llvm-project/llvm/lib/Target/ARM
Rafael Espindola daeafb4c2a Add back r201608, r201622, r201624 and r201625
r201608 made llvm corretly handle private globals with MachO. r201622 fixed
a bug in it and r201624 and r201625 were changes for using private linkage,
assuming that llvm would do the right thing.

They all got reverted because r201608 introduced a crash in LTO. This patch
includes a fix for that. The issue was that TargetLoweringObjectFile now has
to be initialized before we can mangle names of private globals. This is
trivially true during the normal codegen pipeline (the asm printer does it),
but LTO has to do it manually.

llvm-svn: 201700
2014-02-19 17:23:20 +00:00
..
AsmParser Fix the arm assembler so that this malformed instruction: 2014-02-17 21:45:27 +00:00
Disassembler [CMake] Let add_public_tablegen_target responsible to provide dependency to CommonTableGen. 2013-11-28 17:04:04 +00:00
InstPrinter ARM IAS: properly handle expression operands 2014-01-08 03:28:14 +00:00
MCTargetDesc Re-commit: Demote EmitRawText call in AsmPrinter::EmitInlineAsm() and remove hasRawTextSupport() call 2014-02-13 14:44:26 +00:00
TargetInfo Prune redundant dependencies in LLVMBuild.txt. 2013-12-11 00:30:57 +00:00
A15SDOptimizer.cpp Re-sort all of the includes with ./utils/sort_includes.py so that 2014-01-07 11:48:04 +00:00
ARM.h Adding an A15 specific optimization pass for interactions between S/D/Q registers. The pass handles all the required transformations pre-regalloc. 2013-03-15 18:28:25 +00:00
ARM.td Make the ARM ABI selectable via SubtargetFeature. 2014-01-02 13:40:08 +00:00
ARMAsmPrinter.cpp Pass the Mangler by reference. 2014-02-08 14:53:28 +00:00
ARMAsmPrinter.h Remove const_cast for STI when parsing inline asm 2014-02-06 18:19:40 +00:00
ARMBaseInstrInfo.cpp Test commit 2014-01-29 16:01:24 +00:00
ARMBaseInstrInfo.h ARM: decide whether to use movw/movt based on "minsize" attribute. 2013-12-02 14:46:26 +00:00
ARMBaseRegisterInfo.cpp Revert r200095 and r200152. It turns out when compiling with -arch armv7 -mcpu=cortex-m3, the triple would still set iOS as the OS so the hack is still needed. rdar://15984891 2014-02-06 18:51:34 +00:00
ARMBaseRegisterInfo.h ARM: r12 is callee-saved for interrupt handlers 2014-02-10 14:24:23 +00:00
ARMCallingConv.h Fix for 5.5 Parameter Passing --> Stage C: 2013-04-22 13:06:52 +00:00
ARMCallingConv.td Tweak ARM fastcc by adopting these two AAPCS rules: 2014-02-11 23:49:31 +00:00
ARMCodeEmitter.cpp [ARMv8] Add support for the NEON instructions vmaxnm/vminnm. 2013-07-17 13:59:38 +00:00
ARMConstantIslandPass.cpp ARM: Fix crash in ARM backend inside of ARMConstantIslandPass 2013-06-04 17:46:15 +00:00
ARMConstantPoolValue.cpp ARM: Deduplicate ConstantPoolValues. 2013-09-16 10:17:31 +00:00
ARMConstantPoolValue.h ARM: Deduplicate ConstantPoolValues. 2013-09-16 10:17:31 +00:00
ARMExpandPseudoInsts.cpp Remove the useless pseudo instructions VDUPfdf and VDUPfqf, replacing them with patterns to match VDUPLN. 2014-01-20 17:14:48 +00:00
ARMFPUName.def [arm] Add softvfp to supported FPU names. 2014-01-02 15:50:02 +00:00
ARMFPUName.h [arm] Implement eabi_attribute, cpu, and fpu directives. 2013-10-28 17:51:12 +00:00
ARMFastISel.cpp Rename a DebugLoc variable to DbgLoc and a DataLayout to DL. 2014-02-18 22:05:46 +00:00
ARMFeatures.h [Thumbv8] Fix the value of BLXOperandIndex of isV8EligibleForIT 2014-01-23 19:55:33 +00:00
ARMFrameLowering.cpp Generate the DWARF stack frame decode operations in the function prologue for ARM/Thumb functions. 2014-02-14 17:19:07 +00:00
ARMFrameLowering.h Move the eliminateCallFramePseudoInstr method from TargetRegisterInfo 2013-02-21 20:05:00 +00:00
ARMHazardRecognizer.cpp Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
ARMHazardRecognizer.h Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
ARMISelDAGToDAG.cpp Fix known typos 2014-01-24 17:20:08 +00:00
ARMISelLowering.cpp Use 16 byte stack alignment for NaCl on ARM 2014-02-16 18:59:48 +00:00
ARMISelLowering.h ARM: use natural LLVM IR for vshll instructions 2014-02-10 16:20:29 +00:00
