linux-kernel - Re: [RFC] [PATCH 1/2] binary stream format for /proc/stat

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: <CABEgKgrWaOgEOgmdcKPycWH8mQc35QzJ854hrXd+1CSk0kXH5Q@mail.gmail.com>
Date:	Sun, 1 Apr 2012 07:02:41 +0900
From:	Hiroyuki Kamezawa <kamezawa.hiroyuki@...il.com>
To:	Jan Engelhardt <jengelh@...ozas.de>
Cc:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Alexey Dobriyan <adobriyan@...il.com>,
	Ulrich Drepper <drepper@...il.com>
Subject: Re: [RFC] [PATCH 1/2] binary stream format for /proc/stat

2012年4月1日6:36 Jan Engelhardt <jengelh@...ozas.de>:
>
> On Saturday 2012-03-31 23:23, Hiroyuki Kamezawa wrote:
>>>
>>> Rather than adding more obscure formats and fields to /proc(/N)/stat, we
>>> should pursue descriptive lines like in /proc/N/status.
>>> Or, if so desired, sysfs-style splitted attributes where the filename
>>> takes on the description.
>>
>>Yes, I like format like /proc/N/status , /proc/meminfo...which is
>>readable to me.
>>But when I wrote this patch, I didn't want to break current format.
>>
>>My concern on sysfs-style one data per a file is that it will add many
>>open/close and making top/ps/sar slow.
>
> Or you go netlink. That gives you a binary format.
>
> That said however, binary does not always mean faster. (It's curious
> to see that generating and grepping through the text dump of the
> Windows registry is faster than letting regedit.exe search through
> the binary tree.)

I agree. I need to consider more.

I'm now thinking of adding some data filtering feature of /proc contents
may be helpful rather than adding complex format...then we can skip
some heavy calls if allowd.

Thanks,
-Kame
--
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/