| *** n2klb_ is now known as n2klb | 01:10 | |
| T42 | <elros34> you were patching IIRC system/core, that is not part of kernel so flashing hybris-boot (kernel + small script) is not enough. You need to build droid-hal then whole rootfs. BTW you shouldn't build such a outdated sfos minor release | 08:45 |
|---|---|---|
| T42 | <Mattia990> Ah now i understand, I thought apex was part of the kernel | 08:51 |
| T42 | <Mattia990> But why it is outdated? I'm building for saiflsih 5.0 isn't the almost latest? i followed the documentation instructions (re @elros34: you were patching II...) | 08:51 |
| T42 | <elros34> almost latest is 5.0.0.77 | 08:55 |
| T42 | <Teleg3_7> I’ve already tried the new tooling, targets, and SDK platform for version 5.1. Compiling pulseaudio-modules-droid* failed, so I’m still using that package from the fp5 repo. Command: | 10:11 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid-hidl” | 10:11 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid” | 10:11 |
| T42 | <Teleg3_7> Error: | 10:11 |
| T42 | <Teleg3_7> ERROR: The build data file ‘/home/debian13/sfos/src/hybris-20.0/hybris/mw/pulseaudio-modules-droid/aarch64-meego-linux-gnu/meson-private/build.dat’ references functions or classes that do not exist. This likely means it was generated using an older version of Meson. Consider reconfiguring the directory with “meson setup --reconfigure”. | 10:11 |
| T42 | <Teleg3_7> error: Bad exit status from /var/tmp/rpm-tmp.xuM5FR (%build) | 10:11 |
| T42 | <Teleg3_7> | 10:11 |
| T42 | <Teleg3_7> Second error during the final command: rpm/dhd/helpers/build_packages.sh --mic The creation of the sfe*.tar.bz2 files was successful, but it failed while creating the ZIP installer file. The error is: | 10:11 |
| T42 | <Teleg3_7> Running pack scripts ... | 10:11 |
| T42 | <Teleg3_7> /home/debian13/sfos/src/hybris-20.0/SailfishOScommunity-release-5.1.0.11-fog-lineage-20.0 /home/debian13/sfos/src/hybris-20.0 | 10:11 |
| T42 | <Teleg3_7> Info[06/20 00:06:43] : Script returned with a non-zero status, failing. | 10:11 |
| T42 | <Teleg3_7> | 10:11 |
| T42 | <Teleg3_7> Error <creator>[06/20 00:06:46] : Failed to execute %pack script with /bin/bash | 10:11 |
| T42 | <Teleg3_7> | 10:11 |
| T42 | <Teleg3_7> So I added the tar.bz2 file to the ZIP installer file that I had previously created. After installation, the result was the same as when using the older version of the tooling for the SIM cards. Every time SFOS reboots, the SIM card isn’t detected, and this issue can be resolved by tapping the “Restart” button in System Settings → Utilities under “Mobile Network.” After that, both SIM cards are detected, but only one SIM card ico | 10:11 |
| T42 | <Teleg3_7> I’ve already tried the new tooling, targets, and SDK platform for version 5.1. Compiling pulseaudio-modules-droid* failed, so I’m still using that package from the fp5 repo. Command: | 10:12 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid-hidl” | 10:12 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid” | 10:12 |
| T42 | <Teleg3_7> Error: | 10:12 |
| T42 | <Teleg3_7> ERROR: The build data file ‘/home/debian13/sfos/src/hybris-20.0/hybris/mw/pulseaudio-modules-droid/aarch64-meego-linux-gnu/meson-private/build.dat’ references functions or classes that do not exist. This likely means it was generated using an older version of Meson. Consider reconfiguring the directory with “meson setup --reconfigure”. | 10:12 |
| T42 | <Teleg3_7> error: Bad exit status from /var/tmp/rpm-tmp.xuM5FR (%build) | 10:12 |
| T42 | <Teleg3_7> | 10:12 |
| T42 | <Teleg3_7> Second error during the final command: rpm/dhd/helpers/build_packages.sh --mic The creation of the sfe*.tar.bz2 files was successful, but it failed while creating the ZIP installer file. The error is: | 10:12 |
| T42 | <Teleg3_7> Running pack scripts ... | 10:12 |
| T42 | <Teleg3_7> /home/debian13/sfos/src/hybris-20.0/SailfishOScommunity-release-5.1.0.11-fog-lineage-20.0 /home/debian13/sfos/src/hybris-20.0 | 10:12 |