ARMInstrFormats.td ARM: use LLVM IR to represent the vshrn operation 2014-02-10 14:04:07 +00:00
ARMInstrInfo.cpp ARM: Use the PICADD opcode calculated. 2013-09-10 17:21:39 +00:00
ARMInstrInfo.h
ARMInstrInfo.td Fix unsupported addressing mode assertion for pld 2014-01-27 21:39:04 +00:00
ARMInstrNEON.td ARM: remove floating-point patterns for @llvm.arm.neon.vabs 2014-02-13 10:44:30 +00:00
ARMInstrThumb.td ARM: constrain Thumb LDRLIT pseudo-instructions to r0-r7. 2014-01-13 14:19:17 +00:00
ARMInstrThumb2.td ARM: Thumb2 LDR(literal) can target SP. 2014-02-11 20:48:39 +00:00
ARMInstrVFP.td Add ARM fconsts/fconstd aliases for vmov.f32/vmov.f64 2014-01-07 18:19:23 +00:00
ARMJITInfo.cpp
ARMJITInfo.h
ARMLoadStoreOptimizer.cpp ARM: correct liveness flags during ARMLoadStoreOpt 2013-10-10 09:28:20 +00:00
ARMMCInstLower.cpp Move the llvm mangler to lib/IR. 2014-01-07 21:19:40 +00:00
ARMMachineFunctionInfo.cpp
ARMMachineFunctionInfo.h Use 16 byte stack alignment for NaCl on ARM 2014-02-16 18:59:48 +00:00
ARMPerfectShuffle.h
ARMRegisterInfo.cpp Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
ARMRegisterInfo.h Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
ARMRegisterInfo.td Fix known typos 2014-01-24 17:20:08 +00:00
ARMRelocations.h
ARMSchedule.td ARM sched model: Add integer VFP/SIMD instructions on Swift 2013-06-06 20:26:18 +00:00
ARMScheduleA8.td
ARMScheduleA9.td New machine model for cortex-a9. Schedule for resources and latency. 2013-12-28 21:57:05 +00:00
ARMScheduleSwift.td Fix known typos 2014-01-24 17:20:08 +00:00
ARMScheduleV6.td
ARMSelectionDAGInfo.cpp ARM MachO: sort out isTargetDarwin/isTargetIOS/... checks. 2014-01-06 14:28:05 +00:00
ARMSelectionDAGInfo.h Track IR ordering of SelectionDAG nodes 2/4. 2013-05-25 02:42:55 +00:00
ARMSubtarget.cpp Use 16 byte stack alignment for NaCl on ARM 2014-02-16 18:59:48 +00:00
ARMSubtarget.h Unaligned access is supported on ARMv6 and ARMv7 for the NetBSD target. 2014-02-02 21:18:36 +00:00
ARMTargetMachine.cpp GlobalMerge: move "-global-merge" option to the pass itself. 2014-02-18 11:17:29 +00:00
ARMTargetMachine.h Adding simple cast cost to ARM 2013-01-29 23:31:38 +00:00
ARMTargetObjectFile.cpp Add back r201608, r201622, r201624 and r201625 2014-02-19 17:23:20 +00:00
ARMTargetObjectFile.h Add back r201608, r201622, r201624 and r201625 2014-02-19 17:23:20 +00:00
ARMTargetTransformInfo.cpp ARMTTI: We don't have 16 allocatable scalar registers 2014-02-01 18:00:25 +00:00
CMakeLists.txt [CMake] Let add_public_tablegen_target() provide intrinsics_gen, too. 2013-11-28 17:04:31 +00:00
LICENSE.TXT Text files should not be marked executable. 2013-04-30 19:06:15 +00:00
LLVMBuild.txt Add proper dependencies to LLVMBuild.txt in llvm/lib. 2013-12-10 05:39:34 +00:00
MLxExpansionPass.cpp
Makefile
README-Thumb.txt Testing commit access to llvm. Remove two lines of whitespace from the Thumb README. 2013-03-28 18:26:15 +00:00
README-Thumb2.txt
README.txt
Thumb1FrameLowering.cpp Generate the DWARF stack frame decode operations in the function prologue for ARM/Thumb functions. 2014-02-14 17:19:07 +00:00
Thumb1FrameLowering.h Move the eliminateCallFramePseudoInstr method from TargetRegisterInfo 2013-02-21 20:05:00 +00:00
Thumb1InstrInfo.cpp Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
Thumb1InstrInfo.h
Thumb1RegisterInfo.cpp ARM: remove unnecessary state-tracking during frame lowering. 2013-11-04 23:04:15 +00:00
Thumb1RegisterInfo.h Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
Thumb2ITBlockPass.cpp Enable generating legacy IT block for AArch32 2013-11-13 18:29:49 +00:00
Thumb2InstrInfo.cpp ARM: remove unnecessary state-tracking during frame lowering. 2013-11-04 23:04:15 +00:00
Thumb2InstrInfo.h
Thumb2RegisterInfo.cpp Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
Thumb2RegisterInfo.h Don't cache the instruction and register info from the TargetMachine, because 2013-06-07 05:54:19 +00:00
Thumb2SizeReduction.cpp Fix PR 18369: [Thumbv8] asserts due to inconsistent CPSR liveness of IT blocks 2014-01-13 18:47:54 +00:00

