[Bug target/62642] New: x86 rdtsc is moved through barrier

[Bug target/62642] New: x86 rdtsc is moved through barrier

[M8R-ynb11d at mailinator dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

            Bug ID: 62642
           Summary: x86 rdtsc is moved through barrier
           Product: gcc
           Version: 4.8.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: M8R-ynb11d at mailinator dot com

given:

unsigned long long measure(void (*func)(void))
{
    unsigned long long before = __builtin_ia32_rdtsc();
    asm volatile("" ::: "memory");
    func();
    asm volatile("" ::: "memory");
    unsigned long long after = __builtin_ia32_rdtsc();
    return after - before;
}

On x86 linux with -O2, this results in the obviously useless:

measure:
    push    edi
    push    esi
    push    ebx
    call    [DWORD PTR [esp+16]]
    rdtsc
    mov    esi, eax
    mov    edi, edx
    rdtsc
    pop    ebx
    sub    esi, eax
    sbb    edi, edx
    mov    eax, esi
    mov    edx, edi
    pop    esi
    pop    edi
    ret

I can reproduce the problem on 32 bit x86 on Linux and MinGW with 4.8.2 and
4.9.1. (I guess 4.8.0 also exhibits the problem but I don't have that available
for testing, so I set the version to 4.8.2 in the report.)

4.7.x and 4.6.x work correctly, as does 64 bit on both platforms.

If this is not the proper sanctioned way to write this function, I'm all ears
to a better way.  I've tried also adding calls to __builtin_ia32_mfence() which
as I understand it should not be necessary, and it gets even more comical:

        ...
    mfence
    call    [DWORD PTR [esp+16]]
    mfence
    rdtsc
    mov    esi, eax
    mov    edi, edx
    rdtsc
        ...

[Bug target/62642] x86 rdtsc is moved through barrier

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-09-01
     Ever confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
The problem shows up in IRA stage.
New iteration of spill/restore move
+++Costs: overall -65521000, reg -65536000, mem 15000, ld 0, st 0, move 0
+++       move loops 0, new jumps 0
scanning new insn with uid = 22.
moving def of 88 (insn 6 now) deleting insn with uid = 21.
deleting insn with uid = 6.
rescanning insn with uid = 22.
 22) rather than keep unallocated replacement 93

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.8.4

[Bug target/62642] [4.8/4.9/5 Regression] x86 rdtsc is moved through barrier

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ra
             Target|                            |i?86-*-*
           Priority|P3                          |P2
                 CC|                            |vmakarov at gcc dot gnu.org
            Summary|[4.8/4.9 Regression] x86    |[4.8/4.9/5 Regression] x86
                   |rdtsc is moved through      |rdtsc is moved through
                   |barrier                     |barrier

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed on trunk.

[Bug target/62642] [4.8/4.9/5 Regression] x86 rdtsc is moved through barrier

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

--- Comment #3 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
The problem is in a live range splitting optimization to reduce register
pressure.  It was introduced by

https://gcc.gnu.org/ml/gcc-patches/2011-12/msg01531.html

It is easy to switch off for this kind of insns.  I'll submit the patch on
Monday.

[Bug target/62642] [4.8/4.9/5 Regression] x86 rdtsc is moved through barrier

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

--- Comment #4 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
Author: vmakarov
Date: Mon Dec 15 20:04:04 2014
New Revision: 218761

URL: https://gcc.gnu.org/viewcvs?rev=218761&root=gcc&view=rev
Log:
2014-12-15  Vladimir Makarov  <vmakarov@redhat.com>

    PR target/62642
    * ira.c (rtx_moveable_p): Prevent UNSPEC_VOLATILE moves.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/ira.c

[Bug target/62642] [4.8/4.9/5 Regression] x86 rdtsc is moved through barrier

[M8R-ynb11d at mailinator dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

--- Comment #5 from M8R-ynb11d at mailinator dot com ---
I originally put the barriers there in a futile attempt to work around the bug.
 Can anyone tell me whether I actually need them, or whether the intrinsic
carries with it an implicit built-in barrier to prevent reordering?  Ideally
I'd like to write portable code using only intrinsics and not gcc-specific
asm() stuff, so I hope that it's the latter.

[Bug target/62642] [4.8/4.9/5 Regression] x86 rdtsc is moved through barrier

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.8.4                       |4.8.5

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 4.8.4 has been released.

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|                            |5.0
            Summary|[4.8/4.9/5 Regression] x86  |[4.8/4.9 Regression] x86
                   |rdtsc is moved through      |rdtsc is moved through
                   |barrier                     |barrier

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk.

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Vlad, do you think this is backportable to 4.9 and 4.8?

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

--- Comment #10 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Fri Jun 12 11:55:11 2015
New Revision: 224421

URL: https://gcc.gnu.org/viewcvs?rev=224421&root=gcc&view=rev
Log:
        Backported from mainline
        2014-12-15  Vladimir Makarov  <vmakarov@redhat.com>

        PR target/62642
        * ira.c (rtx_moveable_p): Prevent UNSPEC_VOLATILE moves.

Modified:
    branches/gcc-4_8-branch/gcc/ChangeLog
    branches/gcc-4_8-branch/gcc/ira.c

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

--- Comment #9 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Fri Jun 12 11:54:43 2015
New Revision: 224420

URL: https://gcc.gnu.org/viewcvs?rev=224420&root=gcc&view=rev
Log:
        Backported from mainline
        2014-12-15  Vladimir Makarov  <vmakarov@redhat.com>

        PR target/62642
        * ira.c (rtx_moveable_p): Prevent UNSPEC_VOLATILE moves.

Modified:
    branches/gcc-4_9-branch/gcc/ChangeLog
    branches/gcc-4_9-branch/gcc/ira.c

[Bug target/62642] [4.8/4.9 Regression] x86 rdtsc is moved through barrier

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62642

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #11 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.