Even looking at the revision history I couldn't quite piece together why this
cast was ever written in the first place, but I assume it was because of some
change in the inheritance, perhaps this function was reimplemented in a
derived type & this caller was meant to get the base version (& it wasn't
virtual)?
llvm-svn: 162301
this is the index of the operand that failed to match.
Note: This may cause a buildbot failure due to an API mismatch in clang. Should
recover with my next commit to clang.
llvm-svn: 162295
This optimization is really just replacing allocas wholesale with
globals, there is no scalarization.
The underlying motivation for this patch is to simplify the SROA pass
and focus it on splitting and promoting allocas.
llvm-svn: 162271
The getSumForBlock function was quadratic in the number of successors
because getSuccWeight would perform a linear search for an already known
iterator.
This patch was originally committed as r161460, but reverted again
because of assertion failures. Now that duplicate Machine CFG edges have
been eliminated, this works properly.
llvm-svn: 162233
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 allows for better code generation.
Added a new DAGCombine transformation to convert FMAX and FMIN to FMANC and
FMINC, which are commutative.
For example:
movaps %xmm0, %xmm1
movsd LC(%rip), %xmm0
minsd %xmm1, %xmm0
becomes:
minsd LC(%rip), %xmm0
llvm-svn: 162187
Add these transformations to the existing add/sub ones:
(and (select cc, -1, c), x) -> (select cc, x, (and, x, c))
(or (select cc, 0, c), x) -> (select cc, x, (or, x, c))
(xor (select cc, 0, c), x) -> (select cc, x, (xor, x, c))
The selects can then be transformed to a single predicated instruction
by peephole.
This transformation will make it possible to eliminate the ISD::CAND,
COR, and CXOR custom DAG nodes.
llvm-svn: 162176
arithmetic instructions. However, when small data types are used, a truncate
node appears between the SETCC node and the arithmetic operation. This patch
adds support for this pattern.
Before:
xorl %esi, %edi
testb %dil, %dil
setne %al
ret
After:
xorb %dil, %sil
setne %al
ret
rdar://12081007
llvm-svn: 162160
No new tests are added.
All tests in ExecutionEngine/MCJIT that have been failing pass after this patch
is applied (when "make check" is done on a mips board).
Patch by Petar Jovanovic.
llvm-svn: 162135
PEI can't handle the pseudo-instructions. This can be removed when the
pseudo-instructions are replaced by normal predicated instructions.
Fixes PR13628.
llvm-svn: 162130
The previous fix only checked for simple cycles, use a set to catch longer
cycles too.
Drop the broken check from the ObjectSizeOffsetEvaluator. The BoundsChecking
pass doesn't have to deal with invalid IR like InstCombine does.
llvm-svn: 162120
make it more consistent with its intended semantics.
The `linker_private_weak_def_auto' linkage type was meant to automatically hide
globals which never had their addresses taken. It has nothing to do with the
`linker_private' linkage type, which outputs the symbols with a `l' (ell) prefix
among other things.
The intended semantic is more like the `linkonce_odr' linkage type.
Change the name of the linkage type to `linkonce_odr_auto_hide'. And therefore
changing the semantics so that it produces the correct output for the linker.
Note: The old linkage name `linker_private_weak_def_auto' will still parse but
is not a synonym for `linkonce_odr_auto_hide'. This should be removed in 4.0.
<rdar://problem/11754934>
llvm-svn: 162114
multiple edges between two blocks is linear. If the caller is iterating all
edges leaving a BB that would be a square time algorithm. It is more efficient
to have the callers handle that case.
Currently the only callers are:
* GVN: already avoids the multiple edge case.
* Verifier: could only hit this assert when looking at an invalid invoke. Since
it already rejects the invoke, just avoid computing the dominance for it.
llvm-svn: 162113
Increment the MBB iterator at the top of the loop to properly handle the
current (and previous) instructions getting erased.
This fixes PR13625.
llvm-svn: 162099
It is not my plan to duplicate the entire ARM instruction set with
predicated versions. We need a way of representing predicated
instructions in SSA form without requiring a separate opcode.
Then the pseudo-instructions can go away.
llvm-svn: 162061
Select instructions pick one of two virtual registers based on a
condition, like x86 cmov. On targets like ARM that support predication,
selects can sometimes be eliminated by predicating the instruction
defining one of the operands.
