Date: Sun, 19 Sep 2010 22:33:26 -0700 From: Randall Stewart <rrs@lakerest.net> To: Andriy Gapon <avg@freebsd.org> Cc: Lawrence Stewart <lstewart@freebsd.org>, FreeBSD Current <freebsd-current@freebsd.org> Subject: Re: Crash during boot of current (rev 212885) Message-ID: <5B9801DF-2DBB-4203-9274-706276C6A910@lakerest.net> In-Reply-To: <4C96E5D1.20906@freebsd.org> References: <EBB29842-99B9-4912-9E86-B5965F9C4201@lakerest.net> <4C96C772.4030006@freebsd.org> <29093DA9-BB4D-49B5-A20E-EDEDF6B5BAEC@lakerest.net> <4C96E5D1.20906@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Andrly: Ok.. I can do that. I can positively say that when I have a kernel with 212646.. all is well. But a kernel with 212647 crashes as described below... I will ship you the read-elf offlist R On Sep 19, 2010, at 9:40 PM, Andriy Gapon wrote: > on 20/09/2010 06:38 Randall Stewart said the following: >> Hi Lawrence: >> >> I am currently doing a binary search.. >> >> I know that 212660 shows the break. >> >> I am just about to try 212560 ;-) >> >> If that works I will update to 212646 and see if it works.. ;-) > > Randall, > > please also make sure that you have sufficiently recent ld as > described in > UPDATING from 20100915. > I'd be interested to see output of readelf -a -W for your kernel > that crashes. > > >> On Sep 19, 2010, at 7:31 PM, Lawrence Stewart wrote: >> >>> Hiya Randall! >>> >>> On 09/20/10 08:56, Randall Stewart wrote: >>>> Hey all: >>>> >>>> I am now seeing a crash when I boot my Intel (in 64 bit more)... >>>> >>>> Its very early in the boot process.. and thus no crash dump ;-0 >>>> >>>> Its in >>>> >>>> netisr_start_swi() >>>> >>>> When it initializes netisr_mtx with a mtx_init() it crashes saying >>>> that netisr_mtx is unaligned... (the address ddb shows for >>>> netisr_mtx ends >>>> with c ... so it definitely is unaligned... >>>> >>>> Looking at the netisr_workstream structure (where netisr_mtx is) it >>>> appears to be in theory aligned right (follows 2 pointers)... so >>>> did something change the DP_CPU Define stuff to cause us to get >>>> unaligned >>>> access? >>>> >>>> Just curious... If I don't hear from anyone I will start backing >>>> things >>>> out 1 >>>> rev at a time until I find what did it I guess ;-) >>> >>> My guess would be r212647. Try backing that rev out and if it fixes >>> things, hopefully Andriy will have some thoughts on how to fix the >>> problem. Apologies if my guess is a red herring. >>> >>> Cheers, >>> Lawrence >>> >> >> ------------------------------ >> Randall Stewart >> 803-317-4952 (cell) >> > > > -- > Andriy Gapon > _______________________________________________ > freebsd-current@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org > " > ------------------------------ Randall Stewart 803-317-4952 (cell)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5B9801DF-2DBB-4203-9274-706276C6A910>