T42 | <zinstack625> mal: you up? | 09:24 |
---|---|---|
hacker420-M | uh | 09:52 |
hacker420-M | that join list | 09:52 |
hacker420-M | Anyway I was trying to update cedric to sfos 4 and in the process up the base to the official LOS | 09:53 |
hacker420-M | running into this | 09:53 |
hacker420-M | ``` | 09:54 |
hacker420-M | FAILED: ninja: 'out/target/product/cedric/obj_arm/SHARED_LIBRARIES/libqsap_sdk_intermediates/export_includes', needed by 'out/target/product/cedric/obj_arm/SHARED_LIBRARIES/libqsapshim_intermediates/import_includes', missing and no known rule to make it | 09:54 |
hacker420-M | ``` | 09:54 |
T42 | <zinstack625> hacker420-M: can't see that from Telegram side | 09:54 |
T42 | hacker12455 was added by: hacker12455 | 09:54 |
T42 | <hacker12455> @zinstack625 [hacker420-M: can't see that from Telegram side], I can see it | 09:55 |
T42 | <zinstack625> hacker420-M: I just threw out libqsapshim from my device tree, I believe it has something to do with wifi access point | 09:55 |
T42 | <hacker12455> forgot that channel is on tg too | 09:55 |
T42 | <hacker12455> @zinstack625 [hacker420-M: I just threw out libqsapshim from …], alright | 09:56 |
T42 | <hacker12455> FAILED: ninja: 'libhwc2_compat_layer', needed by 'hybris-hal', missing and no known rule to make it | 10:03 |
T42 | <hacker12455> I assume the fixes from hadk-faq regarding hybris-16.0 apply to 17.1 too? | 10:03 |
T42 | <hacker12455> applied | 10:06 |
T42 | <hacker12455> let | 10:06 |
T42 | <hacker12455> *let's see if this passes | 10:06 |
T42 | <hacker12455> ok now it went furthe | 10:07 |
T42 | <hacker12455> *further | 10:07 |
rinigus | mal: latest libhybris fails to build for me: https://github.com/libhybris/libhybris/issues/478 | 10:12 |
rinigus | the one before that, builds fine | 10:13 |
rinigus | (using that for now) | 10:13 |
T42 | <cfb014> Hi! I'm using sailfish 3.4 in my sony z3 compact tablet. Is there any possibility to get sailfish 4? I've tried to update via OTA but it doesn't work | 10:40 |
rinigus | @cfb014: do not run OTA before port is updated. right now we wait for OBS updates, probably the same wait for Z3 port | 10:41 |
T42 | <cfb014> @rinigus [@cfb014: do not run OTA before port is updated …], Ok, thank you for your answer! I'm newbie in ports | 10:42 |
swanux | So, since yesterday I got telnet in post-switch_root halt state (thanks!), and now I can interact with the device. I just don't really know where to begin / how to continue. | 10:43 |
T42 | <zinstack625> swanux: see what's broken in systemctl/dmesg/journal and start fixing stuff you want fixed | 10:44 |
swanux | @zinstack625 You mean var/log/journal? (because what you mentioned doesn't exist, and I got the persistent journal here) | 10:49 |
T42 | <zinstack625> swanux: I meant look in systemctl or dmesg or journalctl | 10:50 |
rinigus | @cfb014: no problem, it's better to ask | 10:50 |
swanux | Ah, ok, I read them as path, thanks | 10:51 |
T42 | <zinstack625> logcat (/system/bin/logcat or /usr/libexec/droid-hybris/system/bin/logcat) can be interesting too | 10:51 |
swanux | The logcat files don't exist | 10:52 |
swanux | Probably they're generated after systemd init, but mine is halted before that to be able to telnet | 10:52 |
T42 | <elros34> logcat is not a regular file but executable binary and there is no point to run it before droid-hal-init starts. First grab fresh dmesg and confirm selinux is disabled at boottime | 10:54 |
T42 | <elros34> If you follow instruction you see after you telnet the you can continue booting process | 10:56 |
swanux | @elros34 Yes, selinux is disabled (followed your github pull request). You mean dmesg on the device or on the pc? | 10:59 |
swanux | Yes, I see the option printed out when I telnet that tells how to continue. | 10:59 |
T42 | <elros34> obviously you want to confirm kernel config on device | 10:59 |
swanux | [ 0.001815] SELinux: Disabled at boot. | 11:00 |
T42 | <elros34> ok, using 'journalctl --no-tail -f' you can get live log | 11:01 |
swanux | Ok, I see it. Now continue boot from another telnet session? | 11:02 |
T42 | <elros34> yes | 11:03 |
swanux | Ok. I got ~2k lines of log. Is there anything special I should look for, or go through every error/exception/NULL/etc? | 11:07 |
T42 | <elros34> start from the top: search for errors, droid mounts, whether droid-hal-init initialize correctly and so on | 11:09 |
swanux | Ok, thank you | 11:11 |
swanux | I collected the "promising" problems here (my comments with # before them): https://pastebin.com/Xbqv5Tnj | 11:29 |
swanux | (They're in the same order as in the log) | 11:30 |
