On 6 March 2013 05:20, Doug Anderson <dianders@xxxxxxxxxxxx> wrote: > Thomas, > > On Mon, Mar 4, 2013 at 6:04 AM, Thomas Abraham > <thomas.abraham@xxxxxxxxxx> wrote: >> >> Ok. I will repost this patch again with pinctrl_1 and pinctrl_2 >> included. I had not included this in the earlier patch since I was not >> sure of the best pin grouping for the camera and c2c interface. > > OK, thanks. I have added pinctrl_1 and pinctrl_2 now. > >>>> --> NOTE: Appears (IIRC) to have incorrect interrupt for pinctrl_3. I >>>> believe that 45 is pinctrl_1. Maybe 3 is 47? >>> >>> I can't verify this, unfortunately. >> >> The documentation does not state this clearly. I will recheck on this >> and correct as needed. > > OK. Let me know if you can't find it. I spent a bunch of time > digging in this area and wrote up my findings, but I certainly could > be wrong. ;) So now I have: pinctrl_0 is SPI[46], pinctrl_1 is SPI[45], pinctrl_2 is SPI[50] and pinctrl_3 is SPI[47]. I am yet to test this and confirm. If you feel these are wrong, could you please let me know. > >> I have posted i2c pinctrl support patch based on LinusW's pin grab by >> device core patch. If that is acceptable, I can post pinctrl support >> patches for other other controllers as well. > > OK, I see that. <https://patchwork.kernel.org/patch/2212731/>. That > particular patch isn't strictly required, right? Even without that > patch I can specify pinctrl for the i2c nodes and it will work > OK--this just removes the old gpio nodes. > > One problem I see with taking this approach for DW_MMC and for SPI is > that it will make bisecting hard. If you land a change to take out > GPIO stuff from dw_mmc and the spi driver then it will make the > transition difficult. exynos5 boards will stop booting until the > device tree support is also landed. ...and it would be an awfully big > patch to add all the device tree stuff in the same patch as the dw_mmc > and spi driver changes... Right, I did not pay attention to the bisecting problem. What I have now is, a set of individual patches that migrate i2c, sdhci, dwmmc, i2s, spi and keypad drivers to pinctrl based pin-configuration. In addition to that, two patches that add default pin-state information for Exynos4 and Exynos5 platforms. Since there are multiple drivers involved, is it okay to keep them as separate patches? Thanks, Thomas. > > > -Doug -- To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html