[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v7 00/16] cpu-exec: Safe work in quiescent state
From: |
Emilio G. Cota |
Subject: |
Re: [Qemu-devel] [PATCH v7 00/16] cpu-exec: Safe work in quiescent state |
Date: |
Wed, 21 Sep 2016 13:29:54 -0400 |
User-agent: |
Mutt/1.5.23 (2014-03-12) |
On Mon, Sep 19, 2016 at 14:50:43 +0200, Paolo Bonzini wrote:
> In addition to fixing some of the issues found by Alex, safe work items
> need not run anymore with a mutex taken. Of course, cpu_exec_start/end
> and start_exclusive/end_exclusive are essentially the read and write
> side of a specialized rwlock, so there is still a lock in disguise looming
> to cause deadlocks; however, it does removes worries about recursive
> locking from CPU list manipulations.
>
> The new patches are 8, 12 and 13. Patch 12 of v6 has been split
> across patch 10 and patch 16.
I really like this patchset! Didn't have time to look at previous
iterations, but the use of the cpu_list_mutex to serialize safe
work is a great win--it simplifies things significantly.
BTW should we take this chance to RCU'ify the CPU list? This series
+ safe cpu_list traversals + cmpxchg changes = MTTCG user-mode would
be in very good shape. (full-system mode still needs work on the TLB
side, me thinks)
Thanks,
E.
- Re: [Qemu-devel] [PATCH 15/16] tcg: Make tb_flush() thread safe, (continued)
[Qemu-devel] [PATCH 16/16] cpus-common: lock-free fast path for cpu_exec_start/end, Paolo Bonzini, 2016/09/19
[Qemu-devel] [PATCH 14/16] cpus-common: Introduce async_safe_run_on_cpu(), Paolo Bonzini, 2016/09/19
Re: [Qemu-devel] [PATCH v7 00/16] cpu-exec: Safe work in quiescent state, no-reply, 2016/09/19
Re: [Qemu-devel] [PATCH v7 00/16] cpu-exec: Safe work in quiescent state, no-reply, 2016/09/19
Re: [Qemu-devel] [PATCH v7 00/16] cpu-exec: Safe work in quiescent state,
Emilio G. Cota <=