[Qemu-devel] Re: [PATCH 1/4] block: clarify the meaning of BDRV_O_NOCACHE

[Stefan Hajnoczi]

On Thu, Mar 17, 2011 at 9:07 AM, Kevin Wolf <address@hidden> wrote:
> Am 16.03.2011 18:00, schrieb Stefan Hajnoczi:
>> On Wed, Mar 16, 2011 at 2:08 PM, Christoph Hellwig <address@hidden> wrote:
>>> On Wed, Mar 16, 2011 at 09:42:37AM +0000, Stefan Hajnoczi wrote:
>>>> -    writethrough = ((flags & BDRV_O_CACHE_MASK) == 0);
>>>> +    writethrough = ((flags & (BDRV_O_CACHE_WB | BDRV_O_NO_FLUSH)) == 0);
>>>
>>> or rather
>>>
>>>        writethrough = ((flags & (BDRV_O_CACHE_WB) != );
>>>
>>> but yes, this code had sneaked in since my initial version.
>>
>> My intention was that if we don't care about honoring flushes then we
>> might as well use Qcow2Cache.  But yes, just checking for cache mode
>> is the clearest.
>
> You mean for a possible writethrough mode with BDRV_O_NO_FLUSH set? Such
> a mode doesn't make a lot of sense to me.

Yeah, I guess you're right, we can never have BDRV_O_NO_FLUSH without
BDRV_O_CACHE_WB today.  I wasn't thinking end-to-end, just looking at
the BDRV_O_* flag bits.

Stefan