RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-26 14:33:02 +02:00
Code Issues Projects Releases Wiki Activity
40,911 Commits 4 Branches 5 Tags 704 MiB
1cfbb25e75
Commit Graph

397 Commits

Author SHA1 Message Date
Chris Lattner
2c3f67f6a7 Implement 64-bit select, bswap, etc. 
llvm-svn: 28935
 2006-06-27 20:14:52 +00:00
Chris Lattner
8569f4042d PPC doesn't have bit converts to/from i64 
llvm-svn: 28932
 2006-06-27 18:40:08 +00:00
Chris Lattner
26f2bd4d4b Implement 64-bit undef, sub, shl/shr, srem/urem 
llvm-svn: 28929
 2006-06-27 18:18:41 +00:00
Chris Lattner
b4a636f966 Use i32 for shift amounts instead of i64. This gets bisort working. 
llvm-svn: 28927
 2006-06-27 17:34:57 +00:00
Chris Lattner
494f476ca7 Implement a bunch of 64-bit cleanliness work. With this, treeadd builds (but 
doesn't work right).

llvm-svn: 28921
 2006-06-27 00:04:13 +00:00
Chris Lattner
cbd4d14b24 Improve PPC64 calling convention support 
llvm-svn: 28919
 2006-06-26 22:48:35 +00:00
Chris Lattner
5fa6e47534 Correct returns of 64-bit values, though they seemed to work before... 
llvm-svn: 28892
 2006-06-21 00:34:03 +00:00
Chris Lattner
81845946ff fix some assumptions that pointers can only be 32-bits. With this, we can 
now compile:

static unsigned long X;
void test1() {
  X = 0;
}

into:

_test1:
        lis r2, ha16(_X)
        li r3, 0
        stw r3, lo16(_X)(r2)
        blr

Totally amazing :)

llvm-svn: 28839
 2006-06-16 21:01:35 +00:00
Chris Lattner
fa884ac11b Rename some subtarget features. A CPU now can *have* 64-bit instructions, 
can in 32-bit mode we can choose to optionally *use* 64-bit registers.

llvm-svn: 28824
 2006-06-16 17:34:12 +00:00
Evan Cheng
32feafd76c Type of extract_element index operand should be iPTR. 
llvm-svn: 28797
 2006-06-15 08:18:06 +00:00
Chris Lattner
b231c3d11c Fix a problem exposed by the local allocator. CALL instructions are not marked 
as using incoming argument registers, so the local allocator would clobber them
between their set and use.  To fix this, we give the call instructions a variable
number of uses in the CALL MachineInstr itself, so live variables understands
the live ranges of these register arguments.

llvm-svn: 28744
 2006-06-10 01:14:28 +00:00
Chris Lattner
31b150e334 Always reserve space for 8 spilled GPRs. GCC apparently assumes that this 
space will be available, even if the callee isn't varargs.

llvm-svn: 28571
 2006-05-30 21:21:04 +00:00
Evan Cheng
de0f25081a Change RET node to include signness information of the return values. i.e. 
RET chain, value1, sign1, value2, sign2, ...

llvm-svn: 28510
 2006-05-26 23:10:12 +00:00
Evan Cheng
4a74dd0c51 CALL node change (arg / sign pairs instead of just arguments). 
llvm-svn: 28462
 2006-05-25 00:57:32 +00:00
Chris Lattner
f604017e47 Patches to make the LLVM sources more -pedantic clean. Patch provided 
by Anton Korobeynikov!  This is a step towards closing PR786.

llvm-svn: 28447
 2006-05-24 17:04:05 +00:00
Chris Lattner
bc3be2ff8a Fix CodeGen/Generic/vector.ll:test_div with altivec. 
llvm-svn: 28445
 2006-05-24 00:15:25 +00:00
Chris Lattner
56862bbd53 Handle SETO* like we handle SET*, restoring behavior after Evan's setcc 
change.  This fixes PowerPC/fnegsel.ll.

llvm-svn: 28443
 2006-05-24 00:06:44 +00:00
Chris Lattner
2208c3214c Make PPC call lowering more aggressive, making the isel matching code simple 
enough to be autogenerated.

llvm-svn: 28354
 2006-05-17 19:00:46 +00:00
Chris Lattner
03c70b7f27 Switch PPC over to a call-selection model where the lowering code creates 
the copyto/fromregs instead of making the PPCISD::CALL selection code create
them.  This vastly simplifies the selection code, and moves the ABI handling
parts into one place.

llvm-svn: 28346
 2006-05-17 06:01:33 +00:00
Chris Lattner
348883611c 3 changes, 2 of which are cleanup one of which changes codegen: 
1. Rearrange code a bit so that the special case doesn't require indenting lots
   of code.
2. Add comments describing PPC calling convention.
3. Only round up to 56-bytes of stack space for an outgoing call if the callee
   is varargs.  This saves a bit of stack space.

llvm-svn: 28342
 2006-05-17 00:15:40 +00:00
Chris Lattner
a36579803f implement passing/returning vector regs to calls, at least non-varargs calls. 
llvm-svn: 28341
 2006-05-16 23:54:25 +00:00
Chris Lattner
b5271a0f4c Instead of implementing LowerCallTo directly, let the default impl produce an 
ISD::CALL node, then custom lower that.  This means that we only have to handle
LEGAL call operands/results, not every possible type.  This allows us to
simplify the call code, shrinking it by about 1/3.

llvm-svn: 28339
 2006-05-16 22:56:08 +00:00
Chris Lattner
40d1eaad0a Simplify the argument counting logic by only incrementing the index. 
llvm-svn: 28335
 2006-05-16 18:58:15 +00:00
Chris Lattner
0ae068ed8f Simplify the dead argument handling code. 
llvm-svn: 28334
 2006-05-16 18:54:32 +00:00
Chris Lattner
fbbe542235 Vector args passed in registers don't reserve stack space. 
llvm-svn: 28333
 2006-05-16 18:51:52 +00:00
Chris Lattner
0a12e343e2 Switch the PPC backend over to using FORMAL_ARGUMENTS for formal argument 
handling.  This makes the lower argument code significantly simpler (we
only need to handle legal argument types).

