#asahi-dev · 2023-10-07 00:04
janneghttps://www.itu.int/rec/T-REC-H.264-200503-S/en is the last sane one, 343 pages#asahi-dev · 2023-10-07 00:01
jannegstart with historic versions without svc and mvc#asahi-dev · 2023-10-07 00:01
jannegdoo not read through the full itu h.264 spec#asahi-dev · 2023-10-07 00:00
jannegeiln: bed time reading through you avd notes#asahi-dev · 2023-10-06 22:46
janneghere is the fw version check. much code for a single check: https://github.com/AsahiLinux/linux/pull/203#asahi-dev · 2023-10-06 22:35
jannegfor macos#asahi-dev · 2023-10-06 22:34
jannegI hope you mean by software either the scaler or the gpu#asahi-dev · 2023-10-06 20:29
jannegkettenis: uncommenting '#define RTKIT_SYSLOG' in config.h makes dcp chatty#asahi-dev · 2023-10-06 20:25
janneglina: does macos use 1 on the m2 mb air?#asahi-dev · 2023-10-06 20:24
jannegeiln: image is better with the same temporal filter settings as macos, i.e. '.temporal_filter = 0,'#asahi-dev · 2023-10-06 20:23
jannegis it intentional that the webcam image is not mirrored by default? or are the applications supposed to handle this?#asahi-dev · 2023-10-06 19:07
jannegm2 with 12.4 firmware is fixed by using the 32-bit CISP_CMD_CH_SET_FILE_LOAD command#asahi-dev · 2023-10-06 10:41
jannegmarcan has or has access to one and I can get access to one as well. I can't remember seeing an user with an imac. the telemetry might tell us how many users with imacs we have#asahi-dev · 2023-10-06 10:02
jannegdo we know if the imac uses the same sensor as the m1 pro/max macbooks? it has a 1080p sensor#asahi-dev · 2023-10-06 10:01
janneglina: can you push the m2 tracer to a branch? I have a partial port with fixed mailbox/doorbell offsets but that wasn't enough#asahi-dev · 2023-10-06 09:58
jannegeiln: yes, very grainy#asahi-dev · 2023-10-06 08:20
jannegI doubt you've configured toolchain/portage configured to use BTI in the first place contrary to what fedora does#asahi-dev · 2023-10-06 07:02
jannegthe 720p sensor is even with calibration data very noisy. could that be caused by a wrong config?#asahi-dev · 2023-10-06 06:59
jannegeiln: calibration upload works with the 13.5 isp firmware so this seems to be a 12.4 incompatibility#asahi-dev · 2023-10-06 06:53
jannegtbm: forgot to git add, fixed. thanks#asahi-dev · 2023-10-06 06:49
jannegleio: https://github.com/AsahiLinux/m1n1/pull/338 makes 1st stage display init reliable for me#asahi-dev · 2023-10-05 22:25
jannegdoes that mean we got the wrong one? can we verify that with the tracer?#asahi-dev · 2023-10-05 22:24
jannegeiln: isp breaks on j493 (m2 macbook pro 13-inch) with calibration data: https://gist.github.com/jannau/9f87fd51f892106316c9e32340c27cae#asahi-dev · 2023-10-05 21:47
janneg'mdelay(100);' after pmgr_adt_power_enable(cfg->dptx_phy) in src/dcp.c fixes it for me#asahi-dev · 2023-10-05 21:26
janneglooks like there's somewhere a delay missing during the dptx setup. on j473 1st stage m1n1 fails to bring the display up#asahi-dev · 2023-10-05 20:31
jannegI think initially yes due to the mtp isses with 13.5. I think since then neither the alarm images nor the boot.bin bundle where refreshed#asahi-dev · 2023-10-05 18:14
jannegyes, needs 13.5. 13.3 will not work since Apple changed the dptx interface in 13.4 or 13.5#asahi-dev · 2023-10-05 15:55
jannegyes, but might need os support since it now depends on 2 power-domains#asahi-dev · 2023-10-05 15:54
jannegkettenis: boot framebuffer support on m2* desktop systems as well#asahi-dev · 2023-10-05 13:41
jannegmarcan: the dapf parsing? that was on v1.4.0, probably fixed now. at least caused no problems with the m1 ultra#asahi-dev · 2023-10-05 07:32
jannegdisplay init works with chainloading#asahi-dev · 2023-10-05 07:28
jannegthe adt dapf parsing is broken on m2 pro mini: https://gist.github.com/jannau/0bbe64c43dd3f4cd51d4fbeed314c3b8#asahi-dev · 2023-10-04 18:09
jannegI think I've seen only connects followed by immediate disconnects. ended up working after plugging the cable a few times#asahi-dev · 2023-10-04 14:52
jannegI wouldn't complain if we can merge the m1n1 m2* dptx support before linux#asahi-dev · 2023-10-04 14:51
jannegalarm builds are to old but barred due to the 13.x requirement, should work without issue with fedora#asahi-dev · 2023-10-04 14:50
janneglaso we could remove the expert mode flag from the 15-inch air. it's just a larger 13-inch#asahi-dev · 2023-10-04 14:49
jannegat least the touchbar suspend fix and the bluetooth fixes but taking other pull requests would make dealing with devicetree changes a little less annoying#asahi-dev · 2023-10-04 14:46
jannegmarcan: can you look into taking kernel fixes and prepare a new tag/release?#asahi-dev · 2023-10-04 07:14
jannegjonmasters: I have https://paste.debian.net/1293960/ in my testing tree but iirc only that fixes detection problems#asahi-dev · 2023-10-02 11:17
jannegkettenis: great, send patches for things which will simplify porting#asahi-dev · 2023-10-02 00:11
jannegprobably complained about that one#asahi-dev · 2023-10-02 00:11
jannegdevm_hwmon_device_register_with_info has still a "macsmc_hwmon" parameter#asahi-dev · 2023-10-02 00:09
jannegI see only complaints about the temp sensors nopt being attached top a thermal zone#asahi-dev · 2023-10-02 00:06
jannega comment over the driver name would be helpful#asahi-dev · 2023-10-02 00:05
jannegoh well#asahi-dev · 2023-10-02 00:04
jannegah, that'll break the pattern in smc_core.c#asahi-dev · 2023-10-02 00:01
jannegI tried to make macsmc_hwmon_populate_info() nicer buit gave up#asahi-dev · 2023-10-02 00:00
jannegI thought it would be a good idea to support adding keys dynamically through sysfs for simple user testing of keys but after seeing how hwmon is probed: no way#asahi-dev · 2023-10-01 23:56
jannegI don't think we can use keys found elsewhere without sanity checks#asahi-dev · 2023-10-01 23:55
jannegchadmed: https://github.com/jannau/linux/tree/macsmc-hwmon has some minor cleanup and shows what I meant with using .dtsi includes for the sensor keys