This is important for the correct lowering of unwind instructions
(which doesn't matter at all) and llvm.eh.resume calls (which does).
llvm-svn: 132291
to load/store i64 values. Since there's no current support to explicitly
declare such restrictions, implement it by using specific hardcoded register
pairs during isel.
llvm-svn: 132248
register allocation dependent and will occasionally break. WIP in the
register allocator to model paired/etc registers.
rdar://9119939
llvm-svn: 132242
The practical effects here are that x86-64 fast-isel can now handle trunc from i8 to i1, and ARM fast-isel can handle many more constructs involving integers narrower than 32 bits (including loads, stores, and many integer casts).
rdar://9437928 .
llvm-svn: 132099
When instructions are deleted, they leave tombstone SlotIndex entries.
The isZeroLength method should ignore these null indexes.
This causes RABasic to sometimes spill a callee-saved register in the
abi-isel.ll test, so don't run that test with -regalloc=basic. Prioritizing
register allocation according to spill weight can cause more registers to be
used.
llvm-svn: 131436
If there is a store after the load node, then there is a chain, which means
that there is another user. Thus, asking hasOneUser would fail. Instead we
ask hasNUsesOfValue on the 'data' value.
llvm-svn: 131183
landing pad as its successor.
SjLj exception handling jumps to the correct landing pad via a switch statement
that's generated right before code-gen. Loosen the constraint in the machine
instruction verifier to allow for this. Note, this isn't the most rigorous check
since we cannot determine where that switch statement came from. But it's
marginally better than turning this check off when SjLj exceptions are used.
<rdar://problem/9187612>
llvm-svn: 130881
model constants which can be added to base registers via add-immediate
instructions which don't require an additional register to materialize
the immediate.
llvm-svn: 130743
successors) and use inverse depth first search to traverse the BBs. However
that doesn't work when the CFG has infinite loops. Simply do a linear
traversal of all BBs work just fine.
rdar://9344645
llvm-svn: 130324
more callee-saved registers and introduce copies. Only allows it if scheduling
a node above calls would end up lessen register pressure.
Call operands also has added ABI restrictions for register allocation, so be
extra careful with hoisting them above calls.
rdar://9329627
llvm-svn: 130245
Fixes Thumb2 ADCS and SBCS lowering: <rdar://problem/9275821>.
t2ADCS/t2SBCS are now pseudo instructions, consistent with ARM, so the
assembly printer correctly prints the 's' suffix.
Fixes Thumb2 adde -> SBC matching to check for live/dead carry flags.
Fixes the internal ARM machine opcode mnemonic for ADCS/SBCS.
Fixes ARM SBC lowering to check for live carry (potential bug).
llvm-svn: 130048
manually and pass all (now) 4 arguments to the mul libcall. Add a new
ExpandLibCall for just this (copied gratuitously from type legalization).
Fixes rdar://9292577
llvm-svn: 129842
- There is a minor semantic change here (evidenced by the test change) for
Darwin triples that have no version component. I debated changing the default
behavior of isOSVersionLT, but decided it made more sense for triples to be
explicit.
llvm-svn: 129802
Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
additional pipeline stall. So it's frequently better to single codegen
vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
vmla + vmla is very bad. But this isn't ideal either:
vmul
vadd
vmla
Instead, we want to expand the second vmla:
vmla
vmul
vadd
Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
faster.
Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.
A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
vmla / vmls will trigger one of the special hazards.
Enable these fp vmlx codegen changes for Cortex-A9.
llvm-svn: 129775
Add a avoidWriteAfterWrite() target hook to identify register classes that
suffer from write-after-write hazards. For those register classes, try to avoid
writing the same register in two consecutive instructions.
This is currently disabled by default. We should not spill to avoid hazards!
The command line flag -avoid-waw-hazard can be used to enable waw avoidance.
llvm-svn: 129772
registers for fast allocation a different way. This has us updating
used registers only when we're using that exact register.
Fixes rdar://9207598
llvm-svn: 129711
the max itself, so it is not easy to write a test case for this, but I added a
test case that would fail if the code in AsmPrinter were removed.
llvm-svn: 129432
Additional fixes:
Do something reasonable for subtargets with generic
itineraries by handle node latency the same as for an empty
itinerary. Now nodes default to unit latency unless an itinerary
explicitly specifies a zero cycle stage or it is a TokenFactor chain.
