<div dir="ltr">Thanks Ben for the review, my reply below,<br><div class="gmail_extra"><br><br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>
><br>
> Bug #1240626<br>
><br>
> Signed-off-by: Alex Wang <<a href="mailto:alexw@nicira.com" target="_blank">alexw@nicira.com</a>><br>
<br>
</div>The new version of tunnel_check_status_change__() uses memset(),<br>
memcpy(), and memcmp() to work with strings. Is there some reason that<br>
one can't use the normal approach for strings, e.g. set byte 0 to '\0'<br>
to make it empty and strcmp() and ovs_strlcpy() to copy?<br></blockquote><div><br></div><div><br></div><div>There is no particular reason. I'll adjust.</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
netdev_vport_route_changed() doesn't appear to use the names in the<br>
shash, so it might make more sense for netdev_get_vports() to just<br>
return an array of pointers to netdevs instead of wasting the extra time<br>
and space on an shash.<br></blockquote><div><br></div><div><br></div><div>Okay, I can do that,</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
There seems to be an assumption that netdev_run() and netdev_wait() are<br>
called single-threaded. I think this might actually be true now, and<br>
the assumption was there before (more subtly), so it's probably not a<br>
big deal.<br></blockquote><div><br></div><div>Yes, there is. </div><div><br></div></div></div></div>