Incidentally, this also implements support for vector argument registers,
so long as they are not on the stack.

llvm-svn: 28331
 2006-05-16 18:18:50 +00:00
Chris Lattner
199f3f6af8 Fit in 80 cols 
llvm-svn: 28311
 2006-05-16 04:20:24 +00:00
Chris Lattner
adcb0582d8 Remove dead var, fix bad override. 
llvm-svn: 28264
 2006-05-12 21:09:57 +00:00
Chris Lattner
e3de67fae2 Fix CodeGen/Generic/2006-04-28-Sign-extend-bool.ll 
llvm-svn: 28017
 2006-04-28 21:56:10 +00:00
Nate Begeman
7ed816f900 JumpTable support! What this represents is working asm and jit support for 
x86 and ppc for 100% dense switch statements when relocations are non-PIC.
This support will be extended and enhanced in the coming days to support
PIC, and less dense forms of jump tables.

llvm-svn: 27947
 2006-04-22 18:53:45 +00:00
Chris Lattner
47a41ae889 Fix a crash on: 
void foo2(vector float *A, vector float *B) {
  vector float C = (vector float)vec_cmpeq(*A, *B);
  if (!vec_any_eq(*A, *B))
    *B = (vector float){0,0,0,0};
  *A = C;
}

llvm-svn: 27808
 2006-04-18 18:28:22 +00:00
Chris Lattner
2bd91746e1 pretty print node name 
llvm-svn: 27806
 2006-04-18 18:05:58 +00:00
Chris Lattner
44ea12c5f8 Implement an important entry from README_ALTIVEC: 
If an altivec predicate compare is used immediately by a branch, don't
use a (serializing) MFCR instruction to read the CR6 register, which requires
a compare to get it back to CR's.  Instead, just branch on CR6 directly. :)

For example, for:
void foo2(vector float *A, vector float *B) {
  if (!vec_any_eq(*A, *B))
    *B = (vector float){0,0,0,0};
}

We now generate:

_foo2:
        mfspr r2, 256
        oris r5, r2, 12288
        mtspr 256, r5
        lvx v2, 0, r4
        lvx v3, 0, r3
        vcmpeqfp. v2, v3, v2
        bne cr6, LBB1_2 ; UnifiedReturnBlock
LBB1_1: ; cond_true
        vxor v2, v2, v2
        stvx v2, 0, r4
        mtspr 256, r2
        blr
LBB1_2: ; UnifiedReturnBlock
        mtspr 256, r2
        blr

instead of:

_foo2:
        mfspr r2, 256
        oris r5, r2, 12288
        mtspr 256, r5
        lvx v2, 0, r4
        lvx v3, 0, r3
        vcmpeqfp. v2, v3, v2
        mfcr r3, 2
        rlwinm r3, r3, 27, 31, 31
        cmpwi cr0, r3, 0
        beq cr0, LBB1_2 ; UnifiedReturnBlock
LBB1_1: ; cond_true
        vxor v2, v2, v2
        stvx v2, 0, r4
        mtspr 256, r2
        blr
LBB1_2: ; UnifiedReturnBlock
        mtspr 256, r2
        blr

This implements CodeGen/PowerPC/vec_br_cmp.ll.

llvm-svn: 27804
 2006-04-18 17:59:36 +00:00
Chris Lattner
e90fdf3b98 Use vmladduhm to do v8i16 multiplies which is faster and simpler than doing 
even/odd halves.  Thanks to Nate telling me what's what.

llvm-svn: 27793
 2006-04-18 04:28:57 +00:00
Chris Lattner
5951b60cb4 Implement v16i8 multiply with this code: 
vmuloub v5, v3, v2
        vmuleub v2, v3, v2
        vperm v2, v2, v5, v4

This implements CodeGen/PowerPC/vec_mul.ll.  With this, v16i8 multiplies are
6.79x faster than before.

Overall, UnitTests/Vector/multiplies.c is now 2.45x faster with LLVM than with
GCC.

Remove the 'integer multiplies' todo from the README file.

llvm-svn: 27792
 2006-04-18 03:57:35 +00:00
Chris Lattner
4d84b56e64 Lower v8i16 multiply into this code: 
li r5, lo16(LCPI1_0)
        lis r6, ha16(LCPI1_0)
        lvx v4, r6, r5
        vmulouh v5, v3, v2
        vmuleuh v2, v3, v2
        vperm v2, v2, v5, v4

where v4 is:
LCPI1_0:                                        ;  <16 x ubyte>
        .byte   2
        .byte   3
        .byte   18
        .byte   19
        .byte   6
        .byte   7
        .byte   22
        .byte   23
        .byte   10
        .byte   11
        .byte   26
        .byte   27
        .byte   14
        .byte   15
        .byte   30
        .byte   31

This is 5.07x faster on the G5 (measured) than lowering to scalar code +
loads/stores.

llvm-svn: 27789
 2006-04-18 03:43:48 +00:00
Chris Lattner
613d7fda64 Custom lower v4i32 multiplies into a cute sequence, instead of having legalize 
scalarize the sequence into 4 mullw's and a bunch of load/store traffic.

This speeds up v4i32 multiplies 4.1x (measured) on a G5.  This implements
PowerPC/vec_mul.ll

llvm-svn: 27788
 2006-04-18 03:24:30 +00:00
Chris Lattner
f2347c31b4 Make sure to check splats of every constant we can, handle splat(31) by 
being a bit more clever, add support for odd splats from -31 to -17.

llvm-svn: 27764
 2006-04-17 18:09:22 +00:00
Chris Lattner
cc4222d95b Teach the ppc backend to use rol and vsldoi to generate splatted constants. 
This implements vec_constants.ll:test_vsldoi and test_rol

llvm-svn: 27760
 2006-04-17 17:55:10 +00:00
Chris Lattner
2d8d6c9feb Make some code more general, adding support for constant formation of several 
new patterns.

llvm-svn: 27754
 2006-04-17 06:58:41 +00:00
Chris Lattner
9dd4ebffca Learn how to make odd splatted constants in range [17,29]. This implements 
PowerPC/vec_constants.ll:test_29.

llvm-svn: 27752
 2006-04-17 06:07:44 +00:00
Chris Lattner
72a67a5b1f Pull some code out into a helper function. 
Effeciently codegen even splats in the range [-32,30].