| T42 | <Teleg3_7> Info[06/20 00:06:43] : Script returned with a non-zero status, failing. | 10:12 |
| T42 | <Teleg3_7> | 10:12 |
| T42 | <Teleg3_7> Error <creator>[06/20 00:06:46] : Failed to execute %pack script with /bin/bash | 10:12 |
| T42 | <Teleg3_7> | 10:12 |
| T42 | <Teleg3_7> So I added the tar.bz2 file to the ZIP installer file that I had previously created. After installation, the result was the same as when using the older version of the tooling for the SIM cards. Every time after SFOS reboots, the SIM card isn’t detected, and this issue can be resolved by tapping the “Restart” button in System Settings → Utilities under “Mobile Network.” After that, both SIM cards are detected, but only one SIM ca | 10:12 |
| T42 | <Teleg3_7> I’ve already tried the new tooling, targets, and SDK platform for version 5.1. Compiling pulseaudio-modules-droid* failed, so I’m still using that package from the fp5 repo. Command: | 10:15 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid-hidl” | 10:15 |
| T42 | <Teleg3_7> sfossdk sh -c “rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/pulseaudio-modules-droid” | 10:15 |
| T42 | <Teleg3_7> Error: | 10:15 |
| T42 | <Teleg3_7> ERROR: The build data file ‘/home/debian13/sfos/src/hybris-20.0/hybris/mw/pulseaudio-modules-droid/aarch64-meego-linux-gnu/meson-private/build.dat’ references functions or classes that do not exist. This likely means it was generated using an older version of Meson. Consider reconfiguring the directory with “meson setup --reconfigure”. | 10:15 |
| T42 | <Teleg3_7> error: Bad exit status from /var/tmp/rpm-tmp.xuM5FR (%build) | 10:15 |
| T42 | <Teleg3_7> | 10:15 |
| T42 | <Teleg3_7> Second error during the final command: rpm/dhd/helpers/build_packages.sh --mic The creation of the sfe*.tar.bz2 files was successful, but it failed while creating the ZIP installer file. The error is: | 10:15 |
| T42 | <Teleg3_7> Running pack scripts ... | 10:15 |
| T42 | <Teleg3_7> /home/debian13/sfos/src/hybris-20.0/SailfishOScommunity-release-5.1.0.11-fog-lineage-20.0 /home/debian13/sfos/src/hybris-20.0 | 10:15 |
| T42 | <Teleg3_7> Info[06/20 00:06:43] : Script returned with a non-zero status, failing. | 10:15 |
| T42 | <Teleg3_7> | 10:15 |
| T42 | <Teleg3_7> Error <creator>[06/20 00:06:46] : Failed to execute %pack script with /bin/bash | 10:15 |
| T42 | <Teleg3_7> | 10:15 |
| T42 | <Teleg3_7> So I added the tar.bz2 file to the ZIP installer file that I had previously created. After installation, the result was the same as when using the older version of the tooling for the SIM cards. Every time after SFOS reboots, the SIM card isn’t detected, and this issue can be resolved by tapping the “Restart” button in System Settings → Utilities under “Mobile Network.” After that, both SIM cards are detected, but only one SIM ca | 10:15 |
| T42 | <elros34> try to remove hybris/mw/pulseaudio-modules-droid and build it again | 10:31 |
| T42 | <Teleg3_7> Yes, now the two pulseaudio-modules-droid packages have been successfully compiled | 11:37 |
| T42 | <elros34> @Teleg3_7 about that zip issue: did you change anything in droid-config? Where is your droid-config repo? | 15:06 |
| T42 | <Mattia990> I'm still having the issue with apex, I verified the patch i applied, then I built everything again | 17:31 |
| T42 | <Mattia990> I flashed boot image and upload the rootfs (re @elros34: you were patching II...) | 17:31 |
| T42 | <elros34> same issue? | 17:32 |
| T42 | <Mattia990> Yes, i don't know if I'm doing something wrong. I applied the patch written in hadk hot and built everything | 17:37 |
| T42 | <Mattia990> www.pastebin.com/nbdAXPbF (re @elros34: same issue?) | 17:37 |
| T42 | <elros34> show git diff in system/core/init | 17:56 |
| T42 | <elros34> btw you still didn't fix metadata | 17:58 |
| T42 | <Mattia990> Metadata? Sorry, what was the problem? (re @elros34: btw you still didn't...) | 17:59 |
| T42 | <elros34> wrong by-name path in fixup-mountpoints | 17:59 |
