Re: [PATCH] virtio-blk: allow toggling host cache between writebackand writethrough

[Paolo Bonzini]

Il 05/07/2012 20:39, Badari Pulavarty ha scritto:
> On Tue, 2012-07-03 at 15:19 +0200, Paolo Bonzini wrote:
>> This patch adds support for the new VIRTIO_BLK_F_CONFIG_WCE feature,
>> which exposes the cache mode in the configuration space and lets the
>> driver modify it.  The cache mode is exposed via sysfs.
>>
>> Even if the host does not support the new feature, the cache mode is
>> visible (thanks to the existing VIRTIO_BLK_F_WCE), but not modifiable.
>>
>> Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
>> ---
> 
> Hi Paolo,
> 
> Curious - What is the host side change to support this ? QEMU would
> close and re-open the device/file with the corresponding flags
> (O_SYNC) ?

QEMU is not using O_SYNC anymore; instead, it manually issues a flush
after each write.  We found this didn't penalize performance (in fact,
for qcow2 metadata writes the fdatasyncs will be more coarse and there
is a small improvement).  So, when you toggle writethrough to writeback
QEMU simply has to stop forcing flushes, and vice versa if you go to
writethrough.

> And also, is there a way to expose cache=none (O_DIRECT) to the guest ?

Not yet.  The main problem is that while we can invent something in
virtio-blk, I'm not sure if there is an equivalent in SCSI for example.

Paolo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/