Teach PeepholeOptimizer to recognize select instructions, and ask the
target to optimize them.
llvm-svn: 162059
where some fact lake a=b dominates a use in a phi, but doesn't dominate the
basic block itself.
This feature could also be implemented by splitting critical edges, but at least
with the current algorithm reasoning about the dominance directly is faster.
The time for running "opt -O2" in the testcase in pr10584 is 1.003 times slower
and on gcc as a single file it is 1.0007 times faster.
llvm-svn: 162023
Without fastcc support, the caller just falls through to CallingConv::C
for fastcc, but callee still uses fastcc, this inconsistency of calling
convention is a problem, and fastcc support can fix it.
llvm-svn: 162013
The ARM select instructions are just predicated moves. If the select is
the only use of an operand, the instruction defining the operand can be
predicated instead, saving one instruction and decreasing register
pressure.
This implementation can turn AND/ORR/EOR instructions into their
corresponding ANDCC/ORRCC/EORCC variants. Ideally, we should be able to
predicate any instruction, but we don't yet support predicated
instructions in SSA form.
llvm-svn: 161994
allocations of executable memory would not be padded
to account for the size of the allocation header.
This resulted in undersized allocations, meaning that
when the allocation was written to later the next
allocation's header would be corrupted.
llvm-svn: 161984
When predicating this instruction:
Rd = ADD Rn, Rm
We need an extra operand to represent the value given to Rd when the
predicate is false:
Rd = ADDCC Rfalse, Rn, Rm, pred
The Rd and Rfalse operands are different registers while in SSA form.
Rfalse is tied to Rd to make sure they get the same register during
register allocation.
Previously, Rd and Rn were tied, but that is not required.
Compare to MOVCC:
Rd = MOVCC Rfalse, Rtrue, pred
llvm-svn: 161955
instruction to something absurdly high, while setting the probability of
branching to the 'unwind' destination to the bare minimum. This should set cause
the normal destination's invoke blocks to be moved closer to the invoke.
PR13612
llvm-svn: 161944
- memcpy size is wrongly truncated into 32-bit and treat 8GB memcpy is
0-sized memcpy
- as 0-sized memcpy/memset is already removed before SimplifyMemTransfer
and SimplifyMemSet in visitCallInst, replace 0 checking with
assertions.
- replace getZExtValue() with getLimitedValue() according to
Eli Friedman
llvm-svn: 161923
reversed. This leads to wrong codegen for float-to-half conversion
intrinsics which are used to support storage-only fp16 type.
NEON variants of same instructions are fine.
llvm-svn: 161907
- FP_EXTEND only support extending from vectors with matching elements.
This results in the scalarization of extending to v2f64 from v2f32,
which will be legalized to v4f32 not matching with v2f64.
- add X86-specific VFPEXT supproting extending from v4f32 to v2f64.
- add BUILD_VECTOR lowering helper to recover back the original
extending from v4f32 to v2f64.
- test case is enhanced to include different vector width.
llvm-svn: 161894
Refactor the TableGen'erated fixed length disassemblmer to use a
table-driven state machine rather than a massive set of nested
switch() statements.
As a result, the ARM Disassembler (ARMDisassembler.cpp) builds much more
quickly and generates a smaller end result. For a Release+Asserts build on
a 16GB 3.4GHz i7 iMac w/ SSD:
Time to compile at -O2 (averaged w/ hot caches):
Previous: 35.5s
New: 8.9s
TEXT size:
Previous: 447,251
New: 297,661
Builds in 25% of the time previously required and generates code 66% of
the size.
Execution time of the disassembler is only slightly slower (7% disassembling
10 million ARM instructions, 19.6s vs 21.0s). The new implementation has
not yet been tuned, however, so the performance should almost certainly
be recoverable should it become a concern.
llvm-svn: 161888
and allow some optimizations to turn conditional branches into unconditional.
This commit adds a simple control-flow optimization which merges two consecutive
basic blocks which are connected by a single edge. This allows the codegen to
operate on larger basic blocks.
rdar://11973998
llvm-svn: 161852
may invalidate its AliasSet because SSAUpdater does not update the AliasSet properly.
