[Bug rtl-optimization/70164] [8/9/10 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[Bug rtl-optimization/70164] [8/9/10 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[clyon at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

Christophe Lyon <clyon at gcc dot gnu.org> changed:

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

--- Comment #24 from Christophe Lyon <clyon at gcc dot gnu.org> ---
With today's trunk, I still see the same problem:

history_expand_line_internal:
        @ args = 0, pretend = 0, frame = 8
        @ frame_needed = 0, uses_anonymous_args = 0
        push    {r0, r1, r2, r4, r5, r6, r7, lr}
        ldr     r3, .L3
        ldr     r6, .L3+4
        ldr     r7, [r3]
        ldr     r3, [r6]       ; ; <--- load of 'hist_verify' onto r3
        movs    r5, r0
        str     r3, [sp, #4]    ; <--- Spill
        movs    r3, #0
        str     r3, [r6]            
        bl      pre_process_line
        ldr     r3, [sp, #4]    ; <--- reload
        movs    r4, r0
        adds    r7, r7, r3
        str     r7, [r6]
        cmp     r5, r0
        bne     .L1
        bl      str_len
        adds    r0, r0, #1
        bl      x_malloc
        movs    r1, r4
        bl      str_cpy
        movs    r4, r0
.L1:
        @ sp needed
        movs    r0, r4
        pop     {r1, r2, r3, r4, r5, r6, r7, pc}
.L4:
        .align  2
.L3:
        .word   a1
        .word   hist_verify

[Bug rtl-optimization/70164] [8/9/10 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

--- Comment #25 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
Sorry, I can not reproduce this.  With today trunk I have for pr45701-1.c (-Os
-mthumb):

history_expand_line_internal:
        @ args = 0, pretend = 0, frame = 0
        @ frame_needed = 0, uses_anonymous_args = 0
        push    {r3, r4, r5, r6, r7, lr}
        mov     r4, r0
        ldr     r5, .L3
        ldr     r3, .L3+4
        ldr     r7, [r5]
        ldr     r6, [r3]
        movs    r3, #0
        str     r3, [r5]
        bl      pre_process_line
        cmp     r4, r0
        add     r6, r6, r7
        str     r6, [r5]
        bne     .L1
        bl      str_len
        adds    r0, r0, #1
        bl      x_malloc
        mov     r1, r4
        pop     {r3, r4, r5, r6, r7, lr}
        b       str_cpy
.L1:
        pop     {r3, r4, r5, r6, r7, pc}

[Bug rtl-optimization/70164] [8/9/10 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[clyon at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

--- Comment #26 from Christophe Lyon <clyon at gcc dot gnu.org> ---
For what CPU did you configure GCC?
With today's trunk I still see the same code as in comment #24.

I can get the same code as you have in comment #25 if I force -mcpu=cortex-a9.

The bug report is about cortex-m0, so you either need to configure GCC
--with-cpu=cortex-m0 or use -mcpu=cortex-m0

[Bug rtl-optimization/70164] [8/9/10 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

--- Comment #27 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
(In reply to Christophe Lyon from comment #26)
> For what CPU did you configure GCC?
> With today's trunk I still see the same code as in comment #24.
> 
> I can get the same code as you have in comment #25 if I force
> -mcpu=cortex-a9.
> 
> The bug report is about cortex-m0, so you either need to configure GCC
> --with-cpu=cortex-m0 or use -mcpu=cortex-m0

Ok.  Thank you.  I've reproduced this with -mcpu=cortex-m0.

[Bug rtl-optimization/70164] [9/10/11/12 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|8.5                         |9.4

--- Comment #28 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 8 branch is being closed.

[Bug rtl-optimization/70164] [9/10/11/12 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.4                         |9.5

--- Comment #29 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9.4 is being released, retargeting bugs to GCC 9.5.

[Bug rtl-optimization/70164] [10/11/12/13 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.5                         |10.4

--- Comment #30 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9 branch is being closed

[Bug rtl-optimization/70164] [10/11/12/13 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.4                        |10.5

--- Comment #31 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 10.4 is being released, retargeting bugs to GCC 10.5.

[Bug rtl-optimization/70164] [11/12/13/14 Regression] Code/performance regression due to poor register allocation on Cortex-M0

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70164

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.5                        |11.5

--- Comment #32 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 10 branch is being closed.