On Tue, Apr 14, 2015 at 8:22 AM, Alexander <[email protected]> wrote:
> On 13 Apr 2015 at 20:51:21, Matthew Ahrens ([email protected]) wrote: > > > > On Mon, Apr 13, 2015 at 10:46 AM, Matthew Ahrens <[email protected]> > wrote: > >> >> >> On Mon, Apr 13, 2015 at 10:40 AM, Andriy Gapon <[email protected]> wrote: >> >>> >>> Matt, >>> >>> is it plausible that dsl_dir_hold(pool, "different-pool/fs", ...) could >>> ever be successful? >>> >> >> No, I don't think so. >> > > To elaborate -- besides verifying that, an assertion would also make it > clear to the reader that we know they are in the same pool. > >> >> > These assertions do not make any sense, as in first case pdd > <http://src.illumos.org/source/s?defs=pdd&project=illumos-gate>->dd_pool > <http://src.illumos.org/source/s?defs=dd_pool&project=illumos-gate> = dp > <http://src.illumos.org/source/s?defs=dp&project=illumos-gate> always, in > any scenario (I looked into this while debugging and Andriy also mentioned > this in redmine). > > I'm not sure I understand that argument. The point of assertions is to declare things that are always true in any scenario. But I'm not going to press the issue. We can rely on the error checking in dsl_dataset/dir_hold(). Code looks good to me, please RTI it. --matt
_______________________________________________ developer mailing list [email protected] http://lists.open-zfs.org/mailman/listinfo/developer
