Commit Graph

11601 Commits

Author SHA1 Message Date
Chris Lattner
83bf2880c3 Initial checkin of a simple loop unroller. This pass is extremely basic and
limited.  Even in it's extremely simple state (it can only *fully* unroll single
basic block loops that execute a constant number of times), it already helps improve
performance a LOT on some benchmarks, particularly with the native code generators.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13028 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-18 05:20:17 +00:00
Chris Lattner
3c85eef81a Make the tail duplication threshold accessible from the command line instead of hardcoded
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13025 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-18 00:52:43 +00:00
Chris Lattner
f57a43da9c Fix a memory leak. We leaked the vector holding the entries in switch tables.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13023 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 23:49:15 +00:00
Chris Lattner
f49090035f New testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13020 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 23:00:51 +00:00
Chris Lattner
3221ad0db7 Add the ability to compute exit values for complex loop using unanalyzable
operations.  This allows us to compile this testcase:

int main() {
        int h = 1;
         do h = 3 * h + 1; while (h <= 256);
        printf("%d\n", h);
        return 0;
}

into this:

int %main() {
entry:
        call void %__main( )
        %tmp.6 = call int (sbyte*, ...)* %printf( sbyte* getelementptr ([4 x sbyte]*  %.str_1, long 0, long 0), int 364 )        ; <int> [#uses=0]
        ret int 0
}

This testcase was taken directly from 256.bzip2, believe it or not.

This code is not as general as I would like.  Next up is to refactor it
a bit to handle more cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13019 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 22:58:41 +00:00
Chris Lattner
20aa098ba6 If the loop executes a constant number of times, try a bit harder to replace
exit values.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13018 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 18:44:09 +00:00
Chris Lattner
7980fb9007 Add the ability to compute trip counts that are only controlled by constants
even if the loop is using expressions that we can't compute as a closed-form.
This allows us to calculate that this function always returns 55:

int test() {
  double X;
  int Count = 0;
  for (X = 100; X > 1; X = sqrt(X), ++Count)
    /*empty*/;
  return Count;
}

And allows us to compute trip counts for loops like:

        int h = 1;
         do h = 3 * h + 1; while (h <= 256);

(which occurs in bzip2), and for this function, which occurs after inlining
and other optimizations:

int popcount()
{
   int x = 666;
  int result = 0;
  while (x != 0) {
    result = result + (x & 0x1);
    x = x >> 1;
  }
  return result;
}

We still cannot compute the exit values of result or h in the two loops above,
which means we cannot delete the loop, but we are getting closer.  Being able to
compute a constant trip count for these two loops will allow us to unroll them
completely though.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13017 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 18:36:24 +00:00
Chris Lattner
4f1134e51a Fix a HUGE pessimization on X86. The indvars pass was taking this
(familiar) function:

int _strlen(const char *str) {
    int len = 0;
    while (*str++) len++;
    return len;
}

And transforming it to use a ulong induction variable, because the type of
the pointer index was left as a constant long.  This is obviously very bad.

The fix is to shrink long constants in getelementptr instructions to intptr_t,
making the indvars pass insert a uint induction variable, which is much more
efficient.

Here's the before code for this function:

int %_strlen(sbyte* %str) {
entry:
        %tmp.13 = load sbyte* %str              ; <sbyte> [#uses=1]
        %tmp.24 = seteq sbyte %tmp.13, 0                ; <bool> [#uses=1]
        br bool %tmp.24, label %loopexit, label %no_exit

no_exit:                ; preds = %entry, %no_exit
***     %indvar = phi uint [ %indvar.next, %no_exit ], [ 0, %entry ]            ; <uint> [#uses=2]
***     %indvar = phi ulong [ %indvar.next, %no_exit ], [ 0, %entry ]           ; <ulong> [#uses=2]
        %indvar1 = cast ulong %indvar to uint           ; <uint> [#uses=1]
        %inc.02.sum = add uint %indvar1, 1              ; <uint> [#uses=1]
        %inc.0.0 = getelementptr sbyte* %str, uint %inc.02.sum          ; <sbyte*> [#uses=1]
        %tmp.1 = load sbyte* %inc.0.0           ; <sbyte> [#uses=1]
        %tmp.2 = seteq sbyte %tmp.1, 0          ; <bool> [#uses=1]
        %indvar.next = add ulong %indvar, 1             ; <ulong> [#uses=1]
        %indvar.next = add uint %indvar, 1              ; <uint> [#uses=1]
        br bool %tmp.2, label %loopexit.loopexit, label %no_exit

loopexit.loopexit:              ; preds = %no_exit
        %indvar = cast uint %indvar to int              ; <int> [#uses=1]
        %inc.1 = add int %indvar, 1             ; <int> [#uses=1]
        ret int %inc.1

