From: Guenter Roeck Date: Wed, 25 Feb 2015 04:18:07 +0000 Subject: Re: [lm-sensors] [PATCH] hwmon: (jc42) Add support for additional IDT temperature sensors Message-Id: <54ED4CFF.8080601@roeck-us.net> List-Id: References: <1423769172-13519-1-git-send-email-linux@roeck-us.net> In-Reply-To: <1423769172-13519-1-git-send-email-linux@roeck-us.net> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: lm-sensors@vger.kernel.org On 02/24/2015 02:19 PM, Jean Delvare wrote: > On Tue, 24 Feb 2015 10:14:04 -0800, Guenter Roeck wrote: >> On Mon, Feb 23, 2015 at 08:52:47PM +0100, Jean Delvare wrote: >>> On Thu, 12 Feb 2015 11:26:12 -0800, Guenter Roeck wrote: >>>> TS3000GB0 has a new device ID (0x2913). Since IDT's datasheets suggest >>>> that the upper 8 bit of the device ID reflect the chip ID and the lower >>>> 8 bit reflect the version number, modify the code to accept all chips >>>> with ID 0x29xx. >>> >>> You did that for all other IDT chips as well. While this is in >>> accordance with the datasheets, this might lead to some false >>> positives, as the driver probes 8 popular I2C addressed and the >>> detection code doesn't check for that many bits anymore. >>> >>> That being said I don't have a strong feeling about it. We can leave it >>> this way for now and revisit it later if false positives are actually >>> reported. >> >> We could check some other registers instead; the limit registers >> have five unused bits. Would it make sense to check those ? > > That makes sense, however every register being read adds up to the > probe time, so we have to be reasonable. Which is why I am generally > fine with leaving the code as is until actual false positives are > reported. But then again I am fine either way so if you prefer to check > the limit registers now, just do that. > I think I'll leave it as is for now. Guenter _______________________________________________ lm-sensors mailing list lm-sensors@lm-sensors.org http://lists.lm-sensors.org/mailman/listinfo/lm-sensors