README.txt

//===---------------------------------------------------------------------===//
// Random ideas for the ARM backend.
//===---------------------------------------------------------------------===//

Reimplement 'select' in terms of 'SEL'.

* We would really like to support UXTAB16, but we need to prove that the
  add doesn't need to overflow between the two 16-bit chunks.

* Implement pre/post increment support.  (e.g. PR935)
* Implement smarter constant generation for binops with large immediates.

A few ARMv6T2 ops should be pattern matched: BFI, SBFX, and UBFX

Interesting optimization for PIC codegen on arm-linux:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43129

//===---------------------------------------------------------------------===//

Crazy idea:  Consider code that uses lots of 8-bit or 16-bit values.  By the
time regalloc happens, these values are now in a 32-bit register, usually with
the top-bits known to be sign or zero extended.  If spilled, we should be able
to spill these to a 8-bit or 16-bit stack slot, zero or sign extending as part
of the reload.

Doing this reduces the size of the stack frame (important for thumb etc), and
also increases the likelihood that we will be able to reload multiple values
from the stack with a single load.

//===---------------------------------------------------------------------===//

The constant island pass is in good shape.  Some cleanups might be desirable,
but there is unlikely to be much improvement in the generated code.

1.  There may be some advantage to trying to be smarter about the initial
placement, rather than putting everything at the end.

2.  There might be some compile-time efficiency to be had by representing
consecutive islands as a single block rather than multiple blocks.

3.  Use a priority queue to sort constant pool users in inverse order of
    position so we always process the one closed to the end of functions
    first. This may simply CreateNewWater.

//===---------------------------------------------------------------------===//

Eliminate copysign custom expansion. We are still generating crappy code with
default expansion + if-conversion.

