#asahi-dev · 2023-09-28 15:56
jannegbut we will still need someone to look at j413/j415#asahi-dev · 2023-09-28 15:56
janneglina: I might look at it tonight and push everything to https://github.com/jannau/linux/tree/isp/t8112#asahi-dev · 2023-09-28 13:09
jannegj`ey: 2h 16min and doesn't crash anymore on m2#asahi-dev · 2023-09-28 11:04
jannegseems to either reuse object files or some caching#asahi-dev · 2023-09-28 11:04
jannegsigh, that was just the headless renderer, now compiling chrome with 57239 build targets#asahi-dev · 2023-09-28 10:52
jannegchromium build takes a while: 37100/41210 and the studio's case is warm. nvme reports 64ºC#asahi-dev · 2023-09-28 09:05
jannegeiln: no linux traces on t6001 (12.3), I just ran it quickly to check what the tracer is supposed to output. I'll grab traces later if nobody beats me to it#asahi-dev · 2023-09-28 07:50
janneglooks like I have to patch fedora's chromium spec to get rid of the bti issues. compare https://www.reddit.com/r/AsahiLinux/comments/16u9fnb/chromium_crashes_with_sigill_on_asahi_fedora_39/#asahi-dev · 2023-09-28 06:29
jannegmy t8112 changes are in https://github.com/jannau/linux/tree/isp/t8112#asahi-dev · 2023-09-28 06:24
jannegt8112 failure is in https://gist.github.com/jannau/139b7599dbae683e41b5dc12aa3ad889#asahi-dev · 2023-09-28 05:52
jannegfor m1n1 we want https://paste.centos.org/view/faa37f06 squashed into lina's changes#asahi-dev · 2023-09-28 05:41
jannegwasn't working on t6001 either. I haven't looked at it, was a test to see what the tracer is supposed to output#asahi-dev · 2023-09-28 05:39
jannegnot yet working, I get an piodma irq error at the point where it should start#asahi-dev · 2023-09-28 05:37
jannegmarcan: t8112 has a different register layout, offset between ISP_MBOX_IRQ_INTERRUPT and ISP_MBOX_IRQ_DOORBELL is 0x430 instead of 0x3f0 so we'll need another reg/resource in the devicetree#asahi-dev · 2023-09-27 16:22
jannegsee the Multi-planar formats section#asahi-dev · 2023-09-27 16:22
janneglina: https://www.linuxtv.org/downloads/v4l-dvb-apis-old/planar-apis.html#asahi-dev · 2023-09-27 08:11
jannegand then simply searched for the commit on lore.kernel.org#asahi-dev · 2023-09-27 08:10
jannegI started with the last asahi-6.3 tag and cherry-picked net/bluetooth commits that went into v6.4 until pairing broke. manual bisect so to speak#asahi-dev · 2023-09-27 07:38
jannegmy plan to have one of the affected users look at the bt issue unfortunately failed#asahi-dev · 2023-09-26 22:45
jannegI'll send the fix/regression mail out tomorrow#asahi-dev · 2023-09-26 22:45
jannegdo we know someone with bcm4377 and ble devices?#asahi-dev · 2023-09-26 22:44
janneghttps://github.com/AsahiLinux/linux/pull/194#asahi-dev · 2023-09-26 22:07
jannegah, there is already https://lore.kernel.org/all/20230823194128.3893244-1-luiz.dentz@gmail.com/#asahi-dev · 2023-09-26 22:01
jannegHCI_LE_PHY_CODED seems to be either not supported or broken#asahi-dev · 2023-09-26 21:56
jannegno, I just looked at the wrong lines after the kernel already broke this#asahi-dev · 2023-09-26 21:53
jannegaccording to my printfs it shouldn't make a difference#asahi-dev · 2023-09-26 21:52
jannegwth, reverting 288c90224eec55 fixes the ble pairing issues on bcm4387 but I don't see why#asahi-dev · 2023-09-25 20:10
jannegthis has become off-topic here quite a while ago let's move to #asahi#asahi-dev · 2023-09-25 20:05
jannegyour problem sounds like pcie is completely broken#asahi-dev · 2023-09-25 20:03
jannegsounds like a kernel config or wrong binary issue on your side#asahi-dev · 2023-09-25 19:59
jannegPCIE_APPLE? any pcie messages in kernel log?#asahi-dev · 2023-09-25 19:57
janneghave you build a 4k or 16k kernel? see CONFIG_ARM64_PAGE_SHIFT#asahi-dev · 2023-09-25 19:55
jannegthat should have fw handling, you could unpack $ESP/vendorfw/firmware.{tar.cpio} to /lib/firmware/vendor/#asahi-dev · 2023-09-25 19:47
jannegwhere is your intrd comming from? is it based on the fedora initrd on the system? You need hooks in there to load and install the firmware in /lib/firmware/vendor. see https://github.com/AsahiLinux/asahi-scripts/tree/main/dracut/modules.d/99asahi-firmware#asahi-dev · 2023-09-25 19:28
jannegthat sounds more like it's things controlled by your initrd and boot kernel cmdline parameters#asahi-dev · 2023-09-25 19:18
jannegregarding the black screen in the beginning. is it possible that your display is slow to show I picture after initializing/configuring it? I have one as well and booting is so fast that the display only shows a picture when linux is already booting#asahi-dev · 2023-09-25 19:16
jannegthe device tree was broken, #192 was due to other oddities in the devicetree not necessary but is now. I can simulate the setup on a m2 pro mini#asahi-dev · 2023-09-25 18:31
janneg`ls -d /sys/devices/platform/soc*/*iommu`#asahi-dev · 2023-09-25 18:30
jannegare you 100% sure you're booting the correct kernel? do you have /sys/module/nvme_apple/ (should be even there if built-in)#asahi-dev · 2023-09-25 18:23
jannegdo you have APPLE_DART built-in? missing log messages for that, not related to nvme though#asahi-dev · 2023-09-25 18:21
jannegleio: pushed changes #180 to fix hdmi turning off without pd_ignore_unused#asahi-dev · 2023-09-25 14:42
jannegother nvme configs? NVME_CORE and BLK_DEV_NVME? the first is probably a dependency#asahi-dev · 2023-09-25 14:26
jannegAre all of NVME_APPLE, APPLE_MBOX, APPLE_RTKIT, APPLE_SART built-in?#asahi-dev · 2023-09-25 14:23
jannegis there anything matching nvme-apple or nvme0 inthe kernel log?#asahi-dev · 2023-09-25 14:21
jannegyou have to mount debugfs yourself in the emergency shell#asahi-dev · 2023-09-25 14:20
jannegis there anything in /sys/kernel/debug/devices_deferred ?#asahi-dev · 2023-09-25 14:16
jannegthat's strange, does it contain anything at all? it should also match on "nvmem"#asahi-dev · 2023-09-25 14:12
jannegleio: there should be a soc@22000.../nvme@3... directory#asahi-dev · 2023-09-25 11:31
jannegthere is https://github.com/alsa-project/alsa-tests/tree/master/python/ucm-validator but after seeing that the alsa-ucm-config repo fails validation I didn't bother looking at it#asahi-dev · 2023-09-25 11:28
jannegI detected the pitfall only via a typo. I deleted random things and it started working aafter I typoed "spk" as "skp"