summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms
Commit message (Collapse)AuthorAgeFilesLines
* Fix a fixme in CondPropagate.cpp by moving a PhiNode optimization intoNate Begeman2005-08-045-51/+4
| | | | | | | | BasicBlock's removePredecessor routine. This requires shuffling around the definition and implementation of hasContantValue from Utils.h,cpp into Instructions.h,cpp llvm-svn: 22664
* Modify how immediates are removed from base expressions to deal with the factChris Lattner2005-08-041-26/+41
| | | | | | | | | | | | | that the symbolic evaluator is not always able to use subtraction to remove expressions. This makes the code faster, and fixes the last crash on 178.galgel. Finally, add a statistic to see how many phi nodes are inserted. On 178.galgel, we get the follow stats: 2562 loop-reduce - Number of PHIs inserted 3927 loop-reduce - Number of GEPs strength reduced llvm-svn: 22662
* * Refactor some code into a new BasedUser::RewriteInstructionToUseNewBaseChris Lattner2005-08-041-16/+48
| | | | | | | | method. * Fix a crash on 178.galgel, where we would insert expressions before PHI nodes instead of into the PHI node predecessor blocks. llvm-svn: 22657
* Fix a case that caused this to crash on 178.galgelChris Lattner2005-08-041-0/+6
| | | | llvm-svn: 22653
* Teach LSR about loop-variant expressions, such as loops like this:Chris Lattner2005-08-041-68/+93
| | | | | | | | | | | | | | for (i = 0; i < N; ++i) A[i][foo()] = 0; here we still want to strength reduce the A[i] part, even though foo() is l-v. This also simplifies some of the 'CanReduce' logic. This implements Transforms/LoopStrengthReduce/ops_after_indvar.ll llvm-svn: 22652
* Remove some more dead code.Nate Begeman2005-08-041-20/+0
| | | | llvm-svn: 22650
* Refactor this code substantially with the following improvements:Chris Lattner2005-08-041-138/+38
| | | | | | | | | | | 1. We only analyze instructions once, guaranteed 2. AnalyzeGetElementPtrUsers has been ripped apart and replaced with something much simpler. The next step is to handle expressions that are not all indvar+loop-invariant values (e.g. handling indvar+loopvariant). llvm-svn: 22649
* refactor some codeChris Lattner2005-08-041-34/+45
| | | | llvm-svn: 22643
* invert to if's to make the logic simplerChris Lattner2005-08-041-14/+11
| | | | llvm-svn: 22641
* When processing outer loops and we find uses of an IV in inner loops, makeChris Lattner2005-08-041-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sure to handle the use, just don't recurse into it. This permits us to generate this code for a simple nested loop case: .LBB_foo_0: ; entry stwu r1, -48(r1) stw r29, 44(r1) stw r30, 40(r1) mflr r11 stw r11, 56(r1) lis r2, ha16(L_A$non_lazy_ptr) lwz r30, lo16(L_A$non_lazy_ptr)(r2) li r29, 1 .LBB_foo_1: ; no_exit.0 bl L_bar$stub li r2, 1 or r3, r30, r30 .LBB_foo_2: ; no_exit.1 lfd f0, 8(r3) stfd f0, 0(r3) addi r4, r2, 1 addi r3, r3, 8 cmpwi cr0, r2, 100 or r2, r4, r4 bne .LBB_foo_2 ; no_exit.1 .LBB_foo_3: ; loopexit.1 addi r30, r30, 800 addi r2, r29, 1 cmpwi cr0, r29, 100 or r29, r2, r2 bne .LBB_foo_1 ; no_exit.0 .LBB_foo_4: ; return lwz r11, 56(r1) mtlr r11 lwz r30, 40(r1) lwz r29, 44(r1) lwz r1, 0(r1) blr instead of this: _foo: .LBB_foo_0: ; entry stwu r1, -48(r1) stw r28, 44(r1) ;; uses an extra register. stw r29, 40(r1) stw r30, 36(r1) mflr r11 stw r11, 56(r1) li r30, 1 li r29, 0 or r28, r29, r29 .LBB_foo_1: ; no_exit.0 bl L_bar$stub mulli r2, r28, 800 ;; unstrength-reduced multiply lis r3, ha16(L_A$non_lazy_ptr) ;; loop invariant address computation lwz r3, lo16(L_A$non_lazy_ptr)(r3) add r2, r2, r3 mulli r4, r29, 800 ;; unstrength-reduced multiply addi r3, r3, 8 add r3, r4, r3 li r4, 1 .LBB_foo_2: ; no_exit.1 lfd f0, 0(r3) stfd f0, 0(r2) addi r5, r4, 1 addi r2, r2, 8 ;; multiple stride 8 IV's addi r3, r3, 8 cmpwi cr0, r4, 100 or r4, r5, r5 bne .LBB_foo_2 ; no_exit.1 .LBB_foo_3: ; loopexit.1 addi r28, r28, 1 ;;; Many IV's with stride 1 addi r29, r29, 1 addi r2, r30, 1 cmpwi cr0, r30, 100 or r30, r2, r2 bne .LBB_foo_1 ; no_exit.0 .LBB_foo_4: ; return lwz r11, 56(r1) mtlr r11 lwz r30, 36(r1) lwz r29, 40(r1) lwz r28, 44(r1) lwz r1, 0(r1) blr llvm-svn: 22640
