linux-kernel - Re: [PATCH] lis3lv02d: add axes knowledge of HP Pavilion dv5 models

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:	Wed, 11 Feb 2009 00:17:21 +0100
From:	Éric Piel <Eric.Piel@...mplin-utc.net>
To:	Giuseppe Bilotta <giuseppe.bilotta@...il.com>
Cc:	Pavel Machek <pavel@...e.cz>, linux-kernel@...r.kernel.org,
	Palatis Tseng <palatis@...il.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] lis3lv02d: add axes knowledge of HP Pavilion dv5 models

Giuseppe Bilotta schreef:
> On Tue, Feb 10, 2009 at 10:39 PM, Giuseppe Bilotta
> <giuseppe.bilotta@...il.com> wrote:
>> On Tue, Feb 10, 2009 at 11:10 AM, Pavel Machek <pavel@...e.cz> wrote:
>>> Yeah, that patch. Not sure what to do next... you could grab lis3
>>> manual and see if you can setup chip by hand to get better results...
>> As Eric pointed out, it looks definitely like an endianness problem,
>> although there are a few strange things which are happening at the
>> same time. Is the lis3 manual available online somewhere, or can I
>> request it to the mfgr?
> 
> 'k, found it on the STMicroelectronics website. Reading the doc about
> LIS3LV02DL I see that WHO_AM_I should return 3Ah, but my sensor
> returns 3Bh. I'll see if I find the correct datasheet for this one
> instead, it might have info about the differences.
Yeah, we accept also 3Bh for the who_am_i (see in hp_accel.c). The 
reporter who had a laptop with such chip never reported any problem and 
I didn't pay attention enough to notice those chips have only 8bits 
precision.

> One thing that I noticed is that (modulo axis inversion) I'm able to
> use the sensor correctly if I set the thing to only use the high byte,
> totally discarding the lower byte:
> 	return *((s8*)(&hi));
> Maybe this sensor needs a different setup to return 12 instead of 8
> bits of information.
So you probably have a lis202dl (which has only two axes of 8bits each) 
or a LIS302DL (3 axes of 8 bits):
http://www.st.com/stonline/products/literature/ds/13624/lis202dl.htm
http://www.st.com/stonline/products/literature/ds/12726/lis302dl.htm

Unfortunately, the WHO_AM_I register returns 3Bh in both cases. I'd tend 
to imagine it's a three axes chip, but you can confirm that by just 
checking if Z has any meaningful value. You can then modify 
lis3lv02d_read_16() to only return the byte as low byte and the high 
byte as FFh if the high bit of the value is set (because I guess they 
are signed integers). You also have to change MDPS_MAX_VAL to 128, to 
get things completely right.

Once we have this sorted, we should add a flag to know if the current 
device is 8 bits or 12 bits, depending on the value of who_am_i. This 
would allow the driver to work fine with both type of chip.

Eric
--
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/