Commit Graph

726 Commits

Author SHA1 Message Date
Evan Cheng
da95a84a11 fold (shl x, 1) -> (add x, x)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25120 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-06 01:06:31 +00:00
Evan Cheng
17c428e23d Support for custom lowering of ISD::RET.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25116 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-06 00:41:43 +00:00
Jim Laskey
abf6d1784b Added initial support for DEBUG_LABEL allowing debug specific labels to be
inserted in the code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25104 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-05 01:25:28 +00:00
Jim Laskey
b2efb853f0 Applied some recommend changes from sabre. The dominate one beginning "let the
pass manager do it's thing."  Fixes crash when compiling -g files and suppresses
dwarf statements if no debug info is present.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25100 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-04 22:28:25 +00:00
Jim Laskey
5bf6f25b4a Add unique id to debug location for debug label use (work in progress.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25096 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-04 15:04:11 +00:00
Jim Laskey
7ca9ab3f5c Change how MachineDebugInfo is fetched.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25089 91177308-0d34-0410-b5e6-96231b3b80d8
2006-01-04 13:42:59 +00:00
Nate Begeman
512beb9465 Make sure to pass the offset into the new node, so that we don't silently
drop it on the floor.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25044 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-30 00:10:38 +00:00
Duraid Madina
f65d917f71 purity++
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25041 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-29 05:59:19 +00:00
Andrew Lenharth
57030e36e4 allow custom lowering to return null for legal results
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25007 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-25 01:07:37 +00:00
Andrew Lenharth
e8f65f1e62 Support Custom lowering of a few more operations.
Alpha needs to custom lower *DIV and *REM


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@25006 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-24 23:42:32 +00:00
Jim Laskey
d6e8d41340 Remove redundant debug locations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24995 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 20:08:28 +00:00
Chris Lattner
c26f7a033f unbreak the build :-/
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24992 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 16:12:20 +00:00
Evan Cheng
f3fd9fe20d Allow custom lowering of LOAD, EXTLOAD, ZEXTLOAD, STORE, and TRUNCSTORE. Not
currently used.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24988 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 07:29:34 +00:00
Chris Lattner
c33baaafb6 Simplify store(bitconv(x)) to store(x). This allows us to compile this:
void bar(double Y, double *X) {
  *X = Y;
}

to this:

bar:
        save -96, %o6, %o6
        st %i1, [%i2+4]
        st %i0, [%i2]
        restore %g0, %g0, %g0
        retl
        nop

instead of this:

bar:
        save -104, %o6, %o6
        st %i1, [%i6+-4]
        st %i0, [%i6+-8]
        ldd [%i6+-8], %f0
        std  %f0, [%i2]
        restore %g0, %g0, %g0
        retl
        nop

on sparcv8.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24983 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 05:48:07 +00:00
Chris Lattner
5710410dea fold (conv (load x)) -> (load (conv*)x).
This allows us to compile this:
void foo(double);
void bar(double *X) { foo(*X); }

To this:

bar:
        save -96, %o6, %o6
        ld [%i0+4], %o1
        ld [%i0], %o0
        call foo
        nop
        restore %g0, %g0, %g0
        retl
        nop

instead of this:

bar:
        save -104, %o6, %o6
        ldd [%i0], %f0
        std %f0, [%i6+-8]
        ld [%i6+-4], %o1
        ld [%i6+-8], %o0
        call foo
        nop
        restore %g0, %g0, %g0
        retl
        nop

on SparcV8.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24982 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 05:44:41 +00:00
Chris Lattner
c8547d856d Fold bitconv(bitconv(x)) -> x. We now compile this:
void foo(double);
void bar(double X) { foo(X); }

to this:

bar:
        save -96, %o6, %o6
        or %g0, %i0, %o0
        or %g0, %i1, %o1
        call foo
        nop
        restore %g0, %g0, %g0
        retl
        nop

instead of this:

