Re: [RFD] posix-timers: CRIU woes

[Pavel Tikhomirov <ptikhomirov@virtuozzo.com>]

On 10.05.2023 16:30, Thomas Gleixner wrote:
> Pavel!
> 
> On Wed, May 10 2023 at 12:36, Pavel Tikhomirov wrote:
>> On 10.05.2023 05:42, Thomas Gleixner wrote:
>>> So because of that half thought out user space ABI we are now up the
>>> regression creek without a paddle, unless CRIU can accomodate to a
>>> different restore mechanism to lift this restriction from the kernel.
>>>
>>> Thoughts?
>>
>> Maybe we can do something similar to /proc/sys/kernel/ns_last_pid?
>> Switch to per-(process->signal) idr based approach with idr_set_cursor
>> to set next id for next posix timer from new sysctl?
> 
> I'm not a fan of such sysctls. We have already too many of them and that
> particular one does not buy much.

Sorry, it was a bad idea, what you suggest below is much better.

> 
> We can simply let timer_create() or a new syscall create a timer at a
> given ID.

Yes this would work for CRIU. (note: in neighbor thread Andrei writes 
about adding a bit to sigevent.sigev_notify to request a timer with a 
specified id, new syscall is also a good option)

> 
> That allows CRIU to restore any checkpointed process no matter which
> kernel version it came from without doing this insane create/delete
> dance.

Yes, for CRIU this kind of API change is a big improvement.

> 
> The downside is that this allows to create stupidly sparse timer IDs
> even for the non CRIU case, which increases per process kernel memory
> consumption and creates slightly more overhead in the signal delivery
> path. The latter is a burden on the process owning the timer and not
> affecting expiry, which is a context stealing operation. The memory part
> needs eventually some thoughts vs. accounting.
> 
> If the 'explicit at ID' option is not used then the ID mechanism is
> optimzied for dense IDs by using the first available ID in a bottom up
> search, which recovers holes created by a timer_delete() operation.

Not sure how kernel memory consumption increases with sparse timer IDs, 
global hashtable (posix_timers_hashtable) is the same size anyway, 
entries in hlists can be distributed differently as hash depends on id 
directly but we have same number of entries. Probably I miss something, 
why do we need dense IDs?

> 
> Thanks,
> 
>          tglx

-- 
Best regards, Tikhomirov Pavel
Senior Software Developer, Virtuozzo.