* Teach loop-reduce to see into nested loops, to pull out immediate valuesChris Lattner2005-08-031-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | pushed down by SCEV. In a nested loop case, this allows us to emit this: lis r3, ha16(L_A$non_lazy_ptr) lwz r3, lo16(L_A$non_lazy_ptr)(r3) add r2, r2, r3 li r3, 1 .LBB_foo_2: ; no_exit.1 lfd f0, 8(r2) ;; Uses offset of 8 instead of 0 stfd f0, 0(r2) addi r4, r3, 1 addi r2, r2, 8 cmpwi cr0, r3, 100 or r3, r4, r4 bne .LBB_foo_2 ; no_exit.1 instead of this: lis r3, ha16(L_A$non_lazy_ptr) lwz r3, lo16(L_A$non_lazy_ptr)(r3) add r2, r2, r3 addi r3, r3, 8 li r4, 1 .LBB_foo_2: ; no_exit.1 lfd f0, 0(r3) stfd f0, 0(r2) addi r5, r4, 1 addi r2, r2, 8 addi r3, r3, 8 cmpwi cr0, r4, 100 or r4, r5, r5 bne .LBB_foo_2 ; no_exit.1 llvm-svn: 22639
* improve debug outputChris Lattner2005-08-031-4/+9
| | | | llvm-svn: 22638
* Move from Stage 0 to Stage 1.Chris Lattner2005-08-031-31/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Only emit one PHI node for IV uses with identical bases and strides (after moving foldable immediates to the load/store instruction). This implements LoopStrengthReduce/dont_insert_redundant_ops.ll, allowing us to generate this PPC code for test1: or r30, r3, r3 .LBB_test1_1: ; Loop li r2, 0 stw r2, 0(r30) stw r2, 4(r30) bl L_pred$stub addi r30, r30, 8 cmplwi cr0, r3, 0 bne .LBB_test1_1 ; Loop instead of this code: or r30, r3, r3 or r29, r3, r3 .LBB_test1_1: ; Loop li r2, 0 stw r2, 0(r29) stw r2, 4(r30) bl L_pred$stub addi r30, r30, 8 ;; Two iv's with step of 8 addi r29, r29, 8 cmplwi cr0, r3, 0 bne .LBB_test1_1 ; Loop llvm-svn: 22635
* Rename IVUse to IVUsersOfOneStride, use a struct instead of a pair toChris Lattner2005-08-031-25/+41
| | | | | | | unify some parallel vectors and get field names more descriptive than "first" and "second". This isn't lisp afterall :) llvm-svn: 22633
* Fix a nasty dangling pointer issue. The ScalarEvolution pass would keep aChris Lattner2005-08-031-1/+3
| | | | | | | | | map from instruction* to SCEVHandles. When we delete instructions, we have to tell it about it. We would run into nasty cases where new instructions were reallocated at old instruction addresses and get the old map values. Bad bad bad :( llvm-svn: 22632
* The correct fix for PR612, which also fixesChris Lattner2005-08-031-2/+12
| | | | | | Transforms/LowerInvoke/2005-08-03-InvokeWithPHIUse.ll llvm-svn: 22628
* When inserting code, make sure not to insert it before PHI nodes. ThisChris Lattner2005-08-031-1/+3
| | | | | | fixes PR612 and Transforms/LowerInvoke/2005-08-03-InvokeWithPHI.ll llvm-svn: 22626
* Fix Transforms/SimplifyCFG/2005-08-03-PHIFactorCrash.ll, a problem thatChris Lattner2005-08-031-2/+3
| | | | | | occurred while bugpointing another testcase llvm-svn: 22621
* Finally, add the required constraint checks to fix ↵Chris Lattner2005-08-031-2/+29
| | | | | | | | Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll the right way llvm-svn: 22615
* Simplify some code, add the correct pred checksChris Lattner2005-08-031-16/+25
| | | | llvm-svn: 22613
* Refactor code out of PropagatePredecessorsForPHIs, turning it into a pure ↵Chris Lattner2005-08-031-37/+36
| | | | | | function with no side-effects llvm-svn: 22612
* use splice instead of remove/insert to avoid some symtab operationsChris Lattner2005-08-031-2/+2
| | | | llvm-svn: 22611
* move two functions up in the file, use SafeToMergeTerminators to eliminateChris Lattner2005-08-031-61/+45
| | | | | | some duplicated code llvm-svn: 22610
* Rip some code out of the main SimplifyCFG function into a subfunction andChris Lattner2005-08-031-78/+72
| | | | | | call it from the only place it is live. No functionality changes. llvm-svn: 22609
* Disable this patch:Chris Lattner2005-08-021-1/+1
| | | | | | | | | http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20050801/027345.html This breaks real programs and only fixes an obscure regression testcase. A real fix is in development. llvm-svn: 22606
* Change a place to use an arbitrary value instead of null, when possibleChris Lattner2005-08-021-3/+3
| | | | llvm-svn: 22605
* Update to use the new MathExtras.h support for log2 computation.Chris Lattner2005-08-021-22/+15