bar:
        save -112, %o6, %o6
        st %i1, [%i6+-4]
        st %i0, [%i6+-8]
        ldd [%i6+-8], %f0
        std %f0, [%i6+-16]
        ld [%i6+-12], %o1
        ld [%i6+-16], %o0
        call foo
        nop
        restore %g0, %g0, %g0
        retl
        nop

on V8.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24981 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 05:37:50 +00:00
Chris Lattner
94683777ae constant fold bits_convert in getNode and in the dag combiner for fp<->int
conversions.  This allows V8 to compiles this:

void %test() {
        call float %test2( float 1.000000e+00, float 2.000000e+00, double 3.000000e+00, double* null )
        ret void
}

into:

test:
        save -96, %o6, %o6
        sethi 0, %o3
        sethi 1049088, %o2
        sethi 1048576, %o1
        sethi 1040384, %o0
        or %g0, %o3, %o4
        call test2
        nop
        restore %g0, %g0, %g0
        retl
        nop

instead of:

test:
        save -112, %o6, %o6
        sethi 0, %o4
        sethi 1049088, %l0
        st %o4, [%i6+-12]
        st %l0, [%i6+-16]
        ld [%i6+-12], %o3
        ld [%i6+-16], %o2
        sethi 1048576, %o1
        sethi 1040384, %o0
        call test2
        nop
        restore %g0, %g0, %g0
        retl
        nop


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24980 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 05:30:37 +00:00
Chris Lattner
232348d66f Fix a pasto
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24973 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 00:52:30 +00:00
Chris Lattner
ed7b5babfb fix a thinko in the bit_convert handling code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24972 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 00:50:25 +00:00
Chris Lattner
35481892da add very simple support for the BIT_CONVERT node
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24970 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-23 00:16:34 +00:00
Chris Lattner
0fcd40f501 remove dead code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24965 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-22 21:16:08 +00:00
Chris Lattner
1ccf26a639 The 81st column doesn't like code in it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24943 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-22 05:23:45 +00:00
Evan Cheng
ff9be11da2 Lefted out TargetLowering::
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24922 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 23:14:54 +00:00
Evan Cheng
3a03ebb377 * Fix a GlobalAddress lowering bug.
* Teach DAG combiner about X86ISD::SETCC by adding a TargetLowering hook.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24921 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 23:05:39 +00:00
Jim Laskey
e81aecbae6 Disengage DEBUG_LOC from non-PPC targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24919 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 20:51:37 +00:00
Evan Cheng
d5781fca4f * Added support for X86 RET with an additional operand to specify number of
bytes to pop off stack.
* Added support for X86 SETCC.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24917 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 20:21:51 +00:00
Chris Lattner
f9dee6a047 make sure to relegalize all cases
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24911 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 19:40:42 +00:00
Chris Lattner
9797c5cc3e enable the gep isel opt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24910 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 19:36:36 +00:00
Chris Lattner
8339736334 fix a bug I introduced that broke recursive expansion of nodes (e.g. scalarizing vectors)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24905 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 18:02:52 +00:00
Chris Lattner
3b841e9f52 Lower ConstantAggregateZero into zeros
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24890 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-21 02:43:26 +00:00
Evan Cheng
7226158d7e Added a hook to print out names of target specific DAG nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24877 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-20 06:22:03 +00:00
Chris Lattner
69a889eb35 Fix a nasty latent bug in the legalizer that was triggered by my patch
last night, breaking crafty and twolf.  Make sure that the newly found
legal nodes are themselves not re-legalized until the next iteration.