T42 | <hacker12455> IMPORTANT: some devices in your Android tree are 64bit targets. If your device is aarch64, | 11:40 |
T42 | <hacker12455> please define droid_target_aarch64 in your .spec, otherwise define droid_target_armv7hl | 11:40 |
T42 | <hacker12455> NOTE: Currently there is no Sailfish OS ARM 64bit target, so leave PORT_ARCH as armv7hl | 11:40 |
T42 | <hacker12455> Mixed builds of 64bit Android+Linux Kernel and 32bit Sailfish OS work just fine. | 11:40 |
T42 | <hacker12455> + exit 1 | 11:40 |
T42 | <hacker12455> error: Bad exit status from /var/tmp/rpm-tmp.FsK0Ky (%build) | 11:40 |
T42 | <hacker12455> in which .spec file do I change that | 11:41 |
T42 | <hacker12455> I set port_arch to aarch64 as that is what hadk recommends | 11:41 |
mal | rinigus: libhybris fix coming very soon | 12:04 |
mal | rinigus: do you want to test a fix | 12:08 |
rinigus | mal: sure. there could be some delay in testing, but will be happy to | 12:12 |
mal | rinigus: https://github.com/libhybris/libhybris/pull/479 | 12:12 |
mal | rinigus: just a trivial build fix | 12:12 |
mal | hmm, actually I can also test it for fp3 | 12:12 |
T42 | <hacker12455> quick question | 12:13 |
T42 | <hacker12455> if I have a 64 bit base | 12:13 |
T42 | <hacker12455> do I set droid_target_armv7hl | 12:13 |
T42 | <hacker12455> or droid_target_aarch64 | 12:13 |
T42 | <hacker12455> ok defining droid_target_aarch64 worked | 12:15 |
T42 | <hacker12455> with port_arch as aarch64 | 12:15 |
mal | @hacker12455 "%define droid_target_aarch64 1" | 12:15 |
T42 | <hacker12455> yea figured that out | 12:15 |
rinigus | mal: spent morning on separate hybris-hal and droidmedia building and packaging. nope, that did not help. camera still gets stuck after shooting video. (Waiting for service 'media.audio_policy') | 12:17 |
mal | rinigus: that new fix for libhybris built fine without tracing enabled | 12:18 |
T42 | <hacker12455> ```+ cp -a ./out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libdl.so ./out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libc.so ./out/soong/.intermediates/art/bui | 12:20 |
T42 | ld/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libm.so /home/kacper/hadk/installroot/usr/libexec/droid-hybris/system/lib | 12:20 |
T42 | <hacker12455> cp: cannot stat './out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libdl.so': No such file or directory | 12:20 |
T42 | <hacker12455> cp: cannot stat './out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libc.so': No such file or directory | 12:20 |
T42 | <hacker12455> cp: cannot stat './out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/lib/bionic/libm.so': No such file or directory``` | 12:20 |
rinigus | @mal: nice! I'll check but later and report back | 12:26 |
mal | @hacker12455 intesresting, I didn't know android 10 base can be done without apex, can you check have at all the path out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex/ | 12:31 |
T42 | <hacker12455> @mal [@hacker12455 intesresting, I didn't know andro …], PlatformSDK [kacper@kacper-pc android_common_com.android.runtime.release]$ ls | 12:32 |
T42 | <hacker12455> apex_manifest.json apex_pubkey | 12:32 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc android_common_com.android.runtime.release]$ pwd | 12:32 |
T42 | <hacker12455> /home/kacper/hadk/out/soong/.intermediates/art/build/apex/com.android.runtime.release/android_common_com.android.runtime.release | 12:32 |
T42 | <hacker12455> there is no image.apex | 12:32 |
mal | ok | 12:32 |
mal | @hacker12455 can you see libdl.so, libm.so and libc.so anywhere in out/, maybe use "find out/ -name libdl.so" | 12:33 |
T42 | <hacker12455> @mal [@hacker12455 can you see libdl.so, libm.so and …], https://bpa.st/2NFHO | 12:34 |
T42 | <hacker12455> libm and libc are also there | 12:35 |
mal | @hacker12455 ok, as an ugly hack for now edit the rpm/dhd/droid-hal-device.inc file and fix fix paths around here https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L604 | 12:37 |
mal | I will see how to make those work better | 12:37 |
T42 | <hacker12455> @mal [@hacker12455 ok, as an ugly hack for now edit …], what path do I put there | 12:38 |
mal | somehow this out/target/product/cedric/system/apex/com.android.runtime.release | 12:38 |
mal | which you saw when using find | 12:39 |
T42 | <hacker12455> art_path=%{android_root}/out/target/product/cedric/system/apex/com.android.runtime.release/ | 12:41 |
T42 | <hacker12455> apex_path=apex/com.android.runtime.release/android_common_com.android.runtime.release/image.apex | 12:41 |
mal | no | 12:41 |
mal | art_path=%{android_root}/out/target/product/cedric/system/ | 12:42 |