//===---------------------------------------------------------------------===//

Eliminate one instruction from:

define i32 @_Z6slow4bii(i32 %x, i32 %y) {
        %tmp = icmp sgt i32 %x, %y
        %retval = select i1 %tmp, i32 %x, i32 %y
        ret i32 %retval
}

__Z6slow4bii:
        cmp r0, r1
        movgt r1, r0
        mov r0, r1
        bx lr
=>

__Z6slow4bii:
        cmp r0, r1
        movle r0, r1
        bx lr

//===---------------------------------------------------------------------===//

Implement long long "X-3" with instructions that fold the immediate in.  These
were disabled due to badness with the ARM carry flag on subtracts.

//===---------------------------------------------------------------------===//

More load / store optimizations:
1) Better representation for block transfer? This is from Olden/power:

	fldd d0, [r4]
	fstd d0, [r4, #+32]
	fldd d0, [r4, #+8]
	fstd d0, [r4, #+40]
	fldd d0, [r4, #+16]
	fstd d0, [r4, #+48]
	fldd d0, [r4, #+24]
	fstd d0, [r4, #+56]

If we can spare the registers, it would be better to use fldm and fstm here.
Need major register allocator enhancement though.

2) Can we recognize the relative position of constantpool entries? i.e. Treat

	ldr r0, LCPI17_3
	ldr r1, LCPI17_4
	ldr r2, LCPI17_5

   as
	ldr r0, LCPI17
	ldr r1, LCPI17+4
	ldr r2, LCPI17+8

   Then the ldr's can be combined into a single ldm. See Olden/power.

Note for ARM v4 gcc uses ldmia to load a pair of 32-bit values to represent a
double 64-bit FP constant:

	adr	r0, L6
	ldmia	r0, {r0-r1}

	.align 2
L6:
	.long	-858993459
	.long	1074318540

3) struct copies appear to be done field by field 
instead of by words, at least sometimes:

struct foo { int x; short s; char c1; char c2; };
void cpy(struct foo*a, struct foo*b) { *a = *b; }

llvm code (-O2)
        ldrb r3, [r1, #+6]
        ldr r2, [r1]
        ldrb r12, [r1, #+7]
        ldrh r1, [r1, #+4]
        str r2, [r0]
        strh r1, [r0, #+4]
        strb r3, [r0, #+6]
        strb r12, [r0, #+7]
gcc code (-O2)
        ldmia   r1, {r1-r2}
        stmia   r0, {r1-r2}

In this benchmark poor handling of aggregate copies has shown up as
having a large effect on size, and possibly speed as well (we don't have
a good way to measure on ARM).

//===---------------------------------------------------------------------===//

* Consider this silly example:

double bar(double x) {  
  double r = foo(3.1);
  return x+r;
}

_bar:
        stmfd sp!, {r4, r5, r7, lr}
        add r7, sp, #8
        mov r4, r0
        mov r5, r1
        fldd d0, LCPI1_0
        fmrrd r0, r1, d0
        bl _foo
        fmdrr d0, r4, r5
        fmsr s2, r0
        fsitod d1, s2
        faddd d0, d1, d0
        fmrrd r0, r1, d0
        ldmfd sp!, {r4, r5, r7, pc}

Ignore the prologue and epilogue stuff for a second. Note 
	mov r4, r0
	mov r5, r1
the copys to callee-save registers and the fact they are only being used by the
fmdrr instruction. It would have been better had the fmdrr been scheduled
before the call and place the result in a callee-save DPR register. The two
mov ops would not have been necessary.

//===---------------------------------------------------------------------===//

Calling convention related stuff:

* gcc's parameter passing implementation is terrible and we suffer as a result:

e.g.
struct s {
  double d1;
  int s1;
};

void foo(struct s S) {
  printf("%g, %d\n", S.d1, S.s1);
}

'S' is passed via registers r0, r1, r2. But gcc stores them to the stack, and
then reload them to r1, r2, and r3 before issuing the call (r0 contains the
address of the format string):

	stmfd	sp!, {r7, lr}
	add	r7, sp, #0
	sub	sp, sp, #12
	stmia	sp, {r0, r1, r2}
	ldmia	sp, {r1-r2}
	ldr	r0, L5
	ldr	r3, [sp, #8]
L2:
	add	r0, pc, r0
	bl	L_printf$stub

Instead of a stmia, ldmia, and a ldr, wouldn't it be better to do three moves?

