[Debian] Re: What version of debian kernel to use?

[Kir Kolyshkin]

It is just checkpointing that is not supported for SPARC. For this arch, 
you have to compile w/out checkpointing feature.

No, unfortunately it's not easy to patch -- at least a few weeks of work 
of our very best developer.

Ola Lundqvist wrote:
> Hi again
>
> Sparc do not seem to be supported anymore...
>
> cpt_context.c
>
> #ifdef CONFIG_X86_64
>         hdr.cpt_os_arch = CPT_OS_ARCH_EMT64;
> #elif defined(CONFIG_X86_32)
>         hdr.cpt_os_arch = CPT_OS_ARCH_I386;
> #elif defined(CONFIG_IA64)
>         hdr.cpt_os_arch = CPT_OS_ARCH_IA64;
> #else
> #error  Arch is not supported
> #endif
>
> Is that easy to patch?
>
> Regards,
>
> // Ola
>
> On Thu, Mar 01, 2007 at 07:19:48PM +0300, Vasily Tarasov wrote:
>   
>> Hello, Ola,
>>
>> This is a fixed patch:
>> http://download.openvz.org/~vass/diff-ovz-2.6.18-028test015-deb-2.gz
>>
>> Thanks,
>> Vasily
>>
>> Ola Lundqvist wrote:
>>     
>>> Privet Vasily (and others)
>>>
>>> On amd64, 486 and sparc architecture I got this build error:
>>>
>>> kernel/sched.c: In function '__activate_task':
>>> kernel/sched.c:1565: warning: implicit declaration of function 've_stop_idle'
>>> kernel/sched.c:1565: error: 've' undeclared (first use in this function)
>>> kernel/sched.c:1565: error: (Each undeclared identifier is reported only once
>>> kernel/sched.c:1565: error: for each function it appears in.)
>>> kernel/sched.c: In function 'deactivate_task':
>>> kernel/sched.c:1718: error: 'pcpu' undeclared (first use in this function)
>>> kernel/sched.c:1735: warning: implicit declaration of function 've_strt_idle'
>>> kernel/sched.c:1735: error: 've' undeclared (first use in this function)
>>> kernel/sched.c:1735: error: 'cpu' undeclared (first use in this function)
>>> kernel/sched.c: In function 'pull_task':
>>> kernel/sched.c:3037: error: 've' undeclared (first use in this function)
>>> kernel/sched.c: In function 'vsched_add_vcpu':
>>> kernel/sched.c:8218: warning: implicit declaration of function 'VE_CPU_STATS'
>>> kernel/sched.c:8218: error: 'struct fairsched_node' has no member named 'owner_env'
>>> kernel/sched.c:8219: error: invalid application of 'sizeof' to incomplete type 'struct ve_cpu_stats' 
>>> kernel/sched.c:8219: warning: passing argument 1 of 'memset' makes pointer from integer without a cast
>>> kernel/sched.c:8221: error: 'struct fairsched_node' has no member named 'owner_env'
>>> make[2]: *** [kernel/sched.o] Fel 1
>>> make[1]: *** [kernel] Fel 2
>>> make[1]: Leaving directory `/home/ola/build/linux-openvz-2.6/linux-source-2.6.18'
>>>
>>>
>>> It would be nice if you could fix that.
>>>
>>> Regards,
>>>
>>> // Ola
>>>
>>>
>>> On Tue, Feb 27, 2007 at 03:50:35PM +0300, Vasily Tarasov wrote:
>>>   
>>>       
>>>> Hej, Ola,
>>>>
>>>> this is OpenVZ patch against 2.6.18.dfsg.1-11 Debian kernel:
>>>> http://download.openvz.org/~vass/patch-ovz-028.015-2.6.18.dfsg.1-11.gz
>>>>
>>>> Vasily
>>>>
>>>> Ola Lundqvist wrote:
>>>>     
>>>>         
>>>>> Privet Vasily
>>>>>
>>>>> On Thu, Feb 22, 2007 at 02:28:52PM +0300, Vasily Tarasov wrote:
>>>>>   
>>>>>       
>>>>>           
>>>>>> Hello, Ola
>>>>>>
>>>>>> I want to clarify one thing.
>>>>>>
>>>>>> linux-source-2.6.18 package is available now in testing and in unstable
>>>>>> repositories.
>>>>>> Last version in testing repository is 2.6.18-7 and last version in
>>>>>> unstable repository is 2.6.18.dfsg.1-11
>>>>>>
>>>>>> As far as I understand testing repository will be transformed to stable
>>>>>> one as soon as Etch will be released. We want to have OpenVZ patch
>>>>>> against Etch kernel, thus I should prepare patch against 2.6.18-7
>>>>>> version, but not against 2.6.18.dfsg.1-11one. Should I?
>>>>>>     
>>>>>>         
>>>>>>             
>>>>> There is a special exception for the kernel currently. In general you
>>>>> are correct but the case today is actually that the 2.6.18.dfsg.1-11
>>>>> will be a part of stable. At least according to the release team.
>>>>>
>>>>> So make the patch against that version.
>>>>>
>>>>> Regards,
>>>>>
>>>>> // Ola
>>>>>
>>>>>   
>>>>>       
>>>>>           
>>>>>> Thank you,
>>>>>> Vasily
>>>>>>
>>>>>>     
>>>>>>         
>>>>>>             
>>     
>
>