Re: [Qemu-devel] [PATCH 3/3] tcg: Avoid bouncing tb_lock between tb_gen_code() and tb_add_jump()

[Paolo Bonzini]

On 08/07/2016 14:32, Sergey Fedorov wrote:
>>> >> I think we can do even better. One option is using a separate tiny lock
>>> >> to protect direct jump set/reset instead of tb_lock.
>> > If you have to use a separate tiny lock, you don't gain anything compared
>> > to the two critical sections, do you?
> If we have a separate lock for direct jump set/reset then we can do fast
> TB lookup + direct jump patching without taking tb_lock at all. How much
> this would reduce lock contention largely depends on the workload we use.

Yeah, it probably would be easy enough that it's hard to object to it
(unlike the other idea below, which I'm not very comfortable with, at
least without seeing patches).

The main advantage would be that this tiny lock could be a spinlock
rather than a mutex.

Paolo

>> The one below is even more complicated.  I'm all for simple lock-free
>> stacks (e.g. QSLIST_INSERT_HEAD_ATOMIC and QSLIST_MOVE_ATOMIC), but lock-free
>> lists are too much, especially if with the complicated first/next mechanism
>> of TCG's chained block lists.
> 
> Direct jump handling code is pretty isolated and self-contained. It
> would require to back out of tb_remove_from_jmp_list() and sprinkle a
> couple of atomic_rcu_read()/atomic_rcu_set() with some comments, I
> think. Maybe it could be easier to justify looking at actual patches.