1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-30 07:22:55 +01:00
Commit Graph

23419 Commits

Author SHA1 Message Date
Chris Lattner
3a3c8682b5 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.

llvm-svn: 26727
2006-03-13 06:51:27 +00:00
Chris Lattner
7821da357e new testcase
llvm-svn: 26726
2006-03-13 06:50:47 +00:00
Chris Lattner
1782f3971d I can't convince myself that this is safe, remove the recursive call.
llvm-svn: 26725
2006-03-13 06:42:16 +00:00
Chris Lattner
9d0ebb55a6 add a couple of missing folds
llvm-svn: 26724
2006-03-13 06:26:26 +00:00
Chris Lattner
3cb122bfa8 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.

llvm-svn: 26723
2006-03-13 06:08:38 +00:00
Chris Lattner
3aff8e6acf Fix a couple of bugs that broke the alpha tester build
llvm-svn: 26722
2006-03-13 05:23:59 +00:00
Chris Lattner
9898674f99 Handle cracked instructions in dispatch group formation.
llvm-svn: 26721
2006-03-13 05:20:04 +00:00
Chris Lattner
ba10d4e4ab Mark instructions that are cracked by the PPC970 decoder as such.
llvm-svn: 26720
2006-03-13 05:15:10 +00:00
Chris Lattner
a278639f29 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.

llvm-svn: 26719
2006-03-12 09:13:49 +00:00
Chris Lattner
a54a9d83c3 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.

llvm-svn: 26718
2006-03-12 09:01:41 +00:00
Chris Lattner
aea53fc36f Chain operands aren't real uses: they don't require the full latency of the
predecessor to finish before they can start.

llvm-svn: 26717
2006-03-12 03:52:09 +00:00
Chris Lattner
7396b82585 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

llvm-svn: 26716
2006-03-12 00:38:57 +00:00
Chris Lattner
2594ddf452 Mark llvm.stacksave as only reading memory, this fixes
Regression/Transforms/InstCombine/stacksaverestore.ll

llvm-svn: 26715
2006-03-12 00:04:28 +00:00
Chris Lattner
62eafde9c1 rename priorityqueue -> availablequeue. When a node is scheduled, remember
which cycle it lands on.

llvm-svn: 26714
2006-03-11 22:44:37 +00:00
Chris Lattner
04ada12cb7 Make CurrCycle a local var instead of an instance var
llvm-svn: 26713
2006-03-11 22:34:41 +00:00
Chris Lattner
6a9861953b Move some methods around so that BU specific code is together, TD specific code
is together, and direction independent code is together.

llvm-svn: 26712
2006-03-11 22:28:35 +00:00
Chris Lattner
40c02c01a5 merge preds/chainpreds -> preds set
merge succs/chainsuccs -> succs set

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

llvm-svn: 26711
2006-03-11 22:24:20 +00:00
Chris Lattner
19b93158c1 blr is a branch too
llvm-svn: 26710
2006-03-11 21:49:49 +00:00
Chris Lattner
916761949b add an example
llvm-svn: 26709
2006-03-11 20:20:40 +00:00
Chris Lattner
115b1be710 add a note
llvm-svn: 26708
2006-03-11 20:17:08 +00:00
Nate Begeman
7c2afcfc02 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.

llvm-svn: 26707
2006-03-11 02:20:46 +00:00
Chris Lattner
8930c740ac fix pasto in generate assertion msg
llvm-svn: 26706
2006-03-11 00:20:47 +00:00
Evan Cheng
02418117e1 Doh!
llvm-svn: 26705
2006-03-11 00:13:10 +00:00
Evan Cheng
57c206232d Added a parameter to control whether Constant::getStringValue() would chop
off the result string at the first null terminator.