mal | apex_path=apex/com.android.runtime.release | 12:42 |
mal | so when you combine those it will become out/target/product/cedric/system/apex/com.android.runtime.release | 12:42 |
T42 | <hacker12455> art_path=%{android_root}/out/target/product/cedric/system/ | 12:42 |
T42 | <hacker12455> apex_path=apex/com.android.runtime.release | 12:42 |
rinigus | mal: libhybris fix worked over here as well | 12:43 |
mal | rinigus: yeah, it's merged now, just prepared a PR to mer-hybris | 12:50 |
T42 | <hacker12455> hm should I just accept building all middleware packages? | 12:50 |
mal | yes | 12:51 |
T42 | <hacker12455> so I can just answer al;l | 12:51 |
T42 | <hacker12455> *all | 12:51 |
rinigus | mal: issue closed as well | 12:59 |
T42 | <hacker12455> Error <creator>[02/27 13:03:03] : URLGrabber error: http://repo.merproject.org/obs/nemo:/devel:/hw:/common/sailfish_latest_aarch64/repodata/repomd.xml - [Errno 14] HTTP Error 404 - Not Found | 13:03 |
T42 | <hacker12455> ah there is only an armv7hl directory | 13:03 |
T42 | <hacker12455> change arch to armv7hl? | 13:05 |
T42 | <hacker12455> or what | 13:05 |
T42 | <elros34> @swanux great that you analyze logs but if you want other to take a look at them then provide full logs. Do you still lose telnet? If yes did you figure out what triggers that? | 13:06 |
mal | @hacker12455 you need to remove the repo from being used | 13:07 |
T42 | <hacker12455> @mal [@hacker12455 you need to remove the repo from …], how? | 13:09 |
swanux | @elros34 Yeah, sure, here it is (it's the full log after continue from telnet - https://pastebin.com/FVzQR5F1). Without the workaround I still loose it (but if I stop the loop after one successful connection it stays that way). I have no idea what triggers that, but I think it has something to do with the thing that prevents proper boot. | 13:12 |
mal | @hacker12455 again an ugly fix edit hybris/mw/community-adaptation/rpm/community-adaptation.inc and comment out the line "adaptation-community-common.ini > $RPM_BUILD_ROOT/%{_datadir}/ssu/features.d/adaptation-community-common.ini" near the end of the file, then run "rpm/dhd/helpers/build_packages.sh --build=hybris/mw/community-adaptation --spec=rpm/community-adaptation-localbuild.spec" | 13:12 |
T42 | <hacker12455> comment out with #? | 13:17 |
mal | yes | 13:17 |
T42 | <hacker12455> %endif | 13:18 |
T42 | <hacker12455> #adaptation-community-common.ini > $RPM_BUILD_ROOT/%{_datadir}/ssu/features.d/adaptation-#community-.ini | 13:18 |
T42 | <hacker12455> %endif | 13:18 |
T42 | <hacker12455> (Photo, 744x72) https://irc.thaodan.de/.imgstore/DSeqcAv9hW.png | 13:18 |
T42 | <zinstack625> I fixed vibrator on my phone (aw869x)! Hooray! | 13:21 |
mal | how | 13:21 |
T42 | <zinstack625> Turns out, it couldn't find the vibrator firmware(it needed that!), because the vendor partition was not yet mounted | 13:22 |
T42 | <elros34> @swanux is that all? Nothing was printed later? When droid-hal-init start you can try to get logcat. Also make sure you have needed files in /usr/libexec/droid-hybris/system/etc/init/ | 13:23 |
T42 | <zinstack625> Building it as module and loading it at runtime allowed it to load its firmware and now it's vibrating like a brand new challenger | 13:23 |
T42 | <zinstack625> Also yesterissue was a long-forgottend systemd 237 bug with cgroups, which was fixed by adding "systemd.legacy_systemd_cgroup_controller=1" to kernel cmdline | 13:25 |
T42 | <zinstack625> Retroactive but works | 13:25 |
T42 | <hacker12455> @mal> @hacker12455 again an ugly fix edit hybris/mw/community-adaptation/rpm/community-adaptation.inc and comment out the line "adaptation-community-common.ini [$RPM_BUILD_ROOT/%{_datadir}/ssu/features.d/ada …], how long should that build be | 13:31 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc hadk]$ rpm/dhd/helpers/build_packages.sh --build=hybris/mw/community-adaptation --spec=rpm/community-adaptation-localbuild.spec | 13:35 |
T42 | <hacker12455> * Building rpm/community-adaptation-localbuild.spec | 13:35 |
T42 | <hacker12455> only thing I am getting | 13:35 |
T42 | <hacker12455> mb2 seems to be running | 13:37 |
mal | it should not take long | 13:37 |
mal | maybe do this change instead https://github.com/mer-hybris/community-adaptation/pull/3/files | 13:37 |
T42 | <hacker12455> did that | 13:41 |
T42 | <hacker12455> barely any cpu usage | 13:41 |
T42 | <hacker12455> building is still stuck | 13:41 |
T42 | <hacker12455> (Photo, 1280x690) https://irc.thaodan.de/.imgstore/0Fh3JPo5pk.png | 13:42 |
mal | show what you did to the .inc file | 13:42 |
T42 | <hacker12455> %if "%{flavour}" == "testing" | 13:43 |
