| Umeaboy | The link to the hot-hadk in the title of this channel is wrong. | 01:05 |
|---|---|---|
| Umeaboy | Gotta go to sleep. Take care. | 01:18 |
| T42 | <A_T_R> journalctl -b —no-pager log - | 02:18 |
| T42 | <A_T_R> https://pastebin.ubuntu.com/p/MwptRjczWG/ | 02:18 |
| T42 | <A_T_R> mount https://pastebin.ubuntu.com/p/74TDdjh8st/ | 02:26 |
| T42 | <郝生淡 %lastname%> Fwd from b3ak5qtf %lastname%: 😺😺😺 | 06:22 |
| T42 | <郝生淡 %lastname%> Garfield Coins (GARF)---Green Dogecoin | 06:22 |
| T42 | <郝生淡 %lastname%> We are a spontaneously established animal protection organization, and we hope to protect our planet by protecting wild animals. | 06:22 |
| T42 | <郝生淡 %lastname%> 🔥🔥🔥🔥 | 06:22 |
| T42 | <郝生淡 %lastname%> GARF details: | 06:22 |
| T42 | <郝生淡 %lastname%> Total amount: 100 Trillion, no team share, no pre-mining | 06:22 |
| T42 | <郝生淡 %lastname%> Contract address:0x6eb7c4b8f4a3ca01062d5b752dea5c629c5f1b15 | 06:22 |
| T42 | <郝生淡 %lastname%> ⚡️20% GARF is used as an airdrop to reward participants who help promote GARF.and you will receive 100 million GARF airdrops. | 06:22 |
| T42 | <郝生淡 %lastname%> Telegram: https://t.me/GarfcoinGlobal4 | 06:22 |
| *** wiktorek141 is now known as wiktorek140 | 06:45 | |
| T42 | <elros34> @A_T_R as you probably already know that you have only mounted system and vendor. Do you have more mount units in /usr/lib/systemd/system/*.mount? Are you sure your device do not use system-as-root? | 07:08 |
| T42 | <A_T_R> my device is SAR | 07:14 |
| T42 | <A_T_R> droid-config-X00TD/sparse at master · SailfishOS-sdm660/droid-config-X00TD · GitHub | 07:17 |
| T42 | <A_T_R> https://github.com/SailfishOS-sdm660/droid-config-X00TD/tree/master/sparse | 07:17 |
| T42 | <A_T_R> droid-config-X00TD/sparse/usr/lib/systemd/system at hybris-17.1 · SailfishOS-sdm660/droid-config-X00TD · GitHub | 07:19 |
| T42 | <A_T_R> https://github.com/SailfishOS-sdm660/droid-config-X00TD/tree/hybris-17.1/sparse/usr/lib/systemd/system | 07:19 |
| T42 | <elros34> you wont find them in droid-config but in droid-hal rpm or on device itself. If device is system as root then your /system is wrong. Change in fstab mount point from /system to / | 07:21 |
| T42 | <A_T_R> fstab - set system-as-root · SailfishOS-sdm660/device_asus_X00TD@d4c992b | 07:27 |
| T42 | <A_T_R> https://github.com/SailfishOS-sdm660/device_asus_X00TD/commit/d4c992b946a319c8b679a0d3acf523ae3329b544 | 07:27 |
| T42 | <elros34> make sure that change is also inluded in out/ and confirm list of mount units in droid-hal-device.log | 08:01 |
| T42 | <A_T_R> so i should build droid-hal-device too after changed anything in device/asus/X00TD ? | 08:11 |
| T42 | <A_T_R> —droid-hal * | 08:11 |
| T42 | <A_T_R> @elros34 THANKS AGAIN —IT'S BOOTED | 08:34 |
| T42 | <A_T_R> but ater ew seconds screen becomes blank | 08:39 |
| T42 | <A_T_R> https://pastebin.ubuntu.com/p/x7ZcYv2Nyp/ | 08:39 |
| T42 | <A_T_R> few* | 08:40 |
| T42 | <elros34> after changing something in device/* you need to make hybris-hal so all changes are include in out, then package these changes to droid-hal | 08:53 |
| *** wiktorek141 is now known as wiktorek140 | 11:01 | |
| *** wiktorek141 is now known as wiktorek140 | 11:20 | |
| T42 | <Verevka86> https://paste.ubuntu.com/p/Y84VsQrm8r/ | 11:44 |
| T42 | <Verevka86> I don’t know how to fix this problem, symlink doesn’t work :-( | 11:44 |
| T42 | <elros34> what about bind mount same as you for sure already do for ld.config? | 11:51 |
| T42 | <Verevka86> @elros34 [what about bind mount same as you for sure alr …], i will check it, thanks | 12:06 |
| mal | PSA: latest gst-droid should build successfully now even for 4.0.1 | 12:47 |
| T42 | eriki73 was added by: eriki73 | 13:14 |
| mighty17 | @elros34, can you help me with halium? I have having the same display issues like sailfishos with halium | 13:59 |
