On Wed, Dec 9, 2020 at 1:16 PM H. Nikolaus Schaller <hns@xxxxxxxxxxxxx> wrote: > > This is also what made me wonder if that is really intended because then > the whole discussion about the cs-gpio-flags and inversion and the fixes > would not have been needed. The current code and fixes are all about > not ignoring the flags... The inversion you witnessed was a bug caused by spi client drivers that simply "plow over" the SPI_CS_HIGH mode flag. This includes the panel driver you're using, see: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/panel/panel-tpo-td028ttec1.c?h=v5.10-rc6#n337 You responded to this inversion bug by inverting your chip select, which made things work again. Now that the bug is gone, you can indicate the correct polarity in your devicetree, i.e. add 'spi-cs-high' for an active-high CS, and leave it out for an active-low CS. Your panel's CS is active-low, so it should not contain spi-cs-high. > Secondly, please imagine some reader of a device tree who finds > > cs-gpios = <&gpio 7 ACTIVE_LOW>; > spi-cs-high; That reader looks at the rules, sees that: - the ACTIVE_LOW is ignored, - presence of spi-cs-high means active-high and concludes this chip-select is active-high.
