live-range. If it does not, we do
not insert it.
Indeed, it will not help the global allocation problem:
- V1 will have the same constraints.
- V1 will have the same interference + possibly the newly added split variable
VS.
- VS will produce an uncoalesceable copy if alive at the same time as V1.
<rdar://problem/15570057>
llvm-svn: 198369
CR logicals (crand, crxor, etc.) on the P7 need to be in the first slot of each
dispatch group. The old itinerary entry was just wrong (but has not mattered
because we don't generate these instructions).
This will matter when, in an upcoming commit, we start generating these
instructions.
llvm-svn: 198359
Several of the 64-bit fixed-point instructions with immediate operands were
using the 32-bit (i32) operand nodes instead of the corresponding 64-bit (i64)
operand definitions (u16imm instead of u16imm64, for example).
This error has had no effect so far, but would have caused type-checking
violations with an upcoming change.
llvm-svn: 198356
As noted in the comment above CodeGenPrepare::OptimizeInst, which aggressively
sinks compares to reduce pressure on the condition register(s), for targets
such as PowerPC with multiple condition registers, this may not be the right
thing to do. This adds an HasMultipleConditionRegisters boolean to TLI, and
CodeGenPrepare::OptimizeInst is skipped when HasMultipleConditionRegisters is
true.
This functionality will be used by the PowerPC backend in an upcoming commit.
Especially when the PowerPC backend starts tracking individual condition
register bits as separate allocatable entities (which will happen in this
upcoming commit), this sinking from CodeGenPrepare::OptimizeInst is
significantly suboptimial.
llvm-svn: 198354
Use an if statement instead of a pair of ternary operators checking
the same condition.
Use a cheap method call rather than returning the local symbol.
llvm-svn: 198351
Even within a multiclass, we had been generating concrete implicit anonymous
defs when parsing values (generally in value lists). This behavior was
incorrect, and led to errors when multiclass parameters were used in the
parameter list of the implicit anonymous def.
If we had some multiclass:
multiclass mc<string n> {
... : SomeClass<SomeOtherClass<n> >
The capture of the multiclass parameter 'n' would not work correctly, and
depending on how the implicit SomeOtherClass was used, either TableGen would
ignore something it shouldn't, or would crash.
To fix this problem, when inside a multiclass, we generate prototype anonymous
defs for implicit anonymous defs (just as we do for explicit anonymous defs).
Within the multiclass, the current record prototype is populated with a node
that is essentially: !cast<SomeOtherClass>(!strconcat(NAME, anon_value_name)).
This is then resolved to the correct concrete anonymous def, in the usual way,
when NAME is resolved during multiclass instantiation.
llvm-svn: 198348
TableGen had been generating a different name for an anonymous multiclass's
NAME for every def in the multiclass. This had an unfortunate side effect: it
was impossible to reference one def within the multiclass from another (in the
parameter list, for example). By making sure we only generate an anonymous name
once per multiclass (which, as it turns out, requires only changing the name
parameter to reference type), we can now concatenate NAME within the multiclass
with a def name in order to generate a reference to that def.
This does not matter so much, in and of itself, but is necessary for a
follow-up commit that will fix variable capturing in implicit anonymous
multiclass defs (and that is important).
llvm-svn: 198340
When widening an IV to remove s/zext, we generally try to eliminate
the original narrow IV. However, LCSSA phi nodes outside the loop were
still using the original IV. Clean this up more aggressively to avoid
redundancy in generated code.
llvm-svn: 198338
When there are cycles in the value graph we have to be careful interpreting
"Value*" identity as "value" equivalence. We interpret the value of a phi node
as the value of its operands.
When we check for value equivalence now we make sure that the "Value*" dominates
all cycles (phis).
%0 = phi [%noaliasval, %addr2]
%l = load %ptr
%addr1 = gep @a, 0, %l
%addr2 = gep @a, 0, (%l + 1)
store %ptr ...
Before this patch we would return NoAlias for (%0, %addr1) which is wrong
because the value of the load is from different iterations of the loop.
Tested on x86_64 -mavx at O3 and O3 -flto with no performance or compile time
regressions.
PR18068
radar://15653794
llvm-svn: 198290
During the years there have been some attempts at figuring out how to
align byval arguments. A look at the commit log suggests that they
were
* Use the ABI alignment.
* When that was not sufficient for x86-64, I added the 's' specification to
DataLayout.
* When that was not sufficient Evan added the virtual getByValTypeAlignment.
* When even that was not sufficient, we just got the FE to add the alignment
to the byval.
This patch is just a simple cleanup that removes my first attempt at fixing the
problem. I also added an AArch64 implementation of getByValTypeAlignment to
make sure this patch is a nop. I also left the 's' parsing for backward
compatibility.
I will send a short email to llvmdev about the change for anyone maintaining
an out of tree target.
llvm-svn: 198287
Venkatraman Govindaraju