qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [RFC PATCH v7 15/21] replay: checkpoints


From: Paolo Bonzini
Subject: Re: [Qemu-devel] [RFC PATCH v7 15/21] replay: checkpoints
Date: Tue, 13 Jan 2015 15:53:01 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.3.0


On 13/01/2015 15:26, Pavel Dovgaluk wrote:
>> > Should you instead distinguish which TimerListGroup is being run?  Then
>> > main loop would checkpoint once for every TimerListGroup (which means
>> > twice).
> Then I'll have to introduce some kind of deterministic ID for them
> and new asynchronous event to be saved instead of common checkpoints.
> Because these calls can be invoked in any order.
> But I have no idea of how to make such a deterministic ID.

There is currently one TimerListGroup per iothread object, and two more.
 You can start by not supporting iothreads, or you can give a
deterministic ID to iothreads.  Thread pools are similar to
TimerListGroups, except there is only one global pool instead of two.

Paolo



reply via email to

[Prev in Thread] Current Thread [Next in Thread]