Re: Build regressions/improvements in v4.4-rc8

[Guenter Roeck <linux@roeck-us.net>]

On 01/04/2016 07:04 AM, Geert Uytterhoeven wrote:
> On Mon, Jan 4, 2016 at 4:01 PM, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>> JFYI, when comparing v4.4-rc8[1] to v4.4-rc7[3], the summaries are:
>>    - build errors: +19/-18
>
>    + /home/kisskb/slave/src/arch/arm/kernel/ftrace.c: error: implicit
> declaration of function 'flush_tlb_all'
> [-Werror=implicit-function-declaration]:  => 93:2
>    + /home/kisskb/slave/src/arch/arm/kernel/patch.c: error:
> 'L_PTE_DIRTY' undeclared (first use in this function):  => 39:2
>    + /home/kisskb/slave/src/arch/arm/kernel/patch.c: error:
> 'L_PTE_MT_WRITEBACK' undeclared (first use in this function):  => 39:2
>    + /home/kisskb/slave/src/arch/arm/kernel/patch.c: error:
> 'L_PTE_PRESENT' undeclared (first use in this function):  => 39:2
>    + /home/kisskb/slave/src/arch/arm/kernel/patch.c: error: 'L_PTE_XN'
> undeclared (first use in this function):  => 39:2
>    + /home/kisskb/slave/src/arch/arm/kernel/patch.c: error:
> 'L_PTE_YOUNG' undeclared (first use in this function):  => 39:2
>
> arm-randconfig
> Seen and report before
>
>    + /tmp/cc5DX198.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 43
>    + /tmp/ccHnSrdb.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 49, 366
>    + /tmp/ccSLqWGf.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 43
>    + /tmp/cch44bTJ.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 378, 49
>    + /tmp/ccjj7cLa.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 43
>    + /tmp/ccsgtMo8.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 41, 403
>    + /tmp/ccxItlIa.s: Error: can't resolve `_start' {*UND* section} -
> `L0^A' {.text section}:  => 43
>
> Various mips.
> Seems like the fix for this fix still doesn't fix everything?
>

Yes, there are more binutils problems. The current kernel code does not correctly
detect binutils versions such as 2.24.90, where the third level version number
is >= 10. See the discussion in [1], and the fix from James Hogan in [2].


If I find the time, I'll extend my test scripts to run some mips builds
with different binutils versions.

Guenter

---
[1] https://patchwork.linux-mips.org/patch/11929/
[2] https://patchwork.linux-mips.org/patch/11931/