This patch teaches SSAUpdater to notify AliasSet that it made changes.
The testcase in PR12901 is too big to be useful and I could not reduce it to a normal size.
rdar://11872059 PR12901
llvm-svn: 161803
Currently, if GetLocation reports that it did not find a valid pointer (this is the case for volatile load/stores),
we ignore the result. This patch adds code to handle the cases where we did not obtain a valid pointer.
rdar://11872864 PR12899
llvm-svn: 161802
It never does anything when running 'make check', and it get's in the
way of updating live intervals in 2-addr.
The hook was originally added to help form IT blocks in Thumb2 code
before register allocation, but the pass ordering has changed since
then, and we run if-conversion after register allocation now.
When the MI scheduler is enabled, there will be no less than two
schedulers between 2-addr and Thumb2ITBlockPass, so this hook is
unlikely to help anything.
llvm-svn: 161794
It is still possible to if-convert if the tail block has extra
predecessors, but the tail phis must be rewritten instead of being
removed.
llvm-svn: 161781
Previously, we used VLD1.32 in all cases, however there are both 16 and 64-bit
accesses being selected, so we need to use an appropriate width load in those
cases.
llvm-svn: 161748
- FCMOV only supports a subset of X86 conditions. Skip boolean
simplification if X86 condition is not valid for FCMOV.
- add a minimal test case for PR13577.
llvm-svn: 161732
FeatureFastUAMem for Nehalem, Westmere and Sandy Bridge.
FeatureFastUAMem is already on if we pass in nehalem or westmere as a command
argument.
rdar: 7252306
llvm-svn: 161717
Detect when there is not enough available ILP, so if-conversion can't
speculate instructions for free.
Compute the lengthening of the critical path when inserting a select
instruction that depends on the condition as well as both sides of the
if.
Reject conversions that would stretch the critical path by more than
half a mispredict penalty.
llvm-svn: 161713
Trace::getResourceLength() computes the number of cycles required to
execute the trace when ignoring data dependencies. The number can be
compared to the critical path to estimate the trace ILP.
Trace::getPHIDepth() computes the data dependency depth of a PHI in a
trace successor that isn't necessarily part of the trace.
llvm-svn: 161711
When a trace ends with a back-edge, include PHIs in the loop header in
the height computations. This makes the critical path through a loop
more accurate by including the latencies of the last instructions in the
loop.
llvm-svn: 161688
- if a boolean test (X86ISD::CMP or X86ISD:SUB) checks a boolean value
generated from X86ISD::SETCC, try to simplify the boolean value
generation and checking by reusing the original EFLAGS with proper
condition code
- add hooks to X86 specific SETCC/BRCOND/CMOV, the major 3 places
consuming EFLAGS
part of patches fixing PR12312
llvm-svn: 161687
When replacing Old with New, it can happen that New is already a
successor. Add the old and new edge weights instead of creating a
duplicate edge.
llvm-svn: 161653
This new attribute is intended to be used by the backend to determine how
the inline asm string should be parsed/printed. This patch adds the
ia_nsdialect attribute and also adds a test case to ensure the IR is
correctly parsed, but there is no functional change at this time.
The standard dialect is assumed to be AT&T. Therefore, this attribute
should only be added to MS-style inline assembly statements, which use
the Intel dialect. If we ever support more dialects we'll need to
add additional state to the attribute.
llvm-svn: 161641
This makes it possible to speed up def_iterator by stopping at the first
use. This makes def_empty() and getUniqueVRegDef() much faster when
there are many uses.
In a +Asserts build, LiveVariables is 100x faster in one case because
getVRegDef() has an assertion that would scan to the end of a
def_iterator chain.
Spill weight calculation is significantly faster (300x in one case)
because isTriviallyReMaterializable() calls MRI->isConstantPhysReg(%RIP)
which calls def_empty(%RIP).
llvm-svn: 161634
Use a more conventional doubly linked list where the Prev pointers form
a cycle. This means it is no longer necessary to adjust the Prev
pointers when reallocating the VRegInfo array.