| T42 | <elros34> you have 30s timeout because of this and bt_firmware | 18:00 |
| T42 | <Mattia990> Here's the diff: www.pastebin.com/VqpT5p2m (re @elros34: show git diff in sys...) | 18:01 |
| T42 | <elros34> and you build hybris-hal and then package droid-hal? | 18:01 |
| T42 | <Mattia990> I built hybris-hal droidmedia | 18:04 |
| T42 | <Mattia990> droid-hal configs mw gg version mic (re @elros34: and you build hybris...) | 18:04 |
| T42 | <elros34> so if you are sure that you have used correct zip file not old one then check if dmesg gives more info | 18:08 |
| T42 | <elros34> did you add printk.devkmsg=on to kernel cmdline? | 18:10 |
| T42 | <Mattia990> No, I'm going to add it. Will that command save log in /sys/fs/pstore ? (re @elros34: did you add printk.d...) | 18:12 |
| T42 | <elros34> no, it disable possible ratelimiting | 18:14 |
| T42 | <elros34> just to be sure please add/change error here https://github.com/LineageOS/android_system_core/blob/lineage-18.1/init/init.cpp#L638 then make hybris-hal, build droid-hal and mic image | 18:17 |
| T42 | <Mattia990> Now there is more stuff: www.pastebin.com/zTEhpxfp (re @elros34: did you add printk.d...) | 18:22 |
| T42 | <elros34> borringssl triggers fatal reaction. IIRC this patch should fix it so please make that change I was talking about. Just to be sure you have flashed lineage 18 now? | 18:30 |
| T42 | <Mattia990> I think yes, i will check | 18:37 |
| T42 | <Mattia990> I fixed the two mount points errors and added the change to that comment in system core init | 18:37 |
| T42 | <Mattia990> Just to be sure, in order to have the change, do I need to run a make clean when I build the bootimage with hybris-hal and droidmedia? (re @elros34: borringssl triggers ...) | 18:37 |
| T42 | <elros34> you never need to run make clean | 18:38 |
| T42 | <elros34> it's even harmful to run make clean. Now after makeing changes to system/core building hybris-hal should show something was rebuild so you know the changes are there | 18:40 |
| T42 | <Teleg3_7> https://github.com/SourceLab081/droid-configs-fog/tree/hybris-20.0 | 21:03 |
| T42 | <Teleg3_7> I made some changes to droid-config on the server but didn't push them to GitHub; for example, I updated the revision for droid-configs-device and set the URLs for the Chum and Storeman repositories to 5.1. (re @elros34: @Teleg3_7 about that...) | 21:03 |
| T42 | <elros34> so you are using custom pack_package-droid-updater, maybe this fail now | 21:05 |
| T42 | <Mattia990> I understand, thank you for the information! | 21:39 |
| T42 | <Mattia990> Now, I fixed mountpoints and add that modification to the apex init comment | 21:39 |
| T42 | <Mattia990> What happen now is the same, after telnet, if I echo "continue" > /init-ctl/stdin after some seconds the phone reboots | 21:39 |
| T42 | <Mattia990> In the journal I don't see apex logs now, I don't know what is the problem now | 21:39 |
| T42 | <Mattia990> www.pastebin.com/Pnuhv3tm (re @elros34: it's even harmful to...) | 21:39 |
| T42 | <elros34> have you tried dmesg -w? | 21:41 |
| T42 | <Mattia990> I didn't try, this is its output: www.pastebin.com/5WWBHk6e (re @elros34: have you tried dmesg...) | 21:45 |
| T42 | <elros34> you don't include full log so I have to ask about details: do you still have printk.devkmsg=on added, do you have peristen journal logging enabled or do you run journalctl --no-tail -f? | 21:48 |
| T42 | <Mattia990> This is journalctl --no-tail -f: www.pastebin.com/Xuiff4Hs | 22:05 |
| T42 | <Mattia990> This is the persistent one: www.pastebin.com/AzKhNBR7 | 22:05 |
| T42 | <Mattia990> And this dmesg -w: www.pastebin.com/MzNKHCi0 (re @elros34: you don't include fu...) | 22:05 |
| T42 | <Mattia990> I'm not seeing what is going wrong | 22:05 |
| T42 | <Mattia990> And the printk is still set to on | 22:06 |
| T42 | <elros34> so if you have persisten logging enabled then journalctl -b-1 doesn't show anything new? | 22:06 |
