From owner-freebsd-hackers Sun Jul 25 22:16:39 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 8DC2E1528A for ; Sun, 25 Jul 1999 22:16:37 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id WAA43000; Sun, 25 Jul 1999 22:15:29 -0700 (PDT) (envelope-from dillon) Date: Sun, 25 Jul 1999 22:15:29 -0700 (PDT) From: Matthew Dillon Message-Id: <199907260515.WAA43000@apollo.backplane.com> To: Bill Fenner Cc: papezik@pvt.net, hackers@FreeBSD.ORG Subject: Re: Squid - a bug in src/sys/kern/uipc_socket.c References: <37976C03.A4A797A7@pvt.net> <199907222056.NAA87639@apollo.backplane.com> <199907260154.SAA18221@windsor.research.att.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG :I wrote it in rev 1.41 and gave it to the squid folks; it turned out :to cause X to fail in unexplained ways so we reverted it. Then I added :PRUS_MORETOCOME in rev 1.50, which was supposed to have fixed the problem. :Let's please not put the hack back in; if PRUS_MORETOCOME is broken :let's fix it instead. : :Is this an observed problem on recent FreeBSD versions, or just something :read in the Squid FAQ? : : Bill Looking at the PRUS_MORETOCOME code again I think it does solve this particular problem, albeit in a somewhat more complex fashion. I can see why the original patch failed - it set the atomic flag unconditionally and blew two cases in the loop. In general, I think it would have been cleaner to solve this sort of thing at the higher level. That is, correcting the original patch rather then introducing the comparitively greater complexity of PRUS_MORETOCOME. But since we would no longer be fixing a 'bug' it's up to you as the author to decide which solution you want. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message