This allows us to codegen <30,30,30,30> as:

        vspltisw v0, 15
        vadduwm v2, v0, v0

instead of as a cp load.

llvm-svn: 27750
 2006-04-17 06:00:21 +00:00
Chris Lattner
5367a73dec Implement a TODO: for any shuffle that can be viewed as a v4[if]32 shuffle, 
if it can be implemented in 3 or fewer discrete altivec instructions, codegen
it as such.  This implements Regression/CodeGen/PowerPC/vec_perf_shuffle.ll

llvm-svn: 27748
 2006-04-17 05:28:54 +00:00
Chris Lattner
d86516991a Implement a TODO: have the legalizer canonicalize a bunch of operations to 
one type (v4i32) so that we don't have to write patterns for each type, and
so that more CSE opportunities are exposed.

llvm-svn: 27731
 2006-04-16 01:37:57 +00:00
Chris Lattner
f4126f0db7 Make the BUILD_VECTOR lowering code much more aggressive w.r.t constant vectors. 
Remove some done items from the todo list.

llvm-svn: 27729
 2006-04-16 01:01:29 +00:00
Chris Lattner
44245f11c3 Fix a crash when faced with a shuffle vector that has an undef in its mask. 
llvm-svn: 27726
 2006-04-15 23:48:05 +00:00
Chris Lattner
5c9d357d7c Allow undef in a shuffle mask 
llvm-svn: 27714
 2006-04-14 23:19:08 +00:00
Chris Lattner
cf80e569f6 Move the rest of the PPCTargetLowering::LowerOperation cases out into 
separate functions, for simplicity and code clarity.

llvm-svn: 27693
 2006-04-14 06:01:58 +00:00
Chris Lattner
aacabea404 Pull the VECTOR_SHUFFLE and BUILD_VECTOR lowering code out into separate 
functions, which makes the code much cleaner :)

llvm-svn: 27692
 2006-04-14 05:19:18 +00:00
Chris Lattner
569ea9c6dd Force non-darwin targets to use a static relo model. This fixes PR734, 
tested by CodeGen/Generic/vector.ll

llvm-svn: 27657
 2006-04-13 17:10:48 +00:00
Chris Lattner
e087b8e321 Add a new way to match vector constants, which make it easier to bang bits of 
different types.

Codegen spltw(0x7FFFFFFF) and spltw(0x80000000) without a constant pool load,
implementing PowerPC/vec_constants.ll:test1.  This compiles:

typedef float vf __attribute__ ((vector_size (16)));
typedef int vi __attribute__ ((vector_size (16)));
void test(vi *P1, vi *P2, vf *P3) {
  *P1 &= (vi){0x80000000,0x80000000,0x80000000,0x80000000};
  *P2 &= (vi){0x7FFFFFFF,0x7FFFFFFF,0x7FFFFFFF,0x7FFFFFFF};
  *P3 = vec_abs((vector float)*P3);
}

to:

_test:
        mfspr r2, 256
        oris r6, r2, 49152
        mtspr 256, r6
        vspltisw v0, -1
        vslw v0, v0, v0
        lvx v1, 0, r3
        vand v1, v1, v0
        stvx v1, 0, r3
        lvx v1, 0, r4
        vandc v1, v1, v0
        stvx v1, 0, r4
        lvx v1, 0, r5
        vandc v0, v1, v0
        stvx v0, 0, r5
        mtspr 256, r2
        blr

instead of (with two constant pool entries):

_test:
        mfspr r2, 256
        oris r6, r2, 49152
        mtspr 256, r6
        li r6, lo16(LCPI1_0)
        lis r7, ha16(LCPI1_0)
        li r8, lo16(LCPI1_1)
        lis r9, ha16(LCPI1_1)
        lvx v0, r7, r6
        lvx v1, 0, r3
        vand v0, v1, v0
        stvx v0, 0, r3
        lvx v0, r9, r8
        lvx v1, 0, r4
        vand v1, v1, v0
        stvx v1, 0, r4
        lvx v1, 0, r5
        vand v0, v1, v0
        stvx v0, 0, r5
        mtspr 256, r2
        blr

GCC produces (with 2 cp entries):

_test:
        mfspr r0,256
        stw r0,-4(r1)
        oris r0,r0,0xc00c
        mtspr 256,r0
        lis r2,ha16(LC0)
        lis r9,ha16(LC1)
        la r2,lo16(LC0)(r2)
        lvx v0,0,r3
        lvx v1,0,r5
        la r9,lo16(LC1)(r9)
        lwz r12,-4(r1)
        lvx v12,0,r2
        lvx v13,0,r9
        vand v0,v0,v12
        stvx v0,0,r3
        vspltisw v0,-1
        vslw v12,v0,v0
        vandc v1,v1,v12
        stvx v1,0,r5
        lvx v0,0,r4
        vand v0,v0,v13
        stvx v0,0,r4
        mtspr 256,r12
        blr

llvm-svn: 27624
 2006-04-12 19:07:14 +00:00
Chris Lattner
ce6e988fa6 Rename get_VSPLI_elt -> get_VSPLTI_elt 
Canonicalize BUILD_VECTOR's that match VSPLTI's into a single type for each
form, eliminating a bunch of Pat patterns in the .td file and allowing us to
CSE stuff more aggressively.  This implements
PowerPC/buildvec_canonicalize.ll:VSPLTI

llvm-svn: 27614
 2006-04-12 17:37:20 +00:00
Chris Lattner
602d86f7af Ensure that zero vectors are always v4i32, which forces them to CSE with 
each other.  This implements CodeGen/PowerPC/vxor-canonicalize.ll

llvm-svn: 27609
 2006-04-12 16:53:28 +00:00
Chris Lattner
e12152a64b Vector function results go into V2 according to GCC. The darwin ABI doc 
doesn't say where they go :-/

llvm-svn: 27579
 2006-04-11 01:38:39 +00:00
Chris Lattner
5d1acb831a Move some return-handling code from lowerarguments to the ISD::RET handling stuff. 
No functionality change.

llvm-svn: 27577
 2006-04-11 01:21:43 +00:00
Chris Lattner
3c6e4a1dc9 properly mark vector selects as expanded to select_cc 
llvm-svn: 27544
 2006-04-08 22:59:15 +00:00
Chris Lattner
2ffa288a23 Add VRRC select support 
llvm-svn: 27543
 2006-04-08 22:45:08 +00:00
Chris Lattner
8234bfe18e Implement PowerPC/CodeGen/vec_splat.ll:spltish to use vsplish instead of a 
constant pool load.

llvm-svn: 27538
 2006-04-08 07:14:26 +00:00
Chris Lattner
e8defcff7d Change the interface to the predicate that determines if vsplti* can be used. 
No functionality changes.

llvm-svn: 27536
 2006-04-08 06:46:53 +00:00
Chris Lattner
a390188fd4 Make sure to return the result in the right type. 
llvm-svn: 27469
 2006-04-06 23:12:19 +00:00
Chris Lattner
c0680ae07e Match vpku[hw]um(x,x). 
Convert vsldoi(x,x) to work the same way other (x,x) cases work.

