Linux 3.8-rc1: compiling problem in perf-event-p6.o

Linux 3.8-rc1: compiling problem in perf-event-p6.o

[werner]

[-- Attachment #1: Type: text/plain, Size: 1250 bytes --]

We are now on -rc3 and someone should correct this, 
finally

This is a regression, it was not before, on 3.6

This messes up any compilation of the whole kernel, it 
results in don't be produced vmlinuz

arch/x86/kernel/cpu/perf_event_p6.o   depends on so much 
things that I don't get it switched off, I suppose it's 
necesary for the most systems


W.Landgraf






=================
The problem continues with 3.8-rc

This is grave, no vmlinuz is produced.


wl

   CC arch/x86/kernel/cpu/perf_event.o
   CC arch/x86/kernel/cpu/perf_event_amd.o
   CC arch/x86/kernel/cpu/perf_event_p6.o
arch/x86/kernel/cpu/perf_event_p6.c:22: error: 
p6_hw_cache_event_ids causes a section type conflict
make[3]: [arch/x86/kernel/cpu/perf_event_p6.o] Error 1 
(ignored)
   CC arch/x86/kernel/cpu/perf_event_knc.o
   CC arch/x86/kernel/cpu/perf_event_p4.o
   CC arch/x86/kernel/cpu/perf_event_intel_lbr.o

================================================
There ocurs a compiling error in perf-event-p6.o , any 
regression, unfortunately I lost the compiling list but I 
think it was any incompatibility / redefinition with 
something else, pls check and correct that, if not already 
done
W.Landgraf
---
---
Professional hosting for everyone - http://www.host.ru

[-- Attachment #2: linux-3.8-rc3-i486-1sys.log.bz2 --]
[-- Type: application/octet-stream, Size: 86077 bytes --]

Re: Linux 3.8-rc1: compiling problem in perf-event-p6.o

[Randy Dunlap]

On 01/15/13 06:50, werner wrote:
> We are now on -rc3 and someone should correct this, finally
> 
> This is a regression, it was not before, on 3.6
> 
> This messes up any compilation of the whole kernel, it results in don't be produced vmlinuz
> 
> arch/x86/kernel/cpu/perf_event_p6.o   depends on so much things that I don't get it switched off, I suppose it's necesary for the most systems
> 
> 
> W.Landgraf
> 
> 
> 
> 
> 
> 
> =================
> The problem continues with 3.8-rc
> 
> This is grave, no vmlinuz is produced.
> 
> 
> wl
> 
>   CC arch/x86/kernel/cpu/perf_event.o
>   CC arch/x86/kernel/cpu/perf_event_amd.o
>   CC arch/x86/kernel/cpu/perf_event_p6.o
> arch/x86/kernel/cpu/perf_event_p6.c:22: error: p6_hw_cache_event_ids causes a section type conflict
> make[3]: [arch/x86/kernel/cpu/perf_event_p6.o] Error 1 (ignored)
>   CC arch/x86/kernel/cpu/perf_event_knc.o
>   CC arch/x86/kernel/cpu/perf_event_p4.o
>   CC arch/x86/kernel/cpu/perf_event_intel_lbr.o
> 
> ================================================
> There ocurs a compiling error in perf-event-p6.o , any regression, unfortunately I lost the compiling list but I think it was any incompatibility / redefinition with something else, pls check and correct that, if not already done
> W.Landgraf

Hi,

I don't see this problem on 3.8-rc1 or -rc3.
Maybe a difference/problem in gcc??


-- 
~Randy

Re: Linux 3.8-rc1: compiling problem in perf-event-p6.o

[werner]

This didn't happen on 3.6  (3.7 I had no time to check 
out), although I continue to use the same c compiler and 
libraries.  So it should be caused by any change of the 
subroutine perf_event_p6  made in the last time.  Also, 
everything else of the kernel compiled, just not that 
subroutine.   Anyway the kernel should compile correctly 
with all recent versions of the c compiler, so one should 
search and fix that.   The maintainer of that subroutine 
should be notified, and check the last changes since 3.6 .

