Commit Graph

13241 Commits

Author SHA1 Message Date
Evan Cheng
9c543b2299 PPC LSR pass should use target lowering hooks.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26743 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:56:51 +00:00
Evan Cheng
c4c6257c1a Added getTargetLowering() to TargetMachine. Refactored targets to support this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26742 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:20:37 +00:00
Evan Cheng
e617b085fe Update
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26741 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:19:10 +00:00
Evan Cheng
30b37b5f29 Add LSR hooks.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26740 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:18:16 +00:00
Evan Cheng
d277f2c669 Added target lowering hooks which LSR consults to make more intelligent
transformation decisions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26738 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:14:23 +00:00
Chris Lattner
872421553e Handle builtins that directly correspond to GCC builtins.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26737 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 23:09:05 +00:00
Chris Lattner
1877ec9b02 For functions that use vector registers, save VRSAVE, mark used
registers, and update it on entry to each function, then restore it on exit.

This compiles:

void func(vfloat *a, vfloat *b, vfloat *c) {
        *a = *b * *c + *c;
}

to this:

_func:
        mfspr r2, 256
        oris r6, r2, 49152
        mtspr 256, r6
        lvx v0, 0, r5
        lvx v1, 0, r4
        vmaddfp v0, v1, v0, v0
        stvx v0, 0, r3
        mtspr 256, r2
        blr

GCC produces this (which has additional stack accesses):

_func:
        mfspr r0,256
        stw r0,-4(r1)
        oris r0,r0,0xc000
        mtspr 256,r0
        lvx v0,0,r5
        lvx v1,0,r4
        lwz r12,-4(r1)
        vmaddfp v0,v0,v1,v0
        stvx v0,0,r3
        mtspr 256,r12
        blr


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26733 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 21:52:10 +00:00
Chris Lattner
ac0f8f2929 make sure dead token factor nodes are removed by the dag combiner.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26731 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 18:37:30 +00:00
Reid Spencer
a136429462 Bring makefile back into compliance with standard by using
$(Echo) instead of @echo


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26730 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 17:57:31 +00:00
Jim Laskey
f4321a3a43 Handle the removal of the debug chain.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26729 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 13:07:37 +00:00
Chris Lattner
2e8a77ff42 remove two implemented items
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26728 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 06:52:22 +00:00
Chris Lattner
947c28935d Fold X+Y -> X|Y when safe. This implements:
Regression/CodeGen/PowerPC/and_add.ll

a case that occurs with dynamic allocas of constant size.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26727 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 06:51:27 +00:00
Chris Lattner
b6b17ffbc6 I can't convince myself that this is safe, remove the recursive call.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26725 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 06:42:16 +00:00
Chris Lattner
79dbea5ab4 add a couple of missing folds
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26724 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 06:26:26 +00:00
Chris Lattner
8f4191d619 For targets with FABS/FNEG support, lower copysign to an integer load,
a select and FABS/FNEG.

This speeds up a trivial (aka stupid) copysign benchmark I wrote from 6.73s
to 2.64s, woo.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26723 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 06:08:38 +00:00
Chris Lattner
64ce964673 Fix a couple of bugs that broke the alpha tester build
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26722 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 05:23:59 +00:00
Chris Lattner
3faad495bc Handle cracked instructions in dispatch group formation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26721 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 05:20:04 +00:00
Chris Lattner
fd97734f36 Mark instructions that are cracked by the PPC970 decoder as such.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26720 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-13 05:15:10 +00:00
Chris Lattner
88d211f823 Several big changes:
1. Use flags on the instructions in the .td file to indicate the PPC970 unit
   type instead of a table in the .cpp file.  Much cleaner.
2. Change the hazard recognizer to build d-groups according to the actual
   algorithm used, not my flawed understanding of it.
3. Model "must be in the first slot" and "must be the only instr in a group"
   accurately.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26719 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-12 09:13:49 +00:00
Chris Lattner
fc3549ee8c Don't advance the hazard recognizer when there are no hazards and no instructions
to be emitted.

Don't add one to the latency of a completed instruction if the latency of the
op is 0.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26718 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-12 09:01:41 +00:00
Chris Lattner
b2215030d6 Chain operands aren't real uses: they don't require the full latency of the
predecessor to finish before they can start.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26717 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-12 03:52:09 +00:00
Chris Lattner
53fbf2a8e8 As a pending queue data structure to keep track of instructions whose
operands have all issued, but whose results are not yet available.  This
allows us to compile:

int G;
int test(int A, int B, int* P) {
   return (G+A)*(B+1);
}

to:

_test:
        lis r2, ha16(L_G$non_lazy_ptr)
        addi r4, r4, 1
        lwz r2, lo16(L_G$non_lazy_ptr)(r2)
        lwz r2, 0(r2)
        add r2, r2, r3
        mullw r3, r2, r4
        blr

instead of this, which has a stall between the lis/lwz:

_test:
        lis r2, ha16(L_G$non_lazy_ptr)
        lwz r2, lo16(L_G$non_lazy_ptr)(r2)
        addi r4, r4, 1
        lwz r2, 0(r2)
        add r2, r2, r3
        mullw r3, r2, r4
        blr