llvm-svn: 27467
 2006-04-06 22:28:36 +00:00
Chris Lattner
a52d88ee89 Add support for matching vmrg(x,x) patterns 
llvm-svn: 27463
 2006-04-06 22:02:42 +00:00
Chris Lattner
300076cbd8 Pattern match vmrg* instructions, which are now lowered by the CFE into shuffles. 
llvm-svn: 27457
 2006-04-06 21:11:54 +00:00
Chris Lattner
2875bb116e Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows the f.e. to 
lower it and LLVM to have one fewer intrinsic.  This implements
CodeGen/PowerPC/vec_shuffle.ll

llvm-svn: 27450
 2006-04-06 18:26:28 +00:00
Chris Lattner
10fa7be550 Compile the vpkuhum/vpkuwum intrinsics into vpkuhum/vpkuwum instead of into 
vperm with a perm mask lvx'd from the constant pool.

llvm-svn: 27448
 2006-04-06 17:23:16 +00:00
Chris Lattner
d1b47b18ed Fix CodeGen/PowerPC/2006-04-05-splat-ish.ll 
llvm-svn: 27439
 2006-04-05 17:39:25 +00:00
Evan Cheng
9e56e97205 Fallthrough to expand if a VECTOR_SHUFFLE cannot be custom lowered. 
llvm-svn: 27433
 2006-04-05 06:09:26 +00:00
Chris Lattner
d1483ca1ad Fix some broken logic that would cause us to codegen {2147483647,2147483647,2147483647,2147483647} as 'vspltisb v0, -1'. 
llvm-svn: 27413
 2006-04-04 22:28:35 +00:00
Chris Lattner
4e99e6dfdd Ask legalize to promote all vector shuffles to be v16i8 instead of having to 
handle all 4 PPC vector types.   This simplifies the matching code and allows
us to eliminate a bunch of patterns.  This also adds cases we were missing,
such as CodeGen/PowerPC/vec_splat.ll:splat_h.

llvm-svn: 27400
 2006-04-04 17:25:31 +00:00
Chris Lattner
0128e4d335 Revert accidentally committed hunks. 
llvm-svn: 27386
 2006-04-03 23:58:04 +00:00
Chris Lattner
57b9e01b3e Make sure to mark unsupported SCALAR_TO_VECTOR operations as expand. 
llvm-svn: 27385
 2006-04-03 23:55:43 +00:00
Chris Lattner
8ba4723c74 Inform the dag combiner that the predicate compares only return a low bit. 
llvm-svn: 27359
 2006-04-02 06:26:07 +00:00
Chris Lattner
da4217646a Custom lower all BUILD_VECTOR's so that we can compile vec_splat_u8(8) into 
"vspltisb v0, 8" instead of a constant pool load.

llvm-svn: 27335
 2006-04-02 00:43:36 +00:00
Chris Lattner
336d6646ab Rearrange code a bit 
llvm-svn: 27306
 2006-03-31 19:52:36 +00:00
Chris Lattner
786f782398 Add, sub and shuffle are legal for all vector types 
llvm-svn: 27305
 2006-03-31 19:48:58 +00:00
Chris Lattner
e3774da014 note to self: *save* file, then check it in 
llvm-svn: 27291
 2006-03-31 06:04:53 +00:00
Chris Lattner
95d358dbdb Implement an item from the readme, folding vcmp/vcmp. instructions with 
identical instructions into a single instruction.  For example, for:

void test(vector float *x, vector float *y, int *P) {
  int v = vec_any_out(*x, *y);
  *x = (vector float)vec_cmpb(*x, *y);
  *P = v;
}

we now generate:

_test:
        mfspr r2, 256
        oris r6, r2, 49152
        mtspr 256, r6
        lvx v0, 0, r4
        lvx v1, 0, r3
        vcmpbfp. v0, v1, v0
        mfcr r4, 2
        stvx v0, 0, r3
        rlwinm r3, r4, 27, 31, 31
        xori r3, r3, 1
        stw r3, 0(r5)
        mtspr 256, r2
        blr

instead of:

_test:
        mfspr r2, 256
        oris r6, r2, 57344
        mtspr 256, r6
        lvx v0, 0, r4
        lvx v1, 0, r3
        vcmpbfp. v2, v1, v0
        mfcr r4, 2
***     vcmpbfp v0, v1, v0
        rlwinm r4, r4, 27, 31, 31
        stvx v0, 0, r3
        xori r3, r4, 1
        stw r3, 0(r5)
        mtspr 256, r2
        blr

Testcase here: CodeGen/PowerPC/vcmp-fold.ll

llvm-svn: 27290
 2006-03-31 06:02:07 +00:00
Chris Lattner
e330741a6c Lower vector compares to VCMP nodes, just like we lower vector comparison 
predicates to VCMPo nodes.

llvm-svn: 27285
 2006-03-31 05:13:27 +00:00
Chris Lattner
a31d719e0a Mark INSERT_VECTOR_ELT as expand 
llvm-svn: 27276
 2006-03-31 01:48:55 +00:00
Nate Begeman
5a82c8ccbd Add a few more altivec intrinsics 
llvm-svn: 27215
 2006-03-28 04:15:58 +00:00
Chris Lattner
ac98e20cc9 Use normal lvx for scalar_to_vector instead of lve*x. They do the exact 
same thing and we have a dag node for the former.

llvm-svn: 27205
 2006-03-28 01:43:22 +00:00
Chris Lattner
d5da541d42 Tblgen doesn't like multiple SDNode<> definitions that map to the sameenum value. Split them into separate enums. 
llvm-svn: 27201
 2006-03-28 00:40:33 +00:00
Nate Begeman
3d518334b9 SelectionDAGISel can now natively handle Switch instructions, in the same 
manner that the LowerSwitch LLVM to LLVM pass does: emitting a binary
search tree of basic blocks.  The new approach has several advantages:
it is faster, it generates significantly smaller code in many cases, and
it paves the way for implementing dense switch tables as a jump table by
handling switches directly in the instruction selector.

This functionality is currently only enabled on x86, but should be safe for
every target.  In anticipation of making it the default, the cfg is now
properly updated in the x86, ppc, and sparc select lowering code.

llvm-svn: 27156
 2006-03-27 01:32:24 +00:00
Chris Lattner
65a455b060 Codegen vector predicate compares. 
llvm-svn: 27151
 2006-03-26 10:06:40 +00:00
Evan Cheng
b17bbf8ccb Remove PPC:isZeroVector, use ISD::isBuildVectorAllZeros instead 
llvm-svn: 27149
 2006-03-26 09:52:32 +00:00
Chris Lattner
57064915a6 Add some basic patterns for other datatypes 
llvm-svn: 27116
 2006-03-25 07:39:07 +00:00
Chris Lattner
0899b16b2d Codegen things like: 
<int -1, int -1, int -1, int -1>
and
 <int 65537, int 65537, int 65537, int 65537>

Using things like:
  vspltisb v0, -1
and:
  vspltish v0, 1

instead of using constant pool loads.

