linux-kernel - Re: [PATCH 8/8] cputime: Safely read cputime of full dynticks CPUs

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]

Date:	Thu, 31 Jan 2013 17:29:50 +0100
From:	Frederic Weisbecker <fweisbec@...il.com>
To:	sedat.dilek@...il.com
Cc:	Ingo Molnar <mingo@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Li Zhong <zhong@...ux.vnet.ibm.com>,
	Namhyung Kim <namhyung.kim@....com>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	Paul Gortmaker <paul.gortmaker@...driver.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH 8/8] cputime: Safely read cputime of full dynticks CPUs

2013/1/31 Sedat Dilek <sedat.dilek@...il.com>:
> I still dunno all parts of dynticks and their correlation.
> AFAICS they seem to be independent, can't say if it is "complete" with
> dynticks-cputime-only.

Dynticks cputime and dynticks printk are both standalone and
independant. In mainline, both printk and cputime accounting rely on
the tick. With these two patchsets we can now use these without the
tick.

But that's not enough to shut down the timer safely. There are still
some other subsystems that depend on the tick: posix cpu timers, perf
events, scheduler, RCU (although RCU is mostly handled now),
timekeeping....
Quite some of these subsystems are handled in my nohz tree
(https://lwn.net/Articles/533677/) which also contains the code that
actually shutdown the tick when possible.

>
> Was that dynticks stuff ever advertised for Linux-Next inclusion?

No, the code is not yet clean and complete enough for -next.

> Or wnt through any "mingo-next" tree?
> AFAICS, better chances these ways!

So indeed, I'm trying to get the things merged by Ingo. And for that I
try to cut my tree into individual features (tickless printk, tickless
cputime accounting, next step is probably posix cpu timers, ...).
Cutting things that way makes them much easier to review and merge. In
the meantime I continue to release my full dynticks tree (such as
3.8-rc4-nohz3): this is where I experiment the next remaining pieces
to handle in the future, it's mostly a draft.
--
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/