#asahi-dev · 2023-12-02 11:12
sveni'm not really sure where it comes from yet#asahi-dev · 2023-12-02 11:12
sventhat would be n=4 then.#asahi-dev · 2023-12-02 11:03
svenLN_TXA_CAL_CTRL_BASE you mean?#asahi-dev · 2023-12-02 11:00
svenand I think the same n is used later for mask32(rx_shm + LN_AUSPMA_RX_SHM_TJ_RXA_SAVOS_CTRL16, LN_TX_CAL_CODE, FIELD_PREP(LN_TX_CAL_CODE, 6));#asahi-dev · 2023-12-02 11:00
svencould be related to the link rate somehow since we already know it at that point#asahi-dev · 2023-12-02 10:57
svenokay, so for me n = 6 always and you see 5 and 7#asahi-dev · 2023-12-02 10:56
svenI think it's ~(-0x1 << n) and n has always been 5 for me. no idea what exactly n is though#asahi-dev · 2023-12-02 10:47
sveni'm trying to figure out where that number comes from#asahi-dev · 2023-12-02 10:47
svenwas that for all your displays?#asahi-dev · 2023-12-02 10:47
svenjanneg: mask32(tx_shm + LN_AUSPMA_TX_SHM_TXA_IMP_REG0, LN_TXA_CAL_CTRL, FIELD_PREP(LN_TXA_CAL_CTRL, 0x3f)); <-- you've seen 0x1f there, haven't you?#asahi-dev · 2023-12-01 13:41
svenonce they are powered up we only care about plug/unplug events, cable orientation and alternate mode changes #asahi-dev · 2023-12-01 13:40
svensounds reasonable, the old tipd chips are mostly standalone with Apple’s firmware #asahi-dev · 2023-11-30 17:06
svennice work :-)#asahi-dev · 2023-11-30 16:51
sven*so#asahi-dev · 2023-11-30 16:51
svenlooks good do far!#asahi-dev · 2023-11-30 16:49
svenoh, you already told me all the interesting bits :)#asahi-dev · 2023-11-30 16:47
svenmaybe it’s part of the soc now#asahi-dev · 2023-11-30 16:43
svendevice mode should work then since it doesn’t need VBUS to be powered #asahi-dev · 2023-11-30 16:38
sven(unless they also changed the PHY)#asahi-dev · 2023-11-30 16:38
sventhere’s a decent chance that device mode is still going to work if you just ignore that thing #asahi-dev · 2023-11-30 16:37
svencan’t find anything public about sn20120x :/#asahi-dev · 2023-11-30 16:35
svenyeah, sounds like a new pd controller#asahi-dev · 2023-11-30 16:34
svenType C controller maybe #asahi-dev · 2023-11-30 16:33
svenSounds like they changed their usb pd controller chips #asahi-dev · 2023-11-30 16:32
svenweird#asahi-dev · 2023-11-30 16:31
svenCan you upload the adt somewhere? (note that iirc the wifi password and possibly other personal data is in there and you may want to remove those)#asahi-dev · 2023-11-30 16:27
svenhuh#asahi-dev · 2023-11-30 12:22
svenhah, nice. my day job is entirely unrelated to Linux so selling anything I do as relevant for them is probably impossible :D#asahi-dev · 2023-11-30 12:18
svendoesn’t help that $work actually wants me to do some work they pay me for as well ;)#asahi-dev · 2023-11-30 12:17
svenprobably *sigh*#asahi-dev · 2023-11-30 12:17
sven*is#asahi-dev · 2023-11-30 12:16
sveni‘ll probably wait for the soc with usb4gen2 or however that even faster thunderbolt thing Intel announced it called :)#asahi-dev · 2023-11-30 12:16
svenhah#asahi-dev · 2023-11-30 12:15
svenvery nice#asahi-dev · 2023-11-30 07:35
svenhah, nice! glad it works now :)#asahi-dev · 2023-11-29 18:30
svenfwiw, most European countries have laws that are somewhat similar to the dmca#asahi-dev · 2023-11-28 16:48
svenI‘m very curious about the root cause for this :)#asahi-dev · 2023-11-28 16:47
sventhe mmu should work with MMC left in whatever state iboot leaves it - at least for m1#asahi-dev · 2023-11-28 16:36
svenim pretty sure the memory allocation functions just do internal bookkeeping and ignore whatever the mmu does #asahi-dev · 2023-11-28 16:09
svencentral scrutinizer, hardware that talks usb pd to send those Apple VDM commands #asahi-dev · 2023-11-28 15:23
svenalso looked for missing barriers but we have a isb inbetween setting up tcr and enabling the mmu #asahi-dev · 2023-11-28 15:19
svenwhat happens if you just write the m1n1 value twice? does that also fix it? :D#asahi-dev · 2023-11-28 15:18
sven*gets#asahi-dev · 2023-11-28 15:18
sveni'm still very confused because we overwrite TCR anyway. what state is left that get cleared by first setting that one to zero?#asahi-dev · 2023-11-28 15:17
svenI think there are keys floating around for some iboots#asahi-dev · 2023-11-28 15:13
svendouble checked that we also overwrite tcr from m1n1 so the previous value really shouldn’t matter :/#asahi-dev · 2023-11-28 15:04
svenit’s still so weird that first writing zero to tcr makes everything work #asahi-dev · 2023-11-28 14:44
svenhm, okay, so probably nothing special then#asahi-dev · 2023-11-28 14:40
svenah, right#asahi-dev · 2023-11-28 14:39
svendo they write to TCR in the chicken bit sequence inside xnu? maybe there's some truly weird hw bug where they have to do that specifically ordered w.r.t some chicken bit