On Tue, Jun 11, 2024, at 8:45 AM, Yaroslav Isakov wrote: > Hello, Mark! > вт, 11 июн. 2024 г. в 14:39, Mark Pearson <mpearson@xxxxxxxxx>: >> >> Hi >> >> On Wed, Mar 20, 2024, at 2:38 PM, Yaroslav Isakov wrote: >> > Hello, Mark! >> > >> > пн, 18 мар. 2024 г. в 20:42, Mark Pearson <mpearson@xxxxxxxxx>: >> >> >> >> >> >> >> >> On Mon, Mar 18, 2024, at 3:22 PM, Yaroslav Isakov wrote: >> >> > Hello, Mark! >> >> > >> >> > пн, 18 мар. 2024 г. в 19:48, Mark Pearson <mpearson@xxxxxxxxx>: >> >> >> >> >> >> On Thu, Feb 22, 2024, at 9:30 AM, Mark Pearson wrote: >> >> >> > Hi, >> >> >> > >> >> >> Just to confirm that I've tested a trial BIOS from the FW team and it fixes the issue (shows up under /sys/class/typec >> >> > Great to hear, thank you! >> >> > >> >> >> If there's anything in particular you'd like me to confirm let me know. >> >> > >> >> > If /sys/class/typec is populated, I think it should be enough... Just >> >> > to double check, is /sys/class/usb_power_delivery populated, too, now? >> >> >> >> Yes. From my system: >> >> >> >> # ls /sys/class/typec >> >> port0 port0-partner port1 >> >> # ls /sys/class/usb_power_delivery >> >> pd0 pd1 pd2 >> >> >> > >> > Great! These are the ones I saw with my workaround, too. >> >> > >> >> > I also did not manage to see anything in /sys/class/usb_role or >> >> > /sys/class/typec_mux, even with my hack - is it expected on AMD, >> >> > because of lack of appropriate mux/role switch drivers? >> >> >> >> I'm not seeing anything under those with this BIOS either >> >> >> >> I checked on an Intel platform on my desk (P14s G3) and those aren't populated there either...so not sure this is AMD specific. >> >> I'm being lazy and not looking at the kernel code - what would we expect to see in those entries? >> > >> > According to >> > https://www.kernel.org/doc/Documentation/ABI/testing/sysfs-class-usb_role, >> > with /sys/class/usb_role, we can switch between host and device modes. >> > My main interest there is typec_displayport driver - to be able to see >> > altmode info (not sure, though, in which /sys directory). With my >> > workaround, I failed to see this driver detecting my dock, so, I was >> > wondering, if it's caused by wrong UCSI version reported, or some >> > other driver missing (DP altmode is working with the doc, just no info >> > from typec_displayport driver) >> > >> > BTW, what is the version of ucsi reported from firmware? I wonder, if >> > it will work with this change >> > https://lore.kernel.org/linux-usb/20240209223824.622869-1-abhishekpandit@xxxxxxxxxxxx/ >> > >> Just as a note - BIOS 1.36 was released with the fix. Available on LVFS. > I tried 1.36 (R2FET56W) when it was released, but it did not fix my > issue. I have it installed, for sure: > sudo dmidecode | grep 1.36 > Version: R2FET56W (1.36 ) > but kernel still sees result of ret = ucsi->ops->read(ucsi, > UCSI_VERSION, &ucsi->version, sizeof(ucsi->version)); as 0 > > I also have latest EC firmware, too (corresponding to BIOS version) > I'll try to reset BIOS to default setting, to check if it will help. Strange - let me check. I did test the trial FW on my system but haven't updated to the release version yet. Thanks for letting me know
