[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[rguenth at gcc dot gnu dot org]

------- Comment #2 from rguenth at gcc dot gnu dot org  2008-01-01 17:42 -------
I see different IL for 4.2 compared to 4.3, is the bug present in 4.3?  Can you
attach the optimized tree dump?


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rguenth at gcc dot gnu dot
                   |                            |org
   Target Milestone|4.3.0                       |4.2.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[mmitchel at gcc dot gnu dot org]

------- Comment #3 from mmitchel at gcc dot gnu dot org  2008-01-02 23:14 -------
Put into WAITING state until the information requested in Comment #2 is
available.


-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #4 from tbm at cyrius dot com  2008-01-03 18:03 -------
Bugzilla wraps the testcase in a way that some commented out is no longer
commented out and so you don't see the segfault.  Here's the testcase again
with proper wrapping:

typedef unsigned short u16;
typedef unsigned char u8;

static void
do_segfault(u8 in_buf[], const u8 out_buf[], const int len)
{
  int i;

  for (i = 0; i < len; i++) {
    //SSVAL(in_buf, 2*i, SVAL(out_buf,2*i)); // more or less just a
                                             // byte-wise memcpy

    asm("nop\n");

    in_buf[2*i] = (   out_buf[2*i] | out_buf[(2*i)+1]<<8  ) & 0xFF;
 // in_buf[2*i] = out_buf[2*i];

    asm("nop\n");

    in_buf[(2*i)+1] =  ( out_buf[2*i] | out_buf[(2*i)+1]<<8 ) >> 8;
 // in_buf[(2*i)+1] = out_buf[(2*i)+1];

    asm("nop\n");
  }
}

int main(int argc, char *argv[])
{
  u8 outbuf[32] = "buffer     ";
  u8 inbuf[32] = "\f";

  asm("nop\n");
  do_segfault(inbuf, outbuf, 12);
  asm("nop\n");

  return 0;
}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #5 from tbm at cyrius dot com  2008-01-03 18:07 -------
Created an attachment (id=14870)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14870&action=view)
optimized tree dump (4.2)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

-- 

tbm at cyrius dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |UNCONFIRMED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #6 from tbm at cyrius dot com  2008-01-03 18:22 -------
(In reply to comment #2)
> I see different IL for 4.2 compared to 4.3, is the bug present in 4.3?  Can you
> attach the optimized tree dump?

I also get a segfault with the testcase and 4.3.0 20070916.

The original program (samba) compiled with 4.2 shows bad code (smbclient
returning 'string length'+1 of the first character of the share name rather
than the share name) whereas it segfaults when compiled with 4.3.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[rguenth at gcc dot gnu dot org]

------- Comment #7 from rguenth at gcc dot gnu dot org  2008-01-03 18:58 -------
The final tree IL looks good, so I suspect the RTL loop optimizer gets this
wrong.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rakdver at gcc dot gnu dot
                   |                            |org
           Keywords|                            |wrong-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[rakdver at gcc dot gnu dot org]

------- Comment #8 from rakdver at gcc dot gnu dot org  2008-01-03 21:23 -------
(In reply to comment #7)
> The final tree IL looks good, so I suspect the RTL loop optimizer gets this
> wrong.
> 

>  add     r1, sp, #56 // upper loop-bound; should have been #12
>  I actually wanted to say 'should have been #24' :-)

This insn is actually correct, r1 = &outbuf + 24.  The missing increment of r4
seems to be the problem.  Post-increment is created for r4, but it disappears
in the combine pass.  Does not seem to be loop optimizer related.


-- 

rakdver at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2008-01-03 21:23:36
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug target/34628] [4.2/4.3 Regression] problems with inlining on ARM

[rguenth at gcc dot gnu dot org]

------- Comment #9 from rguenth at gcc dot gnu dot org  2008-01-09 15:56 -------
Trying someone else - Eric, any clue about post-increment?