Original fixes:
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make the ndoe latency adjustments work, I also
needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.
llvm-svn: 129421
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make these heuristic adjustments to node latency work,
I also needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.
llvm-svn: 129383
induction variable. The preRA scheduler is unaware of induction vars,
so we look for potential "virtual register cycles" instead.
Fixes <rdar://problem/8946719> Bad scheduling prevents coalescing
llvm-svn: 129100
registers that arise from argument shuffling with the soft float ABI. These
instructions are particularly slow on Cortex A8. This fixes one half of
<rdar://problem/8674845>.
llvm-svn: 128759
The rematerialized instruction may require a more constrained register class
than the register being spilled. In the test case, the spilled register has been
inflated to the DPR register class, but we are rematerializing a load of the
ssub_0 sub-register which only exists for DPR_VFP2 registers.
The register class is reinflated after spilling, so the conservative choice is
only temporary.
llvm-svn: 128610
was lowering them to sext / uxt + mul instructions. Unfortunately the
optimization passes may hoist the extensions out of the loop and separate them.
When that happens, the long multiplication instructions can be broken into
several scalar instructions, causing significant performance issue.
Note the vmla and vmls intrinsics are not added back. Frontend will codegen them
as intrinsics vmull* + add / sub. Also note the isel optimizations for catching
mul + sext / zext are not changed either.
First part of rdar://8832507, rdar://9203134
llvm-svn: 128502
isel lowering to fold the zero-extend's and take advantage of no-stall
back to back vmul + vmla:
vmull q0, d4, d6
vmlal q0, d5, d6
is faster than
vaddl q0, d4, d5
vmovl q1, d6
vmul q0, q0, q1
This allows us to vmull + vmlal for:
f = vmull_u8( vget_high_u8(s), c);
f = vmlal_u8(f, vget_low_u8(s), c);
rdar://9197392
llvm-svn: 128444
int tries = INT_MAX;
while (tries > 0) {
tries--;
}
The check should be:
subs r4, #1
cmp r4, #0
bgt LBB0_1
The subs can set the overflow V bit when r4 is INT_MAX+1 (which loop
canonicalization apparently does in this case). cmp #0 would have cleared
it while not changing the N and Z bits. Since BGT is dependent on the V
bit, i.e. (N == V) && !Z, it is not safe to eliminate the cmp #0.
rdar://9172742
llvm-svn: 128179
v2 = bitcast v1
...
v3 = bitcast v2
...
= v3
=>
v2 = bitcast v1
...
= v1
if v1 and v3 are of in the same register class.
bitcast between i32 and fp (and others) are often not nops since they
are in different register classes. These bitcast instructions are often
left because they are in different basic blocks and cannot be
eliminated by dag combine.
rdar://9104514
llvm-svn: 127668
Also more cleanly separate the ARM vs. Thumb functionality. Previously, the
encoding would be incorrect for some Thumb instructions (the indirect calls).
llvm-svn: 127637
Optimize trivial branches in CodeGenPrepare, which often get created from the
lowering of objectsize intrinsics. Unfortunately, a number of tests were relying
on llc not optimizing trivial branches, so I had to add an option to allow them
to continue to test what they originally tested.
This fixes <rdar://problem/8785296> and <rdar://problem/9112893>.
llvm-svn: 127498
lowering of objectsize intrinsics. Unfortunately, a number of tests were relying
on llc not optimizing trivial branches, so I had to add an option to allow them
to continue to test what they originally tested.
This fixes <rdar://problem/8785296> and <rdar://problem/9112893>.
llvm-svn: 127459
The previous codegen for the slow path (when values are in VFP / NEON
registers) was incorrect if the source is NaN.
