> On Dec 3, 2015, at 11:42 PM, Matthew Ahrens <[email protected]> wrote:
>
> libzfs_sendrecv.c: 2718-2719, this comment is inaccurate, you set all props
> every time.
Doh! Fixed the code, didn't fix the comments. I've now addressed that.
> Several of the other large comments you're adding seem wrong, please re-check
> them.
Okay. You mentioned the libzfs one above, the only other large one is the
preamble to extract_delay_props(). I'll revise that too.
> extract_delay_props(), rather than "dontbother", you could see if the nvlist
> is empty and if so destroy it.
Done. (Odd, nvlist_empty() isn't documented in libnvpair or its DDI/9f
equivalent. I should file a bug...)
> zfs_ioctl.c:4296, can you elaborate on what this has to do with errors?
Done.
> zfs_ioctl.c:4302, I think you mean "... AFTER we have merged the delayprops
> into props", but I don't understand why that is relevant.
Rewrote that entirely. That was an unfinished comment.
I've updated the primary webrev:
http://kebe.com/~danmcd/webrevs/4986/
AND added a post-Matt webrev with just these changes:
http://kebe.com/~danmcd/webrevs/4986-post-Matt/
Even though the actual code-changed is VERY small (losing "dontbother" and
using nvlist_empty()), I'll be rerunning zfstest as well.
Thanks!
Dan
_______________________________________________
developer mailing list
[email protected]
http://lists.open-zfs.org/mailman/listinfo/developer