Also, since this functionality exists now, we can reduce number of legalizer
iterations by depending on this behavior instead of having to misuse 'do
another iteration' to get the same effect.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24875 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-20 00:53:54 +00:00
Evan Cheng
898101c15f X86 conditional branch support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24870 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-19 23:12:38 +00:00
Evan Cheng
115c036a4c Print out opcode number if it's an unknown target node.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24869 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-19 23:11:49 +00:00
Chris Lattner
9f8cc69009 Fix a case where the DAG Combiner would accidentally CSE flag-producing nodes,
creating graphs that cannot be scheduled.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24866 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-19 22:21:21 +00:00
Jim Laskey
d845582d4a Amend comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24861 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-19 16:32:26 +00:00
Jim Laskey
de48ee28a3 Create a strong dependency for loads following stores. This will leave a
latency period between the two.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24860 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-19 16:30:13 +00:00
Chris Lattner
e773673d8e Make sure to relegalize new nodes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24843 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 23:54:29 +00:00
Jeff Cohen
18840db533 Keep VC++ happy.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24835 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 22:20:05 +00:00
Chris Lattner
f1a47c3371 More corrections for flagged copyto/from reg
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24828 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 15:36:21 +00:00
Chris Lattner
7310fb11f2 legalize copytoreg and copyfromreg nodes that have flag operands correctly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24826 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 15:27:43 +00:00
Jim Laskey
bd2b621486 Fix a bug Sabre was having where the DAG root was a group. The group dominator
needed to be added to the ordering list, not the first member of the group.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24816 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 04:40:52 +00:00
Jim Laskey
9022ed955a Groups were not emitted if the dominator node and the node in the ordering list
were not the same node.  Ultimately the test was bogus.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24815 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 03:59:21 +00:00
Chris Lattner
a5282d8df2 Simplify code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24806 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-18 01:03:46 +00:00
Chris Lattner
181b7a382f allow custom expansion of BR_CC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24804 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-17 23:46:46 +00:00
Evan Cheng
7df96d6672 X86 lowers SELECT to a cmp / test followed by a conditional move.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24754 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-17 01:21:05 +00:00
Jim Laskey
f5395cee6a Added source file/line correspondence for dwarf (PowerPC only at this point.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24748 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-16 22:45:29 +00:00
Chris Lattner
d0f6d18aa8 Don't create SEXTLOAD/ZEXTLOAD instructions that the target doesn't support
if after legalize.  This fixes IA64 failures.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24725 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-15 19:02:38 +00:00
Chris Lattner
d4771840fd When folding loads into ops, immediately replace uses of the op with the
load.  This reduces number of worklist iterations and avoid missing optimizations
depending on folding of things into sext_inreg nodes (which aren't supported by
all targets).
Tested by Regression/CodeGen/X86/extend.ll:test2


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24712 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-14 19:25:30 +00:00
Chris Lattner
ad25d4e2df Fix the (zext (zextload)) case to trigger, similarly for sign extends.
Allow (zext (truncate)) to apply after legalize if the target supports
AND (which all do).

This compiles
short %foo() {
        %tmp.0 = load ubyte* %X         ; <ubyte> [#uses=1]
        %tmp.3 = cast ubyte %tmp.0 to short             ; <short> [#uses=1]
        ret short %tmp.3
}

to:
_foo:
        movzbl _X, %eax
        ret

instead of:

_foo:
        movzbl _X, %eax
        movzbl %al, %eax
        ret

thanks to Evan for pointing this out.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24709 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-14 19:05:06 +00:00
Chris Lattner
00cb95c9be Fix a miscompilation in crafty due to a recent patch
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24706 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-14 07:58:38 +00:00
Evan Cheng
110dec2000 Fold (zext (load x) to (zextload x).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24702 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-14 02:19:23 +00:00
Chris Lattner
86cb643801 Don't lump the filename and working dir together
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24697 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-13 17:40:33 +00:00
Nate Begeman
d7d746f603 Lowering constant pool entries on ppc exposed a bug in the recently added
ConstantVec legalizing code, which would return constantpool nodes that
were not of the target's pointer type.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24691 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-13 03:03:23 +00:00
Chris Lattner
ac22c83e68 Accept and ignore prefetches for now
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24678 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-12 22:51:16 +00:00
Chris Lattner
58f7963b8b Fix CodeGen/Generic/2005-12-12-ExpandSextInreg.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24677 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-12 22:27:43 +00:00
Chris Lattner
3802c2552f Minor tweak to get isel opt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24663 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-11 09:05:13 +00:00
Nate Begeman
28a6b02626 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24657 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-10 02:36:00 +00:00
Evan Cheng
7038daf342 Added new getNode and getTargetNode variants for X86 stores.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24653 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-10 00:37:58 +00:00
Chris Lattner
5e3c5b4e13 Teach legalize how to promote sext_inreg to fix a problem Andrew pointed
out to me.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24644 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-09 17:32:47 +00:00
Chris Lattner
c78b0b740b improve code insertion in two ways:
1. Only forward subst offsets into loads and stores, not into arbitrary
   things, where it will likely become a load.
2. If the source is a cast from pointer, forward subst the cast as well,
   allowing us to fold the cast away (improving cases when the cast is
   from an alloca or global).

