linux-kernel - Re: [RESEND PATCH] iio: light: add support for TI's opt3001 light sensor

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: <20141002180553.GI7933@saruman>
Date:	Thu, 2 Oct 2014 13:05:53 -0500
From:	Felipe Balbi <balbi@...com>
To:	Felipe Balbi <balbi@...com>
CC:	Michael Welling <mwelling@...e.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Jonathan Cameron <jic23@...nel.org>,
	<linux-iio@...r.kernel.org>, <pmeerw@...erw.net>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [RESEND PATCH] iio: light: add support for TI's opt3001 light
 sensor

Hi,

On Tue, Sep 30, 2014 at 04:22:04PM -0500, Felipe Balbi wrote:
> On Mon, Sep 29, 2014 at 06:41:59PM -0500, Michael Welling wrote:
> > On Mon, Sep 29, 2014 at 05:46:38PM -0500, Felipe Balbi wrote:
> > > Hi,
> > > 
> > > On Mon, Sep 29, 2014 at 05:38:33PM -0500, Michael Welling wrote:
> > > > > > > Alright, this is already ridiculous. Andrew, if I resend the patch can
> > > > > > > you apply it since maintainer has been ignoring this thread anyway ?
> > > > > > > 
> > > > > > 
> > > > > > Do you reall think this is the best way to approach this?
> > > > > 
> > > > > when maintainer doesn't respond for weeks, yeah! Sure it is.
> > > > > 
> > > > > > Perhaps it would be better to explain what each field of the
> > > > > > configuration register does and then we can move on.
> > > > > 
> > > > > perhaps Jonathan could tell me exactly what he wants because I can't
> > > > > handle back-and-forth anymore. Specially when he complains about stuff
> > > > > he asked me to modify himself.
> > > > >
> > > > > > In particular the OPT3001_CONFIGURATION_L field needs to be explained
> > > > > > such that the ABI can be properly applied.
> > > > > > 
> > > > > > Looking at the docs for the Windows demo program the field is associated
> > > > > > with a latch configuration. What does this bit field actually do?
> > > > 
> > > > Still no technical information. Without all the facts how can you expect
> > > > him to tell you what he wants?
> > > 
> > > yeah, because clearly he doesn't know himself, right ?
> > 
> > Could you explain how it works to me then?
> 
> checking how much of the docs I can expose now, gimme a couple days.

alright, here's a snippet of what's on preliminary docs:

Latch field (read or write).
The latch field controls the functionality of the interrupt reporting
mechanisms: the INT pin, the flag high field (FH), and flag low field (FL).
This bit selects the reporting style between a latched window-style comparison
and a transparent hysteresis-style comparison.

0 = The device functions in transparent hysteresis-style comparison operation,
where the three interrupt reporting mechanisms directly reflect the comparison
of the result register with the high- and low-limit registers with no
user-controlled clearing event. See the Interrupt Operation, INT Pin, and
Interrupt Reporting Mechanisms section for further details.

1 = The device functions in latched window-style comparison operation, latching
the interrupt reporting mechanisms until a user-controlled clearing event.

[...]

7.4.2.2 Transparent Hysteresis-Style Comparison Mode
The transparent hysteresis-style comparison mode is typically used when a
single digital signal is desired that indicates whether the input light is
higher than or lower than a light level of interest. If the result register is
higher than the high-limit register for a consecutive number of events set by
the fault count field, the INT line is set to active, the flag high field is
set to 1, and the flag low field is set to 0. If the result register is lower
than the lowlimit register for a consecutive number of events set by the fault
count field, the INT line is set to inactive, the flag low field is set to 1,
and the flag high field is set to 0. The INT pin and flag high and flag low
fields do not change state with configuration reads and writes. The INT pin and
flag fields continually report the appropriate comparison of the light to the
low-limit and high-limit registers. The device does not respond to the SMBus
alert response protocol while in either of the two transparent comparison modes
(configuration register, latch field = 0).

-- 
balbi

Download attachment "signature.asc" of type "application/pgp-signature" (820 bytes)