* Return an aggregate type is even worse:

e.g.
struct s foo(void) {
  struct s S = {1.1, 2};
  return S;
}

	mov	ip, r0
	ldr	r0, L5
	sub	sp, sp, #12
L2:
	add	r0, pc, r0
	@ lr needed for prologue
	ldmia	r0, {r0, r1, r2}
	stmia	sp, {r0, r1, r2}
	stmia	ip, {r0, r1, r2}
	mov	r0, ip
	add	sp, sp, #12
	bx	lr

r0 (and later ip) is the hidden parameter from caller to store the value in. The
first ldmia loads the constants into r0, r1, r2. The last stmia stores r0, r1,
r2 into the address passed in. However, there is one additional stmia that
stores r0, r1, and r2 to some stack location. The store is dead.

The llvm-gcc generated code looks like this:

csretcc void %foo(%struct.s* %agg.result) {
entry:
	%S = alloca %struct.s, align 4		; <%struct.s*> [#uses=1]
	%memtmp = alloca %struct.s		; <%struct.s*> [#uses=1]
	cast %struct.s* %S to sbyte*		; <sbyte*>:0 [#uses=2]
	call void %llvm.memcpy.i32( sbyte* %0, sbyte* cast ({ double, int }* %C.0.904 to sbyte*), uint 12, uint 4 )
	cast %struct.s* %agg.result to sbyte*		; <sbyte*>:1 [#uses=2]
	call void %llvm.memcpy.i32( sbyte* %1, sbyte* %0, uint 12, uint 0 )
	cast %struct.s* %memtmp to sbyte*		; <sbyte*>:2 [#uses=1]
	call void %llvm.memcpy.i32( sbyte* %2, sbyte* %1, uint 12, uint 0 )
	ret void
}

llc ends up issuing two memcpy's (the first memcpy becomes 3 loads from
constantpool). Perhaps we should 1) fix llvm-gcc so the memcpy is translated
into a number of load and stores, or 2) custom lower memcpy (of small size) to
be ldmia / stmia. I think option 2 is better but the current register
allocator cannot allocate a chunk of registers at a time.

A feasible temporary solution is to use specific physical registers at the
lowering time for small (<= 4 words?) transfer size.

* ARM CSRet calling convention requires the hidden argument to be returned by
the callee.

//===---------------------------------------------------------------------===//

We can definitely do a better job on BB placements to eliminate some branches.
It's very common to see llvm generated assembly code that looks like this:

LBB3:
 ...
LBB4:
...
  beq LBB3
  b LBB2

If BB4 is the only predecessor of BB3, then we can emit BB3 after BB4. We can
then eliminate beq and and turn the unconditional branch to LBB2 to a bne.

See McCat/18-imp/ComputeBoundingBoxes for an example.

//===---------------------------------------------------------------------===//

Pre-/post- indexed load / stores:

1) We should not make the pre/post- indexed load/store transform if the base ptr
is guaranteed to be live beyond the load/store. This can happen if the base
ptr is live out of the block we are performing the optimization. e.g.

mov r1, r2
ldr r3, [r1], #4
...

vs.

ldr r3, [r2]
add r1, r2, #4
...

In most cases, this is just a wasted optimization. However, sometimes it can
negatively impact the performance because two-address code is more restrictive
when it comes to scheduling.

Unfortunately, liveout information is currently unavailable during DAG combine
time.

2) Consider spliting a indexed load / store into a pair of add/sub + load/store
   to solve #1 (in TwoAddressInstructionPass.cpp).

3) Enhance LSR to generate more opportunities for indexed ops.

4) Once we added support for multiple result patterns, write indexed loads
   patterns instead of C++ instruction selection code.

5) Use VLDM / VSTM to emulate indexed FP load / store.