This hasn't been fully tested, but does appear to further reduce register
pressure and improve code.  Lets let the testers grind on it a bit. :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24640 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-08 08:00:12 +00:00
Nate Begeman
cc827e60b6 Fix a crash where ConstantVec nodes were being generated with the wrong
type when the target did not support them.  Also teach Legalize how to
expand ConstantVecs.

This allows us to generate

_test:
        lwz r2, 12(r3)
        lwz r4, 8(r3)
        lwz r5, 4(r3)
        lwz r6, 0(r3)
        addi r2, r2, 4
        addi r4, r4, 3
        addi r5, r5, 2
        addi r6, r6, 1
        stw r2, 12(r3)
        stw r4, 8(r3)
        stw r5, 4(r3)
        stw r6, 0(r3)
        blr

For:

void %test(%v4i *%P) {
        %T = load %v4i* %P
        %S = add %v4i %T, <int 1, int 2, int 3, int 4>
        store %v4i %S, %v4i * %P
        ret void
}

On PowerPC.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24633 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-07 19:48:11 +00:00
Chris Lattner
cc2210b4fa Only transform (sext (truncate x)) -> (sextinreg x) if before legalize or
if the target supports the resultant sextinreg


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24632 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-07 18:02:05 +00:00
Chris Lattner
b14ab8a10d Teach the dag combiner to turn a truncate/sign_extend pair into a sextinreg
when the types match up.  This allows the X86 backend to compile:

sbyte %toggle_value(sbyte* %tmp.1) {
        %tmp.2 = load sbyte* %tmp.1
        ret sbyte %tmp.2
}

to this:

_toggle_value:
        mov %EAX, DWORD PTR [%ESP + 4]
        movsx %EAX, BYTE PTR [%EAX]
        ret

instead of this:

_toggle_value:
        mov %EAX, DWORD PTR [%ESP + 4]
        movsx %EAX, BYTE PTR [%EAX]
        movsx %EAX, %AL
        ret

noticed in Shootout/objinst.

-Chris


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24630 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-07 07:11:03 +00:00
Nate Begeman
8cfa57b1b4 Teach the SelectionDAG ISel how to turn ConstantPacked values into
constant nodes with vector types.  Also teach the asm printer how to print
ConstantPacked constant pool entries.  This allows us to generate altivec
code such as the following, which adds a vector constantto a packed float.

LCPI1_0:  <4 x float> < float 0.0e+0, float 0.0e+0, float 0.0e+0, float 1.0e+0 >
        .space  4
        .space  4
        .space  4
        .long   1065353216      ; float 1
        .text
        .align  4
        .globl  _foo
_foo:
        lis r2, ha16(LCPI1_0)
        la r2, lo16(LCPI1_0)(r2)
        li r4, 0
        lvx v0, r4, r2
        lvx v1, r4, r3
        vaddfp v0, v1, v0
        stvx v0, r4, r3
        blr

For the llvm code:

void %foo(<4 x float> * %a) {
entry:
  %tmp1 = load <4 x float> * %a;
  %tmp2 = add <4 x float> %tmp1, < float 0.0, float 0.0, float 0.0, float 1.0 >
  store <4 x float> %tmp2, <4 x float> *%a
  ret void
}


