On 11/11/19 11:27 AM, Eric Biggers wrote: ... >>>> The same leak happens later in the function too. How about this instead: >>> >>> whoops yes it does. I kind of hate "retval = command_usage" but seeing the >>> memset of the key on the way out it's probably prudent to have one common >>> exit point after the function gets started. >>> >>> Care to send this as a formal patch? >> >> <interprets silence as a "no"> ;) >> >> I'll just incorporate your fixes as an addendum to my patch, then. >> >> -Eric > > Sorry, I didn't receive this because I was dropped from Cc, and I'm not > currently subscribed to linux-xfs. The patch you committed looks fine, thanks. Oh, I'm sorry about that, my mistake. Thanks, -Eric