| mighty17 | and i cant understand how to fix it, you helped me do it in sfos so im asking | 13:59 |
| T42 | <elros34> Copy changes you made on sfos to halium:) I have never build halium. IIRC it was about setting HYBRIS_EGLPLATFORM=fbdev, EGL_PLATFORM=hwcomposer and making sure that you have permissions to some /dev/ nodes | 14:39 |
| mighty17 | Correct indeed, but halium doesn't have droid-hal or smth like that so the env variables idk how to set in build | 15:11 |
| mighty17 | Next the thing was to change /dev/dsscomp to system graphics 0660 | 15:11 |
| mighty17 | Sfos was android 6 with 3.0 kernel but halium is android 7 with 3.4 kernel (and different sources) | 15:12 |
| mighty17 | here's the strace for test_hwcomposer https://pastebin.ubuntu.com/p/8pFbHx4gb9/ | 15:32 |
| mighty17 | @elros34 ^^^ | 15:33 |
| mighty17 | and this is in logcat | 15:39 |
| * mighty17 < https://matrix.org/_matrix/media/r0/download/matrix.org/RUtRvIUIwSJPZRIkqBFSlzmB/message.txt > | 15:39 | |
| * mighty17 < https://matrix.org/_matrix/media/r0/download/matrix.org/QiKPVXzcwDwUqQDnHmFquPYE/message.txt > | 17:01 | |
| T42 | <elros34> @mighty17 my neither as I never built it but I suggest chmod a+rw /dev/* and then run minimer instead test_hwcomposer with mentioned env variables. | 17:41 |
| mighty17 | @elros34, does halium support minimer?? | 17:56 |
| mighty17 | `zypper in qt5-qtdeclarative-qmlscene` this part, what does it do (and maybe how to replicate it) | 18:14 |
| mighty17 | ok got it nvm | 18:18 |
| * mighty17 < https://matrix.org/_matrix/media/r0/download/matrix.org/GmxlDLZwcJitNSnOVhbtXeEZ/message.txt > | 18:45 | |
| mighty17 | ugh @elros34 | 18:45 |
| * mighty17 < https://matrix.org/_matrix/media/r0/download/matrix.org/FapVmNwfZPgFzcUytUOLIPUc/message.txt > | 18:46 | |
| * mighty17 < https://matrix.org/_matrix/media/r0/download/matrix.org/KipvvMTDsIDrSBelzBHtxmsb/message.txt > | 18:46 | |
| mal | rinigus: ping | 19:00 |
| rinigus | mal: pong | 19:00 |
| mal | rinigus: which soc does tama have? | 19:01 |
| rinigus | mal: sd845 | 19:02 |
| mal | rinigus: ok, related to hall sensor maybe try to apply this kind of change for your kernel https://github.com/sonyxperiadev/kernel-defconfig/pull/98/files | 19:02 |
| mal | rinigus: didn't you use 4.9 kernel on that? | 19:04 |
| rinigus | mal: thanks. we use 4.9 for aosp9 port; 4.14 for aosp10 | 19:05 |
| rinigus | I saw few days ago that you pushed hall sensor updates - so would be good to test. | 19:06 |
| piggz | mal: Thaodan: did obs ever get fixed aarch64? | 19:06 |
| mal | rinigus: also check that devicetree has the needed entry for hall sensor | 19:06 |
| mal | piggz: not yet at least | 19:07 |
| mal | maybe I should ask again | 19:07 |
| rinigus | mal: how do you check that? | 19:07 |
| piggz | mal: tbuilder build then! | 19:07 |
| T42 | <elros34> @mighty17 you are missing qt5-qpa-hwcomposer-plugin | 19:07 |
| rinigus | piggz: sure, no need to wait :) | 19:07 |
| mal | rinigus: looks for "rohm,bu520x1nvx" in dts* files | 19:08 |
| mal | rinigus: on 4.14 kernel tama dts has that | 19:09 |
| rinigus | mal: let's see... | 19:09 |
| mal | rinigus: and also check your defconfig for that CONFIG_INPUT_BU520X1NVX | 19:10 |
| piggz | rinigus: what the logic for the package version? seems it took the git tag, not the version from the spec right? | 19:12 |
| rinigus | piggz: it was supposed to take from spec, but I have a feeling something changed in 4.1. haven't looked into it yet, just noticed while working on pure maps | 19:14 |
| piggz | rinigus: this was a 4.0 target | 19:15 |
| rinigus | hmm, on platformsdk? | 19:15 |
| piggz | yes | 19:15 |
| piggz | not sure what release of the psdk is | 19:15 |
| piggz | PlatformSDK [piggz@linux-f1uu advanced-camera-build]$ ssu re | 19:16 |