This implements CodeGen/PowerPC/vec_splat.ll:splat_imm_i{32|16}.

llvm-svn: 27106
 2006-03-25 06:12:06 +00:00
Chris Lattner
303dc30593 Disable the i32->float G5 optimization. It is unsafe, as documented in the 
comment.

This fixes 177.mesa, and McCat/09-vor with the td scheduler.

llvm-svn: 27060
 2006-03-24 07:53:47 +00:00
Chris Lattner
ba4966c16c add support for using vxor to build zero vectors. This implements 
Regression/CodeGen/PowerPC/vec_zero.ll

llvm-svn: 27059
 2006-03-24 07:48:08 +00:00
Chris Lattner
f84f3bf95b When possible, custom lower 32-bit SINT_TO_FP to this: 
_foo2:
        extsw r2, r3
        std r2, -8(r1)
        lfd f0, -8(r1)
        fcfid f0, f0
        frsp f1, f0
        blr

instead of this:

_foo2:
        lis r2, ha16(LCPI2_0)
        lis r4, 17200
        xoris r3, r3, 32768
        stw r3, -4(r1)
        stw r4, -8(r1)
        lfs f0, lo16(LCPI2_0)(r2)
        lfd f1, -8(r1)
        fsub f0, f1, f0
        frsp f1, f0
        blr

This speeds up Misc/pi from 2.44s->2.09s with LLC and from 3.01->2.18s
with llcbeta (16.7% and 38.1% respectively).

llvm-svn: 26943
 2006-03-22 05:30:33 +00:00
Chris Lattner
31a93c7740 These targets don't support EXTRACT_VECTOR_ELT, though, in time, X86 will. 
llvm-svn: 26930
 2006-03-21 20:51:05 +00:00
Chris Lattner
a498dd25d9 remove dead variable 
llvm-svn: 26907
 2006-03-20 22:37:23 +00:00
Chris Lattner
978628896b Fix a couple of bugs in permute/splat generate, thanks to Nate for actually 
figuring these out! :)

llvm-svn: 26904
 2006-03-20 18:26:51 +00:00
Chris Lattner
dc3605efdb Add support for generating vspltw, instead of a vperm instruction with a 
constant pool load.  This generates significantly nicer code for splats.

When tblgen gets bugfixed, we can remove the custom selection code.

llvm-svn: 26898
 2006-03-20 06:51:10 +00:00
Chris Lattner
8faa2cf693 Implement PPC::isSplatShuffleMask and PPC::getVSPLTImmediate. 
llvm-svn: 26897
 2006-03-20 06:37:44 +00:00
Chris Lattner
4b7aa59bbc fix duplicate definition errors 
llvm-svn: 26896
 2006-03-20 06:33:01 +00:00
Chris Lattner
0e56cf0d94 Custom lower arbitrary VECTOR_SHUFFLE's to VPERM. 
TODO: leave specific ones as VECTOR_SHUFFLE's and turn them into specialized
operations like vsplt*

llvm-svn: 26887
 2006-03-20 01:53:53 +00:00
Chris Lattner
789570bafb Custom lower SCALAR_TO_VECTOR into lve*x. 
llvm-svn: 26868
 2006-03-19 06:55:52 +00:00
Chris Lattner
80f9f7138a PPC doesn't have SCALAR_TO_VECTOR 
llvm-svn: 26865
 2006-03-19 06:17:19 +00:00
Chris Lattner
1bd0aaf2b8 rename these nodes 
llvm-svn: 26848
 2006-03-19 01:13:28 +00:00
Nate Begeman
42736d46b2 Remove BRTWOWAY* 
Make the PPC backend not dependent on BRTWOWAY_CC and make the branch
selector smarter about the code it generates, fixing a case in the
readme.

llvm-svn: 26814
 2006-03-17 01:40:33 +00:00
Evan Cheng
7ec94f2ff7 Added getTargetLowering() to TargetMachine. Refactored targets to support this. 
llvm-svn: 26742
 2006-03-13 23:20:37 +00:00
Chris Lattner
6b0947c277 Copysign needs to be expanded everywhere. Note that Alpha and IA64 should 
implement copysign as a native op if they have it.

llvm-svn: 26541
 2006-03-05 05:08:37 +00:00
Chris Lattner
137c02aa60 Compile this: 
void foo(float a, int *b) { *b = a; }

to this:

_foo:
        fctiwz f0, f1
        stfiwx f0, 0, r4
        blr

instead of this:

_foo:
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        stw r2, 0(r4)
        blr

This implements CodeGen/PowerPC/stfiwx.ll, and also incidentally does the
right thing for GCC bugzilla 26505.

llvm-svn: 26447
 2006-03-01 05:50:56 +00:00
Chris Lattner
6ba976b557 Use a target-specific dag-combine to implement CodeGen/PowerPC/fp-int-fp.ll. 
llvm-svn: 26445
 2006-03-01 04:57:39 +00:00
Evan Cheng
ebe1f272b7 Vector op lowering. 
llvm-svn: 26438
 2006-03-01 01:11:20 +00:00
Evan Cheng
305141c1ba - Added option -relocation-model to set relocation model. Valid values include static, pic, 
dynamic-no-pic, and default.
PPC and x86 default is dynamic-no-pic for Darwin, pic for others.
- Removed options -enable-pic and -ppc-static.

llvm-svn: 26315
 2006-02-22 20:19:42 +00:00
Chris Lattner
6bb2c3e9cd split register class handling from explicit physreg handling. 
llvm-svn: 26308
 2006-02-22 00:56:39 +00:00
Chris Lattner
a124432746 Updates to match change of getRegForInlineAsmConstraint prototype 
llvm-svn: 26305
 2006-02-21 23:11:00 +00:00
Evan Cheng
c4cda2b89a Moved PICEnabled to include/llvm/Target/TargetOptions.h 
llvm-svn: 26272
 2006-02-18 00:08:58 +00:00
Chris Lattner
9db5a1fd19 Switch to using getCALLSEQ_START instead of using our own creation calls 
llvm-svn: 26142
 2006-02-13 08:55:29 +00:00
Chris Lattner
abaa15898f Implement getConstraintType for PPC. 
llvm-svn: 26042
 2006-02-07 20:16:30 +00:00
Chris Lattner
50af3c6eb7 Add the simple PPC integer constraints 
llvm-svn: 26027
 2006-02-07 00:47:13 +00:00
Nate Begeman
0be60963bd Fix some of the stuff in the PPC README file, and clean up legalization 
of the SELECT_CC, BR_CC, and BRTWOWAY_CC nodes.