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24616 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-06 06:18:55 +00:00
Chris Lattner
c88d8e944d Fix the #1 code quality problem that I have seen on X86 (and it also affects
PPC and other targets).  In a particular, consider code like this:

struct Vector3 { double x, y, z; };
struct Matrix3 { Vector3 a, b, c; };
double dot(Vector3 &a, Vector3 &b) {
   return a.x * b.x  +  a.y * b.y  +  a.z * b.z;
}
Vector3 mul(Vector3 &a, Matrix3 &b) {
   Vector3 r;
   r.x = dot( a, b.a );
   r.y = dot( a, b.b );
   r.z = dot( a, b.c );
   return r;
}
void transform(Matrix3 &m, Vector3 *x, int n) {
   for (int i = 0; i < n; i++)
      x[i] = mul( x[i], m );
}

we compile transform to a loop with all of the GEP instructions for indexing
into 'm' pulled out of the loop (9 of them).  Because isel occurs a bb at a time
we are unable to fold the constant index into the loads in the loop, leading to
PPC code that looks like this:

LBB3_1: ; no_exit.preheader
        li r2, 0
        addi r6, r3, 64        ;; 9 values live across the loop body!
        addi r7, r3, 56
        addi r8, r3, 48
        addi r9, r3, 40
        addi r10, r3, 32
        addi r11, r3, 24
        addi r12, r3, 16
        addi r30, r3, 8
LBB3_2: ; no_exit
        lfd f0, 0(r30)
        lfd f1, 8(r4)
        fmul f0, f1, f0
        lfd f2, 0(r3)        ;; no constant indices folded into the loads!
        lfd f3, 0(r4)
        lfd f4, 0(r10)
        lfd f5, 0(r6)
        lfd f6, 0(r7)
        lfd f7, 0(r8)
        lfd f8, 0(r9)
        lfd f9, 0(r11)
        lfd f10, 0(r12)
        lfd f11, 16(r4)
        fmadd f0, f3, f2, f0
        fmul f2, f1, f4
        fmadd f0, f11, f10, f0
        fmadd f2, f3, f9, f2
        fmul f1, f1, f6
        stfd f0, 0(r4)
        fmadd f0, f11, f8, f2
        fmadd f1, f3, f7, f1
        stfd f0, 8(r4)
        fmadd f0, f11, f5, f1
        addi r29, r4, 24
        stfd f0, 16(r4)
        addi r2, r2, 1
        cmpw cr0, r2, r5
        or r4, r29, r29
        bne cr0, LBB3_2 ; no_exit

uh, yuck.  With this patch, we now sink the constant offsets into the loop, producing
this code:

LBB3_1: ; no_exit.preheader
        li r2, 0
LBB3_2: ; no_exit
        lfd f0, 8(r3)
        lfd f1, 8(r4)
        fmul f0, f1, f0
        lfd f2, 0(r3)
        lfd f3, 0(r4)
        lfd f4, 32(r3)       ;; much nicer.
        lfd f5, 64(r3)
        lfd f6, 56(r3)
        lfd f7, 48(r3)
        lfd f8, 40(r3)
        lfd f9, 24(r3)
        lfd f10, 16(r3)
        lfd f11, 16(r4)
        fmadd f0, f3, f2, f0
        fmul f2, f1, f4
        fmadd f0, f11, f10, f0
        fmadd f2, f3, f9, f2
        fmul f1, f1, f6
        stfd f0, 0(r4)
        fmadd f0, f11, f8, f2
        fmadd f1, f3, f7, f1
        stfd f0, 8(r4)
        fmadd f0, f11, f5, f1
        addi r6, r4, 24
        stfd f0, 16(r4)
        addi r2, r2, 1
        cmpw cr0, r2, r5
        or r4, r6, r6
        bne cr0, LBB3_2 ; no_exit

This is much nicer as it reduces register pressure in the loop a lot.  On X86,
this takes the function from having 9 spilled registers to 2.  This should help
some spec programs on X86 (gzip?)

