#asahi-dev · 2024-09-28 06:13
chadmed_if the tip sense interrupt has fired, wait a bit before detecting#asahi-dev · 2024-09-28 06:13
chadmed_im thinking of just adding a 1 to 2 second delay in the IRQ thread#asahi-dev · 2024-09-28 06:12
chadmed_yeah if the hotplug detect runs and returns before youve pushed the plug in all the way youre stuffed and you have to replug it #asahi-dev · 2024-09-28 06:06
chadmed_yeah msleep()s it is -.-#asahi-dev · 2024-09-28 06:03
chadmed_most of the realtek codecs etc will wait a very, very long time between the plug event and actually trying to figure out what type of thing you plugged in #asahi-dev · 2024-09-28 06:03
chadmed_unfortunately i think the pc as a platform has mostly the right idea here#asahi-dev · 2024-09-28 06:02
chadmed_the issue is between the tip sense interrupt and starting the detection routine#asahi-dev · 2024-09-28 06:02
chadmed_unfortunately making the debounce time longer on the actual tip sense doesnt help#asahi-dev · 2024-09-28 06:02
chadmed_see https://github.com/AsahiLinux/linux/issues/294#asahi-dev · 2024-09-28 06:01
chadmed_LarstiQ: theres enough room for error that people will hit it#asahi-dev · 2024-09-28 03:09
chadmed_you can reliably cause it to misdetect a headset by getting the plug timing juuuust wrong#asahi-dev · 2024-09-28 02:50
chadmed_or worse yet, more msleep()s....#asahi-dev · 2024-09-28 02:50
chadmed_the fallback values in the driver are quite slow, but maybe we need slower#asahi-dev · 2024-09-28 02:50
chadmed_i wonder if folks' spurious headset misdetections are related#asahi-dev · 2024-09-28 02:50
chadmed_hm we're not setting the cs42l42 tip sense debounce or anything in the DT#asahi-dev · 2024-09-28 00:08
chadmed_oh man the cs42l42 is a bit of a beast#asahi-dev · 2024-09-01 07:47
chadmed_is sommelier presenting a 1280x720 x11 server to its clients a known issue?#asahi-dev · 2024-09-01 07:44
chadmed_edid stuff works nicely, both my dell 2721dgfs have builtin profiles which almost perfectly match the display-p3 icc#asahi-dev · 2024-08-31 06:33
chadmed_jannau: all good, thought there might be something :)#asahi-dev · 2024-08-30 23:52
chadmed_(assuming it even does without anything v4l2 wired up) #asahi-dev · 2024-08-30 23:52
chadmed_leio: might be worth testing the dcp one with whatever version of gnome is supposed to use that functionality? #asahi-dev · 2024-08-30 23:51
chadmed_is there anything majorly wrong with the hwmon and dcp PRs that i still need to address? #asahi-dev · 2024-08-09 10:18
chadmed_finally got REW to play nice with a sane version of openjdk#asahi-dev · 2024-08-07 11:56
chadmed_smc reports them as doing ~6000 rpm each#asahi-dev · 2024-08-07 11:56
chadmed_so i guess the macos "high power" mode might just do what it says on the tin - clamp the fans to ~3500-4000 rpm and just unclamp that if you enable it#asahi-dev · 2024-08-07 11:54
chadmed_perfect#asahi-dev · 2024-08-07 11:53
chadmed_looks like j314 just throttles if you clamp the fan speed#asahi-dev · 2024-08-06 05:27
chadmed_trying to write 0/nothing outside of manual mode will also return -EINVAL #asahi-dev · 2024-08-06 05:27
chadmed_once a fan is in manual mode, it will respond to any request for a valid speed within its min/max range. writing 0/nothing will take it out of manual mode and any attempt to write a value outside of its safe range will return -EINVAL and do nothing#asahi-dev · 2024-08-06 05:26
chadmed_pushed some changes to how the fans are handled when in manual mode#asahi-dev · 2024-08-06 03:28
chadmed_so i think we definitely need to change the behaviour such that any out of bounds write just brings us back to auto control#asahi-dev · 2024-08-06 03:26
chadmed_the SMC doesnt actually limit the fan speed itself to the min and max :) #asahi-dev · 2024-08-06 03:25
chadmed_well uhh thats fun#asahi-dev · 2024-08-05 22:51
chadmed_based on latest asahi-wip so has all the required dt stuff too#asahi-dev · 2024-08-05 22:50
chadmed_https://github.com/chadmed/linux/tree/hwmon-fanset#asahi-dev · 2024-08-05 22:49
chadmed_maz: the branch that patch came from has jannau's write f32 in there, it's the previous commit#asahi-dev · 2024-07-08 21:54
chadmed_ugh that just disables xdg stuff... #asahi-dev · 2024-07-08 21:53
chadmed_ella-0: i build it with USE="-X" on gentoo and it _seems_ to do what it's supposed to#asahi-dev · 2024-06-25 02:15
chadmed_what benefit is there to fixing the system timer frequency to such a high value?#asahi-dev · 2024-06-24 22:44
chadmed_ill pull your changes in and fix it all up tonight. thanks for looking at it :) #asahi-dev · 2024-06-24 22:43
chadmed_yeah i was going to do that anyway since we're splitting stuff out #asahi-dev · 2024-06-24 22:18
chadmed_hopefully once (if?) the driver is merged we can add it to the yaks and others can help out here#asahi-dev · 2024-06-24 22:18
chadmed_jannau: not a problem at all. i will rip out the smc key stuff in the MR and replace it with just the keys i know exist on the m2 max studio for now (later) and we can sort out other machines as we figure them out#asahi-dev · 2024-05-26 23:07
chadmed_could you even do something with the usb-pd commands?#asahi-dev · 2024-05-15 04:46
chadmed_kwin's cursor code mostly calls into QCursor#asahi-dev · 2024-05-15 04:46
chadmed_j`ey: looks like qt handles the cursor https://doc.qt.io/qt-6/qcursor.html so this might be a qt bug #asahi-dev · 2024-05-13 23:41
chadmed_so all we do for NFP is check if any of the samples are not finite (rust has a nice method on its float types for this) and then set them to 0 in the output buffer#asahi-dev · 2024-05-13 23:41
chadmed_anyway with lv2 we can ignore time. we just get a buffer of samples, iterate over them and do something to it, then fill up an output buffer#asahi-dev · 2024-05-13 23:39
chadmed_(it's me btw im the stupid moron)#asahi-dev · 2024-05-13 23:37
chadmed_thankfully its very rare but if some stupid moron is for example trying to write a psychoacoustic bass enhancer and screws up the signal processing path... it can cause some unpleasantness :p