The test changes are required because the register allocation hint is
using the use-list order to break ties.
llvm-svn: 161633
Register MachineOperands are kept in linked lists accessible via MRI's
reg_iterator interfaces. The linked list management was handled partly
by MachineOperand methods, partly by MRI methods.
Move all of the list management into MRI, delete
MO::AddRegOperandToRegInfo() and MO::RemoveRegOperandFromRegInfo().
Be more explicit about handling the cases where an MRI pointer isn't
available.
llvm-svn: 161632
the register info for getEncodingValue. This builds on the
small patch of yesterday to set HWEncoding in the register
file.
One (deprecated) use was turned into a hard number to avoid
needing register info in the old JIT.
llvm-svn: 161628
This new API will be used by clang to parse ms-style inline asms.
One goal of this project is to use this style of inline asm for targets other
then x86. Therefore, this API needs to be implemented for non-x86 targets at
some point in the future.
llvm-svn: 161624
This patch corrects the definition of umlal/smlal instructions and adds support
for matching them to the ARM dag combiner.
Bug 12213
Patch by Yin Ma!
llvm-svn: 161581
This way of using getNextOperandForReg() was unlikely to work as
intended. We don't give any guarantees about the order of operands in
the use-def chains, so looking only at operands following a given
operand in the chain doesn't make sense.
llvm-svn: 161542
We filter out MachineLoop back-edges during the trace-building PO
traversals, but it is possible to have CFG cycles that aren't natural
loops, and MachineLoopInfo doesn't include such cycles.
Use a standard visited set to detect such CFG cycles, and completely
ignore them when picking traces.
llvm-svn: 161532
There are situations where inline ASM may want to change the section -- for
instance, to create a variable in the .data section. However, it cannot do this
without (potentially) restoring to the wrong section. E.g.:
asm volatile (".section __DATA, __data\n\t"
".globl _fnord\n\t"
"_fnord: .quad 1f\n\t"
".text\n\t"
"1:" :::);
This may be wrong if this is inlined into a function that has a "section"
attribute. The user should use `.pushsection' and `.popsection' here instead.
The addition of `.previous' is added for completeness.
<rdar://problem/12048387>
llvm-svn: 161477
We perform the following:
1> Use SUB instead of CMP for i8,i16,i32 and i64 in ISel lowering.
2> Modify MachineCSE to correctly handle implicit defs.
3> Convert SUB back to CMP if possible at peephole.
Removed pattern matching of (a>b) ? (a-b):0 and like, since they are handled
by peephole now.
rdar://11873276
llvm-svn: 161462
We can't rematerialize a PIC base after register allocation anyway, and
scanning physreg use-def chains is very expensive in a function with
many calls.
<rdar://problem/12047515>
llvm-svn: 161461
The getSumForBlock function was quadratic in the number of successors
because getSuccWeight would perform a linear search for an already known
iterator.
llvm-svn: 161460
multiple scalar promotions on a single loop. This also has the effect of
preserving the order of stores sunk out of loops, which is aesthetically
pleasing, and it happens to fix the testcase in PR13542, though it doesn't
fix the underlying problem.
llvm-svn: 161459
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
An unsigned value converted to floating-point will always be greater than
a negative constant. Unfortunately InstCombine reversed the check so that
unsigned values were being optimized to always be greater than all positive
floating-point constants. <rdar://problem/12029145>
llvm-svn: 161452
A target index operand looks a lot like a constant pool reference, but
it is completely target-defined. It contains the 8-bit TargetFlags, a
32-bit index, and a 64-bit offset. It is preserved by all code generator
passes.
TargetIndex operands can be used to carry target-specific information in
cases where immediate operands won't suffice.
llvm-svn: 161441
Compare the critical paths of the two traces through an if-conversion
candidate. If the difference is larger than the branch brediction
penalty, reject the if-conversion. If would never pay.
llvm-svn: 161433
a use or a BB, but it is inline in the handling of the invoke instruction.
This patch refactors it so that it can be used in other cases. For example, in
define i32 @f(i32 %x) {
bb0:
%cmp = icmp eq i32 %x, 0
br i1 %cmp, label %bb2, label %bb1
bb1:
br label %bb2
bb2:
%cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
%foo = add i32 %cond, %x
ret i32 %foo
}
GVN should be able to replace %x with 0 in any use that is dominated by the
true edge out of bb0. In the above example the only such use is the one in
the phi.
llvm-svn: 161429
and "instruction address -> file/line" lookup.