loopexit:               ; preds = %entry
        ret int 0
}


Here's the after code:

int %_strlen(sbyte* %str) {
entry:
        %inc.02 = getelementptr sbyte* %str, uint 1             ; <sbyte*> [#uses=1]
        %tmp.13 = load sbyte* %str              ; <sbyte> [#uses=1]
        %tmp.24 = seteq sbyte %tmp.13, 0                ; <bool> [#uses=1]
        br bool %tmp.24, label %loopexit, label %no_exit

no_exit:                ; preds = %entry, %no_exit
***     %indvar = phi uint [ %indvar.next, %no_exit ], [ 0, %entry ]            ; <uint> [#uses=3]
        %indvar = cast uint %indvar to int              ; <int> [#uses=1]
        %inc.0.0 = getelementptr sbyte* %inc.02, uint %indvar           ; <sbyte*> [#uses=1]
        %inc.1 = add int %indvar, 1             ; <int> [#uses=1]
        %tmp.1 = load sbyte* %inc.0.0           ; <sbyte> [#uses=1]
        %tmp.2 = seteq sbyte %tmp.1, 0          ; <bool> [#uses=1]
        %indvar.next = add uint %indvar, 1              ; <uint> [#uses=1]
        br bool %tmp.2, label %loopexit, label %no_exit

loopexit:               ; preds = %entry, %no_exit
        %len.0.1 = phi int [ 0, %entry ], [ %inc.1, %no_exit ]          ; <int> [#uses=1]
        ret int %len.0.1
}


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13016 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 18:16:10 +00:00
Chris Lattner
f50af088f1 Even if there are not any induction variables in the loop, if we can compute
the trip count for the loop, insert one so that we can canonicalize the exit
condition.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13015 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-17 18:08:33 +00:00
Chris Lattner
f5b9eb37a9 Add support for evaluation of exp/log/log10/pow
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13011 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 22:35:33 +00:00
Chris Lattner
657253119e Fix some really nasty dominance bugs that were exposed by my patch to
make the verifier more strict.  This fixes building zlib


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13002 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 18:08:07 +00:00
Misha Brukman
ea51deb18a Fix retriving parent Function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@13001 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 17:37:12 +00:00
Misha Brukman
ef2b0db858 Fit comment into 80 cols.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12996 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 17:13:52 +00:00
Brian Gaeke
e773b869de Regenerated using autoconf-2.57.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12995 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 17:13:44 +00:00
Brian Gaeke
dbaf981345 Refactor external benchmark checking stuff into one hairy
macro-to-bind-them-all, called EXTERNAL_BENCHMARK().


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12994 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 17:13:33 +00:00
Misha Brukman
dec8cfb43c Add idea about a disassembler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12993 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 16:55:30 +00:00
Brian Gaeke
f9a5f780bf Switch to including <iostream> for compatibility with gcc-3.0.x (Debian).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12990 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 16:28:33 +00:00
Misha Brukman
fe22af620b * Fix capitalization of PICk
* Wrap long lines to 80 cols


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12988 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 16:20:07 +00:00
Brian Gaeke
c598517c24 Include <cmath> for compatibility with gcc 3.0.x (the system compiler on
Debian.)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12986 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 15:57:32 +00:00
Brian Gaeke
5a509c8225 Include <string> for compatibility with gcc 3.0.x (the system compiler on
Debian.)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12985 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 15:57:14 +00:00
Brian Gaeke
d994db00b2 As a part of the bootstrapping process, the top-level tools-only target
should not build projects.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12984 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 15:57:02 +00:00
Misha Brukman
98eb6264d4 Assert if deleting BasicBlock before removing it from Function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12983 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 15:47:21 +00:00
Misha Brukman
bf6059300f Assert if Instruction is being deleted before being removed from BasicBlock.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12982 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 15:46:43 +00:00
Chris Lattner
265c87a6ef Remove libraries that have no reason to be here, and keep breaking the nightly tester because their makefiles do not have the right dependencies!!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12981 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 14:12:36 +00:00
Chris Lattner
006118fe8c Fix some of the strange CBE-only failures that happened last night.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12980 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 06:03:17 +00:00
Chris Lattner
e1f0cf179f Make sure to check for a very bad class of errors: an instruction
that does not dominate all of its users, but is in the same basic block as
its users.  This class of error is what caused the mysterious CBE only
failures last night.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12979 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 05:51:47 +00:00
Chris Lattner
bf3d2e2683 Bugpoint was not correctly capturing stderr! This caused it to "find" bugs
that didn't exist, missing the ones that do :(