T42 | <hacker12455> sed -e "s|@FLAVOUR@|testing|g" \ | 13:43 |
T42 | <hacker12455> %else | 13:43 |
T42 | <hacker12455> sed -e "s|@FLAVOUR@|devel|g" \ | 13:43 |
T42 | <hacker12455> -e "s|sailfishos_%%(release)|sailfish_latest_%%(arch)|g" \ | 13:43 |
T42 | <hacker12455> %endif | 13:43 |
T42 | <hacker12455> %ifnarch aarch64 | 13:43 |
T42 | <hacker12455> adaptation-community-common.ini > $RPM_BUILD_ROOT/%{_datadir}/ssu/features.d/adaptation-community-common.ini | 13:43 |
T42 | <hacker12455> %endif | 13:43 |
T42 | <hacker12455> %endif | 13:43 |
mal | try killing the sb2 and mb2 processes and try again | 13:45 |
T42 | <hacker12455> just to be sure | 13:46 |
T42 | <hacker12455> rpm/dhd/helpers/build_packages.sh --build=hybris/mw/community-adaptation --spec=rpm/community-adaptation-localbuild.spec | 13:46 |
T42 | <hacker12455> ran it again | 13:47 |
T42 | <hacker12455> nothing new | 13:49 |
T42 | <hacker12455> when I control+c it | 13:51 |
T42 | <hacker12455> * Building rpm/community-adaptation-localbuild.spec | 13:51 |
T42 | <hacker12455> ^CBuilding target platforms: aarch64-meego-linux-gnu | 13:51 |
T42 | <hacker12455> Building for target aarch64-meego-linux-gnu | 13:51 |
T42 | <hacker12455> Executing(%install): /bin/sh -e /var/tmp/rpm-tmp.Z6pyNL | 13:51 |
T42 | <hacker12455> + umask 022 | 13:51 |
T42 | <hacker12455> + cd /home/kacper/hadk/hybris/mw/community-adaptation | 13:51 |
T42 | <hacker12455> + '[' /home/kacper/hadk/hybris/mw/community-adaptation/installroot '!=' / ']' | 13:51 |
T42 | <hacker12455> + rm -rf /home/kacper/hadk/hybris/mw/community-adaptation/installroot | 13:51 |
T42 | <hacker12455> ++ dirname /home/kacper/hadk/hybris/mw/community-adaptation/installroot | 13:51 |
T42 | <hacker12455> + mkdir -p /home/kacper/hadk/hybris/mw/community-adaptation | 13:51 |
T42 | <hacker12455> + mkdir /home/kacper/hadk/hybris/mw/community-adaptation/installroot | 13:51 |
T42 | <hacker12455> + LANG=C | 13:51 |
T42 | <hacker12455> + export LANG | 13:51 |
T42 | <hacker12455> + unset DISPLAY | 13:51 |
T42 | <hacker12455> + mkdir -p /home/kacper/hadk/hybris/mw/community-adaptation/installroot//usr/share/community-adaptation/ | 13:51 |
mal | looks like there is something wrong in the changes, trying to fix it | 13:54 |
mal | ah, found the issue | 13:55 |
swanux | @elros34 There is nothing after it, just before. Nothing printed later because at the end I loose telnet (when mer loader becomes Mi 8 in dmesg - battery light starts working) - that's why I can't get logcat after droid hal init, because there's no telnet at the time. If you mean the .rc files from the faq, then I have them | 13:57 |
mal | @hacker12455 check the new changes here https://github.com/mer-hybris/community-adaptation/pull/3/files | 13:57 |
T42 | <hacker12455> that worked | 13:58 |
T42 | <hacker12455> except | 13:59 |
T42 | <hacker12455> Error <creator>[02/27 13:59:02] : URLGrabber error: http://repo.merproject.org/obs/nemo:/devel:/hw:/common/sailfish_latest_aarch64/repodata/repomd.xml - [Errno 14] HTTP Error 404 - Not Found | 13:59 |
T42 | <hacker12455> still have that | 13:59 |
T42 | <hacker12455> unless I need to do something else | 13:59 |
T42 | <hacker12455> let me just run —configs | 14:00 |
swanux | @hacker12455 Refer here (it worked for me): https://irclogs.sailfishos.org/logs/%23sailfishos-porters/%23sailfishos-porters.2021-01-06.log.html | 14:01 |
swanux | (ctrl+f the link in the log) | 14:02 |
mal | swanux: my fix is the proper fix | 14:03 |
swanux | mal No doubt, the one I shared is just a workaround. just in case | 14:04 |
T42 | <hacker12455> now I just need to manually run mic | 14:04 |
mal | @hacker12455 there should not be need to run mic manually with my fix | 14:05 |
T42 | <hacker12455> well | 14:06 |
mal | if it worked as it should | 14:06 |
T42 | <hacker12455> it does not work | 14:06 |
T42 | <hacker12455> ran rpm/dhd/helpers/build_packages.sh --configs afterwards | 14:06 |
T42 | <hacker12455> [02/27 14:07:34] : mic 1.0.2 (Sailfish OS 4.0.1.48 (Koli)) | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : Substitute macro variable @ARCH@ with aarch64 | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : Substitute macro variable @RELEASE@ with 4.0.1.48 | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : Substitute macro variable @EXTRA_NAME@ with -my1 | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : Retrieving repo metadata: | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : filename /var/tmp/mic/cache/adaptation-community-cedric-4.0.1.48/7ec3fd8ecfd04f7300168e38ff36e918a523ae2915d6c373470beefdc9843d4e-primary.xml.gz gunzipped | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : filename is now /var/tmp/mic/cache/adaptation-community-cedric-4.0.1.48/7ec3fd8ecfd04f7300168e38ff36e918a523ae2915d6c373470beefdc9843d4e-primary.xml | 14:07 |