llvm-svn: 25875
 2006-02-01 07:19:44 +00:00
Evan Cheng
f115c17f23 Allow the specification of explicit alignments for constant pool entries. 
llvm-svn: 25855
 2006-01-31 22:23:14 +00:00
Chris Lattner
892fe31362 add info about the inline asm register constraints for PPC 
llvm-svn: 25853
 2006-01-31 19:20:21 +00:00
Nate Begeman
7a83bb4285 Codegen 
bool %test(int %X) {
  %Y = seteq int %X, 13
  ret bool %Y
}

as

_test:
        addi r2, r3, -13
        cntlzw r2, r2
        srwi r3, r2, 5
        blr

rather than

_test:
        cmpwi cr7, r3, 13
        mfcr r2
        rlwinm r3, r2, 31, 31, 31
        blr

This has very little effect on most code, but speeds up analyzer 23% and
mason 11%

llvm-svn: 25848
 2006-01-31 08:17:29 +00:00
Chris Lattner
2ab7f21d20 Functions that are lazily streamed in from the .bc file are *not* external. 
This fixes llvm-test/SingleSource/UnitTests/2006-01-29-SimpleIndirectCall.c
and PR704

llvm-svn: 25793
 2006-01-29 20:49:17 +00:00
Chris Lattner
4c6384a412 Now that OpActions is big enough, we can specify actions for vector types 
llvm-svn: 25784
 2006-01-29 08:41:37 +00:00
Chris Lattner
00e9ba5526 disable this for now 
llvm-svn: 25778
 2006-01-29 07:31:33 +00:00
Chris Lattner
6c4173145e Request expansion of ConstantVec nodes. 
llvm-svn: 25773
 2006-01-29 06:32:58 +00:00
Chris Lattner
5f0a3df176 Targets all now request ConstantFP to be legalized into TargetConstantFP. 
'fpimm' in .td files is now TargetConstantFP.

llvm-svn: 25771
 2006-01-29 06:26:08 +00:00
Chris Lattner
d2eb15f467 Fix a bug in my elimination of ISD::CALL this morning. PPC now has to 
provide the expansion for i64 calls itself

llvm-svn: 25735
 2006-01-28 07:33:03 +00:00
Chris Lattner
6a5d2450a3 Use PPCISD::CALL instead of ISD::CALL 
llvm-svn: 25717
 2006-01-27 23:34:02 +00:00
Chris Lattner
59a4f3f637 Make llvm.frame/returnaddr not crash on ppc 
llvm-svn: 25710
 2006-01-27 22:25:06 +00:00
Nate Begeman
d2c6fbef4a Remove TLI.LowerReturnTo, and just let targets custom lower ISD::RET for 
the same functionality.  This addresses another piece of bug 680.  Next,
on to fixing Alpha VAARG, which I broke last time.

llvm-svn: 25696
 2006-01-27 21:09:22 +00:00
Evan Cheng
7da6daadc7 Set SchedulingForLatency to be the default scheduling preference for all. 
llvm-svn: 25607
 2006-01-25 18:52:42 +00:00
Nate Begeman
c29fac7fce First part of bug 680: 
Remove TLI.LowerVA* and replace it with SDNodes that are lowered the same
way as everything else.

llvm-svn: 25606
 2006-01-25 18:21:52 +00:00
Evan Cheng
c368fb710f Default scheduling preference is SchedulingForLatency. 
llvm-svn: 25603
 2006-01-25 09:15:54 +00:00
Chris Lattner
23da2dc2ac Don't assert on 'select_cc SETUO' 
llvm-svn: 25423
 2006-01-18 19:42:35 +00:00
Chris Lattner
4d2c4cb7a7 Use the default impl of DYNAMIC_STACKALLOC, allowing us to delete some code. 
llvm-svn: 25334
 2006-01-15 09:02:48 +00:00
Nate Begeman
85b2dc0c4e bswap implementation 
llvm-svn: 25312
 2006-01-14 03:14:10 +00:00
Chris Lattner
c615ed65ac implement stacksave/stackrestore on PPC 
llvm-svn: 25277
 2006-01-13 17:52:03 +00:00
Chris Lattner
80fed2d66e expand unsupported stacksave/stackrestore nodes 
llvm-svn: 25272
 2006-01-13 02:42:53 +00:00
Nate Begeman
cff96008ac Add bswap, rotl, and rotr nodes 
Add dag combiner code to recognize rotl, rotr
Add ppc code to match rotl

Targets should add rotl/rotr patterns if they have them

llvm-svn: 25222
 2006-01-11 21:21:00 +00:00
Chris Lattner
5a59c88c17 Fix calls that need to store values in stack slots, to not copy the stack 
pointer.  This allows us to emit stuff like this:

        li r10, 0
        stw r10, 56(r1)
        or r3, r10, r10
        or r4, r10, r10
        or r5, r10, r10
        or r6, r10, r10
        or r7, r10, r10
        or r8, r10, r10
        or r9, r10, r10
        bl L_bar$stub

instead of this:

        or r2, r1, r1     ;; Extraneous copy.
        li r10, 0
        stw r10, 56(r2)
        or r3, r10, r10
        or r4, r10, r10
        or r5, r10, r10
        or r6, r10, r10
        or r7, r10, r10
        or r8, r10, r10
        or r9, r10, r10
        bl L_bar$stub

wowness.

llvm-svn: 25221
 2006-01-11 19:55:07 +00:00
Chris Lattner
6b8fb2e9bc Dead FP arguments still use an incoming FP reg. This fixes 
Regression/CodeGen/PowerPC/2006-01-11-darwin-fp-argument.ll, which was
distilled from a miscompilation in 252.eon.

llvm-svn: 25217
 2006-01-11 18:21:25 +00:00
Chris Lattner
a59d6394d2 Give PPCISD:: nodes legible names in dumps. 
llvm-svn: 25166
 2006-01-09 23:52:17 +00:00
Chris Lattner
d583872b9e linkonce symbols have an extra indirection, just like weak ones do. This fixes 
Prolangs-C++/family and Prolangs-C++/primes.

llvm-svn: 25119
 2006-01-06 01:04:03 +00:00
Jim Laskey
41b3ee3c4f Had expand logic backward. 
llvm-svn: 25105
 2006-01-05 01:47:43 +00:00
Jim Laskey
5eddaee9f3 Added initial support for DEBUG_LABEL allowing debug specific labels to be 
inserted in the code.

llvm-svn: 25104
 2006-01-05 01:25:28 +00:00
Nate Begeman
e1fdcf5a20 Remove a fixme 
llvm-svn: 25045
 2005-12-30 00:11:07 +00:00
Nate Begeman
96c7e22231 Fix one of the things in the todo file, and get a bit closer to folding 
constant offsets from statics into the address arithmetic.