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12978 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 05:35:58 +00:00
Chris Lattner
cd01ae5c70 Fix Inline/2004-04-15-InlineDeletesCall.ll
Basically we were using SimplifyCFG as a huge sledgehammer for a simple
optimization.  Because simplifycfg does so many things, we can't use it
for this purpose.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12977 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-16 05:17:59 +00:00
Misha Brukman
180e8af758 Add note about easier way to debug tests in the llvm tree.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12972 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 21:01:21 +00:00
Misha Brukman
f4e4feb619 Add note about passing arguments to program being debugged.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12970 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 20:49:32 +00:00
Chris Lattner
6f352f3075 New testcase that Brian provided which crashes the inliner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12969 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 20:45:45 +00:00
Chris Lattner
d244057a48 Fix a bug in the previous checkin: if the exit block is not the same as
the back-edge block, we must check the preincremented value.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12968 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 20:26:22 +00:00
Brian Gaeke
b75a316758 Give SparcV9CodeEmitter a head-of-file comment and a PassName.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12967 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 20:23:13 +00:00
Misha Brukman
e803682f9a If we're going to use tabs, use them consistently. Maybe doxygen will find the
@parameter line documentation that way, too.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12966 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:38:48 +00:00
Misha Brukman
f804e2c587 Removed obsolete doxygen options (they were blank anyway).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12965 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:33:49 +00:00
Misha Brukman
234b3ec879 The "best" of both worlds: readable C++ comments and valid HTML For doxygen.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12964 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:30:15 +00:00
Misha Brukman
70c5836631 Don't use invalid HTML in doxygen comments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12963 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:24:55 +00:00
Misha Brukman
7a46e4c62b Don't use invalid HTML in a doxygen comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12962 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:23:45 +00:00
Chris Lattner
59fdaeeae8 Change the canonical induction variable that we insert.
Instead of producing code like this:

Loop:
  X = phi 0, X2
  ...

  X2 = X + 1
  if (X != N-1) goto Loop

We now generate code that looks like this:

Loop:
  X = phi 0, X2
  ...

  X2 = X + 1
  if (X2 != N) goto Loop

This has two big advantages:
  1. The trip count of the loop is now explicit in the code, allowing
     the direct implementation of Loop::getTripCount()
  2. This reduces register pressure in the loop, and allows X and X2 to be
     put into the same register.

As a consequence of the second point, the code we generate for loops went
from:

.LBB2:  # no_exit.1
	...
        mov %EDI, %ESI
        inc %EDI
        cmp %ESI, 2
        mov %ESI, %EDI
        jne .LBB2 # PC rel: no_exit.1

To:

.LBB2:  # no_exit.1
	...
        inc %ESI
        cmp %ESI, 3
        jne .LBB2 # PC rel: no_exit.1

... which has two fewer moves, and uses one less register.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12961 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:21:43 +00:00
Chris Lattner
92020faa2c add some helpful methods. Rearrange #includes to proper order
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12960 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:16:02 +00:00
Chris Lattner
e725cb0d5a Add some helpful methods
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12959 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:15:40 +00:00
Chris Lattner
0a7f98c80d Factor a bunch of classes out into a public header
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12958 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:07:24 +00:00
Chris Lattner
4950e88e0f Publically export all of these classes from the ScalarEvolutions.cpp file
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12957 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 15:06:59 +00:00
Chris Lattner
0aa84fdd04 Unbreak the build
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12956 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-15 14:17:43 +00:00
Chris Lattner
e9fc96e7cc Implement a FIXME: if we're going to insert a cast, we might as well only
insert it once!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12955 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 22:01:22 +00:00
John Criswell
53b54be5fc Remove code to adjust the iterator for llvm.readio and llvm.writeio.
The iterator is pointing at the next instruction which should not disappear
when doing the load/store replacement.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12954 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 21:27:56 +00:00
Brian Gaeke
8121fcd260 Fix typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12953 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 21:21:56 +00:00
Chris Lattner
16011e6201 This is a trivial tweak to the addrec insertion code: insert the increment
at the bottom of the loop instead of the top.  This reduces the number of
overlapping live ranges a lot, for example, eliminating a spill in an important
loop in 183.equake with linear scan.

I still need to make the exit comparison of the loop use the post-incremented
version of this variable, but this is an easy first step.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12952 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 21:11:25 +00:00
Brian Gaeke
50872d5385 Add a TargetData to the PassManager regardless of the TargetMachine.
This should unbreak the Sparc JIT again.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12949 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 17:45:52 +00:00
Brian Gaeke
325297142c Add a copy constructor for TargetData.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@12948 91177308-0d34-0410-b5e6-96231b3b80d8
2004-04-14 17:45:48 +00:00