Date: Sat, 4 Jun 2011 18:33:59 -0400 From: Attilio Rao <attilio@freebsd.org> To: Nathan Whitehorn <nwhitehorn@freebsd.org> Cc: freebsd-current@freebsd.org, freebsd-stable@freebsd.org, Andriy Gapon <avg@freebsd.org> Subject: Re: [poll / rfc] kdb_stop_cpus Message-ID: <BANLkTinJHTW9UEh8F2WFmy65uhmete%2B1wQ@mail.gmail.com> In-Reply-To: <4DE8FD83.6030503@freebsd.org> References: <4DE8FA2E.4030202@FreeBSD.org> <4DE8FD83.6030503@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
2011/6/3 Nathan Whitehorn <nwhitehorn@freebsd.org>: > On 06/03/11 10:13, Andriy Gapon wrote: >> >> I wonder if anybody uses kdb_stop_cpus with non-default value. >> If, yes, I am very interested to learn about your usecase for it. >> >> I think that the default kdb behavior is the correct one, so it doesn't >> make sense >> to have a knob to turn on incorrect behavior. >> But I may be missing something obvious. >> >> The comment in the code doesn't really satisfy me: >> /* >> =C2=A0* Flag indicating whether or not to IPI the other CPUs to stop the= m on >> =C2=A0* entering the debugger. =C2=A0Sometimes, this will result in a de= adlock as >> =C2=A0* stop_cpus() waits for the other cpus to stop, so we allow it to = be >> =C2=A0* disabled. =C2=A0In order to maximize the chances of success, use= a hard >> =C2=A0* stop for that. >> =C2=A0*/ >> >> The hard stop should be sufficiently mighty. >> Yes, I am aware of supposedly extremely rare situations where a deadlock >> could >> happen even when using hard stop. =C2=A0But I'd rather fix that than hav= e this >> switch. >> >> Oh, the commit message (from 2004) explains it: >>> >>> Add a new sysctl, debug.kdb.stop_cpus, which controls whether or not we >>> attempt to IPI other cpus when entering the debugger in order to stop >>> them while in the debugger. =C2=A0The default remains to issue the stop= ; >>> however, that can result in a hang if another cpu has interrupts disabl= ed >>> and is spinning, since the IPI won't be received and the KDB will wait >>> indefinitely. =C2=A0We probably need to add a timeout, but this is a us= eful >>> stopgap in the mean time. >> >> But that was before we started using hard stop in this context (in 2009)= . > > Some non-x86 platforms (e.g. PPC) don't support real NMIs, and so this st= ill > applies. Well, if I get Andriy's proposal right, he just wants to trim off the possibility to not stop the CPUs on entering KDB. I'm not entirely sure why there is a sysctl for disabling that and I really don't want it. Note that the missing of the NMI/privileged Interrupt is not going to be a factor on this request, unless you are worried a lot by the easy deadlock that a normal stop operation may lead. If that is the case, I think that the upcoming work on skipping locking during KDB/panic entering is going to help a lot for this case. At that point removing the possibility to turn off CPU stopping will be a good idea, IMHO. Attilio --=20 Peace can only be achieved by understanding - A. Einstein
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BANLkTinJHTW9UEh8F2WFmy65uhmete%2B1wQ>