The new codegen uses NEON vbsl instruction to copy the sign bit. e.g.
vmov.i32 d1, #0x80000000
vbsl d1, d2, d0
If NEON is not available, it uses integer instructions to copy the sign bit.
rdar://9034702
llvm-svn: 126295
The i64_buildvector test in this file relies on the alignment of i64 and
f64 types being the same, which is true for Darwin but not AAPCS.
llvm-svn: 125525
This
define float @foo(float %x, float %y) nounwind readnone {
entry:
%0 = tail call float @copysignf(float %x, float %y) nounwind readnone
ret float %0
}
Was compiled to:
vmov s0, r1
bic r0, r0, #-2147483648
vmov s1, r0
vcmpe.f32 s0, #0
vmrs apsr_nzcv, fpscr
it lt
vneglt.f32 s1, s1
vmov r0, s1
bx lr
This fails to copy the sign of -0.0f because it's lost during the float to int
conversion. Also, it's sub-optimal when the inputs are in GPR registers.
Now it uses integer and + or operations when it's profitable. And it's correct!
lsrs r1, r1, #31
bfi r0, r1, #31, #1
bx lr
rdar://8984306
llvm-svn: 125357
The vld1-lane, vld1-dup and vst1-lane instructions do not yet support using
post-increment versions, but all the rest of the NEON load/store instructions
should be handled now.
llvm-svn: 125014
the load, then it may be legal to transform the load and store to integer
load and store of the same width.
This is done if the target specified the transformation as profitable. e.g.
On arm, this can transform:
vldr.32 s0, []
vstr.32 s0, []
to
ldr r12, []
str r12, []
rdar://8944252
llvm-svn: 124708
1. Fixed ARM pc adjustment.
2. Fixed dynamic-no-pic codegen
3. CSE of pc-relative load of global addresses.
It's now enabled by default for Darwin.
llvm-svn: 123991
DAG. Disable using "-disable-sched-cycles".
For ARM, this enables a framework for modeling the cpu pipeline and
counting stalls. It also activates several heuristics to drive
scheduling based on the model. Scheduling is inherently imprecise at
this stage, and until spilling is improved it may defeat attempts to
schedule. However, this framework provides greater control over
tuning codegen.
Although the flag is not target-specific, it should have very little
affect on the default scheduler used by x86. The only two changes that
affect x86 are:
- scheduling a high-latency operation bumps the current cycle so independent
operations can have their latency covered. i.e. two independent 4
cycle operations can produce results in 4 cycles, not 8 cycles.
- Two operations with equal register pressure impact and no
latency-based stalls on their uses will be prioritized by depth before height
(height is irrelevant if no stalls occur in the schedule below this point).
llvm-svn: 123971
flags. They are still not enable in this revision.
Added TargetInstrInfo::isZeroCost() to fix a fundamental problem with
the scheduler's model of operand latency in the selection DAG.
Generalized unit tests to work with sched-cycles.
llvm-svn: 123969
value, the "add pc" must be CSE'ed at the same time. We could follow the same
approach as T2 by adding pseudo instructions that combine the ldr + "add pc".
But the better approach is to use movw + movt (which I will enable soon), so
I'll leave this as a TODO.
llvm-svn: 123949
TargetInstrInfo:
Change produceSameValue() to take MachineRegisterInfo as an optional argument.
When in SSA form, targets can use it to make more aggressive equality analysis.
Machine LICM:
1. Eliminate isLoadFromConstantMemory, use MI.isInvariantLoad instead.
2. Fix a bug which prevent CSE of instructions which are not re-materializable.
3. Use improved form of produceSameValue.
ARM:
1. Teach ARM produceSameValue to look pass some PIC labels.
2. Look for operands from different loads of different constant pool entries
which have same values.
3. Re-implement PIC GA materialization using movw + movt. Combine the pair with
a "add pc" or "ldr [pc]" to form pseudo instructions. This makes it possible
to re-materialize the instruction, allow machine LICM to hoist the set of
instructions out of the loop and make it possible to CSE them. It's a bit
hacky, but it significantly improve code quality.
4. Some minor bug fixes as well.
With the fixes, using movw + movt to materialize GAs significantly outperform the
load from constantpool method. 186.crafty and 255.vortex improved > 20%, 254.gap
and 176.gcc ~10%.
llvm-svn: 123905
with an invalid type then split the result and perform the overflow check
normally.