| T42 | <Mattia990> I think no, this is the output: www.pastebin.com/FbkCXHAb | 22:10 |
| T42 | <elros34> ok so you need mask and strace -f -s256 droid-hal-init | 22:13 |
| T42 | <Mattia990> During droid-hal-init I see again the apex error: paste.opensuse.org/pastes/a447b8385609 (re @elros34: ok so you need mask ...) | 22:26 |
| T42 | <elros34> boringssl doesnt fail anymore. So what have you added in that line L638 in init.cpp? | 22:32 |
| T42 | <Mattia990> I can see it in the log, I put Error: umount ... failed | 22:33 |
| T42 | <Mattia990> It is not in the log i sent in the paste, but I searched in my local log and there is | 22:33 |
| T42 | <elros34> so still rebots now? to fastboot mode or simply reboots | 22:34 |
| T42 | <Mattia990> It simply reboots when I run the droid-hal-init | 22:35 |
| T42 | <elros34> anything in strace? | 22:35 |
| T42 | <elros34> mask also user@100000 for now | 22:36 |
| T42 | <Mattia990> I masked also user | 22:41 |
| T42 | <Mattia990> This is strace: paste.opensuse.org/pastes/4d7b34f6d9bb | 22:41 |
| T42 | <Mattia990> And this is journalctl: paste.opensuse.org/pastes/832b15b2934f | 22:41 |
| T42 | <Mattia990> It still reboots after droid hal init | 22:41 |
| T42 | <elros34> is this reall whole strace? Why don't you include full command you are using? | 22:45 |
| T42 | <Mattia990> Because the console isn't permitting me it, it seems limited with the number of lines I can see (re @elros34: is this reall whole ...) | 22:47 |
| T42 | <elros34> increase buffer to infinity | 22:47 |
| T42 | <Mattia990> Thank you, I did it | 22:56 |
| T42 | <Mattia990> paste.opensuse.org/pastes/6638347bcea9 (re @elros34: increase buffer to i...) | 22:56 |
| T42 | <elros34> hm what about /proc/last_kmsg or /sys/fs/pstore/console-ramoops*? There must be something. | 22:59 |
| T42 | <Mattia990> Both are empty (re @elros34: hm what about /proc/...) | 23:01 |
| T42 | <Mattia990> I mean, there aren't files inside pstore and nothing in last_kmsg | 23:01 |
| T42 | <elros34> strange. Do you have all your kernel changes uploade? | 23:02 |
| T42 | <elros34> you could try also to run logcat after stracing droid-hal-init but you must be fast. BTW not sure if it's still relevant but when stracing do you first run all other commands from /usr/bin/droid/droid-hal-startup.sh? | 23:05 |
| T42 | <Mattia990> No, I haven't uploaded changes on git | 23:07 |
| T42 | <Mattia990> But basically I added the initramfs change and nothing else. | 23:07 |
| T42 | <Mattia990> Modified the defconfig and added that line for log in cmdline (re @elros34: strange. Do you have...) | 23:07 |
| T42 | <elros34> that defconfig changes I would like to see. Probably not related but did you add /etc/selinux files like in instruction? | 23:08 |
| T42 | <Mattia990> No, that i didn't do. Could be the problem? | 23:12 |
| T42 | <Mattia990> This is the defconfig: paste.opensuse.org/pastes/70adbeaf4c2a (re @elros34: that defconfig chang...) | 23:12 |
| T42 | <elros34> yeah but I want to see diff/your changes | 23:13 |
| T42 | <elros34> that is why uploading to github is most convenient | 23:14 |
| T42 | <elros34> I don't like this CONFIG_PANIC_TIMEOUT=-1 | 23:16 |
| T42 | <Mattia990> The problem is that that defconfig doesn't exist on the original github. | 23:21 |
| T42 | <Mattia990> It was pointing to a fail that doesn't exist, so I copied the default defconfig, added cepheus defconfig and with the mer script checked what was wrong (re @elros34: that is why uploadin...) | 23:21 |
| T42 | <Mattia990> I will try to remove it (re @elros34: I don't like this CO...) | 23:21 |
| T42 | <elros34> errr and you sure its correct => same as on lineage? | 23:28 |
| T42 | <elros34> dont remove CONFIG_PANIC_TIMEOUT change it to some positive number | 23:29 |
| T42 | <elros34> remind me why did you choose this particular base? | 23:30 |
| T42 | <elros34> isn't this similar device https://github.com/SailfishOS-vayu? | 23:31 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!