#asahi-dev · 2022-01-21 11:11
povikit will fail at gpiod_get, but that will do the right thing#asahi-dev · 2022-01-21 11:11
povikah right!#asahi-dev · 2022-01-21 11:11
povikthat will fail at gpiod_get, no?#asahi-dev · 2022-01-21 11:10
poviki like this#asahi-dev · 2022-01-21 11:10
povikseems it would work for suspend too#asahi-dev · 2022-01-21 11:09
povikwell if there's no other operation to suspending the codecs that may work too#asahi-dev · 2022-01-21 11:09
povikif we ignore that and add SHARED flag to the shutdown gpio, then we can register it with all codecs and the first to probe will power all up#asahi-dev · 2022-01-21 11:08
poviki meant the problem will be with suspend#asahi-dev · 2022-01-21 11:06
povik*can't#asahi-dev · 2022-01-21 11:06
povikgiven there are more codecs#asahi-dev · 2022-01-21 11:05
povikwe can even handle it being reset-gpio at the moment :-p#asahi-dev · 2022-01-21 11:04
povikreset-gpio has been there since the beginning#asahi-dev · 2022-01-21 11:01
povik(shutdown-gpio)#asahi-dev · 2022-01-21 11:00
povikhttps://lore.kernel.org/all/20200923132600.10652-1-dmurphy@ti.com/#asahi-dev · 2022-01-21 11:00
poviki don't see a cover letter for this, this is where it was added#asahi-dev · 2022-01-21 10:57
povikwhat is reset_gpio for, then?#asahi-dev · 2022-01-21 10:55
povikhttps://www.ti.com/lit/ds/symlink/tas2770.pdf#asahi-dev · 2022-01-21 10:55
povikthere's only one reset-like input on the chip, and that's SDZ#asahi-dev · 2022-01-21 10:55
povikhm, i don't know what's that about#asahi-dev · 2022-01-21 10:39
povikyeah, but they don't depend on each other, so there's no ordering#asahi-dev · 2022-01-21 10:36
povik:)#asahi-dev · 2022-01-21 10:33
povikhappens in some bindings, i read in rob's post#asahi-dev · 2022-01-21 10:33
povikbecause of how ACTIVE_LOW/ACTIVE_HIGH is declared for this pin, the logic is inverted#asahi-dev · 2022-01-21 10:33
povikthis deassert#asahi-dev · 2022-01-21 10:33
povik> gpiod_set_value_cansleep(tas2770->reset_gpio, 1);#asahi-dev · 2022-01-21 10:33
povikthis is assert#asahi-dev · 2022-01-21 10:33
povik> gpiod_set_value_cansleep(tas2770->reset_gpio, 0);#asahi-dev · 2022-01-21 10:32
povikhm?#asahi-dev · 2022-01-21 09:56
povikthis is: https://lore.kernel.org/lkml/20191030120440.3699-1-peter.ujfalusi@ti.com/#asahi-dev · 2022-01-21 09:56
poviksorry, that wasn't the root#asahi-dev · 2022-01-21 09:55
povik^ this didn't go in but has discussion of the available approaches#asahi-dev · 2022-01-21 09:55
povikhttps://lore.kernel.org/lkml/20191030120440.3699-3-peter.ujfalusi@ti.com/#asahi-dev · 2022-01-21 09:54
povikbut i don't particularly like it#asahi-dev · 2022-01-21 09:54
povikwell i can say the codecs depend on i2c which depends on pinctrl which sets the hogs at probe time#asahi-dev · 2022-01-21 09:50
povikthat doesn't help then#asahi-dev · 2022-01-21 09:50
povik:(#asahi-dev · 2022-01-21 09:47
povikmaybe i should just try it out#asahi-dev · 2022-01-21 09:46
povikthere's bindings/pinctrl/pincfg-node.yaml, which has the options i need, don't know where i can use them though#asahi-dev · 2022-01-21 09:37
povikcan it be shared between devices?#asahi-dev · 2022-01-21 09:37
povikcan i use that to set a level on a pin?#asahi-dev · 2022-01-21 09:37
poviklooking at that#asahi-dev · 2022-01-21 09:31
povikis there a way to make sure the hog is applied before the devices are probed?#asahi-dev · 2022-01-21 09:31
povikso there will probably need to be a gpio-hog on that#asahi-dev · 2022-01-21 09:31
povikthe speaker amp chips on macbooks share a reset line#asahi-dev · 2022-01-18 23:02
poviki guess i am missing yamllint or something and it silently skips that check#asahi-dev · 2022-01-18 23:02
povikdarn, the same as last time but i couldn't replicate it locally#asahi-dev · 2022-01-18 21:54
poviki hope it's not just that the bot sleeps#asahi-dev · 2022-01-18 21:54
povikthat was my greatest fear#asahi-dev · 2022-01-18 21:54
povikno complains from rob's binding bot on the nco series, i call this a success#asahi-dev · 2022-01-18 20:17
povikyeah, maybe you don't even need to touch the mca