T42 | <hacker12455> Info[02/27 14:07:34] : /var/tmp/mic/cache/adaptation-community-cedric-4.0.1.48/69d739fed1f2227a55dc6f3540e7f4255f7501e61f5cec30ed564f7396a8084a-patterns.xml checksum f5c595dc4eb4bb6095ef6a4f309275e05a3bd4639b83e0a4fca32be8e04f94e0 matches cache | 14:07 |
T42 | <hacker12455> Error <creator>[02/27 14:07:34] : URLGrabber error: http://repo.merproject.org/obs/nemo:/devel:/hw:/common/sailfish_latest_aarch64/repodata/repomd.xml - [Errno 14] HTTP Error 404 - Not Found | 14:07 |
mal | at least it did what it's supposed to do in my test build, are you sure it installed the new package to local target | 14:08 |
T42 | <birdzhang> @hacker12455 run sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in droid-config-$DEVICE | 14:10 |
mal | also try running sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in --force community-adaptation-localbuild | 14:10 |
mal | before that other command above droid-config | 14:10 |
T42 | <hacker12455> well ran the community-adaptation command first | 14:11 |
T42 | <hacker12455> then the droid config one | 14:12 |
T42 | <hacker12455> afterwards build_packages.sh —mic | 14:12 |
T42 | <hacker12455> and same error | 14:12 |
mal | do you see any files when you run sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R ls -l /usr/share/ssu/features.d/ | 14:14 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc hadk]$ sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R ls -l /usr/share/ssu/features.d/ | 14:16 |
T42 | <hacker12455> total 8 | 14:16 |
T42 | <hacker12455> -rw-r--r-- 1 kacper system 313 Feb 27 14:11 adaptation-community-common.ini | 14:16 |
T42 | <hacker12455> -rw-r--r-- 1 kacper system 0 Feb 27 14:11 adaptation-community.ini | 14:16 |
T42 | <hacker12455> -rw-r--r-- 1 kacper system 473 Feb 27 14:11 customer-jolla-sdk.ini | 14:16 |
mal | hmm, why is the adaptation-community-common.ini file still there if you correctly modified the .inc file | 14:17 |
T42 | <hacker12455> also output of the above commands | 14:17 |
T42 | <hacker12455> https://bpa.st/4QPZM | 14:17 |
mal | are you sure you correctly edited the hybris/mw/community-adaptation/rpm/community-adaptation.inc and the build worked | 14:18 |
T42 | <hacker12455> entire file | 14:19 |
T42 | <hacker12455> https://bpa.st/5EXWC | 14:19 |
T42 | <hacker12455> reran the build again | 14:20 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc hadk]$ rpm/dhd/helpers/build_packages.sh --build=hybris/mw/community-adaptation --spec=rpm/community-adaptation-localbuild.spec | 14:20 |
T42 | <hacker12455> * Building rpm/community-adaptation-localbuild.spec | 14:20 |
T42 | <hacker12455> * Building successful, adding packages to repo | 14:20 |
T42 | <hacker12455> Retrieving repository 'adaptation-community-common' metadata ..........................................................................................................................................................................[done] | 14:20 |
T42 | <hacker12455> Repository 'apps' is up to date. | 14:20 |
T42 | <hacker12455> Repository 'customer-jolla' is up to date. | 14:20 |
T42 | <hacker12455> Repository 'hotfixes' is up to date. | 14:20 |
T42 | <hacker12455> Repository 'jolla' is up to date. | 14:20 |
T42 | <hacker12455> Retrieving repository 'local-cedric-hal' metadata .....................................................................................................................................................................................[done] | 14:20 |
T42 | <hacker12455> Building repository 'local-cedric-hal' cache ..........................................................................................................................................................................................[done] | 14:20 |
T42 | <hacker12455> All repositories have been refreshed. | 14:20 |
T42 | <hacker12455> * Building of community-adaptation finished successfully | 14:20 |
mal | so manually check the contents of the rpm file it created | 14:20 |
T42 | <hacker12455> where did it output it? | 14:21 |
mal | droid-local-repo/$DEVICE/community-adaptation/ | 14:21 |
T42 | <hacker12455> (Photo, 694x187) https://irc.thaodan.de/.imgstore/h9NNwN1dJt.png | 14:22 |
T42 | <hacker12455> only have adaptation-community.ini.in | 14:22 |
mal | so the file is not there | 14:23 |
mal | then why is it in your target | 14:23 |
