Date: Fri, 12 Jul 2002 02:40:21 -0700 (PDT) From: Don Lewis <dl@catspoiler.org> To: hsu@FreeBSD.org Cc: dl-freebsd@catspoiler.org, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/netinet udp_usrreq.c Message-ID: <200207120940.g6C9eLwr013336@gw.catspoiler.org> In-Reply-To: <0GZ400HU8PNLLG@mta7.pltn13.pbi.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12 Jul, Jeffrey Hsu wrote: > > Comparing udp_getcred() with tcp_getcred(), the missing lock looked like > > an oversight. > > Okay, I see the confusion. I had the same question when I reviewed Jennifer's > patch, but decided to err on the side of conservatism and allow it. I actually > believe that lock is unnecessary in tcp_getcred(). > > > Isn't it possible for inp->inp_socket to get modified by > > another thread in either case unless inp is locked? > > Unless I missed a code path, I don't see how that can happen. Yeah, you may be correct. I was worried about udp_detach()->in_pcbdetach() path, but that looks like it is protected by the INP_INFO lock. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200207120940.g6C9eLwr013336>