[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140210170144.GO9987@twins.programming.kicks-ass.net>
Date: Mon, 10 Feb 2014 18:01:44 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: "Norton, Scott J" <scott.norton@...com>
Cc: "Long, Wai Man" <waiman.long@...com>,
"Low, Jason" <jason.low2@...com>,
"mingo@...hat.com" <mingo@...hat.com>,
"paulmck@...ux.vnet.ibm.com" <paulmck@...ux.vnet.ibm.com>,
"torvalds@...ux-foundation.org" <torvalds@...ux-foundation.org>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"riel@...hat.com" <riel@...hat.com>,
"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
"Bueso, Davidlohr" <davidlohr@...com>,
"hpa@...or.com" <hpa@...or.com>,
"andi@...stfloor.org" <andi@...stfloor.org>,
"Chandramouleeswaran, Aswin" <aswin@...com>,
"Vinod, Chegu" <chegu_vinod@...com>
Subject: Re: [RFC][PATCH v2 5/5] mutex: Give spinners a chance to
spin_on_owner if need_resched() triggered while queued
On Thu, Feb 06, 2014 at 08:10:02PM +0000, Norton, Scott J wrote:
> > How do people run this AIM7 piece of shit? I let it run for over an hour
> > and it generated exactly 0 numbers, it just sits there eating cpu-time
> > and creating a racket from my pantry.
>
> ./reaim -s100 -e2000 -t -j100 -i100 -f workfile.high_systime
>
> The reaim.config file contains:
>
> FILESIZE 10k
> POOLSIZE 1m
> DISKDIR /t0
> DISKDIR /t1
> DISKDIR /t2
> DISKDIR /t3
> DISKDIR /t4
> DISKDIR /t5
> DISKDIR /t6
> DISKDIR /t7
> DISKDIR /t8
> DISKDIR /t9
> DISKDIR /t10
> DISKDIR /t11
> DISKDIR /t12
> DISKDIR /t13
> DISKDIR /t14
> DISKDIR /t15
>
> The way Longman uses this is to create 16 ramdisk filesystems through
> /dev/ram* and then mount those filesystems to the /t* directories.
> Although you could run it through a regular filesystem also. It will use
> whatever you place in the reaim.config file as DISKDIR. You can specify
> one or more DISKDIR directories.
OK, and we're back to creating a racket but not producing useful
numbers; how long is that crap supposed to run before it gives a number?
Surely it can produce a useful number after a few minutes of runtime..
Letting it run for hours is just a waste of time and money.
Note that I'm running this on a WSM-EP with (2*6*2) 24 CPUs and 24G of ram.
--
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/
Powered by blists - more mailing lists