Fixes the 32-bit parts of rdar://8622122 and rdar://8774702.
llvm-svn: 123864
movw r0, :lower16:(L_foo$non_lazy_ptr-(LPC0_0+4))
movt r0, :upper16:(L_foo$non_lazy_ptr-(LPC0_0+4))
LPC0_0:
add r0, pc, r0
It's not yet enabled by default as some tests are failing. I suspect bugs in
down stream tools.
llvm-svn: 123619
Also fix an off-by-one in SelectionDAGBuilder that was preventing shuffle
vectors from being translated to EXTRACT_SUBVECTOR.
Patch by Tim Northover.
The test changes are needed to keep those spill-q tests from testing aligned
spills and restores. If the only aligned stack objects are spill slots, we
no longer realign the stack frame. Prior to this patch, an EXTRACT_SUBVECTOR
was legalized by loading from the stack, which created an aligned frame index.
Now, however, there is nothing except the spill slot in the stack frame, so
I added an aligned alloca.
llvm-svn: 122995
If the basic block containing the BCCi64 (or BCCZi64) instruction ends with
an unconditional branch, that branch needs to be deleted before appending
the expansion of the BCCi64 to the end of the block.
llvm-svn: 122521
Type legalization splits up i64 values into pairs of i32 values, which leads
to poor quality code when inserting or extracting i64 vector elements.
If the vector element is loaded or stored, it can be treated as an f64 value
and loaded or stored directly from a VPR register. Use the pre-legalization
DAG combiner to cast those vector elements to f64 types so that the type
legalizer won't mess them up. Radar 8755338.
llvm-svn: 122319
may be called. If the entry block is empty, the insertion point iterator will be
the "end()" value. Calling ->getParent() on it (among others) causes problems.
Modify materializeFrameBaseRegister to take the machine basic block and insert
the frame base register at the beginning of that block. (It's very similar to
what the code does all ready. The only difference is that it will always insert
at the beginning of the entry block instead of after a previous materialization
of the frame base register. I doubt that that matters here.)
<rdar://problem/8782198>
llvm-svn: 122104
BUILD_VECTOR operands where the element type is not legal. I had previously
changed this code to insert TRUNCATE operations, but that was just wrong.
llvm-svn: 122102
Alignments smaller than the total size of the memory being loaded or stored,
unless the alignment is 8 bytes, are not allowed. Add tests for this, too.
llvm-svn: 121506
Added test to check bl __aeabi_read_tp gets emitted properly for ELF/ASM
as well as ELF/OBJ (including fixup)
Also added support for ELF::R_ARM_TLS_IE32
llvm-svn: 121312
vpush instructions to save / restore VFP / NEON registers like this:
vpush {d8,d10,d11}
vpop {d8,d10,d11}
vpush and vpop do not allow gaps in the register list.
rdar://8728956
llvm-svn: 121197
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
additional pipeline stall. So it's frequently better to single codegen
vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
vmla + vmla is very bad. But this isn't ideal either:
vmul
vadd
vmla
Instead, we want to expand the second vmla:
vmla
vmul
vadd
Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
faster.
Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.
A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
vmla / vmls will trigger one of the special hazards.
Work in progress, only A+B are enabled.
llvm-svn: 120960
Lifted adjustFixupValue() from Darwin for sharing w ELF.
Test added
TODO:
refactor ELFObjectWriter::RecordRelocation more.
Possibly share more code with Darwin?
Lots more relocations...
llvm-svn: 120534
legalization time. Since at legalization time there is no mapping from
SDNode back to the corresponding LLVM instruction and the return
SDNode is target specific, this requires a target hook to check for
eligibility. Only x86 and ARM support this form of sibcall optimization
right now.
rdar://8707777
llvm-svn: 120501
We need to check if the individual vector elements are sign/zero-extended
values. For now this only handles constants values. Radar 8687140.
llvm-svn: 120034
state. Previously Thumb2 would restore sp from fp like this:
mov sp, r7
sub, sp, #4
If an interrupt is taken after the 'mov' but before the 'sub', callee-saved
registers might be clobbered by the interrupt handler. Instead, try
restoring directly from sp:
add sp, #4
Or, if necessary (with VLA, etc.) use a scratch register to compute sp and
then restore it:
sub.w r4, r7, #8
mov sp, r7
rdar://8465407
llvm-svn: 119977
