llvm-6502/lib/Target
2006-03-09 21:51:28 +00:00
..
Alpha these are copies too 2006-03-09 18:18:51 +00:00
CBackend Add #line support for CBE. 2006-03-08 19:31:15 +00:00
IA64 Add support for 'special' llvm globals like debug info and static ctors/dtors. 2006-03-09 06:14:35 +00:00
PowerPC Change the interface for getting a target HazardRecognizer to be more clean. 2006-03-08 04:25:59 +00:00
Skeleton Remove the skeleton target, it doesn't produce useful code and there are 2006-02-16 23:14:50 +00:00
Sparc Add support for 'special' llvm globals like debug info and static ctors/dtors. 2006-03-09 06:14:35 +00:00
SparcV8 Remove the SparcV8 backend. It has been renamed to be the Sparc backend. 2006-02-05 06:33:29 +00:00
SparcV9 Adjust to MachineConstantPool interface change: instead of keeping a 2006-02-09 04:46:04 +00:00
X86 Add option -enable-x86-lsr to enable x86 loop strength reduction pass. 2006-03-09 21:51:28 +00:00
Makefile DONT_BUILD_RELINKED is gone and implied by BUILD_ARCHIVE now 2005-10-24 02:26:13 +00:00
MRegisterInfo.cpp Finegrainify namespacification 2006-02-01 18:10:56 +00:00
README.txt add a note 2006-03-09 20:13:21 +00:00
SubtargetFeature.cpp Improve compatibility with VC2005, patch by Morten Ofstad! 2006-01-26 20:41:32 +00:00
Target.td Split the valuetypes out of Target.td into ValueTypes.td 2006-03-03 01:55:26 +00:00
TargetData.cpp Implement a new InvalidateStructLayoutInfo method and add some comments 2006-01-14 00:07:34 +00:00
TargetFrameInfo.cpp Eliminate all remaining tabs and trailing spaces. 2005-07-27 06:12:32 +00:00
TargetInstrInfo.cpp Convert tabs to spaces 2005-04-22 17:54:37 +00:00
TargetMachine.cpp - Added option -relocation-model to set relocation model. Valid values include static, pic, 2006-02-22 20:19:42 +00:00
TargetMachineRegistry.cpp 1. Use SubtargetFeatures in llc/lli. 2005-09-01 21:38:21 +00:00
TargetSchedInfo.cpp Convert tabs to spaces 2005-04-22 17:54:37 +00:00
TargetSchedule.td Add a default NoItinerary class for targets to use. 2006-01-27 01:41:38 +00:00
TargetSelectionDAG.td relax fcopysign 2006-03-09 17:47:22 +00:00
TargetSubtarget.cpp Eliminate all remaining tabs and trailing spaces. 2005-07-27 06:12:32 +00:00

Target Independent Opportunities:

===-------------------------------------------------------------------------===

FreeBench/mason contains code like this:

static p_type m0u(p_type p) {
  int m[]={0, 8, 1, 2, 16, 5, 13, 7, 14, 9, 3, 4, 11, 12, 15, 10, 17, 6};
  p_type pu;
  pu.a = m[p.a];
  pu.b = m[p.b];
  pu.c = m[p.c];
  return pu;
}

We currently compile this into a memcpy from a static array into 'm', then
a bunch of loads from m.  It would be better to avoid the memcpy and just do
loads from the static array.

===-------------------------------------------------------------------------===

Get the C front-end to expand hypot(x,y) -> llvm.sqrt(x*x+y*y) when errno and
precision don't matter (ffastmath).  Misc/mandel will like this. :)

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

Solve this DAG isel folding deficiency:

int X, Y;

void fn1(void)
{
  X = X | (Y << 3);
}

compiles to

fn1:
	movl Y, %eax
	shll $3, %eax
	orl X, %eax
	movl %eax, X
	ret

The problem is the store's chain operand is not the load X but rather
a TokenFactor of the load X and load Y, which prevents the folding.

There are two ways to fix this:

1. The dag combiner can start using alias analysis to realize that y/x
   don't alias, making the store to X not dependent on the load from Y.
2. The generated isel could be made smarter in the case it can't
   disambiguate the pointers.

Number 1 is the preferred solution.

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

DAG combine this into mul A, 8:

int %test(int %A) {
  %B = mul int %A, 8  ;; shift
  %C = add int %B, 7  ;; dead, no demanded bits.
  %D = and int %C, -8 ;; dead once add is gone.
  ret int %D
}

This sort of thing occurs in the alloca lowering code and other places that
are generating alignment of an already aligned value.

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

Turn this into a signed shift right in instcombine:

int f(unsigned x) {
  return x >> 31 ? -1 : 0;
}

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25600
http://gcc.gnu.org/ml/gcc-patches/2006-02/msg01492.html

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

We should reassociate:
int f(int a, int b){ return a * a + 2 * a * b + b * b; }
into:
int f(int a, int b) { return a * (a + 2 * b) + b * b; }
to eliminate a multiply.

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

On targets with expensive 64-bit multiply, we could LSR this:

for (i = ...; ++i) {
   x = 1ULL << i;

into:
 long long tmp = 1;
 for (i = ...; ++i, tmp+=tmp)
   x = tmp;

This would be a win on ppc32, but not x86 or ppc64.

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

Shrink: (setlt (loadi32 P), 0) -> (setlt (loadi8 Phi), 0)

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

Reassociate is missing this:

int test(int X, int Y) {
 return (X+X+Y+Y);  // (X+Y) << 1;
}

it needs to turn the shifts into multiplies to get it.

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

These two functions should generate the same code on big-endian systems:

int g(int *j,int *l)  {  return memcmp(j,l,4);  }
int h(int *j, int *l) {  return *j - *l; }

this could be done in SelectionDAGISel.cpp, along with other special cases,
for 1,2,4,8 bytes.

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