linux-kernel - Re: [PATCH -v2] use per cpu data for single cpu ipi calls

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.DEB.1.10.0901291311160.27527@gandalf.stny.rr.com>
Date:	Thu, 29 Jan 2009 13:11:52 -0500 (EST)
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
cc:	Peter Zijlstra <peterz@...radead.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Rusty Russell <rusty@...tcorp.com.au>, npiggin@...e.de,
	Ingo Molnar <mingo@...e.hu>,
	Thomas Gleixner <tglx@...utronix.de>,
	Arjan van de Ven <arjan@...radead.org>,
	jens.axboe@...cle.com
Subject: Re: [PATCH -v2] use per cpu data for single cpu ipi calls


On Thu, 29 Jan 2009, Linus Torvalds wrote:

> 
> 
> On Thu, 29 Jan 2009, Steven Rostedt wrote:
> > 
> > Actually, we are locking against the destination CPU.
> 
> Oh.
> 
> THAT'S JUST INCOMPETENT.

Or lack of sleep ;-)

> 
> What the *fuck* is the point of having per-CPU data, and then using it for 
> the wrong CPU?
> 
> Stop doing that idiocy. Put the per-cpu data on the senders side, and stop 
> the idiocy. You are going to get cross-CPU cacheline bouncing anyway, 
> there's no way to avoid it, but as long as you do it on the wrong CPU's 
> local data, you're missing the whole POINT of having per-cpu data in the 
> first place.
> 
> But yeah, that explains the locking. One stupid design mistake leads to 
> another.
> 
> And are you really sure it cannot be called from within interrupts? I'm 
> finding a lot of callers of smp_call_function_single(), and while I 
> couldn't find any that look like interrupts, I also couldn't find any 
> indication that it never happens.

No, the solution that Peter gave on top of mine looks like something we 
can all be happy with.

-- Steve

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/