[Bug target/23448] New: .p2align before a jump instruction

[Bug target/23448] New: .p2align before a jump instruction

[hjl at lucon dot org]

Gcc 4.0 and 4.1 generate .p2align before a jump instruction. minloc1_8_r8.o
in libgfortran has codes like

        movl    $1, 12(%ecx)
        .p2align 4,,2
        jmp     .L19

-- 
           Summary: .p2align before a jump instruction
           Product: gcc
           Version: 4.0.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: hjl at lucon dot org
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[hjl at lucon dot org]

------- Additional Comments From hjl at lucon dot org  2005-08-17 21:46 -------
Created an attachment (id=9522)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=9522&action=view)
A testcase for gcc 4.0

Here is the testcase for gcc 4.0. x.s is generated with "-O2". x86-64
has the similar problem.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 21:54 -------
Not a bug, it is aligning the loop:
.L5:
        incl    %edx
        cmpl    %edx, %ecx
        je      .L6
        incl    %edx
        cmpl    %edx, %ecx
        .p2align 4,,5
        jne     .L5

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 21:57 -------
And next time don't attach a tar file as it is much harder to get at the testcase.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[hjl at lucon dot org]

------- Additional Comments From hjl at lucon dot org  2005-08-17 23:02 -------
Were you suggesting

.L5:
        incl    %edx
        cmpl    %edx, %ecx
        je      .L6
        incl    %edx
        cmpl    %edx, %ecx
        jne     .L5

was slower? Where does this information come from?

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 23:07 -------
(note 81 50 85 NOTE_INSN_LOOP_END)

(note 85 81 105 [bb 6] NOTE_INSN_BASIC_BLOCK)

(insn 105 85 91 (unspec_volatile [
            (const_int 4 [0x4])
        ] 68) -1 (nil)
    (nil))



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 23:11 -------
  if (TARGET_FOUR_JUMP_LIMIT && optimize && !optimize_size)
    ix86_avoid_jump_misspredicts ();

/* Some CPU cores are not able to predict more than 4 branch instructions in
   the 16 byte window.  */
const int x86_four_jump_limit = m_PPRO | m_ATHLON_K8 | m_PENT4 | m_NOCONA;


So this is not a bug.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 23:13 -------
The alignment is so the stupid processor (yes stupid) will not mis predict the jump.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448

[Bug target/23448] .p2align before a jump instruction

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-17 23:53 -------
from the gcc-patches (since the archives look broken):

looking on recent copy of Intel optimization manual, it has the same
hint as AMD manual about 4 jumps per cache line.
I did SPEC run on the P4 and there is no change except for bzip2 that
improves by about 3%, that is quite expected as the scenario where 5
jumps happens to be in same window is very rare.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23448