From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Holler Date: Sun, 24 May 2015 20:03:12 +0200 Subject: [U-Boot] [PATCH v2] wandboard: Add board revision detection support In-Reply-To: <55620A81.6030700@ahsoftware.de> References: <1432247045-17518-1-git-send-email-festevam@gmail.com> <87siaob018.fsf@aikidev.net> <5560AA8A.8010601@ahsoftware.de> <55618121.4020300@denx.de> <55620A81.6030700@ahsoftware.de> Message-ID: <55621260.6070008@ahsoftware.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Am 24.05.2015 um 19:29 schrieb Alexander Holler: > Am 24.05.2015 um 16:27 schrieb Fabio Estevam: >> Hi Stefano and Alexander, >> >> On Sun, May 24, 2015 at 4:43 AM, Stefano Babic wrote: >> >>>> Wouldn't it be better to just enable CONFIG_CMD_GPIO and then change >>>> the >>>> boot-script in the config to something like "if gpio ..." instead of >>>> adding something special? >>>> >>>> Assuming the gpio command works on imx, which I haven't tested or >>>> looked >>>> up. >>> >>> gpio works - this is really a good idea, moving the check into the >>> script. Fabio, what do you mind ? >> >> I think the idea is good, thanks. >> >> I wanted to keep consistency with the mx6cuboxi implementation (which >> was based on TI's implementation suggested by Tom during the review of >> the mx6cuboxi patches). >> >> Also, the gpio script idea would work fine for selecting the dtb file, >> but not inside checkboard() function, where I print the board revision >> name. > > Printing the board revision in the script is as easy as selecting the dtb. > >> Other aspect I thought is the fact that in case we have another >> revision of the board in the future, I think that C code is more >> flexible for handling it. > > Not really. Then it it would need again a patch for the C source. Using > the gpio command one could just change the check even by just changing > uEnv.txt. Look at how long it now needed until someone did this patch > (your patch) for u-boot. > >> So I like the idea of gpio script, but I would prefer to keep the >> current implementation if possible due to the reasons stated above. > > I would suggest to change the stuff for mx6cuboxi to use the gpio > command too instead of taking the same (imho wrong) approach. > > But enough said from me, I don't really care. ;) Hmm, just one comment more. If the gpio command would be enabled, it would even be possible to reset the BRCM- WLAN and Bluetooth modules by just adding some stuff to uEnv.txt. So at least WLAN would reliable work even after a reboot or reset, without the need for the rfkill-driver. Also the wandboard-rfkill-driver has more advantages, e.g. the rfkill. ;) For those which don't know it, currently, without the wandboard-rfkill-driver, WLAN works only once after power up, but not after the reboot. Thats because the HW doesn't reset the brcm on reset, so after a reboot or reset uploading the firmware fails (because it is already uploaded due to the missing reset of the module). > Regards, > > Alexander Holler > _______________________________________________ > U-Boot mailing list > U-Boot at lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot