On 10/24/2011 11:36 AM, Kevin Wolf wrote:
> You're changing the API and asking for possibly non-trivial changes in
> all protocol drivers, in order to accomodate semantics that all format
> drivers potentially could desire. So I wonder if the problem is simply
> that the current API is not expressive enough.
Can you think of any cases where a caller would want to invoke
bdrv_flush, but not bdrv_fsync? (The other way round it makes even less
sense)
I'm talking about the internal driver API only. The external API is
fine as is.
Paolo
