Commit Graph

551 Commits

Author SHA1 Message Date
Nate Begeman
b7f6ef12f6 Allow custom lowered FP_TO_SINT ops in the check for whether a larger
FP_TO_SINT is preferred to a larger FP_TO_UINT.  This seems to be begging
for a TLI.isOperationCustom() helper function.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23992 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-25 23:47:25 +00:00
Chris Lattner
a158eee313 Clear a bit in this file that was causing a miscompilation of 178.galgel.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23980 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-25 18:57:30 +00:00
Andrew Lenharth
2a2de66db2 add TargetExternalSymbol
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23886 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-23 03:40:17 +00:00
Chris Lattner
e9936d14b8 BuildSDIV and BuildUDIV only work for i32/i64, but they don't check that
the input is that type, this caused a failure on gs on X86 last night.
Move the hard checks into Build[US]Div since that is where decisions like
this should be made.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23881 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-22 18:50:15 +00:00
Chris Lattner
c8597ca698 add a case missing from the dag combiner that exposed the failure on
2005-10-21-longlonggtu.ll.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23875 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-21 21:23:25 +00:00
Nate Begeman
4d38567689 Fix a typo in the dag combiner, so that this can work on i64 targets
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23856 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-21 01:51:45 +00:00
Nate Begeman
405e3ecb56 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23853 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-21 00:02:42 +00:00
Nate Begeman
c6a454e8d5 Fix a couple bugs in the const div stuff where we'd generate MULHS/MULHU
for types that aren't legal, and fail a divisor is less than zero
comparison, which would cause us to drop a subtract.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23846 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-20 17:45:03 +00:00
Chris Lattner
f75f2a0a02 don't use llabs with apparently VC++ doesn't have
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23845 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-20 17:01:00 +00:00
Nate Begeman
6957523b9d 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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23835 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-20 02:15:44 +00:00
Nate Begeman
5dc897b0e4 Teach Legalize how to do something with EXTRACT_ELEMENT when the type of
the pair of elements is a legal type.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23804 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-19 00:06:56 +00:00
Nate Begeman
4a95945fa5 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).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23802 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 23:23:37 +00:00
Chris Lattner
d222f6ab67 Fix Generic/2005-10-18-ZeroSizeStackObject.ll by not requesting a zero
sized stack object if either the array size or the type size is zero.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23801 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 22:14:06 +00:00
Chris Lattner
2dfa8192ab remove hack
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23797 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 22:11:42 +00:00
Chris Lattner
40c62d5069 Fold (select C, load A, load B) -> load (select C, A, B). This happens quite
a lot throughout many programs.  In particular, specfp triggers it a bunch for
constant FP nodes when you have code like  cond ? 1.0 : -1.0.

If the PPC ISel exposed the loads implicit in pic references to external globals,
we would be able to eliminate a load in cases like this as well:

%X = external global int
%Y = external global int
int* %test4(bool %C) {
        %G = select bool %C, int* %X, int* %Y
        ret int* %G
}

Note that this breaks things that use SrcValue's (see the fixme), but since nothing
uses them yet, this is ok.

Also, simplify some code to use hasOneUse() on an SDOperand instead of hasNUsesOfValue directly.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23781 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 06:04:22 +00:00
Nate Begeman
a0e221dc75 Implement some feedback from Chris re: constant canonicalization
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23777 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 00:28:13 +00:00
Nate Begeman
419f8b62f7 Legalize BUILD_PAIR appropriately for upcoming 64 bit PowerPC work.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23776 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-18 00:27:41 +00:00
Nate Begeman
11af4eaa6d fold fmul X, +2.0 -> fadd X, X;
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23774 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-17 20:40:11 +00:00
Chris Lattner
854077d3a5 add a trivial fold
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23764 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-17 01:07:11 +00:00
Chris Lattner
750dbd5950 Fix this logic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23756 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-15 22:35:40 +00:00
Chris Lattner
85d63bbff7 Add a case we were missing that was causing us to fail CodeGen/PowerPC/rlwinm.ll:test3
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23755 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-15 22:18:08 +00:00
Chris Lattner
8136cdae60 Use getExtLoad here instead of getNode, as extloads produce two values. This
fixes a legalize failure on SPASS for itanium.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23747 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-15 20:24:07 +00:00
Nate Begeman
216def8ecf fold sext_in_reg, sext_in_reg where both have the same VT. This was
popping up in Fourinarow.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23722 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-14 01:29:07 +00:00
Nate Begeman
5054f16212 Relax the checking on zextload generation a bit, since as sabre pointed out
you could be AND'ing with the result of a shift that shifts out all the
bits you care about, in addition to a constant.