| | | | | | Patch contributed by Jim Laskey! llvm-svn: 22592
* Like the comment says, do not insert cast instructions before phi nodesChris Lattner2005-08-021-0/+4
| | | | llvm-svn: 22586
* This code was very close, but not quite right. It did not take intoChris Lattner2005-08-021-3/+10
| | | | | | | | consideration the case where a reference in an unreachable block could occur. This fixes Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll, something I ran into while bugpoint'ing another pass. llvm-svn: 22584
* add a comment, make a check more lenientChris Lattner2005-08-021-8/+10
| | | | llvm-svn: 22581
* Simplify for loop, clear a per-loop map after processing each loopChris Lattner2005-08-021-1/+2
| | | | llvm-svn: 22580
* Add a commentChris Lattner2005-08-021-0/+10
| | | | | | | Make LSR ignore GEP's that have loop variant base values, as we currently cannot codegen them llvm-svn: 22576
* Fix an iterator invalidation problemChris Lattner2005-08-021-1/+3
| | | | llvm-svn: 22575
* ConstantInt::get only works for arguments < 128.Chris Lattner2005-08-011-2/+6
| | | | | | | | | | SimplifyLibCalls probably has to be audited to make sure it does not make this mistake elsewhere. Also, if this code knows that the type will be unsigned, obviously one arm of this is dead. Reid, can you take a look into this further? llvm-svn: 22566
* Keep tabs and trailing spaces out.Jeff Cohen2005-07-301-23/+23
| | | | llvm-svn: 22565
* Fix VC++ build problems.Jeff Cohen2005-07-301-0/+1
| | | | llvm-svn: 22564
* Ack, typoNate Begeman2005-07-301-1/+1
| | | | llvm-svn: 22560
* Commit a new LoopStrengthReduce pass that can use scalar evolutions andNate Begeman2005-07-301-179/+517
| | | | | | | | target data to decide which loop induction variables to strength reduce and how to do so. This work is mostly by Chris Lattner, with tweaks by me to get it working on some of MultiSource. llvm-svn: 22558
* Break SCEVExpander out of IndVarSimplify into its own .h/.cpp file so thatNate Begeman2005-07-301-238/+1
| | | | | | other passes may use it. llvm-svn: 22557
* Eliminate all remaining tabs and trailing spaces.Jeff Cohen2005-07-279-181/+181
| | | | llvm-svn: 22523
* ConvertibleToGEP always returns 0, remove some old crufty code whichChris Lattner2005-07-264-369/+0
| | | | | | is actually dead because of this! llvm-svn: 22515
* Do not let MaskedValueIsZero consider undef to be zero, for reasonsChris Lattner2005-07-201-2/+8
| | | | | | | | explained in the comment. This fixes UnitTests/2003-09-18-BitFieldTest on darwin llvm-svn: 22483
* When transforming &A[i] < &A[j] -> i < j, make sure to perform the comparisonChris Lattner2005-07-181-4/+11
| | | | | | as a signed compare. This patch may fix PR597, but is correct in any case. llvm-svn: 22465
* Fix a problem that instcombine would hit when dealing with unreachable code.Chris Lattner2005-07-071-5/+29
| | | | | | | | | Because the instcombine has to scan the entire function when it starts up to begin with, we might as well do it in DFO so we can nuke unreachable code. This fixes: Transforms/InstCombine/2005-07-07-DeadPHILoop.ll llvm-svn: 22348
* Fix PR590 and Transforms/Mem2Reg/2005-06-30-ReadBeforeWrite.ll.Chris Lattner2005-06-301-19/+65
| | | | | | | | The optimization for locally used allocas was not safe for allocas that were read before they were written. This change disables that optimization in that case. llvm-svn: 22318
* Doh! Forgot to LLVMify the style.John Criswell2005-06-291-2/+0
| | | | llvm-svn: 22312
* Basic fix for PR#591; don't convert an fprintf() to an fwrite() if thereJohn Criswell2005-06-291-0/+9
| | | | | | | | | is a mismatch in their character type pointers (i.e. fprintf() prints an array of ubytes while fwrite() takes an array of sbytes). We can probably do better than this (such as casting the ubyte to an sbyte). llvm-svn: 22310
* add a debug typeChris Lattner2005-06-241-0/+1
| | | | llvm-svn: 22277
* prevent va_arg from being hoisted from a loopAndrew Lenharth2005-06-201-1/+1
| | | | llvm-svn: 22265
* prevent DCE of vaarg intrinsics. This should take care of most regressionsAndrew Lenharth2005-06-191-2/+0
| | | | llvm-svn: 22263
OpenPOWER on IntegriCloud