linux-kernel - Re: [PATCH] x86, kdump: Set crashkernel

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:	Mon, 11 Mar 2013 14:15:21 -0700
From:	Yinghai Lu <yinghai@...nel.org>
To:	Vivek Goyal <vgoyal@...hat.com>
Cc:	"Eric W. Biederman" <ebiederm@...ssion.com>,
	"H. Peter Anvin" <hpa@...or.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...nel.org>,
	WANG Chao <chaowang@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] x86, kdump: Set crashkernel_low automatically

On Mon, Mar 11, 2013 at 2:06 PM, Vivek Goyal <vgoyal@...hat.com> wrote:
> On Mon, Mar 11, 2013 at 01:57:41PM -0700, Yinghai Lu wrote:
>> On Mon, Mar 11, 2013 at 1:45 PM, Vivek Goyal <vgoyal@...hat.com> wrote:
>> > In my experience, trying to keep foot-print small has kind of been a
>> > losing battle.
>> >
>> > - People want more functionality in second kernel, want to dump to more
>> >   complicated IO stacks and that requires pulling in more drivers,
>> >   more libraries, more daemons, more user space tools and what not.
>> >
>> > - Now we use dracut generated initramfs and it has been growing in size.
>> >   Now systemd has been pulled in too.
>> >
>> > - Drivers keep on increasing their memory usage.
>>
>> If the dump file will be only put one place, why should all the drivers
>> for all devices get loaded?
>> for example: dump will be on disk with one scsi controller, can you only
>> load driver that contoller? and forget about all other storage controller and
>> network etc drivers.
>
> We do try to optimize things this way. Only include drivers as needed. But
> a single driver might be handling lots of cards and then try to bring
> up all the cards in second kernel.

Can you disable auto_probe at first? and then use /sys bind driver to specific
device.

>
> Just fixed some issues with one driver where around 40MB extra memory
> was benig consumed because of multiqueue support. Just bunch of allocation
> in driver. And we used kernel command line to disable it.

Yes that the is one problem, we should have some facility or helpers to reject
 those invalid request. and driver should reduce their foot print step by step.
for example, we only have 64M, but first driver want to alloc 32M, it
should be just rejected.
--
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/