[PATCH] checkpoint/restart of robust futex lists

[Oren Laadan]

Serge E. Hallyn wrote:
> Quoting Oren Laadan (orenl at cs.columbia.edu):
>>
>> Serge E. Hallyn wrote:
>>> IIUC (from Matt and Dave), after this patch, you might need something
>>> for PI futexes, but otherwise non-contended cases "just work" because
>>> there is no kernel involvement.
>> That's what I thought. But I also thought that a checkpoint would
>> fail anyway as soon as it hits the futex-file-descriptor. Or am I
>> missing something ?
> 
> Yes - FUTEX_FD no longer exists :)

Heh .. need to catch up !

Oren.