Also, I always compile the WHOLE kernel, not only changed 
subroutines. Perhaps in any other subroutine was changed 
something of global variables in any other subroutine, 
what is now incompatible with subroutine perf_event_p6

W.Landgraf



=============================================
On Wed, 16 Jan 2013 17:26:33 -0800
  Randy Dunlap <rdunlap@infradead.org> wrote:
> On 01/15/13 06:50, werner wrote:
>> We are now on -rc3 and someone should correct this, 
>>finally
>> 
>> This is a regression, it was not before, on 3.6
>> 
>> This messes up any compilation of the whole kernel, it 
>>results in don't be produced vmlinuz
>> 
>> arch/x86/kernel/cpu/perf_event_p6.o   depends on so much 
>>things that I don't get it switched off, I suppose it's 
>>necesary for the most systems
>> 
>> 
>> W.Landgraf
>> 
>> 
>> 
>> 
>> 
>> 
>> =================
>> The problem continues with 3.8-rc
>> 
>> This is grave, no vmlinuz is produced.
>> 
>> 
>> wl
>> 
>>   CC arch/x86/kernel/cpu/perf_event.o
>>   CC arch/x86/kernel/cpu/perf_event_amd.o
>>   CC arch/x86/kernel/cpu/perf_event_p6.o
>> arch/x86/kernel/cpu/perf_event_p6.c:22: error: 
>>p6_hw_cache_event_ids causes a section type conflict
>> make[3]: [arch/x86/kernel/cpu/perf_event_p6.o] Error 1 
>>(ignored)
>>   CC arch/x86/kernel/cpu/perf_event_knc.o
>>   CC arch/x86/kernel/cpu/perf_event_p4.o
>>   CC arch/x86/kernel/cpu/perf_event_intel_lbr.o
>> 
>> ================================================
>> There ocurs a compiling error in perf-event-p6.o , any 
>>regression, unfortunately I lost the compiling list but I 
>>think it was any incompatibility / redefinition with 
>>something else, pls check and correct that, if not 
>>already done
>> W.Landgraf
> 
> Hi,
> 
> I don't see this problem on 3.8-rc1 or -rc3.
> Maybe a difference/problem in gcc??
> 
> 
> -- 
> ~Randy
> 
> 

"werner" <w.landgraf@ru.ru>
---
Professional hosting for everyone - http://www.host.ru

Re: Linux 3.8-rc1: compiling problem in perf-event-p6.o

[werner]

That should be corrected, even if it's a problem of the CC 
version / libries, or not,   compilation can't depend on 
this, code can't be so non-standard that it depends on the 
compiler version.

wl




=====================================

The problem continues with 3.8-rc

This is grave, no vmlinuz is produced.


wl

   CC arch/x86/kernel/cpu/perf_event.o
   CC arch/x86/kernel/cpu/perf_event_amd.o
   CC arch/x86/kernel/cpu/perf_event_p6.o
arch/x86/kernel/cpu/perf_event_p6.c:22: error: 
p6_hw_cache_event_ids causes a section type conflict
make[3]: [arch/x86/kernel/cpu/perf_event_p6.o] Error 1 
(ignored)
   CC arch/x86/kernel/cpu/perf_event_knc.o
   CC arch/x86/kernel/cpu/perf_event_p4.o
   CC arch/x86/kernel/cpu/perf_event_intel_lbr.o
---
Professional hosting for everyone - http://www.host.ru

Re: Linux 3.8-rc1: compiling problem in perf-event-p6.o

[Linus Torvalds]

On Fri, Mar 1, 2013 at 7:08 AM, werner <w.landgraf@ru.ru> wrote:
> That should be corrected, even if it's a problem of the CC version /
> libries, or not,   compilation can't depend on this, code can't be so
> non-standard that it depends on the compiler version.

The __devinit is removed in current git already. If it still does
something like that, then the compiler is too buggy to be used. It's a
plain static "u64" array, nothing more.

          Linus

Linux 3.8-rc1: compiling problem in perf-event-p6.o

[werner]

The problem continues with 3.8-rc

This is grave, no vmlinuz is produced.


wl

   CC      arch/x86/kernel/cpu/perf_event.o
   CC      arch/x86/kernel/cpu/perf_event_amd.o
   CC      arch/x86/kernel/cpu/perf_event_p6.o
arch/x86/kernel/cpu/perf_event_p6.c:22: error: 
p6_hw_cache_event_ids causes a section type conflict
make[3]: [arch/x86/kernel/cpu/perf_event_p6.o] Error 1 
(ignored)
   CC      arch/x86/kernel/cpu/perf_event_knc.o
   CC      arch/x86/kernel/cpu/perf_event_p4.o
   CC      arch/x86/kernel/cpu/perf_event_intel_lbr.o

================================================
There ocurs a compiling error in perf-event-p6.o , any 
regression, unfortunately I lost the compiling list but I 
think it was any incompatibility / redefinition with 
something else, pls check and correct that, if not already 
done
W.Landgraf
---
---
Professional hosting for everyone - http://www.host.ru

Linux 3.8-rc1: compiling problem in perf-event-p6.o

[werner]

There ocurs a compiling error in perf-event-p6.o  , any 
regression, unfortunately I lost the compiling list but I 
think it was any incompatibility / redefinition with 
something else, pls check and correct that, if not already 
done
W.Landgraf
---
Professional hosting for everyone - http://www.host.ru