//===---------------------------------------------------------------------===//

Implement support for some more tricky ways to materialize immediates.  For
example, to get 0xffff8000, we can use:

mov r9, #&3f8000
sub r9, r9, #&400000

//===---------------------------------------------------------------------===//

We sometimes generate multiple add / sub instructions to update sp in prologue
and epilogue if the inc / dec value is too large to fit in a single immediate
operand. In some cases, perhaps it might be better to load the value from a
constantpool instead.

//===---------------------------------------------------------------------===//

GCC generates significantly better code for this function.

int foo(int StackPtr, unsigned char *Line, unsigned char *Stack, int LineLen) {
    int i = 0;

    if (StackPtr != 0) {
       while (StackPtr != 0 && i < (((LineLen) < (32768))? (LineLen) : (32768)))
          Line[i++] = Stack[--StackPtr];
        if (LineLen > 32768)
        {
            while (StackPtr != 0 && i < LineLen)
            {
                i++;
                --StackPtr;
            }
        }
    }
    return StackPtr;
}

//===---------------------------------------------------------------------===//

This should compile to the mlas instruction:
int mlas(int x, int y, int z) { return ((x * y + z) < 0) ? 7 : 13; }

//===---------------------------------------------------------------------===//

At some point, we should triage these to see if they still apply to us:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19598
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18560
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27016

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11831
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11826
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11825
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11824
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11823
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11820
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10982

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10242
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9831
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9760
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9759
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9703
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9702
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9663

http://www.inf.u-szeged.hu/gcc-arm/
http://citeseer.ist.psu.edu/debus04linktime.html

//===---------------------------------------------------------------------===//

gcc generates smaller code for this function at -O2 or -Os:

void foo(signed char* p) {
  if (*p == 3)
     bar();
   else if (*p == 4)
    baz();
  else if (*p == 5)
    quux();
}

llvm decides it's a good idea to turn the repeated if...else into a
binary tree, as if it were a switch; the resulting code requires -1 
compare-and-branches when *p<=2 or *p==5, the same number if *p==4
or *p>6, and +1 if *p==3.  So it should be a speed win
(on balance).  However, the revised code is larger, with 4 conditional 
branches instead of 3.

More seriously, there is a byte->word extend before
each comparison, where there should be only one, and the condition codes
are not remembered when the same two values are compared twice.

//===---------------------------------------------------------------------===//

More LSR enhancements possible:

1. Teach LSR about pre- and post- indexed ops to allow iv increment be merged
   in a load / store.
2. Allow iv reuse even when a type conversion is required. For example, i8
   and i32 load / store addressing modes are identical.


//===---------------------------------------------------------------------===//

This:

int foo(int a, int b, int c, int d) {
  long long acc = (long long)a * (long long)b;
  acc += (long long)c * (long long)d;
  return (int)(acc >> 32);
}

Should compile to use SMLAL (Signed Multiply Accumulate Long) which multiplies 
two signed 32-bit values to produce a 64-bit value, and accumulates this with 
a 64-bit value.

We currently get this with both v4 and v6:

_foo:
        smull r1, r0, r1, r0
        smull r3, r2, r3, r2
        adds r3, r3, r1
        adc r0, r2, r0
        bx lr

//===---------------------------------------------------------------------===//

This:
        #include <algorithm>
        std::pair<unsigned, bool> full_add(unsigned a, unsigned b)
        { return std::make_pair(a + b, a + b < a); }
        bool no_overflow(unsigned a, unsigned b)
        { return !full_add(a, b).second; }

Should compile to:

_Z8full_addjj:
	adds	r2, r1, r2
	movcc	r1, #0
	movcs	r1, #1
	str	r2, [r0, #0]
	strb	r1, [r0, #4]
	mov	pc, lr

_Z11no_overflowjj:
	cmn	r0, r1
	movcs	r0, #0
	movcc	r0, #1
	mov	pc, lr

not:

__Z8full_addjj:
        add r3, r2, r1
        str r3, [r0]
        mov r2, #1
        mov r12, #0
        cmp r3, r1
        movlo r12, r2
        str r12, [r0, #+4]
        bx lr
__Z11no_overflowjj:
        add r3, r1, r0
        mov r2, #1
        mov r1, #0
        cmp r3, r0
        movhs r1, r2
        mov r0, r1
        bx lr

//===---------------------------------------------------------------------===//

Some of the NEON intrinsics may be appropriate for more general use, either
as target-independent intrinsics or perhaps elsewhere in the ARM backend.
Some of them may also be lowered to target-independent SDNodes, and perhaps
some new SDNodes could be added.