Instead of plain collection of rows, debug line table for compilation unit is now
treated as the number of row ranges, describing sequences (series of contiguous machine
instructions). The sequences are not always listed in the order of increasing
address, so previously used std::lower_bound() sometimes produced wrong results.
Now the instruction address lookup consists of two stages: finding the correct
sequence, and searching for address in range of rows for this sequence.
llvm-svn: 161414
We give a bonus for every argument because the argument setup is not needed
anymore when the function is inlined. With this patch we interpret byval
arguments as a compact representation of many arguments. The byval argument
setup is implemented in the backend as an inline memcpy, so to model the
cost as accurately as possible we take the number of pointer-sized elements
in the byval argument and give a bonus of 2 instructions for every one of
those. The bonus is capped at 8 elements, which is the number of stores
at which the x86 backend switches from an expanded inline memcpy to a real
memcpy. It would be better to use the real memcpy threshold from the backend,
but it's not available via TargetData.
This change brings the performance of c-ray in line with gcc 4.7. The included
test case tries to reproduce the c-ray problem to catch regressions for this
benchmark early, its performance is dominated by the inline decision of a
specific call.
This only has a small impact on most code, more on x86 and arm than on x86_64
due to the way the ABI works. When building LLVM for x86 it gives a small
inline cost boost to virtually any function using StringRef or STL allocators,
but only a 0.01% increase in overall binary size. The size of gcc compiled by
clang actually shrunk by a couple bytes with this patch applied, but not
significantly.
llvm-svn: 161413
instsimplify+inline strategy.
The crux of the problem is that instsimplify was reasonably relying on
an invariant that is true within any single function, but is no longer
true mid-inline the way we use it. This invariant is that an argument
pointer != a local (alloca) pointer.
The fix is really light weight though, and allows instsimplify to be
resiliant to these situations: when checking the relation ships to
function arguments, ensure that the argumets come from the same
function. If they come from different functions, then none of these
assumptions hold. All credit to Benjamin Kramer for coming up with this
clever solution to the problem.
llvm-svn: 161410
Previously, MBP essentially aligned every branch target it could. This
bloats code quite a bit, especially non-looping code which has no real
reason to prefer aligned branch targets so heavily.
As Andy said in review, it's still a bit odd to do this without a real
cost model, but this at least has much more plausible heuristics.
Fixes PR13265.
llvm-svn: 161409
If the result of a common subexpression is used at all uses of the candidate
expression, CSE should not increase the live range of the common subexpression.
rdar://11393714 and rdar://11819721
llvm-svn: 161396
initialize fields of the class that it used.
The result was nonsense code.
Before:
0000000000000000 <foo>:
0: 00441100 0x441100
4: 03e00008 jr ra
8: 00000000 nop
After:
0000000000000000 <foo>:
0: 00041000 sll v0,a0,0x0
4: 03e00008 jr ra
8: 00000000 nop
llvm-svn: 161377
This allows codegen passes to query properties like
InstrItins->SchedModel->IssueWidth. It also ensure's that
computeOperandLatency returns the X86 defaults for loads and "high
latency ops". This should have no significant impact on existing
schedulers because X86 defaults happen to be the same as global
defaults.
llvm-svn: 161370
were using a class defined for 32 bit instructions and
thus the instruction was for addiu instead of daddiu.
This was corrected by adding the instruction opcode as a
field in the base class to be filled in by the defs.
llvm-svn: 161359
These 2 relocations gain access to the
highest and the second highest 16 bits
of a 64 bit object.
R_MIPS_HIGHER %higher(A+S)
The %higher(x) function is [ (((long long) x + 0x80008000LL) >> 32) & 0xffff ].
R_MIPS_HIGHEST %highest(A+S)
The %highest(x) function is [ (((long long) x + 0x800080008000LL) >> 48) & 0xffff ].
llvm-svn: 161348
The MFTB instruction itself is being phased out, and its functionality
is provided by MFSPR. According to the ISA docs, using MFSPR works on all known
chips except for the 601 (which did not have a timebase register anyway)
and the POWER3.
