merging an lsl #2 that has multiple uses on A9. This shift is free, so there is
no problem merging it in multiple places. Other unprofitable shifts will not be
merged.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@141247 91177308-0d34-0410-b5e6-96231b3b80d8
Math is hard, and isScaledConstantInRange() always returned false for
negative constants. It was doing unsigned division of negative numbers
before casting back to signed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@140425 91177308-0d34-0410-b5e6-96231b3b80d8
Add the predicate operand to the instructions. Update the back end
accordingly where the instructions are used. Restrict the SP operands
to actually only be SP, as otherwise these break assembly parsing for the
normal instruction variants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138445 91177308-0d34-0410-b5e6-96231b3b80d8
Refactor STR[B] pre and post indexed instructions to use addressing modes for
memory operands, which is necessary for assembly parsing and is more consistent
with the rest of the memory instruction definitions. Make some incremental
progress on refactoring away the mega-operand addrmode2 along the way, which
is nice.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136978 91177308-0d34-0410-b5e6-96231b3b80d8
Encode the width operand as it encodes in the instruction, which simplifies
the disassembler and the encoder, by using the imm1_32 operand def. Add a
diagnostic for the context-sensitive constraint that the width must be in
the range [1,32-lsb].
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136264 91177308-0d34-0410-b5e6-96231b3b80d8
sink them into MC layer.
- Added MCInstrInfo, which captures the tablegen generated static data. Chang
TargetInstrInfo so it's based off MCInstrInfo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134021 91177308-0d34-0410-b5e6-96231b3b80d8
This is intended to support using REG_SEQUENCE SDNode's with type MVT::untyped, and is part of the long road to eliminating some of the hacks we currently use to support register pairs and other strange constraints, particularly on ARM NEON.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133178 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@132248 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129775 91177308-0d34-0410-b5e6-96231b3b80d8
have their low bits set to zero. This allows us to optimize
out explicit stack alignment code like in stack-align.ll:test4 when
it is redundant.
Doing this causes the code generator to start turning FI+cst into
FI|cst all over the place, which is general goodness (that is the
canonical form) except that various pieces of the code generator
don't handle OR aggressively. Fix this by introducing a new
SelectionDAG::isBaseWithConstantOffset predicate, and using it
in places that are looking for ADD(X,CST). The ARM backend in
particular was missing a lot of addressing mode folding opportunities
around OR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125470 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125014 91177308-0d34-0410-b5e6-96231b3b80d8
These operations are expanded to pairs of loads or stores, and the first one
uses the address register update to produce the address for the second one.
So far, the second load/store has also updated the address register, just
for convenience, since that output has never been used. In anticipation of
actually supporting post-increment updates for these operations, this changes
the non-updating operations to use a non-updating load/store for the second
instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125013 91177308-0d34-0410-b5e6-96231b3b80d8
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%.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123905 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123619 91177308-0d34-0410-b5e6-96231b3b80d8
earlyclobber stuff. This should fix PRs 2313 and 8157.
Unfortunately, no testcase, since it'd be dependent on register
assignments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122663 91177308-0d34-0410-b5e6-96231b3b80d8
instruction based on the t_addrmode_s# mode and what it returned. There is some
obvious badness to this. In particular, it's hard to do MC-encoding when the
instruction may change out from underneath you after the t_addrmode_s# variable
is finally resolved.
The solution is to revert a long-ago change that merged the reg/reg and reg/imm
versions. There is the addition of several new addressing modes. They no longer
have extraneous operands associated with them. I.e., if it's reg/reg we don't
have to have a dummy zero immediate tacked on to the SDNode.
There are some obvious cleanups here, which will happen shortly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121747 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121506 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@120960 91177308-0d34-0410-b5e6-96231b3b80d8