llvm-svn: 26704
2006-03-10 23:52:03 +00:00
Chris Lattner
f176c9ecac remove some dead code
llvm-svn: 26703
2006-03-10 23:14:22 +00:00
Chris Lattner
f80c0a7188 remove ShouldEmitDebugFunctions, a hack to support llvm-db via the jit
llvm-svn: 26702
2006-03-10 22:49:05 +00:00
Chris Lattner
02e6c29ad2 scrape out bits of llvm-db
llvm-svn: 26701
2006-03-10 22:48:19 +00:00
Chris Lattner
ec8860221c Remove UnixLocalInferiorProcess: debugging via the JIT isn't a good idea.
Perhaps llvm-db will turn into a read debugger someday.

llvm-svn: 26700
2006-03-10 22:39:48 +00:00
Chris Lattner
12786a1616 Simplify this testcase
llvm-svn: 26699
2006-03-10 22:32:18 +00:00
Chris Lattner
11613abcee reorder these to make it work with static libraries
llvm-svn: 26698
2006-03-10 21:01:34 +00:00
Chris Lattner
2370965e55 teach the JIT to encode vector registers
llvm-svn: 26697
2006-03-10 20:19:50 +00:00
Chris Lattner
d6bc58d086 Fix another broken intrinsic.
llvm-svn: 26696
2006-03-10 18:01:03 +00:00
Chris Lattner
9e05c75c4a weak globals on darwin require an extra load, breaking this test
llvm-svn: 26695
2006-03-10 17:55:10 +00:00
Chris Lattner
228fa5bd76 Fix incorrect definitions of these intrinsics, which broke a bunch of
stuff last night.

llvm-svn: 26694
2006-03-10 17:48:34 +00:00
Chris Lattner
0a64d261e0 Move simple-selector-specific types to the simple selector.
llvm-svn: 26693
2006-03-10 07:51:18 +00:00
Chris Lattner
2024573f47 Simplify the interface to the schedulers, to not pass the selected heuristicin.
llvm-svn: 26692
2006-03-10 07:49:12 +00:00
Chris Lattner
40b1b16c56 Simplify the interface to the schedulers, to not pass the selected heuristic
in.

llvm-svn: 26691
2006-03-10 07:48:52 +00:00
Chris Lattner
da5f77e3cf Move some simple-sched-specific instance vars to the simple scheduler.
llvm-svn: 26690
2006-03-10 07:42:02 +00:00
Chris Lattner
ec2c5aa0bb prune #includes
llvm-svn: 26689
2006-03-10 07:37:35 +00:00
Chris Lattner
808cc02983 move some simple scheduler methods into the simple scheduler
llvm-svn: 26688
2006-03-10 07:35:21 +00:00
Chris Lattner
89a5a946f5 Make EmitNode take a SDNode instead of a NodeInfo*
llvm-svn: 26687
2006-03-10 07:28:36 +00:00
Chris Lattner
3f870d581e Move the VRBase field from NodeInfo to being a separate, explicit, map.
llvm-svn: 26686
2006-03-10 07:25:12 +00:00
Chris Lattner
f1be1182f0 Store VRBase in a map, not in NodeInfo.
llvm-svn: 26685
2006-03-10 07:24:45 +00:00
Chris Lattner
a316d82d2a no need to build groups anymore
llvm-svn: 26684
2006-03-10 07:15:58 +00:00
Chris Lattner
e23afcfd6d Create SUnits directly from the SelectionDAG.
llvm-svn: 26683
2006-03-10 07:13:32 +00:00
Chris Lattner
ed528a5652 Push PrepareNodeInfo/IdentifyGroups down the inheritance hierarchy
llvm-svn: 26682
2006-03-10 06:34:51 +00:00
Chris Lattner
6ef56b998f make some methods protected instead of private
llvm-svn: 26681
2006-03-10 06:30:11 +00:00
Chris Lattner
920325db0a 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

llvm-svn: 26680
2006-03-10 05:51:05 +00:00
Jeff Cohen
3fca15ca78 Put intrinsics.gen in its proper place.
llvm-svn: 26679
2006-03-10 04:36:01 +00:00
Chris Lattner
1bb3ae8dae add an aggregate method for reinserting scheduled nodes, add a callback for
priority impls that want to be notified when a node is scheduled

llvm-svn: 26678
2006-03-10 04:32:49 +00:00