Thanks to Adhemerval Zanella for pointing this out!
llvm-svn: 161346
On PPC64, this can be done with a simple TableGen pattern.
To enable this, I've added the (otherwise missing) readcyclecounter
SDNode definition to TargetSelectionDAG.td.
llvm-svn: 161302
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
This option runs LiveIntervals before TwoAddressInstructionPass which
will eventually learn to exploit and update the analysis.
Eventually, LiveIntervals will run before PHIElimination, and we can get
rid of LiveVariables.
llvm-svn: 161270
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
The 'unused' state of a value number can be represented as an invalid
def SlotIndex. This also exposed code that shouldn't have been looking
at unused value VNInfos.
llvm-svn: 161258
The only real user of the flag was removeCopyByCommutingDef(), and it
has been switched to LiveIntervals::hasPHIKill().
All the code changed by this patch was only concerned with computing and
propagating the flag.
llvm-svn: 161255
The VNInfo::HAS_PHI_KILL is only half supported. We precompute it in
LiveIntervalAnalysis, but it isn't properly updated by live range
splitting and functions like shrinkToUses().
It is only used in one place: RegisterCoalescer::removeCopyByCommutingDef().
This patch changes that function to use a new LiveIntervals::hasPHIKill()
function that computes the flag for a given value number.
llvm-svn: 161254
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
This just provides a way to look up a LibFunc::Func enum value for a
function name. Alphabetize the enums and function names so we can use a
binary search.
llvm-svn: 161231
The "findUsedStructTypes" method is very expensive to run. It needs to be
optimized so that LTO can run faster. Splitting this method out of the Module
class will help this occur. For instance, it can keep a list of seen objects so
that it doesn't process them over and over again.
llvm-svn: 161228
Now that TableGen supports references to NAME w/o it being explicitly
referenced in the definition's own name, use that to simplify
assembly InstAlias definitions in multiclasses.
llvm-svn: 161218
Add more comments and use early returns to reduce nesting in isLoadFoldable.
Also disable folding for V_SET0 to avoid introducing a const pool entry and
a const pool load.
rdar://10554090 and rdar://11873276
llvm-svn: 161207
Previously, def NAME values were only populated, and references to NAME
resolved, when NAME was referenced in the 'def' entry of the multiclass
sub-entry. e.g.,
multiclass foo<...> {
def prefix_#NAME : ...
}
It's useful, however, to be able to reference NAME even when the default
def name is used. For example, when a multiclass has 'def : Pat<...>'
or 'def : InstAlias<...>' entries which refer to earlier instruction
definitions in the same multiclass. e.g.,
multiclass myMulti<RegisterClass rc> {
def _r : myI<(outs rc:$d), (ins rc:$r), "r $d, $r", []>;
def : InstAlias<\"wilma $r\", (!cast<Instruction>(NAME#\"_r\") rc:$r, rc:$r)>;
}
llvm-svn: 161198
Whenever both instruction depths and instruction heights are known in a
block, it is possible to compute the length of the critical path as
max(depth+height) over the instructions in the block.
The stored live-in lists make it possible to accurately compute the
length of a critical path that bypasses the current (small) block.
llvm-svn: 161197
LiveRangeEdit::eliminateDeadDefs() can delete a dead instruction that
reads unreserved physregs. This would leave the corresponding regunit
live interval dangling because we don't have shrinkToUses() for physical
registers.
Fix this problem by turning the instruction into a KILL instead of
deleting it. This happens in a landing pad in
test/CodeGen/X86/2012-05-19-CoalescerCrash.ll:
%vreg27<def,dead> = COPY %EDX<kill>; GR32:%vreg27
becomes:
KILL %EDX<kill>
An upcoming fix to the machine verifier will catch problems like this by
verifying regunit live intervals.
This fixes PR13498. I am not including the test case from the PR since
we already have one exposing the problem once the verifier is fixed.
llvm-svn: 161182
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.
This patch is a rework of r160919 and was tested on clang self-host on my local
machine.
rdar://10554090 and rdar://11873276
llvm-svn: 161152
The height on an instruction is the minimum number of cycles from the
instruction is issued to the end of the trace. Heights are computed for
all instructions in and below the trace center block.