T42 | <elros34> @swanux: do you have your droid repos uploaded? You need to somehow get logcat to figure out why vndservicemanager fails. Try systemctl mask droid-hal-init, reboot then start droid-hal-init same as /usr/bin/droid/droid-hal-startup.sh starts it and logcat at then same time. | 14:24 |
T42 | <hacker12455> @mal [then why is it in your target], good question | 14:25 |
T42 | <hacker12455> oops just saw the don't use direct reply | 14:25 |
T42 | <hacker12455> ran | 14:25 |
T42 | <hacker12455> sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in --force community-adaptation-localbuild | 14:25 |
T42 | <hacker12455> and reran mic | 14:25 |
T42 | <hacker12455> same error | 14:25 |
mal | is the file still there? | 14:27 |
T42 | <hacker12455> yep | 14:27 |
T42 | <elros34> @swanux btw have you tried whether masking usb-moded prevents losing telnet as suggested some time ago? | 14:27 |
mal | @hacker12455 what does this return sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R rpm -ql community-adaptation-localbuild | 14:29 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc hadk]$ sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R rpm -ql community-adaptation-localbuild | 14:29 |
T42 | <hacker12455> /usr | 14:29 |
T42 | <hacker12455> /usr/share | 14:29 |
T42 | <hacker12455> /usr/share/community-adaptation | 14:29 |
T42 | <hacker12455> /usr/share/community-adaptation/adaptation-community.ini.in | 14:29 |
T42 | <hacker12455> /usr/share/ssu | 14:29 |
T42 | <hacker12455> /usr/share/ssu/features.d | 14:29 |
mal | also please do not paste many lines here, usually anything more than 3 lines should go to pastebin or some similar service | 14:29 |
T42 | <hacker12455> alright | 14:30 |
mal | @hacker12455 what does this return sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R rpm -qf /usr/share/ssu/features.d/adaptation-community-common.ini | 14:30 |
T42 | <hacker12455> file /usr/share/ssu/features.d/adaptation-community-common.ini is not owned by any package | 14:30 |
mal | hmm, so sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R rm /usr/share/ssu/features.d/adaptation-community-common.ini | 14:31 |
T42 | <hacker12455> anything to run after that command | 14:31 |
T42 | <hacker12455> or go straight to mic | 14:32 |
mal | run this first build_packages.sh -c | 14:32 |
swanux | @elros34 Here are my repos (https://github.com/swanux/droid-hal-version-dipper-16, https://github.com/swanux/droid-hal-dipper-16, https://github.com/swanux/droid-config-dipper-16). How deas exactly usb-moded mask | 14:32 |
swanux | (accidental enter) masking work? | 14:33 |
T42 | <elros34> it will prevents the service from starting. | 14:34 |
T42 | <hacker12455> works now | 14:34 |
T42 | <elros34> do i same as for droid-hal-init so: systemctl mask usb-moded | 14:34 |
T42 | <hacker12455> well | 14:34 |
T42 | <hacker12455> Error <creator>[02/27 14:34:40] : Unable to find package: patterns-sailfish-device-configuration-cedric | 14:34 |
T42 | <elros34> @swanux I do not see *.rc files from faq in your droid-config. | 14:36 |
swanux | @elros34 Ok, I masked both, I'm about to try the manual init (usb-moded mask didn't help with the telnet stability, still needs loop). They're no there because I pushed them later manually, but now they're there. (just not in the repo yet) | 14:38 |
swanux | BTW, how do I run logcat at the same time? (as logcat command not found) Or journalctl will be fine? | 14:39 |
T42 | <elros34> zinstack625 gives you previously path to logcat. '/sbin/droid-hal-init &' then run logcat | 14:41 |
T42 | <hacker12455> hm | 14:42 |
T42 | <hacker12455> why am I getting a pattern error if my patterns are defined in the .spec file | 14:42 |
T42 | <hacker12455> %include droid-configs-device/droid-configs.inc | 14:42 |
T42 | <hacker12455> %include patterns/patterns-sailfish-device-adaptation-cedric.inc | 14:42 |
T42 | <hacker12455> %include patterns/patterns-sailfish-device-configuration-cedric.inc | 14:42 |
T42 | <elros34> Can you find patterns rpms in droid-local-repo? | 14:43 |
T42 | <hacker12455> (Photo, 401x189) https://irc.thaodan.de/.imgstore/9wfW9A6Y9S.png | 14:43 |
mal | how new is the config repo for your build? | 14:45 |
mal | @hacker12455 what exact error do you get? | 14:45 |
T42 | <hacker12455> @mal [@hacker12455 what exact error do you get?], Error <creator>[02/27 14:41:55] : Unable to find package: patterns-sailfish-device-configuration-cedric | 14:46 |
swanux | @elros34 Th | 14:48 |
swanux | There may be some problem with logcat https://pastebin.com/kaKWCHdG | 14:48 |
T42 | <elros34> @hacker12455 are you sure that you use 4.0.1.48 aarch64 sdk target? | 14:48 |
