linux-kernel - Re: 2.6.20-rc6 pb

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, 29 Jan 2007 10:55:48 +0100 (CET)
From:	Jiri Kosina <jikos@...os.cz>
To:	Soeren Sonnenburg <kernel@....de>
cc:	Linux Kernel <linux-kernel@...r.kernel.org>,
	linux-usb-devel <linux-usb-devel@...ts.sourceforge.net>
Subject: Re: 2.6.20-rc6 pb_fnmode regression

On Sat, 27 Jan 2007, Soeren Sonnenburg wrote:

> I realized that any setting to /sys/module/usbhid/parameters/pb_fnmode 
> is just ignored until the machine does a suspend-resume cycle.

Hi Soeren,

I would probably not call this a regression, as this has been always 
there, as far as I can see.

> I've added a printk in drivers/usb/input/hid-core.c (which is the only 
> place where hid->pb_fnmode is set) and indeed only on module load ( in 
> my case usbhid is compiled into the kernel - so on kernel boot) any 
> change to hid>pb_fnmode is done. Adding a printk to hidinput_pb_event() 
> in drivers/hid/hid-input.c says the same: hid->pb_fnmode cannot be 
> changed on the fly anymore... HOWEVER: If I s2ram the machine 
> hid->pb_fnmode is initialized with the value I put into 
> /sys/module/usbhid/parameters/pb_fnmode . As I have no idea how/whether 
> sysfs works/is possible I now hope someone more knowledgable than me can 
> resolve this issue!

Changing module parameter values through sysfs is not a very nice idea, 
because the change of the value is indeed silent - the driver is not 
notified in any way, that the value has changed. So the driver should take 
care of it by itself, which is not a nice thing.

The fact that during suspend/resume cycle it works is caused by the fact 
that all the hid devices are reinitialized, and therefore the 
hid->pb_fnmode is reassigned a new value, which has eventually been 
changed through sysfs.

I would rather be inclined to just make the 
/sys/module/usbhid/parameters/pb_fnmode read-only (which is what most of 
the drivers do anyway), to avoid this kind of confusion.

Do you have really any strong use-case, when setting the parameter during 
runtime would be much more useful than just do it during modprobe or 
rmmod/modprobe cycle?

Thanks,

-- 
Jiri Kosina
-
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/