For example, maximum, minimum, and absolute value operations are well-defined
and standard operations, both for vector and scalar types.

The current NEON-specific intrinsics for count leading zeros and count one
bits could perhaps be replaced by the target-independent ctlz and ctpop
intrinsics.  It may also make sense to add a target-independent "ctls"
intrinsic for "count leading sign bits".  Likewise, the backend could use
the target-independent SDNodes for these operations.

ARMv6 has scalar saturating and halving adds and subtracts.  The same
intrinsics could possibly be used for both NEON's vector implementations of
those operations and the ARMv6 scalar versions.

//===---------------------------------------------------------------------===//

Split out LDR (literal) from normal ARM LDR instruction. Also consider spliting
LDR into imm12 and so_reg forms. This allows us to clean up some code. e.g.
ARMLoadStoreOptimizer does not need to look at LDR (literal) and LDR (so_reg)
while ARMConstantIslandPass only need to worry about LDR (literal).

//===---------------------------------------------------------------------===//

Constant island pass should make use of full range SoImm values for LEApcrel.
Be careful though as the last attempt caused infinite looping on lencod.

//===---------------------------------------------------------------------===//

Predication issue. This function:   

extern unsigned array[ 128 ];
int     foo( int x ) {
  int     y;
  y = array[ x & 127 ];
  if ( x & 128 )
     y = 123456789 & ( y >> 2 );
  else
     y = 123456789 & y;
  return y;
}

compiles to:

_foo:
	and r1, r0, #127
	ldr r2, LCPI1_0
	ldr r2, [r2]
	ldr r1, [r2, +r1, lsl #2]
	mov r2, r1, lsr #2
	tst r0, #128
	moveq r2, r1
	ldr r0, LCPI1_1
	and r0, r2, r0
	bx lr

It would be better to do something like this, to fold the shift into the
conditional move:

	and r1, r0, #127
	ldr r2, LCPI1_0
	ldr r2, [r2]
	ldr r1, [r2, +r1, lsl #2]
	tst r0, #128
	movne r1, r1, lsr #2
	ldr r0, LCPI1_1
	and r0, r1, r0
	bx lr

it saves an instruction and a register.

//===---------------------------------------------------------------------===//

It might be profitable to cse MOVi16 if there are lots of 32-bit immediates
with the same bottom half.

//===---------------------------------------------------------------------===//

Robert Muth started working on an alternate jump table implementation that
does not put the tables in-line in the text.  This is more like the llvm
default jump table implementation.  This might be useful sometime.  Several
revisions of patches are on the mailing list, beginning at:
http://lists.cs.uiuc.edu/pipermail/llvmdev/2009-June/022763.html

//===---------------------------------------------------------------------===//

Make use of the "rbit" instruction.

//===---------------------------------------------------------------------===//

Take a look at test/CodeGen/Thumb2/machine-licm.ll. ARM should be taught how
to licm and cse the unnecessary load from cp#1.

//===---------------------------------------------------------------------===//

The CMN instruction sets the flags like an ADD instruction, while CMP sets
them like a subtract. Therefore to be able to use CMN for comparisons other
than the Z bit, we'll need additional logic to reverse the conditionals
associated with the comparison. Perhaps a pseudo-instruction for the comparison,
with a post-codegen pass to clean up and handle the condition codes?
See PR5694 for testcase.

//===---------------------------------------------------------------------===//

Given the following on armv5:
int test1(int A, int B) {
  return (A&-8388481)|(B&8388480);
}

We currently generate:
	ldr	r2, .LCPI0_0
	and	r0, r0, r2
	ldr	r2, .LCPI0_1
	and	r1, r1, r2
	orr	r0, r1, r0
	bx	lr

We should be able to replace the second ldr+and with a bic (i.e. reuse the
constant which was already loaded).  Not sure what's necessary to do that.