| piggz | WARNING: ssu.ini does not seem to be writable. Setting values might not work. | 19:16 |
| piggz | Device release is currently: 4.0.1.48 | 19:16 |
| rinigus | mal: yes, 4.9 has that defined in dts for tama | 19:16 |
| rinigus | so, would have to look into it and enable in config | 19:16 |
| rinigus | piggz: for me, in docker version is taken from SPEC. in platformsdk, from git tag | 19:17 |
| rinigus | let me look into why it is so | 19:17 |
| rinigus | piggz: would it fit if shadow builds get git tag? otherwise the one from the spec. this is when tested on platformsdk | 19:30 |
| piggz | i have shadow turned off | 19:30 |
| rinigus | piggz: good choice for ports, as far as I have seen | 19:31 |
| piggz | this is just apps | 19:32 |
| Umeaboy | Hi!I'm trying to shorten to HADK to actually contain only the commands that I shoulö perform without having to read the pre info all the time. When is it fitting to run the patch command before building HAL? Here's what I've done so far: https://ghostbin.co/paste/9n27 | 19:33 |
| Umeaboy | I want to make sure I don't miss anything again. | 19:34 |
| rinigus | piggz: then I don't know. it looks a bit random to me, but I am probably missing something. | 19:36 |
| rinigus | piggz: see mb2 --help towards the end, -x and -X options. | 19:37 |
| rinigus | I don't use any of them in generated makefile | 19:37 |
| piggz | rinigus: doesnt bother me greatly ... just a little messy ;) | 19:40 |
| T42 | <b100dian> mal: looks like a typo here https://github.com/mer-hybris/hybris-patches/blob/hybris-17.1/hardware/libhardware/0002-hybris-Search-for-libraries-first-from-usr-libexec-d.patch#L34 | 19:41 |
| T42 | <b100dian> is that /hw/ override needed actually? | 19:42 |
| mal | @b100dian where? | 19:42 |
| T42 | <b100dian> the HAL_LIBRARY_PATH3 is used and then tested against HAL_LIBRARY_PATH4 | 19:43 |
| rinigus | piggz: it is, I agree. let me fix it | 19:44 |
| T42 | <b100dian> I was debugging yesterday my hwcomposer service but mine.. loads libs from vndk-sp-29 | 19:45 |
| rinigus | piggz: update tbuilder and try again. should be fixed now | 19:46 |
| mal | @b100dian do you have some files in that folder lib/hw or lib64/hw on your device | 19:48 |
| Umeaboy | Can someone please change the Install repo link in the HADK to the proper page please? | 19:49 |
| Umeaboy | It points to https://source.android.com/setup/build/downloading#installing-repo while it should point to https://source.android.com/setup/develop#installing-repo | 19:50 |
| T42 | <b100dian> mal: only two, android.hidl.memory@1.0-impl.so and audio.a2dp.default.so | 19:51 |
| mal | @b100dian ok, not sure how critical those are, what issue are you trying to fix now? | 19:52 |
| T42 | <b100dian> still at stack corruption graphics.composer service. I've started to scp in symbolicated libraries and trying to gdb | 19:52 |
| T42 | <b100dian> so far it seems that only services using sphal (same-process HAL) are affected | 19:53 |
| T42 | <b100dian> which is display and camera:) | 19:53 |
| mal | @b100dian ok, on fp3 /usr/libexec/droid-hybris/system/lib64/hw doesn't have anything in it (neither does lib/hw) | 19:54 |
| piggz | rinigus: thx | 19:54 |
| Umeaboy | mal: Can you add a hybris-17.1 branch on https://github.com/mer-hybris/local_manifests/ ? | 19:55 |
| T42 | <b100dian> mal: thanks. I'm cleanly building the system/ now, maybe I should not have either. I may have made switches between building with the option "BUILD_WITHOUT_VENDOR" on and off and there may be leftover files | 19:55 |
| T42 | <b100dian> mal: do you also have this output as @A_T_R has? vndksupport: Loading /vendor/lib64/hw/gralloc.sdm660.so from current namespace instead of sphal namespace. | 19:57 |
| mal | Umeaboy: ok, but almost nobody uses that local_manifests repo | 19:57 |
