llvm/lib/Target/ARM
Chris Lattner b80610cd13 REmove dead fields of TAI.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@77820 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-01 22:40:22 +00:00
..
AsmPrinter switch off of 'Section' onto MCSection. We're not properly using 2009-07-31 18:48:30 +00:00
TargetInfo Factor commonality in triple match routines into helper template for registering 2009-07-26 05:03:33 +00:00
ARM.h Add new helpers for registering targets. 2009-07-25 06:49:55 +00:00
ARM.td Add fake v7 itineraries for now. 2009-07-21 18:54:14 +00:00
ARMAddressingModes.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
ARMBaseInstrInfo.cpp Workaround a couple of Darwin assembler bugs. 2009-08-01 06:13:52 +00:00
ARMBaseInstrInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
ARMBaseRegisterInfo.cpp When fp is not eliminated, instructions with T2_i12 modes will be changed to T2_i8 ones. Take that into consideration when determining stack size limit for reserving register scavenging slot. 2009-07-30 23:29:25 +00:00
ARMBaseRegisterInfo.h When fp is not eliminated, instructions with T2_i12 modes will be changed to T2_i8 ones. Take that into consideration when determining stack size limit for reserving register scavenging slot. 2009-07-30 23:29:25 +00:00
ARMBuildAttrs.h
ARMCallingConv.td
ARMCodeEmitter.cpp More migration to raw_ostream, the water has dried up around the iostream hole. 2009-07-25 00:23:56 +00:00
ARMConstantIslandPass.cpp Workaround a couple of Darwin assembler bugs. 2009-08-01 06:13:52 +00:00
ARMConstantPoolValue.cpp
ARMConstantPoolValue.h no really, I can spell! 2009-07-21 23:36:01 +00:00
ARMFrameInfo.h
ARMInstrFormats.td Fix Thumb2 function call isel. Thumb1 and Thumb2 should share the same 2009-08-01 00:16:10 +00:00
ARMInstrInfo.cpp - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
ARMInstrInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
ARMInstrInfo.td Make sure Thumb2 uses the right call instructions. 2009-07-29 21:26:42 +00:00
ARMInstrNEON.td Change Neon VLDn intrinsics to return multiple values instead of really 2009-07-29 16:39:22 +00:00
ARMInstrThumb2.td Workaround a couple of Darwin assembler bugs. 2009-08-01 06:13:52 +00:00
ARMInstrThumb.td Fix Thumb2 function call isel. Thumb1 and Thumb2 should share the same 2009-08-01 00:16:10 +00:00
ARMInstrVFP.td Model fpscr to prevent fcmped / fcmpezs etc from being deleted. 2009-07-20 02:12:31 +00:00
ARMISelDAGToDAG.cpp Split t2MOVCCs since some assemblers do not recognize mov shifted register alias with predicate. 2009-08-01 01:43:45 +00:00
ARMISelLowering.cpp Fix Thumb2 function call isel. Thumb1 and Thumb2 should share the same 2009-08-01 00:16:10 +00:00
ARMISelLowering.h Optimize Thumb2 jumptable to use tbb / tbh when all the offsets fit in byte / halfword. 2009-07-29 02:18:14 +00:00
ARMJITInfo.cpp More migration to raw_ostream, the water has dried up around the iostream hole. 2009-07-25 00:23:56 +00:00
ARMJITInfo.h
ARMLoadStoreOptimizer.cpp
ARMMachineFunctionInfo.h
ARMRegisterInfo.cpp
ARMRegisterInfo.h
ARMRegisterInfo.td Add VFP3 D registers to the DPR register class. 2009-07-29 23:03:41 +00:00
ARMRelocations.h
ARMSchedule.td Add fake v7 itineraries for now. 2009-07-21 18:54:14 +00:00
ARMScheduleV6.td Fix comment. 2009-07-21 23:54:22 +00:00
ARMScheduleV7.td Add fake v7 itineraries for now. 2009-07-21 18:54:14 +00:00
ARMSubtarget.cpp Fix Thumb2 function call isel. Thumb1 and Thumb2 should share the same 2009-08-01 00:16:10 +00:00
ARMSubtarget.h
ARMTargetAsmInfo.cpp REmove dead fields of TAI. 2009-08-01 22:40:22 +00:00
ARMTargetAsmInfo.h REmove dead fields of TAI. 2009-08-01 22:40:22 +00:00
ARMTargetMachine.cpp remove dead code. 2009-07-27 19:00:33 +00:00
ARMTargetMachine.h
CMakeLists.txt
Makefile
README-Thumb2.txt Change Thumb2 jumptable codegen to one that uses two level jumps: 2009-07-25 00:33:29 +00:00
README-Thumb.txt Another TODO. 2009-07-25 00:39:37 +00:00
README.txt add a random codegen deficiency. 2009-07-30 16:08:58 +00:00
Thumb1InstrInfo.cpp - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
Thumb1InstrInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
Thumb1RegisterInfo.cpp tADDrSPI doesn't have a predicate operand, but tADDhirr and tADDi3 have. 2009-07-28 07:38:35 +00:00
Thumb1RegisterInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
Thumb2InstrInfo.cpp Optimize Thumb2 jumptable to use tbb / tbh when all the offsets fit in byte / halfword. 2009-07-29 02:18:14 +00:00
Thumb2InstrInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
Thumb2ITBlockPass.cpp
Thumb2RegisterInfo.cpp - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00
Thumb2RegisterInfo.h - More refactoring. This gets rid of all of the getOpcode calls. 2009-07-28 05:48:47 +00:00

//===---------------------------------------------------------------------===//
// 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)
* Coalesce stack slots!
* Implement smarter constant generation for binops with large immediates.

* Consider materializing FP constants like 0.0f and 1.0f using integer 
  immediate instructions then copy to FPU.  Slower than load into FPU?

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

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.

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

We currently compile abs:
int foo(int p) { return p < 0 ? -p : p; }

into:

_foo:
        rsb r1, r0, #0
        cmn r0, #1
        movgt r1, r0
        mov r0, r1
        bx lr

This is very, uh, literal.  This could be a 3 operation sequence:
  t = (p sra 31); 
  res = (p xor t)-t

Which would be better.  This occurs in png decode.

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

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 FLDM / FSTM 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 register scavenging work:

1. Use the register scavenger to track frame index materialized into registers
   (those that do not fit in addressing modes) to allow reuse in the same BB.
2. Finish scavenging for Thumb.

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

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.

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

ARM::MOVCCr is commutable (by flipping the condition). But we need to implement
ARMInstrInfo::commuteInstruction() to support it.

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

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).

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

We need to fix constant isel for ARMv6t2 to use MOVT.

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

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.

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