Date: Sun, 23 Jun 2002 23:13:11 +0300 From: Giorgos Keramidas <keramida@ceid.upatras.gr> To: Kris Kennaway <kris@obsecurity.org> Cc: John Polstra <jdp@FreeBSD.ORG>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/i386/i386 in_cksum.c src/sys/i386/include in_cksum.h Message-ID: <20020623201311.GB36974@hades.hell.gr> In-Reply-To: <20020622212912.A42681@xor.obsecurity.org> References: <200206222235.g5MMZrd58098@freefall.freebsd.org> <20020622212912.A42681@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2002-06-22 21:29 +0000, Kris Kennaway wrote: > On Sat, Jun 22, 2002 at 03:35:53PM -0700, John Polstra wrote: > > jdp 2002/06/22 15:35:53 PDT > > > > Modified files: > > sys/i386/i386 in_cksum.c > > sys/i386/include in_cksum.h > > Log: > > Fix several bugs in the i386 asm statements used to speed up Internet > > checksumming. These bugs could possibly cause bad code to be > > generated at elevated optimization levels. > > I'm pretty sure I've seen this bug in action with -O2 kernels (years > ago) - good catch! I've seen checksums work fine with -O but somehow fail (giving a checksum value of 0xfffe in tcpdump output) when -O2 or anything higher is used. Definitely a very good piece of work! - Giorgos 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?20020623201311.GB36974>