Re: [PATCH 6/7] libxl: fork: Provide ..._always_selective_reap

[Jim Fehlig <jfehlig@suse.com>]

Ian Jackson wrote:
> Jim Fehlig writes ("Re: [PATCH 6/7] libxl: fork: Provide ..._always_selective_reap"):
>   
>> I don't see the assert, regardless of how I call setmode(), when
>> changing this hunk to
>>
>> @@ -264,11 +264,11 @@ static bool chldmode_ours(libxl_ctx *ctx, bool
>> creating)
>>  {
>>      switch (ctx->childproc_hooks->chldowner) {
>>      case libxl_sigchld_owner_libxl:
>> +    case libxl_sigchld_owner_libxl_always_selective_reap:
>>          return creating || !LIBXL_LIST_EMPTY(&ctx->children);
>>      case libxl_sigchld_owner_mainloop:
>>          return 0;
>>      case libxl_sigchld_owner_libxl_always:
>> -    case libxl_sigchld_owner_libxl_always_selective_reap:
>>          return 1;
>>      }
>>      abort();
>>     
>
> I should say: that that works is just luck, I think.

That's what I suspected :).  And I also suspect my luck would run out
once I throw 10's of domains with many concurrent operations in the mix.

> I have a better fix.
>   

Ok, thanks!

Regards,
Jim