T42 | <hacker12455> (Photo, 671x146) https://irc.thaodan.de/.imgstore/NgA5SiA2rI.png | 14:49 |
T42 | <hacker12455> used this target | 14:49 |
T42 | <elros34> @swanux did you run it after starting droid-hal-init or just when you are in init-debug script? ALso update your repositories | 14:49 |
T42 | <hacker12455> PlatformSDK [kacper@kacper-pc hadk]$ sdk-assistant list | 14:50 |
T42 | <hacker12455> SailfishOS-4.0.1.48 | 14:50 |
T42 | <hacker12455> └─motorola-cedric-aarch64 | 14:50 |
swanux | @elros34 I ran /sbin/droid-hal-init but still can't run logcat. (yeah, I'll add the rc files to them as well) | 14:52 |
T42 | <elros34> @hacker12455 so can you find that pattern via: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper se -s patterns-sailfish-device-configuration-cedric or any other package in your local repository? | 14:53 |
swanux | Which is strange as ls shows the file there | 14:53 |
T42 | <hacker12455> S | Name | Type | Version | Arch | Repository | 14:54 |
T42 | <hacker12455> --+-----------------------------------------------+---------+----------------+---------+----------------- | 14:54 |
T42 | <hacker12455> | patterns-sailfish-device-configuration-cedric | package | 1-202102271441 | aarch64 | local-cedric-hal | 14:54 |
T42 | <hacker12455> so yeah | 15:03 |
T42 | <hacker12455> I can find the pattern | 15:03 |
T42 | <hacker12455> I can find all listed patterns | 15:05 |
T42 | <hacker12455> so why mic complains I have no idea | 15:07 |
T42 | <hacker12455> any ideas? | 15:17 |
T42 | <hacker12455> interesting | 15:22 |
T42 | <hacker12455> https://bpa.st/CURFY | 15:22 |
T42 | <hacker12455> if I try to force an install | 15:22 |
T42 | <hacker12455> do I need jolla-devicelock-plugin-encpartition | 15:23 |
T42 | <hacker12455> if I try to force patterns-sailfish-device-adaptation-cedric | 15:26 |
T42 | <hacker12455> https://bpa.st/BPNTG | 15:26 |
T42 | <elros34> so can zypper find jolla-devicelock-daemon-encsfa in target? | 15:27 |
T42 | <elros34> and jolla-devicelock-daemon-encpartition | 15:27 |
T42 | <hacker12455> I mean if it asks to uninstall it | 15:28 |
T42 | <hacker12455> then I assume yes | 15:28 |
T42 | <hacker12455> | jolla-devicelock-daemon-encsfa | package | 0.2.46-1.12.17.jolla | aarch64 | jolla | 15:29 |
T42 | <hacker12455> i | jolla-devicelock-daemon-encpartition | package | 0.2.46-1.12.17.jolla | aarch64 | jolla | 15:29 |
T42 | <hacker12455> @elros34 | 15:49 |
T42 | <zinstack625> @elros34, @mal, how do I get jolla store again? | 15:59 |
T42 | <elros34> @hacker12455 I don't know. Based on your output build_packages.sh -c should fail for you. WHy do you have droid-config-cedric-policy-settings even installed? REmove it and try again | 15:59 |
T42 | <elros34> @zinstack625 not sure what do you mean by again but faq have short instruction about that | 16:01 |
T42 | <hacker12455> it does not seem to be installed | 16:01 |
T42 | <hacker12455> the status window is empty | 16:02 |
T42 | <hacker12455> the package came from local-cedric-hal | 16:02 |
T42 | <zinstack625> @elros34 yeah, about instructions, they insist I must be on OBS, yet every time I ask here about it, noone seems to care | 16:02 |
T42 | <elros34> @hacker12455 right I mixed up. Did you create fresh patterns or you used some old one? | 16:03 |
T42 | <hacker12455> patterns should be fresh | 16:04 |
T42 | <elros34> @zinstack625 OBS is about to be closed so that can be reason why your reqests are ignored but it's not hard requirement for jolla store access | 16:04 |
T42 | <zinstack625> @elros34, then I think I'm ready, here's my ssu s: https://pastebin.com/TukZ5dSV | 16:05 |
T42 | <hacker12455> is that ok? | 16:06 |
T42 | <hacker12455> Requires: droid-config-cedric-policy-settings | 16:06 |
T42 | <hacker12455> in patterns-sailfish-device-adaptation-cedric | 16:06 |
T42 | <elros34> should be fine | 16:06 |
T42 | <hacker12455> so build_packages.sh -c passes | 16:07 |
rinigus | mal: looked into AOSP10 handling of camera. on SFOS, mini* services are 32bit. on AOSP10, camera service is 32 but audio is 64. also all missing libs for audiopolicy are on device at /system/lib64. | 16:08 |
rinigus | small summary at https://github.com/sailfishos-sony-tama/main/issues/115#issuecomment-787073277 | 16:08 |
rinigus | and below. logcat on camera closing in AOSP https://dpaste.org/Ut02; SFOS https://dpaste.org/iSw9 | 16:09 |
rinigus | mal: but it doesn't reply to a question why do I need MINIMEDIA_AUDIOPOLICYSERVICE_ENABLE | 16:10 |
T42 | <elros34> @hacker12455 maybe installing patterns in target was bad idea. | 16:10 |
T42 | <hacker12455> maybe | 16:10 |
T42 | <hacker12455> so what now? | 16:11 |
