Thursday, 2021-10-14

rinigusThaodan:: I remember it was rather trivial, but would have to look up details. Should be in the log of this channel, somewhere spring 2104:17
T42<adampigg> hi Melissa , what is your favourite sailfish app?06:38
T42<edp_17> Hi all! I am trying to build a package that requires cmake(Qt5Core) and cmake(Qt5Network) but I don't know their name to install in Platform SDK. Any idea how they can/should be installed?06:54
T42<edp_17> Nevermind. I finally was able to build it on device. 😃07:58
kquote03[m]Hello, I'm finally able to complete a build. It turns out that the sed error was because i did not put a space before the last '.09:14
kquote03[m]now i have another doubt (sorry), where do i edit the kernel's .config? I've built custom linux kernels before, so this step shouldn't be alien to me. I ran `mer_verify_kernel_config` and it listed some (many) changes, but where do I apply these changes to? Do I run `make menuconfig` on $ANDROID_ROOT/device/$VENDOR/$DEVICE ?09:17
T42<elros34> kquote03[m] you need to search defconfig file name used by your device in your device repo, then find that file in kernel/*/*/arch/*/configs/. Also do not follow all mer_verify_kernel_config_advices, they are simply wrong. Read this: collabedit.com/kr9xx. You can use menuconfig if you want  like this: https://pastebin.com/vXgWvv2u, remember to adjust ARCH and defconfig file name to match your device10:03
kquote03[m]T42: you are the man!10:04
ThaodanHas anyone experienced /odm/bin/sscrpcd crashing and restarting on boot? I can't really make sense of it. I tried to search in Xperiadev bugtracker but did not found anything really14:06
ThaodanThe daemon restarts like this every 25ms14:06
Thaodanhttps://paste.opensuse.org/5513513814:06
piggzmal: can you create me a devel and testing project for volla:yggdrasilx pls14:38
malpiggz: I gave you permissions for the whole volla projects, you can create the rest yourself14:54
piggzthx14:55
T42<CD> help?18:41
T42<CD> im trying to use a pixel 318:41
T42<CD> and oem unlock18:41
T42<CD> it is carrier unlocked18:41
T42<CD> i went to verizon18:41
T42<CD> i have tried backdating18:41
T42<CD> no go18:41
T42<CD> i have tried everything i can online18:41
T42<CD> oem stays greyed out18:41
T42<CD> pulling my hair out18:42
Thaodanrinigus: About https://github.com/sailfishos-sony-tama/main/issues/135 did you also got messages like this in dmesg https://paste.opensuse.org/77833850?20:36
ThaodanJust SM845 instead of SM8150.20:36
T42<b100dian> Thaodan: pushed fix for dbus regression in ngfd PR21:50
T42<b100dian> btw, thanks for pointing out libngf-qt feedback, I am celebrating 10 years with keyboard vibra ;)21:52
T42<b100dian> This was fairly new, btw, @elros34 https://github.com/sailfishos/libngf-qt/tree/master/feedback21:52
Thaodanthanks I will look into it21:55
T42<elros34> indeed I have looked at the whole sourcecs which are old21:55
Thaodancurrently fixing audioroutinng21:55
T42<b100dian> is  /usr/share/ngfd/events.d/feedback_press_weak.ini  and more something I should expect soon to land upstream?22:00
T42<b100dian> I had to create it for the test22:00
ThaodanWhat was the reasony you did not use __s32 for custom len has in kernel?23:35
T42<b100dian> I can switch back to __s16. I think you have __s16 too https://github.com/mer-hybris/android_kernel_sony_msm/blob/hybris-sony-aosp/LA.UM.7.1.r1/drivers/input/misc/qti-haptics.c#L84123:39
T42<b100dian> actually this https://github.com/mer-hybris/android_kernel_sony_msm/blob/hybris-sony-aosp/LA.UM.7.1.r1/include/uapi/linux/input.h#L39623:40
T42<b100dian> anyway, I can switch back to __s16, I have some ideas to improve the PR, but the point was to allocate enough for the kernel to copy 3x16bits for now. The ms/ns fields are not input, but (maybe) output23:41
ThaodanI mean this https://github.com/mer-hybris/android_kernel_sony_msm/blob/hybris-sony-aosp/LA.UM.7.1.r1/include/uapi/linux/input.h#L39523:42
T42<b100dian> ah, len23:42
T42<b100dian> where did I use int for len23:43
T42<b100dian> one flaw is that I declare the custom_len as 8, whereas it's in s16 records (so it should be 3, or 4).23:46
Thaodanhttps://github.com/Thaodan/ngfd/commit/60a99e7ea5ee46b020c4779275b75e6df3d85aab#23:46
ThaodanI would keep as close to the kernel as possible23:47
T42<b100dian> yes, I'll update that. So the int was just for the reading from the config file, I'll change it. Actually that is not a current commit, the situation right now is "8", but that's also incorrect. Let's talk on the PR if you want me to remember tomorrow:)23:49
Thaodannp It's late for me too :)23:50
T42<b100dian> Also.. how did the regression of dbus notifying back go?23:50
ThaodanI use that tree atm: https://github.com/Thaodan/ngfd/commits/constant_thao23:50
Thaodanexcept some complaints from the kernel about the effects before Vibra upload effect: Invalid argument23:52
Thaodan everything is fine23:52
T42<b100dian> this is the thing botched https://github.com/sailfishos-on-tucana/ngfd/commit/bc0258405e7b61f93c2e16d40c07b3ce8c2456e023:52
ThaodanSo I think its just the size of custom len23:52
ThaodanAh didn't noticed that23:53
Thaodanwhat exactly breaks because of it?23:53
T42<b100dian> the limit reached message from dbus23:53
T42<b100dian> is because I 'refactored' some field copying to memcpy but failed to noticed these were not copied23:54
T42<b100dian> actually, if your kernel is as dump as mine, and it has global state mutated on effects upload/erase, it might be worth to look if the custom data actually sends back the milliseconds to play, and add that as a timeout for effect complete23:54
T42<b100dian> so no upload/erase happens while an effect is playing23:54
T42<b100dian> so no 'invalid argument' or whatnot23:54
T42<b100dian> that's for tomorrow:)23:54
Thaodaninteresting23:58
Thaodanon what kernel are you?23:58

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!