[Bug rtl-optimization/21223] New: Output register variable is reloaded into the wrong register for asm

[Bug rtl-optimization/21223] New: Output register variable is reloaded into the wrong register for asm

[drow at gcc dot gnu dot org]

Compile this test with an ARM-targeted compiler.  Using -O2, it works out OK;
but for -O0 or -O1 it doesn't.  I originally encountered the bug in -O2 compiled
code, but a more complex testcase is probably necessary for that.

int bar(int *mem)
{
  return ({ register int a_oldval asm ("r0");
            register int a_tmp asm ("r3");
            int a_oldval2 = 3;
            __asm__ __volatile__ ("# Should be r0 - %0\n\t"
                                  "# Should be r3 - %1"
                                  : "=&r" (a_oldval), "=&r" (a_tmp)
                                  : "r" (a_oldval2));
            a_tmp; });
}

Everything is OK up until reload.  Then this happens:

Spilling for insn 19.
Using reg 2 for reload 0
Spilling for insn 19.
Using reg 2 for reload 0

Reloads for insn # 19
Reload 0: reload_out (SI) = (reg/v:SI 3 r3 [ a_tmp ])
        GENERAL_REGS, RELOAD_OTHER (opnum = 1)
        reload_out_reg: (reg/v:SI 3 r3 [ a_tmp ])
        reload_reg_rtx: (reg:SI 2 r2)

and the set of a_tmp is replaced by a set of r2.  I can't figure out why
the reload was created.

-- 
           Summary: Output register variable is reloaded into the wrong
                    register for asm
           Product: gcc
           Version: 4.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: rtl-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: drow at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
GCC target triplet: arm-none-linux-gnueabi


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

[Bug rtl-optimization/21223] Output register variable is reloaded into the wrong register for asm

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-04-25 20:54 -------
Confirmed, also on powerpc-darwin. Weird.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
 GCC target triplet|arm-none-linux-gnueabi      |arm-none-linux-gnueabi,
                   |                            |powerpc-darwin
           Keywords|                            |wrong-code
      Known to fail|                            |3.3.3 4.1.0
   Last reconfirmed|0000-00-00 00:00:00         |2005-04-25 20:54:01
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

[Bug rtl-optimization/21223] Output register variable is reloaded into the wrong register for asm

[drow at gcc dot gnu dot org]

------- Additional Comments From drow at gcc dot gnu dot org  2005-04-25 21:07 -------
The reload was created because the earlyclobber operand conflicts with
a_oldval2, which is placed in r3 by local-alloc.  I think I see why...

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

[Bug middle-end/21223] Output register variable is reloaded into the wrong register for asm

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-04-25 21:09 -------
And here is an x86 testcase:
int bar(int *mem)
{
  return ({ register int a_oldval asm ("eax");
            register int a_tmp asm ("ebx");
            int a_oldval2 = 3;
            __asm__ __volatile__ ("# Should be eax - %0\n\t"
                                  "# Should be ebx - %1\n\t"
                                  "# unknown - %2"
                                  : "=&r" (a_oldval), "=&r" (a_tmp)
                                  : "r" (a_oldval2));
            a_tmp; });
}

Which shows this fails at -O2 and above too.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|rtl-optimization            |middle-end
 GCC target triplet|arm-none-linux-gnueabi,     |
                   |powerpc-darwin              |
      Known to fail|3.3.3 4.1.0                 |3.3.3 4.1.0 2.95.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

[Bug middle-end/21223] Output register variable is reloaded into the wrong register for asm

[drow at gcc dot gnu dot org]

------- Additional Comments From drow at gcc dot gnu dot org  2005-04-25 21:12 -------
The problem is in reg_is_set or its caller.

Because the output is an earlyclobber operand, it needs to be marked as born
before the insn, to conflict with its inputs.  I do not see an obvious way to
get from note_stores to the recog alternative.

I reproduced this bug in 3.4.3 (csl-arm-branch) also.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

[Bug middle-end/21223] Output register variable is reloaded into the wrong register for asm

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=21223

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=87600,
                   |                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=86939,
                   |                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=87899
      Known to fail|                            |4.9.0, 8.1.0
   Target Milestone|---                         |9.0
             Status|NEW                         |RESOLVED
      Known to work|                            |9.1.0

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
So this is fully fixed in GCC 9+. LRA fixes the problems that reload had.  (LRA
had a similar bug but had been fixes see the linked bug reports).