T42 | <elros34> I am not sure. You could add patterns-sailfish-device-adaptation-cedric to %packages section in your ks file and then run mic manually to see whether it will complain about missing packages | 16:12 |
T42 | <elros34> and show full output from mic | 16:13 |
rinigus | mal: also added list of media services running on AOSP10 in the camera issue linked above. there is audio, but not media.audio | 16:15 |
T42 | <hacker12455> do I run mic like this? | 16:16 |
T42 | <hacker12455> sudo mic create fs --arch=$PORT_ARCH \ | 16:16 |
T42 | <hacker12455> --tokenmap=$tokenmap \ | 16:16 |
T42 | <hacker12455> --record-pkgs=name,url \ | 16:16 |
T42 | <hacker12455> --outdir=$imgname \ | 16:16 |
T42 | <hacker12455> --pack-to=sfe-$DEVICE-$RELEASE"$EXTRA_NAME".tar.bz2 \ | 16:16 |
T42 | <hacker12455> "$ANDROID_ROOT"/$ks | 16:16 |
T42 | <elros34> if you exported all that variabales then yes otherwise: https://pastebin.com/2RFg4jYY | 16:18 |
T42 | <hacker12455> https://bpa.st/6H5SU | 16:20 |
T42 | <hacker12455> %packages | 16:20 |
T42 | <hacker12455> patterns-sailfish-device-configuration-cedric | 16:20 |
T42 | <hacker12455> patterns-sailfish-device-adaptation-cedric | 16:20 |
T42 | <hacker12455> %end | 16:20 |
T42 | <elros34> looks like you are missing local repo in ks file | 16:21 |
T42 | <hacker12455> https://bpa.st/B66VK | 16:24 |
T42 | <elros34> do you have %define community_adaptation 1 in droid-config-$DEVICE.spec? | 16:26 |
T42 | <hacker12455> yes | 16:27 |
T42 | <elros34> so I don't know, add set -x to build_packages.sh and run it with --mic argument | 16:31 |
T42 | <hacker12455> rpm/dhd/helpers/build_packages.sh --mic set -x | 16:34 |
T42 | <hacker12455> or before mic | 16:34 |
T42 | <elros34> no, add set -x after #!/bin/bash in script | 16:35 |
T42 | <hacker12455> ah | 16:37 |
T42 | <hacker12455> #!/bin/bash | 16:38 |
T42 | <hacker12455> set -x | 16:38 |
T42 | <hacker12455> like that | 16:38 |
T42 | <elros34> sure | 16:38 |
T42 | <hacker12455> https://bpa.st/3OWII | 16:39 |
T42 | <elros34> ah so messing with community-adaptation probably breaks that | 16:42 |
T42 | <hacker12455> welp | 16:43 |
T42 | <elros34> add local repo and use mic manuly: repo --name=adaptation-community-$DEVICE-@RELEASE@ --baseurl=file:///$ANDROID_ROOT/droid//droid-local-repo/$DEVICE. Expand env variables | 16:44 |
T42 | <hacker12455> Error <creator>[02/27 16:48:06] : URLGrabber error: can't find file ///$ANDROID_ROOT/droid-local-repo/$DEVICE/repodata/repomd.xml | 16:48 |
T42 | <elros34> evaluate env variables before inserting that line in ks file | 16:49 |
T42 | <hacker12455> they are set properly | 16:50 |
T42 | <elros34> but they are not expanded in ks file do it yourself | 16:51 |
T42 | <hacker12455> ah that | 16:51 |
T42 | <hacker12455> ok now it is going | 16:52 |
T42 | <hacker12455> it finished | 17:08 |
T42 | <hacker12455> whether it boots we will know later | 17:08 |
T42 | <zinstack625> May I please get store access? https://pastebin.com/TukZ5dSV | 17:40 |
mal | @zinstack62 as hint for future, the UID there is your imei number, you should not share it | 17:44 |
mal | @zinstack625 ^ | 17:44 |
T42 | <zinstack625> mal: will never do again, just this once :) | 17:47 |
mal | @zinstack625 the store instructions in faq said you just need to verify the UID is correct and matched the actual imei of your device and then inform pketo on this channel with the line "Device model: moto g(7) (river / river)" and saying UID was verified | 17:50 |
T42 | <zinstack625> mal: ok thanks | 17:52 |
T42 | <zinstack625> pketo: ^ UID is indeed, correct | 17:53 |
pketo | remind me on monday :) | 18:15 |
T42 | <zinstack625> pketo: ok, thanks | 18:15 |
piggz | mal: any thoughts on this error building gecko on 486 https://paste.mozilla.org/sQgOonPw | 19:58 |
piggz | new tooling adn target installed | 19:58 |
T42 | <zinstack625> Seems like broken toolchain to me | 20:13 |
mal | piggz: do you have python-M2Crypto installed? | 20:13 |
piggz | mal: i do now....trying again | 20:16 |
piggz | mal: same error | 20:16 |
mal | piggz: did you properly clean the gecko source from any old build fragments before trying that? | 20:21 |
piggz | mal: https://paste.mozilla.org/PgBuAexd | 20:28 |
piggz | mal: the gecko build seems to create its own python environment | 20:47 |
piggz | seems like all of the virtualenv setup fails | 20:52 |
mal | piggz: strange why it worked in arm for you | 21:05 |
piggz | yup | 21:05 |
piggz | mal: fancy setting up an i486 target, cloning and trying? | 21:05 |
mal | I would need to setup a new target, my i486 target is newer | 21:10 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!