qemu-devel
[Top][All Lists]
Advanced

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

Re: [Qemu-devel] [V2 0/25] Async threading for VirtFS using glib threads


From: Stefan Hajnoczi
Subject: Re: [Qemu-devel] [V2 0/25] Async threading for VirtFS using glib threads & coroutines.
Date: Wed, 18 May 2011 11:39:59 +0100

On Tue, May 17, 2011 at 8:43 PM, Venkateswararao Jujjuri (JV)
<address@hidden> wrote:

I sent comments on individual patches.  This is turning out nice, many
of the PDU handlers are now straight-line code that can be followed
easily.

> o Redesigned to use bh as per Stefan's suggestion.
>  This made the code very simple but is little less performant compared to V1.
>  Anthony suggested to go-in with cleaner code and design (This version) and
>  deal with the performance later.
>  Just to put in perspective:
>  Sequential Writes of creating 1GB files using ffsb
>  o Write size 8k
>  With bh: 66.9MB/sec
>  Without bh (marshalling routines): 74.9 MB/sec
>
>  o Write size 128k
>  With bh: 117MB/sec
>  Without bh (marshalling routines): 122MB/sec

It would be interesting to hack in a post-yield callback which is
invoked after yield transfers control back to the calling coroutine.
This avoids setting up a BH and notifying the iothread to invoke it.
I'm just suggesting this as an experiment to see if the BH mechanism
causes the performance degradation, not as a real solution.

Stefan



reply via email to

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