llvm-svn: 24999
 2005-12-24 01:00:15 +00:00
Chris Lattner
8b2bd265f4 make sure bit_converts are expanded 
llvm-svn: 24978
 2005-12-23 05:13:35 +00:00
Chris Lattner
3c4a3dd86e Simplify some code by using BIT_CONVERT 
llvm-svn: 24974
 2005-12-23 00:59:59 +00:00
Nate Begeman
a114534620 Pattern-match return. Includes gross hack! 
llvm-svn: 24874
 2005-12-20 00:26:01 +00:00
Nate Begeman
9c7dce88b5 Convert load/store over to being pattern matched 
llvm-svn: 24871
 2005-12-19 23:25:09 +00:00
Nate Begeman
a0e26b25f4 Add support for TargetConstantPool nodes to the dag isel emitter, and use 
them in the PPC backend, to simplify some logic out of Select and
SelectAddr.

llvm-svn: 24657
 2005-12-10 02:36:00 +00:00
Chris Lattner
68a0fed879 Use new PPC-specific nodes to represent shifts which require the 6-bit 
amount handling that PPC provides.  These are generated by the lowering code
and prevents the dag combiner from assuming (rightfully) that the shifts
don't only look at 5 bits.  This fixes a miscompilation of crafty with
the new front-end.

llvm-svn: 24615
 2005-12-06 02:10:38 +00:00
Chris Lattner
90a87cc2f1 Fix Regression/CodeGen/PowerPC/2005-11-30-vastart-crash.ll 
llvm-svn: 24547
 2005-11-30 20:40:54 +00:00
Nate Begeman
9a781caa20 Hook up one type, v4f32, to the VR RegisterClass for now. 
llvm-svn: 24517
 2005-11-29 08:17:20 +00:00
Chris Lattner
47feb1ecbb No targets support line number info yet. 
llvm-svn: 24513
 2005-11-29 06:16:21 +00:00
Chris Lattner
b6af6ea0d4 add an option to generate completely non-pic code, corresponding to what 
gcc -static produces on PPC.  This is used for building kexts and other things.

With this, materializing the address of a global looks like:

        lis r2, ha16(L_H$non_lazy_ptr)
        la r3, lo16(L_H$non_lazy_ptr)(r2)

we're still emitting stubs for functions, which is wrong.  That is next.

llvm-svn: 24399
 2005-11-17 18:55:48 +00:00
Chris Lattner
7666d8192c Fix a bug that resistor on IRC hit where we tried to create token factor 
nodes of load results, not of their chain results.

llvm-svn: 24398
 2005-11-17 18:30:17 +00:00
Chris Lattner
4581404290 Enable global address legalization, fixing a todo and allowing the removal 
of some code.  This exposes the implicit load from the stubs to the DAG, allowing
them to be optimized by the dag combiner.  It also moves darwin specific stuff
out of the isel into the legalizer, and allows more to be moved to the .td file.

llvm-svn: 24397
 2005-11-17 18:26:56 +00:00
Chris Lattner
c46c672eba Use the right accessor to create this node 
llvm-svn: 24394
 2005-11-17 17:51:38 +00:00
Chris Lattner
8d04987a39 Add an initial hack at legalizing GlobalAddress into the appropriate nodes 
on Darwin to remove smarts from the isel.  This is currently disabled by
default (uncomment setOperationAction(ISD::GlobalAddress to enable it).
tblgen needs to become smarter about tglobaladdr nodes and bigger patterns
needed to be added to the .td file.  However, we can currently emit stuff like
this:  :)

        li r2, lo16(L_x$non_lazy_ptr)
        lis r3, ha16(L_x$non_lazy_ptr)
        lwzx r2, r3, r2

The obvious improvements will follow.

llvm-svn: 24390
 2005-11-17 07:30:41 +00:00
Chris Lattner
7fdf96ed96 When lowering direct calls, lower them to use a targetglobaladress directly 
instead of a globaladdress.  This has no effect on the generated code at all.

llvm-svn: 24386
 2005-11-17 05:56:14 +00:00
Chris Lattner
355e7583ab Fix an assert compiling MallocBench/gs 
llvm-svn: 24017
 2005-10-26 18:01:11 +00:00
Nate Begeman
c3ab5792e9 Correctly Expand or Promote FP_TO_UINT based on the capabilities of the 
machine.  This allows us to generate great code for i32 FP_TO_UINT now on
targets with 64 bit extensions.

llvm-svn: 23993
 2005-10-25 23:48:36 +00:00
Chris Lattner
8a53051fb1 Expose the fextend on the DAG instead of doing it in the matcher 
llvm-svn: 23986
 2005-10-25 20:54:57 +00:00
Nate Begeman
6c42f509bc Invert the TargetLowering flag that controls divide by consant expansion. 
Add a new flag to TargetLowering indicating if the target has really cheap
  signed division by powers of two, make ppc use it.  This will probably go
  away in the future.
Implement some more ISD::SDIV folds in the dag combiner
Remove now dead code in the x86 backend.

llvm-svn: 23853
 2005-10-21 00:02:42 +00:00
Nate Begeman
dc1a2a1f19 Move the target constant divide optimization up into the dag combiner, so 
that the nodes can be folded with other nodes, and we can not duplicate
code in every backend.  Alpha will probably want this too.

llvm-svn: 23835
 2005-10-20 02:15:44 +00:00
Nate Begeman
ee581735d9 Add the ability to lower return instructions to TargetLowering. This 
allows us to lower legal return types to something else, to meet ABI
requirements (such as that i64 be returned in two i32 regs on Darwin/ppc).

llvm-svn: 23802
 2005-10-18 23:23:37 +00:00
Nate Begeman
5dbcab491a Do the right thing and enable 64 bit regs under the control of a subtarget 
option.  Currently the only way to enable this is to specify the
64bitregs mattr flag.  It is never enabled by default on any config yet.

llvm-svn: 23779
 2005-10-18 00:56:42 +00:00
Nate Begeman
b0e319a7c7 First bits of 64 bit PowerPC stuff, currently disabled. A lot of this is 
purely mechanical.

llvm-svn: 23778
 2005-10-18 00:28:58 +00:00
Nate Begeman
723637974b More PPC32 -> PPC changes, as well as merging some classes that were 
redundant after the change.

llvm-svn: 23759
 2005-10-16 05:39:50 +00:00
Chris Lattner
d3946bbea6 Rename PPC32*.h to PPC*.h 
This completes the grand PPC file renaming

llvm-svn: 23745
 2005-10-14 23:59:06 +00:00
Chris Lattner
f1c1042780 fix an f32/f64 type mismatch 
llvm-svn: 23587
 2005-10-02 06:37:13 +00:00
Chris Lattner
50411b1026 Modify the ppc backend to use two register classes for FP: F8RC and F4RC. 
These are used to represent float and double values, and the two regclasses
contain the same physical registers.

llvm-svn: 23577
 2005-10-01 01:35:02 +00:00
Chris Lattner
9a2fb006e4 Add FP versions of the binary operators, keeping the int and fp worlds seperate. 
llvm-svn: 23506
 2005-09-28 22:29:58 +00:00
Chris Lattner
4bc65cb445 Darwin, like many BSD systems, has a setjmp/longjmp which saves the signal mask 
on setjmp calls and restores it on longjmp calls (both of which require syscalls).

