[Bug target/52838] New: [x32] missed optimization for pointer return value

[Bug target/52838] New: [x32] missed optimization for pointer return value

[bartoldeman at users dot sourceforge.net]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

             Bug #: 52838
           Summary: [x32] missed optimization for pointer return value
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: bartoldeman@users.sourceforge.net


The program test.c:

extern void *foo(void);
extern void bar(void*);

void test(void)
{
  bar(foo());
}

when compiled with
gcc-4.7 -mx32 -Os -S test.c
produces:
    .file    "test.c"
    .text
    .globl    test
    .type    test, @function
test:
.LFB0:
    .cfi_startproc
    pushq    %rax
    .cfi_def_cfa_offset 16
    call    foo
    popq    %rdx
    .cfi_def_cfa_offset 8
    movq    %rax, %rdi
    jmp    bar
    .cfi_endproc
.LFE0:
    .size    test, .-test
    .ident    "GCC: (Debian 4.7.0-1) 4.7.0"
    .section    .note.GNU-stack,"",@progbits

Here "movq %rax, %rdi" could be replaced by "movl %eax, %edi", saving one
prefix byte 0x48.

[Bug rtl-optimization/52838] [x32] missed optimization for pointer return value

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2012-04-04
                 CC|                            |ebotcazou at gcc dot
                   |                            |gnu.org
          Component|target                      |rtl-optimization
     Ever Confirmed|0                           |1

--- Comment #1 from Uros Bizjak <ubizjak at gmail dot com> 2012-04-04 19:06:57 UTC ---
This looks like a combine problem:

(insn 8 6 9 2 (set (reg/f:SI 59 [ D.1705 ])
        (subreg/s/u:SI (reg:DI 60) 0)) pr52838.c:6 64 {*movsi_internal}
     (expr_list:REG_DEAD (reg:DI 60)
        (nil)))

(insn 9 8 10 2 (set (reg/f:DI 62 [ D.1705 ])
        (zero_extend:DI (reg/f:SI 59 [ D.1705 ]))) pr52838.c:6 112
{*zero_extendsidi2_rex64}
     (expr_list:REG_DEAD (reg/f:SI 59 [ D.1705 ])
        (nil)))

This gets transformed by combine pass:


Trying 8 -> 9:
Successfully matched this instruction:
(set (reg/f:DI 62 [ D.1705 ])
    (reg:DI 60))
deferring deletion of insn with uid = 8.
modifying insn i3     9 r62:DI=r60:DI
      REG_DEAD: r60:DI

into:

(note 8 6 9 2 NOTE_INSN_DELETED)

(insn 9 8 10 2 (set (reg/f:DI 62 [ D.1705 ])
        (reg:DI 60)) pr52838.c:6 62 {*movdi_internal_rex64}
     (expr_list:REG_DEAD (reg:DI 60)
        (nil)))

But, but ... high 32 bits are not cleared anymore!

Confirmed as a rtl optimization issue, adding CCs for their opinion.

[Bug rtl-optimization/52838] [x32] missed optimization for pointer return value

[ebotcazou at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

--- Comment #2 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2012-04-04 21:41:02 UTC ---
> This looks like a combine problem:
> 
> (insn 8 6 9 2 (set (reg/f:SI 59 [ D.1705 ])
>         (subreg/s/u:SI (reg:DI 60) 0)) pr52838.c:6 64 {*movsi_internal}
>      (expr_list:REG_DEAD (reg:DI 60)
>         (nil)))
> 
> (insn 9 8 10 2 (set (reg/f:DI 62 [ D.1705 ])
>         (zero_extend:DI (reg/f:SI 59 [ D.1705 ]))) pr52838.c:6 112
> {*zero_extendsidi2_rex64}
>      (expr_list:REG_DEAD (reg/f:SI 59 [ D.1705 ])
>         (nil)))
> 
> This gets transformed by combine pass:
> 
> 
> Trying 8 -> 9:
> Successfully matched this instruction:
> (set (reg/f:DI 62 [ D.1705 ])
>     (reg:DI 60))
> deferring deletion of insn with uid = 8.
> modifying insn i3     9 r62:DI=r60:DI
>       REG_DEAD: r60:DI
> 
> into:
> 
> (note 8 6 9 2 NOTE_INSN_DELETED)
> 
> (insn 9 8 10 2 (set (reg/f:DI 62 [ D.1705 ])
>         (reg:DI 60)) pr52838.c:6 62 {*movdi_internal_rex64}
>      (expr_list:REG_DEAD (reg:DI 60)
>         (nil)))
> 
> But, but ... high 32 bits are not cleared anymore!

The existence of (subreg/s/u:SI (reg:DI 60) 0) means that (reg:DI 60) is known
to be a zero-extended 32-bit value, so the optimization is valid (and useful).

[Bug rtl-optimization/52838] [x32] missed optimization for pointer return value

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl.tools at gmail dot com

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2012-04-05 00:46:13 UTC ---
(In reply to comment #2)
> The existence of (subreg/s/u:SI (reg:DI 60) 0) means that (reg:DI 60) is known
> to be a zero-extended 32-bit value, so the optimization is valid (and useful).

Can we somehow keep the knowledge that it is a zero-extended 32-bit value?
It is useful for encoding purpose.

[Bug rtl-optimization/52838] [x32] missed optimization for pointer return value

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ubizjak at gmail dot com

--- Comment #4 from Uros Bizjak <ubizjak at gmail dot com> 2012-04-05 12:36:28 UTC ---
(In reply to comment #2)

> The existence of (subreg/s/u:SI (reg:DI 60) 0) means that (reg:DI 60) is known
> to be a zero-extended 32-bit value, so the optimization is valid (and useful).

Please note similar problem with sign_extend, PR 52876.

[Bug rtl-optimization/52838] [x32] missed optimization for pointer return value

[ebotcazou at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

--- Comment #5 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2012-04-05 16:16:41 UTC ---
> Can we somehow keep the knowledge that it is a zero-extended 32-bit value?
> It is useful for encoding purpose.

After combine has run, this seems hard.  So the easiest approach might be to
disable the combination by tweaking the costs at -Os.

[Bug target/52838] [x32] missed optimization for pointer return value

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52838

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|rtl-optimization            |target
             Target|                            |x86_64
           Severity|normal                      |enhancement
           Keywords|                            |missed-optimization