The method for computing heights is different from the depth
computation. As we visit instructions in the trace bottom-up, heights of
used instructions are pushed upwards. This way, we avoid scanning long
use lists, looking for uses in the current trace.
At each basic block boundary, a list of live-in registers and their
minimum heights is saved in the trace block info. These live-in lists
are used when restarting depth computations on a trace that
converges with an already computed trace. They will also be used to
accurately compute the critical path length.
llvm-svn: 161138
Since the llvm::sys::fs::map_file_pages() support function it relies on
is not yet implemented on Windows, the unit tests for FileOutputBuffer
are currently conditionalized to run only on unix.
llvm-svn: 161099
No new test case is added.
This patch makes test JITTest.FunctionIsRecompiledAndRelinked pass on mips
platform.
Patch by Petar Jovanovic.
llvm-svn: 161098
instructions that decrement and increment the stack pointer before and after a
call when the function does not have a reserved call frame.
llvm-svn: 161093
MipsSEFrameLowering.
Implement MipsSEFrameLowering::hasReservedCallFrame. Call frames will not be
reserved if there is a call with a large call frame or there are variable sized
objects on the stack.
llvm-svn: 161090
The frame object which points to the dynamically allocated area will not be
needed after changes are made to cease reserving call frames.
llvm-svn: 161076
Assuming infinite issue width, compute the earliest each instruction in
the trace can issue, when considering the latency of data dependencies.
The issue cycle is record as a 'depth' from the beginning of the trace.
This is half the computation required to find the length of the critical
path through the trace. Heights are next.
llvm-svn: 161074
arguments to the stack in MipsISelLowering::LowerCall, use stack pointer and
integer offset operands rather than frame object operands.
llvm-svn: 161068
single-precision load and store.
Also avoid selecting LUXC1 and SUXC1 instructions during isel. It is incorrect
to map unaligned floating point load/store nodes to these instructions.
llvm-svn: 161063
One motivating example is to sink an instruction from a basic block which has
two successors: one outside the loop, the other inside the loop. We should try
to sink the instruction outside the loop.
rdar://11980766
llvm-svn: 161062
We are extending live ranges, so kill flags are not accurate. They
aren't needed until they are recomputed after RA anyway.
<rdar://problem/11950722>
llvm-svn: 161023
We branch to the successor with higher edge weight first.
Convert from
je LBB4_8 --> to outer loop
jmp LBB4_14 --> to inner loop
to
jne LBB4_14
jmp LBB4_8
PR12750
rdar: 11393714
llvm-svn: 161018
This lets traces include the final iteration of a nested loop above the
center block, and the first iteration of a nested loop below the center
block.
We still don't allow traces to contain backedges, and traces are
truncated where they would leave a loop, as seen from the center block.
llvm-svn: 161003
Empty macro arguments at the end of the list should be as-if not specified at
all, but those in the middle of the list need to be kept so as not to screw
up the positional numbering. E.g.:
.macro foo
foo_-bash___:
nop
.endm
foo 1, 2, 3, 4
foo 1, , 3, 4
Should create two labels, "foo_1_2_3_4" and "foo_1__3_4".
rdar://11948769
llvm-svn: 161002
When computing a trace, all the candidates for pred/succ must have been
visited. Filter out back-edges first, though. The PO traversal ignores
them.
Thanks to Andy for spotting this in review.
llvm-svn: 160995
where the other_half of the movt and movw relocation entries needs to get set
and only with the 16 bits of the other half.
rdar://10038370
llvm-svn: 160978
This is a cleaned up version of the isFree() function in
MachineTraceMetrics.cpp.
Transient instructions are very unlikely to produce any code in the
final output. Either because they get eliminated by RegisterCoalescing,
or because they are pseudo-instructions like labels and debug values.
llvm-svn: 160977
A->isPredecessor(B) is the same as B->isSuccessor(A), but it can
tolerate a B that is null or dangling. This shouldn't happen normally,
but it it useful for verification code.
llvm-svn: 160968
Machine CSE and other optimizations can remove instructions so folding
is possible at peephole while not possible at ISel.
rdar://10554090 and rdar://11873276
llvm-svn: 160919
It is possible that an instruction can use and update EFLAGS.