This is currently only enabled with -enable-gep-isel-opt to allow perf testing
tonight.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24606 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-05 07:10:48 +00:00
Chris Lattner
d67b3a8bf7 dbg.stoppoint returns a value, don't forget to init it
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24583 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-03 18:50:48 +00:00
Andrew Lenharth
cde0f5cfe7 bah, must generate all results
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24574 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-02 06:08:08 +00:00
Andrew Lenharth
49c709f891 cycle counter fix
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24573 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-02 04:56:24 +00:00
Chris Lattner
fe14b34d83 Don't remove two operand, two result nodes from the binary ops map. These
should come from the arbitrary ops map.

This fixes Regression/CodeGen/PowerPC/2005-12-01-Crash.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24571 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-01 23:14:50 +00:00
Chris Lattner
9ad17c9c9a Promote line and column number information for our friendly 64-bit targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24568 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-01 18:21:35 +00:00
Chris Lattner
c5e6c649b5 This is a bugfix for SelectNodeTo. In certain situations, we could be
selecting a node and use a mix of getTargetNode() and SelectNodeTo.  Because
SelectNodeTo didn't check the CSE maps for a preexisting node and didn't insert
its result into the CSE maps, we would sometimes miss a CSE opportunity.

This is extremely rare, but worth fixing for completeness.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24565 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-01 18:00:57 +00:00
Nate Begeman
6510b22cec Support multiple ValueTypes per RegisterClass, needed for upcoming vector
work.  This change has no effect on generated code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24563 91177308-0d34-0410-b5e6-96231b3b80d8
2005-12-01 04:51:06 +00:00
Chris Lattner
eb19e40efb Make SelectNodeTo return N
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24548 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 22:45:14 +00:00
Chris Lattner
c85a9f37e9 CALLSEQ_START/END nodes don't get memoized, do not add them in when
replaceAllUses'ing.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24539 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 18:20:52 +00:00
Andrew Lenharth
ae35575957 At long last, you can say that f32 isn't supported for setcc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24537 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 17:12:26 +00:00
Nate Begeman
f43a3ca26d First chunk of actually generating vector code for packed types. These
changes allow us to generate the following code:

_foo:
        li r2, 0
        lvx v0, r2, r3
        vaddfp v0, v0, v0
        stvx v0, r2, r3
        blr

for this llvm:

void %foo(<4 x float>* %a) {
entry:
        %tmp1 = load <4 x float>* %a
        %tmp2 = add <4 x float> %tmp1, %tmp1
        store <4 x float> %tmp2, <4 x float>* %a
        ret void
}


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24534 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 08:22:07 +00:00
Andrew Lenharth
5b5b8c2755 add support for custom lowering SINT_TO_FP
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24531 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 06:43:03 +00:00
Reid Spencer
6ff7240a5c Fix a problem with llvm-ranlib that (on some platforms) caused the archive
file to become corrupted due to interactions between mmap'd memory segments
and file descriptors closing. The problem is completely avoiding by using
a third temporary file.

Patch provided by Evan Jones


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24527 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 05:21:10 +00:00
Evan Cheng
14229bb636 Fixed a bug introduced by my last commit: TargetGlobalValues should key on
GlobalValue * and index pair. Update getGlobalAddress() for symmetry.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24524 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 02:49:21 +00:00
Evan Cheng
61ca74bc3a Added an index field to GlobalAddressSDNode so it can represent X+12, etc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24523 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-30 02:04:11 +00:00
Chris Lattner
36ce69195e Add support for a new STRING and LOCATION node for line number support, patch
contributed by Daniel Berlin, with a few cleanups here and there by me.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24515 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-29 06:21:05 +00:00
Nate Begeman
6a648614e8 Add the majority of the vector machien value types we expect to support,
and make a few changes to the legalization machinery to support more than
16 types.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24511 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-29 05:45:29 +00:00
Nate Begeman
ab48be3772 Check in code to scalarize arbitrarily wide packed types for some simple
vector operations (load, add, sub, mul).