Also, move over an add/sub_parts fold from legalize to the dag combiner,
where it works for things other than constants.  Woot!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23720 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-14 01:12:21 +00:00
Chris Lattner
24edbb7a6c Fix the trunc(load) case, finally allowing crafty and povray to pass
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23718 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 22:10:05 +00:00
Chris Lattner
f98840531a Fix some bugs in (sext (load x))
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23717 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 21:52:31 +00:00
Chris Lattner
9ad8481cfb When ExpandOp'ing a [SZ]EXTLOAD, make sure to remember that the chain
is also legal.  Add support for ExpandOp'ing raw EXTLOADs too.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23716 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 21:44:47 +00:00
Chris Lattner
4c8f8f0481 Implement PromoteOp for *EXTLOAD, allowing MallocBench/gs to Legalize
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23715 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 20:07:41 +00:00
Nate Begeman
bfd65a05c9 Fix the remaining DAGCombiner issues pointed out by sabre. This should fix
the remainder of the failures introduced by my patch last night.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23714 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 18:34:58 +00:00
Chris Lattner
67a44cd3aa Fix a minor bug in the dag combiner that broke pcompress2 and some other
tests.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23713 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 18:16:34 +00:00
Nate Begeman
144ff660e6 Add support to Legalize for expanding i64 sextload/zextload into hi and lo
parts. This should fix the crafty and signed long long unit test failure
on x86 last night.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23711 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 17:15:37 +00:00
Jim Laskey
53c523c00e Inhibit instructions from being pushed before function calls. This will
minimize unnecessary spilling.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23710 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 16:44:00 +00:00
Nate Begeman
ded4963ab9 Move some Legalize functionality over to the DAGCombiner where it belongs.
Kill some dead code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23706 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-13 03:11:28 +00:00
Nate Begeman
765784ad76 Fix a potential bug with two combine-to's back to back that chris pointed
out, where after the first CombineTo() call, the node the second CombineTo
wishes to replace may no longer exist.

Fix a very real bug with the truncated load optimization on little endian
targets, which do not need a byte offset added to the load.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23704 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-12 23:18:53 +00:00
Nate Begeman
3df4d525c5 More cool stuff for the dag combiner. We can now finally handle things
like turning:

_foo:
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        rlwinm r3, r2, 0, 16, 31
        blr

into
_foo:
        fctiwz f0,f1
        stfd f0,-8(r1)
        lhz r3,-2(r1)
        blr

Also removed an unncessary constraint from sra -> srl conversion, which
should take care of hte only reason we would ever need to handle sra in
MaskedValueIsZero, AFAIK.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23703 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-12 20:40:40 +00:00
Jim Laskey
fab66f6900 Finally committing to the new scheduler. Still -sched=none by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23702 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-12 18:29:35 +00:00
Chris Lattner
a179ab3016 Fix a powerpc crash on CodeGen/Generic/llvm-ct-intrinsics.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23694 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-11 17:56:34 +00:00
Chris Lattner
05b57433e6 Add a canonicalization that got lost, fixing PowerPC/fold-li.ll:SUB
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23693 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-11 06:07:15 +00:00
Chris Lattner
04ecf6d65f clean up some corner cases
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23692 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 23:00:08 +00:00
Chris Lattner
87514ca04c Implement trivial DSE. If two stores are neighbors and store to the same
location, replace them with a new store of the last value.  This occurs
in the same neighborhood in 197.parser, speeding it up about 1.5%


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23691 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 22:31:19 +00:00
Chris Lattner
01a2202138 Add support for CombineTo, allowing the dag combiner to replace nodes with
multiple results.

Use this support to implement trivial store->load forwarding, implementing
CodeGen/PowerPC/store-load-fwd.ll.  Though this is the most simple case and
can be extended in the future, it is still useful.  For example, it speeds
up 197.parser by 6.2% by avoiding an LSU reject in xalloc:

        stw r6, lo16(l5_end_of_array)(r2)
        addi r2, r5, -4
        stwx r5, r4, r2
-       lwzx r5, r4, r2
-       rlwinm r5, r5, 0, 0, 30
        stwx r5, r4, r2
        lwz r2, -4(r4)
        ori r2, r2, 1


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23690 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 22:04:48 +00:00
Nate Begeman
07ed417df2 Teach the DAGCombiner several new tricks, teaching it how to turn
sext_inreg into zext_inreg based on the signbit (fires a lot), srem into
urem, etc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23688 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 21:26:48 +00:00
Chris Lattner
d739075cbb Fix comment
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 16:52:03 +00:00
Chris Lattner
bba9aa3475 Add ISD::ADD to MaskedValueIsZero
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23685 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 16:51:40 +00:00
Chris Lattner
3892560680 This function is now dead
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23684 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 16:49:22 +00:00
Chris Lattner
af21d55aee Enable Nate's excellent DAG combiner work by default. This allows the
removal of a bunch of ad-hoc and crufty code from SelectionDAG.cpp.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23682 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-10 16:47:10 +00:00
Chris Lattner
a19cea061e add a todo for something I noticed
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23679 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-09 22:59:08 +00:00
Chris Lattner
ee899e6bfc (X & Y) & C == 0 if either X&C or Y&C are zero
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23678 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-09 22:12:36 +00:00
Chris Lattner
089c25ccb3 When emiting a CopyFromReg and the source is already a vreg, do not bother
creating a new vreg and inserting a copy: just use the input vreg directly.

This speeds up the compile (e.g. about 5% on mesa with a debug build of llc)
by not adding a bunch of copies and vregs to be coallesced away.  On mesa,
for example, this reduces the number of intervals from 168601 to 129040
going into the coallescer.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23671 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-09 05:58:56 +00:00
Nate Begeman
f845b4563a Lo and behold, the last bits of SelectionDAG.cpp have been moved over.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23665 91177308-0d34-0410-b5e6-96231b3b80d8
2005-10-08 00:29:44 +00:00