//===---------------------------------------------------------------------===//

The code generated for bswap on armv4/5 (CPUs without rev) is less than ideal:

int a(int x) { return __builtin_bswap32(x); }

a:
	mov	r1, #255, 24
	mov	r2, #255, 16
	and	r1, r1, r0, lsr #8
	and	r2, r2, r0, lsl #8
	orr	r1, r1, r0, lsr #24
	orr	r0, r2, r0, lsl #24
	orr	r0, r0, r1
	bx	lr

Something like the following would be better (fewer instructions/registers):
	eor     r1, r0, r0, ror #16
	bic     r1, r1, #0xff0000
	mov     r1, r1, lsr #8
	eor     r0, r1, r0, ror #8
	bx	lr

A custom Thumb version would also be a slight improvement over the generic
version.

//===---------------------------------------------------------------------===//

Consider the following simple C code:

void foo(unsigned char *a, unsigned char *b, int *c) {
 if ((*a | *b) == 0) *c = 0;
}

currently llvm-gcc generates something like this (nice branchless code I'd say):

       ldrb    r0, [r0]
       ldrb    r1, [r1]
       orr     r0, r1, r0
       tst     r0, #255
       moveq   r0, #0
       streq   r0, [r2]
       bx      lr

Note that both "tst" and "moveq" are redundant.

//===---------------------------------------------------------------------===//

When loading immediate constants with movt/movw, if there are multiple
constants needed with the same low 16 bits, and those values are not live at
the same time, it would be possible to use a single movw instruction, followed
by multiple movt instructions to rewrite the high bits to different values.
For example:

  volatile store i32 -1, i32* inttoptr (i32 1342210076 to i32*), align 4,
  !tbaa
!0
  volatile store i32 -1, i32* inttoptr (i32 1342341148 to i32*), align 4,
  !tbaa
!0

is compiled and optimized to:

    movw    r0, #32796
    mov.w    r1, #-1
    movt    r0, #20480
    str    r1, [r0]
    movw    r0, #32796    @ <= this MOVW is not needed, value is there already
    movt    r0, #20482
    str    r1, [r0]

//===---------------------------------------------------------------------===//

Improve codegen for select's:
if (x != 0) x = 1
if (x == 1) x = 1

ARM codegen used to look like this:
       mov     r1, r0
       cmp     r1, #1
       mov     r0, #0
       moveq   r0, #1

The naive lowering select between two different values. It should recognize the
test is equality test so it's more a conditional move rather than a select:
       cmp     r0, #1
       movne   r0, #0

Currently this is a ARM specific dag combine. We probably should make it into a
target-neutral one.

//===---------------------------------------------------------------------===//

Optimize unnecessary checks for zero with __builtin_clz/ctz.  Those builtins
are specified to be undefined at zero, so portable code must check for zero
and handle it as a special case.  That is unnecessary on ARM where those
operations are implemented in a way that is well-defined for zero.  For
example:

int f(int x) { return x ? __builtin_clz(x) : sizeof(int)*8; }

should just be implemented with a CLZ instruction.  Since there are other
targets, e.g., PPC, that share this behavior, it would be best to implement
this in a target-independent way: we should probably fold that (when using
"undefined at zero" semantics) to set the "defined at zero" bit and have
the code generator expand out the right code.


//===---------------------------------------------------------------------===//

Clean up the test/MC/ARM files to have more robust register choices.

R0 should not be used as a register operand in the assembler tests as it's then
not possible to distinguish between a correct encoding and a missing operand
encoding, as zero is the default value for the binary encoder.
e.g.,
    add r0, r0  // bad
    add r3, r5  // good

Register operands should be distinct. That is, when the encoding does not
require two syntactical operands to refer to the same register, two different
registers should be used in the test so as to catch errors where the
operands are swapped in the encoding.
e.g.,
    subs.w r1, r1, r1 // bad
    subs.w r1, r2, r3 // good