Re: [RFC PATCH 0/7] libxl: fork: Selective reaping

[Jim Fehlig <jfehlig@suse.com>]

Ian Jackson wrote:
> Ian Jackson writes ("[RFC PATCH 0/7] libxl: fork: Selective reaping"):
>   
>> libvirt reaps its children synchronously and has no central pid
>> registry and no dispatch mechanism.  libxl does have a pid registry so
>> can provide a selective reaping facility, but that is not currently exposed.
>>
>> NB that I have compiled this series but I have NOT EXECUTED IT.
>> The most plausible test environment is a suitably modified libvirt.
>>
>>  1/7 libxl: fork: Break out checked_waitpid
>>  2/7 libxl: fork: Break out childproc_reaped_ours
>>  3/7 libxl: fork: Clarify docs for libxl_sigchld_owner
>>  4/7 libxl: fork: assert that chldmode is right
>>  5/7 libxl: fork: Provide libxl_childproc_sigchld_occurred
>>  6/7 libxl: fork: Provide ..._always_selective_reap
>>  7/7 libxl: fork: Provide LIBXL_HAVE_SIGCHLD_SELECTIVE_REAP
>>     
>
> I should say, to Jim: I think that with this series applied, simply
> having libvirt pass libxl_sigchld_owner_libxl_always_selective_reap
> should be sufficient for everything to work.
>   

Ian, thanks a lot for these patches!  They certainly make life much
easier in the libvirt libxl driver :).  See 6/7 for a few questions.

Regards,
Jim