qemu-devel
[Top][All Lists]
Advanced

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

Re: [Qemu-devel] [PATCH] sheepdog: implement direct write semantics


From: Paolo Bonzini
Subject: Re: [Qemu-devel] [PATCH] sheepdog: implement direct write semantics
Date: Fri, 11 Jan 2013 04:34:13 -0500 (EST)


----- Messaggio originale -----
> Da: "Liu Yuan" <address@hidden>
> A: "Paolo Bonzini" <address@hidden>
> Cc: "Kevin Wolf" <address@hidden>, "Stefan Hajnoczi" <address@hidden>, 
> address@hidden, "MORITA Kazutaka"
> <address@hidden>
> Inviato: Venerdì, 11 gennaio 2013 10:04:23
> Oggetto: Re: [Qemu-devel] [PATCH] sheepdog: implement direct write semantics
> 
> On 01/11/2013 05:00 PM, Paolo Bonzini wrote:
> > That's not correct.  Try hdparm with an IDE disk, or sysfs with a
> > SCSI disk (recent kernels will also support sysfs with an IDE or
> > virtio-blk disk) and you'll see that it changes.
> 
> Okay, at least at startup, even with cache=writehtough set, I saw
> bdrv_enable_write_cache() returns true from sd_open(). So you mean
> this will be reset later (after sd_open())?

It is always true for the protocol.  It is not always true for the format.

This is correct.  qcow2 can do some writes in writeback mode, but they
will always be followed by a flush before the write is reported to the
guest.  It is a bit faster, and does not lose any correctness.

Paolo



reply via email to

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