Date: Sun, 24 Jul 2005 18:59:40 -0400 From: Edwin <edwin@verolan.com> To: freebsd-hackers@freebsd.org Cc: Edwin <edwin@verolan.com>, Max Laier <max@love2party.net>, Simon 'corecode' Schubert <corecode@fs.ei.tum.de> Subject: Re: help w/panic under heavy load - 5.4 Message-ID: <20050724225940.GA28529@asx01.verolan.com> In-Reply-To: <200507242128.41197.max@love2party.net> References: <20050719034215.GB20752@asx01.verolan.com> <20050724141911.GA26240@asx01.verolan.com> <5482ba855efad71b135ae3108352ce39@fs.ei.tum.de> <200507242128.41197.max@love2party.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Max Laier (max@love2party.net) wrote: > > Edwin, what do you have for CFLAGS? Can you try to downgrade to "-O" for now > so that we have a better chance to get a full view? > Max, I have no CFLAGS or COPTFLAGS in /etc/make.conf - this was a basic kern-developer install on a blank PC. The only thing that's a little different about the box that i use to compile is that it's a dual processor machine - but no -j# options used in compilation of the kernel. the compile is proceding with the following as an example output from make/cc $ grep netinet /tmp/make.DEBUG1.output |grep fastfwd cc -c -O -pipe -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -fformat-extensions -std=c99 -g -nostdinc -I- -I. -I/usr/src/sys -I/usr/src/sys/contrib/dev/ acpica -I/usr/src/sys/contrib/altq -I/usr/src/sys/contrib/ipfilter -I/usr/src/sys/contrib/pf -I/usr/src/sys/contrib /dev/ath -I/usr/src/sys/contrib/dev/ath/freebsd -I/usr/src/sys/contrib/ngatm -I/usr/src/sys/dev/twa -D_KERNEL -incl ude opt_global.h -fno-common -finline-limit=8000 --param inline-unit-growth=100 --param large-function-growth=1000 -mno-align-long-strings -mpreferred-stack-boundary=2 -mno-mmx -mno-3dnow -mno-sse -mno-sse2 -ffreestanding -Werror /usr/src/sys/netinet/ip_fastfwd.c $ are you referring to the -fformat-extensions, -fno-common and -finline...etc optimizations as well? or just the -O v. -O2/-O3/-Os one? If yes to the -f* optimizations - besides commenting out parts of the makefiles - is there a 'normal' way to disable them? FWIW - I also had (I think) the same problem with the 5.3 release - but I never worked it out - just other things on my plate, so I don't believe it's a recent code change (ie. 5.4 timeframe) if it does turn out to be a code change. it also has something to do with the load on the box - I'm testing with small udp packets (using iperf) - if I step up the size - I have to step up the bandwidth in order to cause the panic.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050724225940.GA28529>