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26716 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-12 00:38:57 +00:00
Chris Lattner
8469031622 rename priorityqueue -> availablequeue. When a node is scheduled, remember
which cycle it lands on.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26714 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 22:44:37 +00:00
Chris Lattner
c1c078c170 Make CurrCycle a local var instead of an instance var
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26713 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 22:34:41 +00:00
Chris Lattner
7d82b00048 Move some methods around so that BU specific code is together, TD specific code
is together, and direction independent code is together.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26712 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 22:28:35 +00:00
Chris Lattner
309cf8a713 merge preds/chainpreds -> preds set
merge succs/chainsuccs -> succs set

This has no functionality change, simplifies the code, and reduces the size
of sunits.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26711 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 22:24:20 +00:00
Chris Lattner
9c2c38674a blr is a branch too
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26710 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 21:49:49 +00:00
Chris Lattner
74cfb7d7b3 add an example
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26709 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 20:20:40 +00:00
Chris Lattner
c20995e070 add a note
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26708 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 20:17:08 +00:00
Nate Begeman
442b32b5c5 Fix PR681 by using the standard Lengauer and Tarjan algorithm for dominator
set construction, rather than intersecting various std::sets.  This reduces
the memory usage for the testcase in PR681 from 496 to 26MB of ram on my
darwin system, and reduces the runtime from 32.8 to 0.8 seconds on a
2.5GHz G5.  This also enables future code sharing between Dom and PostDom
now that they share near-identical implementations.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26707 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 02:20:46 +00:00
Evan Cheng
2352ce9cb6 Doh!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26705 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-11 00:13:10 +00:00
Evan Cheng
0937103368 Added a parameter to control whether Constant::getStringValue() would chop
off the result string at the first null terminator.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26704 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 23:52:03 +00:00
Chris Lattner
7907b6950b remove some dead code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26703 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 23:14:22 +00:00
Chris Lattner
7d74d1145f scrape out bits of llvm-db
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26701 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 22:48:19 +00:00
Chris Lattner
89b192690a Remove UnixLocalInferiorProcess: debugging via the JIT isn't a good idea.
Perhaps llvm-db will turn into a read debugger someday.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26700 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 22:39:48 +00:00
Chris Lattner
e928a72772 teach the JIT to encode vector registers
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26697 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 20:19:50 +00:00
Chris Lattner
a34b6f8713 Move simple-selector-specific types to the simple selector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26693 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:51:18 +00:00
Chris Lattner
20a4921791 Simplify the interface to the schedulers, to not pass the selected heuristicin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26692 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:49:12 +00:00
Chris Lattner
2f5806c2b3 Move some simple-sched-specific instance vars to the simple scheduler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26690 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:42:02 +00:00
Chris Lattner
1e433c59e0 prune #includes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26689 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:37:35 +00:00
Chris Lattner
e76074ab89 move some simple scheduler methods into the simple scheduler
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26688 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:35:21 +00:00
Chris Lattner
8c7ef0599c Make EmitNode take a SDNode instead of a NodeInfo*
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26687 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:28:36 +00:00
Chris Lattner
df3750642a Move the VRBase field from NodeInfo to being a separate, explicit, map.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26686 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:25:12 +00:00
Chris Lattner
8d41651da7 no need to build groups anymore
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26684 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:15:58 +00:00
Chris Lattner
b2042e31a7 Create SUnits directly from the SelectionDAG.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26683 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 07:13:32 +00:00
Chris Lattner
be24e5996c Push PrepareNodeInfo/IdentifyGroups down the inheritance hierarchy
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26682 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 06:34:51 +00:00
Chris Lattner
da4ff69153 Teach the latency scheduler some new tricks. In particular, to break ties,
keep track of a sense of "mobility", i.e. how many other nodes scheduling one
node will free up.  For something like this:

float testadd(float *X, float *Y, float *Z, float *W, float *V) {
  return (*X+*Y)*(*Z+*W)+*V;
}

For example, this makes us schedule *X then *Y, not *X then *Z.  The former
allows us to issue the add, the later only lets us issue other loads.

This turns the above code from this:

_testadd:
        lfs f0, 0(r3)
        lfs f1, 0(r6)
        lfs f2, 0(r4)
        lfs f3, 0(r5)
        fadds f0, f0, f2
        fadds f1, f3, f1
        lfs f2, 0(r7)
        fmadds f1, f0, f1, f2
        blr

into this:

_testadd:
        lfs f0, 0(r6)
        lfs f1, 0(r5)
        fadds f0, f1, f0
        lfs f1, 0(r4)
        lfs f2, 0(r3)
        fadds f1, f2, f1
        lfs f2, 0(r7)
        fmadds f1, f1, f0, f2
        blr


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26680 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 05:51:05 +00:00
Chris Lattner
f83a47d905 add an aggregate method for reinserting scheduled nodes, add a callback for
priority impls that want to be notified when a node is scheduled


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26678 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 04:32:49 +00:00
Jeff Cohen
9630d271c5 Fix VC++ build breakage.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26676 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-10 03:57:45 +00:00
Chris Lattner
aeebe7f208 use autogenerated side-effect information
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@26673 91177308-0d34-0410-b5e6-96231b3b80d8
2006-03-09 22:38:10 +00:00