This allows us to codegen:
void %foo(<4 x float> * %a) {
entry:
  %tmp1 = load <4 x float> * %a;
  %tmp2 = add <4 x float> %tmp1, %tmp1
  store <4 x float> %tmp2, <4 x float> *%a
  ret void
}

on ppc as:
_foo:
        lfs f0, 12(r3)
        lfs f1, 8(r3)
        lfs f2, 4(r3)
        lfs f3, 0(r3)
        fadds f0, f0, f0
        fadds f1, f1, f1
        fadds f2, f2, f2
        fadds f3, f3, f3
        stfs f0, 12(r3)
        stfs f1, 8(r3)
        stfs f2, 4(r3)
        stfs f3, 0(r3)
        blr


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24484 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-22 18:16:00 +00:00
Nate Begeman
4ef3b817fe Rather than attempting to legalize 1 x float, make sure the SD ISel never
generates it.  Make MVT::Vector expand-only, and remove the code in
Legalize that attempts to legalize it.

The plan for supporting N x Type is to continually epxand it in ExpandOp
until it gets down to 2 x Type, where it will be scalarized into a pair of
scalars.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24482 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-22 01:29:36 +00:00
Chris Lattner
308575be66 Legalize MERGE_VALUES, expand READCYCLECOUNTER correctly, so it doesn't
break control dependence.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24437 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-20 22:56:56 +00:00
Andrew Lenharth
f70e30b41d The first patch of X86 support for read cycle counter
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24429 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-20 21:32:07 +00:00
Chris Lattner
6bf234c4a8 more progress towards bug 291 being finished. Patch by Owen Anderson,
HAVE_GV case fixed up by me.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24428 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-20 03:45:52 +00:00
Chris Lattner
b67eb9131c Unbreak codegen of bools. This should fix the llc/jit/llc-beta failures
from last night.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24427 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 18:40:42 +00:00
Chris Lattner
3aa2c74e19 Improve Selection DAG printer portability. Patch by Owen Anderson!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24425 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 07:44:09 +00:00
Chris Lattner
44fa764355 Teach the graph viewer to handle register operands that are zero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24421 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 06:58:46 +00:00
Chris Lattner
b89175feff Silence a bogus warning
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24420 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 05:51:46 +00:00
Chris Lattner
0fb094fd60 Add some method variants, patch by Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24418 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 01:44:53 +00:00
Nate Begeman
5fbb5d2459 Teach LLVM how to scalarize packed types. Currently, this only works on
packed types with an element count of 1, although more generic support is
coming.  This allows LLVM to turn the following code:

void %foo(<1 x float> * %a) {
entry:
  %tmp1 = load <1 x float> * %a;
  %tmp2 = add <1 x float> %tmp1, %tmp1
  store <1 x float> %tmp2, <1 x float> *%a
  ret void
}

Into:

_foo:
        lfs f0, 0(r3)
        fadds f0, f0, f0
        stfs f0, 0(r3)
        blr


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24416 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-19 00:36:38 +00:00
Nate Begeman
e21ea61588 Split out the shift code from visitBinary.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24412 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-18 07:42:56 +00:00
Chris Lattner
0c8fbe33a4 Allow targets to custom legalize leaf nodes like GlobalAddress.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24387 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-17 06:41:44 +00:00
Chris Lattner
b9debbf54c Teach legalize about targetglobaladdress
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24385 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-17 05:52:24 +00:00
Chris Lattner
b1a5a5c4c0 when debugging lower dbg intrinsics to calls
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24377 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-16 07:22:30 +00:00
Jeff Cohen
06d9b4ab24 Fix operator precedence bug caught by VC++.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24318 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-12 00:59:01 +00:00
Andrew Lenharth
8b91c77385 added a chain output
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@24306 91177308-0d34-0410-b5e6-96231b3b80d8
2005-11-11 22:48:54 +00:00