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ebotcazou at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #10 from ebotcazou at gcc dot gnu dot org  2008-01-09 16:34 -------
Will take a look.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|ebotcazou at gcc dot gnu dot|
                   |org                         |
         AssignedTo|unassigned at gcc dot gnu   |ebotcazou at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
          Component|target                      |rtl-optimization
   Last reconfirmed|2008-01-03 21:23:36         |2008-01-09 16:34:46
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #11 from ebotcazou at gcc dot gnu dot org  2008-01-10 14:03 -------
Created an attachment (id=14907)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14907&action=view)
Lightly tested fix.

I'll give it a whirl on IA-64 but it would be nice to test it on ARM too.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[jakub at gcc dot gnu dot org]

-- 

jakub at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #12 from tbm at cyrius dot com  2008-01-11 22:01 -------
(In reply to comment #11)
> Created an attachment (id=14907)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14907&action=view) [edit]
> Lightly tested fix.
> 
> I'll give it a whirl on IA-64 but it would be nice to test it on ARM too.

I'm away from my machine for another week.  Paul, do you think you could
bootstrap and regtest the patch on ARM?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #13 from tbm at cyrius dot com  2008-01-21 12:31 -------
(In reply to comment #11)
> Created an attachment (id=14907)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14907&action=view) [edit]
> Lightly tested fix.
> 
> I'll give it a whirl on IA-64 but it would be nice to test it on ARM too.

Eric, was your testing successful?  I'll give it a go on ARM now.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #14 from ebotcazou at gcc dot gnu dot org  2008-01-21 14:25 -------
> Eric, was your testing successful?

Yes, everything is clean on the IA-64.

> I'll give it a go on ARM now.

Thanks in advance.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[tbm at cyrius dot com]

------- Comment #15 from tbm at cyrius dot com  2008-01-22 10:49 -------
I've now successfully bootstrapped and regtested this change with trunk on
ARM.  I enabled c,c++,fortran,objc,obj-c++,treelang.

There are no regressions and the segfault is gone.

Eric, can you please submit the patch?

Thanks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #16 from ebotcazou at gcc dot gnu dot org  2008-01-22 11:39 -------
> I've now successfully bootstrapped and regtested this change with trunk on
> ARM.  I enabled c,c++,fortran,objc,obj-c++,treelang.
> 
> There are no regressions and the segfault is gone.

Great!

> Eric, can you please submit the patch?

I can even directly install it. :-)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #17 from ebotcazou at gcc dot gnu dot org  2008-01-22 22:28 -------
Subject: Bug 34628

Author: ebotcazou
Date: Tue Jan 22 22:27:47 2008
New Revision: 131744

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=131744
Log:
        PR rtl-optimization/34628
        * combine.c (try_combine): Stop and undo after the first combination
        if an autoincrement side-effect on the first insn has effectively
        been lost.


Added:
    trunk/gcc/testsuite/gcc.c-torture/execute/20080122-1.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/combine.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #18 from ebotcazou at gcc dot gnu dot org  2008-01-22 22:29 -------
Subject: Bug 34628

Author: ebotcazou
Date: Tue Jan 22 22:29:04 2008
New Revision: 131745

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=131745
Log:
        PR rtl-optimization/34628
        * combine.c (try_combine): Stop and undo after the first combination
        if an autoincrement side-effect on the first insn has effectively
        been lost.


Added:
    branches/gcc-4_2-branch/gcc/testsuite/gcc.c-torture/execute/20080122-1.c
Modified:
    branches/gcc-4_2-branch/gcc/ChangeLog
    branches/gcc-4_2-branch/gcc/combine.c
    branches/gcc-4_2-branch/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628

[Bug rtl-optimization/34628] [4.2/4.3 Regression] problems with inlining on ARM

[ebotcazou at gcc dot gnu dot org]

------- Comment #19 from ebotcazou at gcc dot gnu dot org  2008-01-22 22:32 -------
.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2008-
                   |                            |01/msg01049.html
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34628