When checking the safety, we should check the usage of EFLAGS first before
declaring it is safe to optimize due to the update.
llvm-svn: 160912
A value number is a PHI def if and only if it begins at a block
boundary. This can be derived from the def slot, a separate flag is not
necessary.
llvm-svn: 160893
This option replaces the existing live interval computation with one
based on LiveRangeCalc.cpp. The new algorithm does not depend on
LiveVariables, and it can be run at any time, before or after leaving
SSA form.
llvm-svn: 160892
This can happen as long as the instruction is not reachable. Instcombine does generate these unreachable malformed selects when doing RAUW
llvm-svn: 160874
The (COPY_TO_REGCLASS GR32:$src, VR128) pattern looks odd, but
copyPhysReg does the right thing with it. (The old pattern would
eventually produce the same cross-class copy).
llvm-svn: 160830
The SUBREG_TO_REG instruction has magic semantics asserting that the
source value was defined by an instruction that cleared the high half of
the register. Those semantics are never actually exploited for xmm
registers.
llvm-svn: 160818
These idempotent sub-register indices don't do anything --- They simply
map XMM registers to themselves. They no longer affect register classes
either since the SubRegClasses field has been removed from Target.td.
This patch replaces XMM->XMM EXTRACT_SUBREG and INSERT_SUBREG patterns
with COPY_TO_REGCLASS patterns which simply become COPY instructions.
The number of IMPLICIT_DEF instructions before register allocation is
reduced, and that is the cause of the test case changes.
llvm-svn: 160816
This is still a work in progress.
Out-of-order CPUs usually execute instructions from multiple basic
blocks simultaneously, so it is necessary to look at longer traces when
estimating the performance effects of code transformations.
The MachineTraceMetrics analysis will pick a typical trace through a
given basic block and provide performance metrics for the trace. Metrics
will include:
- Instruction count through the trace.
- Issue count per functional unit.
- Critical path length, and per-instruction 'slack'.
These metrics can be used to determine the performance limiting factor
when executing the trace, and how it will be affected by a code
transformation.
Initially, this will be used by the early if-conversion pass.
llvm-svn: 160796
It is redundant; RegisterCoalescer will do the remat if it can't eliminate
the copy. Collected instruction counts before and after this. A few extra
instructions are generated due to spilling but it is normal to see these kinds
of changes with almost any small codegen change, according to Jakob.
This also fixed rdar://11830760 where xor is expected instead of movi0.
llvm-svn: 160749
When a live range splits into multiple connected components, we would
arbitrarily assign <undef> uses to component 0. This is wrong when the
use is tied to a def that gets assigned to a different component:
%vreg69<def> = ADD8ri %vreg68<undef>, 1
The use and def must get the same virtual register.
Fix this by assigning <undef> uses to the same component as the value
defined by the instruction, if any:
%vreg69<def> = ADD8ri %vreg69<undef>, 1
This fixes PR13402. The PR has a test case which I am not including
because it is unlikely to keep exposing this behavior in the future.
llvm-svn: 160739
of an array element (rather than at the beginning of the element) and extended
into the next element, then the load from the second element was being handled
wrong due to incorrect updating of the notion of which byte to load next. This
fixes PR13442. Thanks to Chris Smowton for reporting the problem, analyzing it
and providing a fix.
llvm-svn: 160711
The long branch pass (fixed in r160601) no longer uses the global base register
to compute addresses of branch destinations, so it is not necessary to reserve
a slot on the stack.
llvm-svn: 160703
if Condition Is Met instuctions that was not correctly determining the target
instruction.
So for a jne rel32 instruction:
% cat x.s
.byte 0x0f, 0x85, 0x09, 0x00, 0x00, 0x00
% as x.s
it was incorrectly deterining the target:
% otool -q -tv a.out
a.out:
(__TEXT,__text) section
0000000000000000 jne 0xd
and with the fix it gets this correct as:
% otool -q -tv a.out
a.out:
(__TEXT,__text) section
0000000000000000 jne 0xf
rdar://11505997
llvm-svn: 160694
are targeting an ELF platform. Only fold gs-relative (and fs-relative) loads
if it is actually sensible to do so for the target platform.
This fixes PR13438.
llvm-svn: 160687
Richard Smith