06:24 <Mis012[m]> vkoul: would you have time to test the rest of the stuff?

07:06 <sibis> calebccff: if the rmtfs driver we do give access to modem ... and the first call in mpss is to take back control from modem

07:07 <sibis> calebccff: so remote would be false there

07:20 <sibis> calebccff: sry my bad we give control to modem so that mba can wipe it out in the first call ...

07:21 <sibis> calebccff: anyway looks like your mpss_region is off ... https://gitlab.com/sdm845-mainline/linux/-/commit/2b7bd430a4c9a229b11f14db0b49ed0a3ae2e2fe#3cd53727eae4137d1f02722e1c994e002ceee3a6_50_92 mpss region starts out at 0x8e400000 ... and modem start region doesn't change across fw but the size can so try correcting that

07:27 <sibis> calebccff: nvm looks like you already tried fixing it to 0x8e00000 but it should be 0x8e000000 instead

09:25 <calebccff> sibis: urgh the zero!! no way >,< thanks for catching that I'll give it a go

09:59 <calebccff> sibis: that got me further, now a later call to xfer_ownership fails... followed shortly by rmtfs causing a stall: https://p.calebs.dev/440bfa@raw

10:11 <sibis> calebccff: havn't seen a device in the wild that just rejects that smc call with that particular set of params... what was the last working kernel that pixel 3 was tested on?

10:12 <calebccff> sibis: 5.17 iirc, however I don't know if anyone had wifi working

10:36 <sibis> calebccff: Can't really find any valid diff between lnext and 5.17 that could cause this ...

10:37 <calebccff> sibis: welp, I had commented out one of the calls and didn't notice i hadn't reverted that change, sorry!

10:37 <calebccff> Modem boots now but immediately crashes, over and over again

10:38 <sibis> calebccff: lol ok .. crash signature?

10:38 <calebccff> sibis: "fatal error without message"

10:39 <calebccff> so, the best kind? aha

10:39 <calebccff> https://p.calebs.dev/f04201@raw

10:44 <sibis> calebccff: can you add a -v to rmtfs to see if there is any action on that front?

10:47 <calebccff> sibis: hmmm... [RMTFS] rmtfs shared memory not large enough for allocation request 0x200000 vs 0x1000

10:50 <calebccff> ah, think i see the issue

10:51 <calebccff> the rmtfs upper guard region had the wrong address

10:52 <calebccff> yay, modem boots

10:53 <sibis> -b