This makes the calls REALLY slow.  Use _setjmp/_longjmp instead.  This speeds up
hexxagon from 120.31s to 15.68s: from 5.53x slower than GCC to 28% faster than GCC.

llvm-svn: 23482
 2005-09-27 22:18:25 +00:00
Chris Lattner
16700b9beb Change the arg lowering code to use copyfromreg from vregs associated 
with incoming arguments instead of the pregs themselves.  This fixes
the scheduler from causing problems by moving a copyfromreg for an argument
to after a select_cc node (now it can, and bad things won't happen).

llvm-svn: 23334
 2005-09-13 19:33:40 +00:00
Chris Lattner
766369cf45 Remove some dead vectors 
llvm-svn: 23329
 2005-09-13 18:47:49 +00:00
Chris Lattner
00e9278551 PowerPC cannot truncstore i1 natively 
llvm-svn: 23304
 2005-09-10 00:21:06 +00:00
Nate Begeman
718cae4eba Implement i64<->fp using the fctidz/fcfid instructions on PowerPC when we 
are allowed to generate 64-bit-only PowerPC instructions for 32 bit hosts,
such as the PowerPC 970.

This speeds up 189.lucas from 81.99 to 32.64 seconds.

llvm-svn: 23250
 2005-09-06 22:03:27 +00:00
Chris Lattner
813a73e8e9 Decouple fsqrt from gpul optimizations, implementing fsqrt.ll. 
Remove the -enable-gpopt option which is subsumed by feature flags.

llvm-svn: 23218
 2005-09-02 18:33:05 +00:00
Chris Lattner
6f9e01aa94 Restore this patch now that the latent bug has been fixed 
llvm-svn: 23209
 2005-09-02 01:24:55 +00:00
Chris Lattner
a58ee78b78 Revert the previous patch which causes a mysterious regression in toast. 
llvm-svn: 23207
 2005-09-02 00:47:05 +00:00
Chris Lattner
983190ce4c Implement small-arguments.ll:test3 by teaching the DAG optimizer that 
the results of calls to functions returning small values are properly
sign/zero extended.

llvm-svn: 23198
 2005-09-01 23:44:32 +00:00
Chris Lattner
914a0dbba1 Move FCTIWZ handling out of the instruction selectors and into legalization, 
getting them out of the business of making stack slots.

llvm-svn: 23180
 2005-08-31 21:09:52 +00:00
Chris Lattner
173c3fb3e5 Move SHL,SHR i64 -> legalizer 
llvm-svn: 23178
 2005-08-31 20:23:54 +00:00
Chris Lattner
353d5c167f lower sra_parts on the dag, implementing it for the dag isel, and exposing 
the ops to dag optimization.

llvm-svn: 23176
 2005-08-31 19:09:57 +00:00
Nate Begeman
c28c33f5a4 Enable generation of AssertSext and AssertZext in the PPC backend. 
llvm-svn: 23168
 2005-08-31 01:58:39 +00:00
Chris Lattner
a31670b930 Fix a bug in my patch for legalizing to fsel. It cannot handle seteq/setne, 
which I failed to include when I moved the code over.  This fixes
MallocBench/gs.

llvm-svn: 23140
 2005-08-30 00:45:18 +00:00
Chris Lattner
14f50b0df9 Fix some really strange indentation that xcode likes to use. 
no xcode, this is not right:

   if (!foo) break;
     X;

llvm-svn: 23138
 2005-08-30 00:19:00 +00:00
Chris Lattner
f470259b50 implement SELECT_CC fully for the DAG->DAG isel! 
llvm-svn: 23101
 2005-08-26 21:23:58 +00:00
Chris Lattner
04b88ca768 Make fsel emission work with both the pattern and dag-dag selectors, by 
giving it a non-instruction opcode.  The dag->dag selector used to not
select the operands of the fsel, because it thought that whole tree was
already selected.

llvm-svn: 23091
 2005-08-26 20:25:03 +00:00
Chris Lattner
007e1e07e3 implement the other half of the select_cc -> fsel lowering, which handles 
when the RHS of the comparison is 0.0.  Turn this on by default.

llvm-svn: 23083
 2005-08-26 17:36:52 +00:00
Chris Lattner
dae9a4899f add initial support for converting select_cc -> fsel in the legalizer 
instead of in the backend.  This currently handles fsel cases with registers,
but doesn't have the 0.0 and -0.0 optimization enabled yet.

Once this is finished, special hack for fp immediates can go away.

llvm-svn: 23075
 2005-08-26 00:52:45 +00:00
Nate Begeman
0f60221f2e Remove option to make SetCC illegal on PowerPC after long discussion with 
Chris.  This will be accomplished through correctly modeling CR's and
subregs.

llvm-svn: 23056
 2005-08-25 20:01:10 +00:00
Nate Begeman
a8ec868347 Ack, typo 
llvm-svn: 22981
 2005-08-23 05:45:10 +00:00
Nate Begeman
0ac8bb2fe1 Add an option to make SetCC illegal as a beta option 
llvm-svn: 22979
 2005-08-23 05:42:36 +00:00
Jim Laskey
cdde9cec23 Make UINT_TO_FP and SINT_TO_FP use generic expansion. 
llvm-svn: 22815
 2005-08-17 00:40:22 +00:00
Chris Lattner
9d651a0e3c updates for changes in nodes 
llvm-svn: 22808
 2005-08-16 21:58:15 +00:00
Nate Begeman
f6b6378f23 Implement BR_CC and BRTWOWAY_CC. This allows the removal of a rather nasty 
fixme from the PowerPC backend.  Emit slightly better code for legalizing
select_cc.

llvm-svn: 22805
 2005-08-16 19:49:35 +00:00
Chris Lattner
a4c5954c52 Pull the LLVM -> DAG lowering code out of the pattern selector so that it 
can be shared with the DAG->DAG selector.

llvm-svn: 22799
 2005-08-16 17:14:42 +00:00