Date: Sat, 11 Apr 2009 22:03:58 +0100 From: Andrew Brampton <brampton+freebsd-hackers@gmail.com> To: freebsd-hackers@freebsd.org Cc: Robert Watson <rwatson@freebsd.org> Subject: Re: FreeBSD memguard + spinlocks Message-ID: <d41814900904111403y6732cdfeme24904e8d1ed735a@mail.gmail.com> In-Reply-To: <alpine.BSF.2.00.0904111607410.19879@fledge.watson.org> References: <d41814900904101616n73c30e05t3c56a6cd8d9b6c0@mail.gmail.com> <alpine.BSF.2.00.0904111607410.19879@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
2009/4/11 Robert Watson <rwatson@freebsd.org>: > On Sat, 11 Apr 2009, Andrew Brampton wrote: > > Your understanding is mostly right. =C2=A0The missing bit is this: there = are two > kinds of interrupt contexts -- fast/filter interrupt handlers, which borr= ow > the stack and execution context of the kernel thread they preempt, and > interrupt threads, which get their own complete thread context. > > Fast interrupt handlers are allowed unlock to acquire spinlocks so as to > avoid deadlock because of the borrowed context. =C2=A0This means they can= 't > perform any sort of sleep, or acquire any locks that might sleep, since t= he > thread they've preempted may hold conflicting locks, or be the one that > would have woken up the sleep that the handler performed. =C2=A0Almost no= code > will run in fast handlers -- perhaps checking some device registers, doin= g > work on a lockless or spinlock-protected queue, and waking up a worker > thread. > > This is why, BTW, spin locks disable interrupt: they need to control > preemption by other interrupt handlers to avoid deadlock, but they are no= t > intended for use except when either in the scheduler, in a few related IP= I > contexts, or when synchronizing between normal kernel code and a fast > handler. > > Full interrupt thread contexts are permitted to perform short lock sleeps= , > such as those performed when contending default mutexes, rwlocks, and > rmlocks. They are permitted to invoke kernel services such as malloc(9), > UMA(9), the network stack, etc, as long as they use M_NOWAIT and don't > invoke msleep(9) or similar unbounded sleeps -- again to avoid the > possibility of deadlocks, since you don't want an interrupt thread sleepi= ng > waiting for an event that only it can satisfy. > > So the first question, really, is whether you are or mean to be using > fast/filter interrupt handler. =C2=A0Device drivers will never call memor= y > allocation, free, etc, from there, but will defer it to an ithread using = the > filter mechanism in 8.x, or to a task queue or other worker in 7.x and > earlier. =C2=A0If you're using a regular INTR_MPSAFE ithread, you should = be able > to use only default mutexes (a single atomic operation if uncontended) > without disabling interrupts, etc. > > Robert N M Watson > Computer Laboratory > University of Cambridge > Thanks very much for your detailed reply. I'm slowly understanding how everything in FreeBSD fits together, and I appreciate your help. I've been given a project to take over, and all of the design decisions were made before I started working on it, thus I'm playing catch up. One of the decisions was to implement their own version of a spin lock, which literally looks something like this: lock_aquire() { critical_enter(); while (! lockHeld ) {} lockHeld++; } This was actually the code tripping up MemGuard, as it is inside a critical section, which MemGuard is unable to sleep within. This is all running inside a kthread_create thread (I'm unsure of the proper name of this type of thread). Anyway, that is why I also asked about a lighter weight spin lock (perhaps similar to this one). I tempted to replace this custom spinlock with the standard MTX_DEF, however I'm unsure of its impact. The custom spin lock seems quick and light to acquire, and it does not concern me that a interrupt can potentially interrupt the code. On a related note, if you change the lock in memguard to a MTX_SPIN, it panics the kernel during boot. So that is not an option :) I was only using memguard because I suspected memory being used after it was freed. However, I think I will either change my locks to MTX_DEF or live without memguard. I realise I've not really asked any questions, but I would be grateful for any insights anyone may have. Andrew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d41814900904111403y6732cdfeme24904e8d1ed735a>