| T42 | <b100dian> oh, there's a repo for manifests! | 19:58 |
| Umeaboy | mal: That part should be removed from the HADK then. | 19:58 |
| mal | @b100dian I also have those messages | 19:59 |
| T42 | <b100dian> mal: I think that means that your device is not treble? launched with android 8 or ..? I'm looking at this table https://source.android.com/devices/architecture/vndk/linker-namespace#linker-namespace-isolation | 20:03 |
| mal | @b100dian it could also be some configuration issue, not sure, but this device was originally released with android 9 and now has android 10 | 20:04 |
| mal | Umeaboy: it's still possible to use that, not in anyway popular though | 20:05 |
| T42 | <ruspartisan> Does anybody know, do I need | 20:12 |
| T42 | <ruspartisan> Does anybody know, do i need arm64 to run pinephone image, assuming i have arm32 kernel? Or kernel and cpu must be 64 bits? | 20:13 |
| T42 | <ruspartisan> @adampigg ? | 20:14 |
| T42 | <ruspartisan> I've recompiled mesa and i no longer see errors about missing dri files, but there's still no gui | 20:15 |
| T42 | <adampigg> current build is a 32bit userspace, | 20:15 |
| T42 | <ruspartisan> So 32 bit arm and kernel should be fine? | 20:16 |
| T42 | <adampigg> yes | 20:18 |
| T42 | <ruspartisan> Ok, will try some more. Thank you | 20:19 |
| piggz_ | rinigus: next tbuilder issue | 20:40 |
| piggz_ | ive added another target | 20:41 |
| piggz_ | built packages have differnt version numbers, which jolla harbour doesnt allow | 20:41 |
| piggz_ | how do i bring them into sync? | 20:41 |
| piggz_ | deleting build/ helped | 21:14 |
| T42 | <leha155> @mal [@leha155 either replace CONFIG_SECURITY_SELINU …], Hello, I rebuilt the kernel and removed the line "CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE = 0" But I still have a black screen instead of the GUI. My log has the following errors | 22:16 |
| T42 | <leha155> "Feb 11 20:28:15 MiMix2s lipstick [5231]: library" libGLESv2_adreno.so "not found | 22:16 |
| T42 | <leha155> Feb 11 20:28:15 MiMix2s lipstick [5231]: library "eglSubDriverAndroid.so" not found | 22:16 |
| T42 | <leha155> Feb 11 20:28:15 MiMix2s lipstick [5231]: library "libandroidicu.so" not found " | 22:16 |
| T42 | <leha155> although these files are present in / vendor / lib64 / egl. Could this be the reason? | 22:16 |
| T42 | <leha155> journal log https://pastebin.ubuntu.com/p/kvxy4jRxXp/ | 22:16 |
| T42 | <leha155> strace -f test_hwcomposer output https://pastebin.ubuntu.com/p/FDKQjV39gf/ | 22:16 |
| mal | @leha155 something strange in the strace, "/dev/kgsl-3d0", O_RDWR|O_SYNC) = -1 ENOENT (No such file or directory) | 22:36 |
| mal | it fails after that | 22:36 |
| mal | @leha155 looks like firmware loading fails, can you find a630_sqe.fw anywhere on the device? | 22:42 |
| mal | do you have these in your defconfig: | 22:56 |
| mal | CONFIG_FW_LOADER=y | 22:56 |
| mal | CONFIG_FW_LOADER_USER_HELPER=y | 22:56 |
| mal | CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y | 22:56 |
| Umeaboy | mal: Why isn't the part of applying hybris patches mentioned in the HADK? Is there a speciai reason for it? | 23:18 |
| Umeaboy | I mean, you can't build HAL successfully without it. | 23:18 |
| Umeaboy | If you tell users to use the HADK to port Sailfish to a new device that part is vital. | 23:19 |
| mal | partially because the example there is for older base | 23:21 |
| Umeaboy | mal: OK, so for Lineage-17.1 it's not needed anymore? | 23:45 |
| Umeaboy | I want to know if I can remove it from my shortened HADK. | 23:46 |
| Umeaboy | As I wrote earlier I'm trying to find out where I have done wrong in the past so that I can debug it easier. :) | 23:46 |
| Umeaboy | I don't know if Lineage 17.1 is to be considered much older. 18.1 has just been released. | 23:55 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!