| Umeaboy | Anyone around to answer? I have just retried syncing with --fetch-submodules and there is still 3 repos failing to fetch from. | 06:09 |
|---|---|---|
| Umeaboy | Sorry..... 4 repos. | 06:09 |
| Umeaboy | About the SHA1. | 06:09 |
| Umeaboy | Where do I need to make the change locally? | 06:09 |
| T42 | <b100dian> Thaodan: the newer one is what I use | 08:21 |
| Umeaboy | I'll save the output....... I have to sleep a little while. | 09:14 |
| Thaodan | b100dian: When did you built it, which ref? | 13:04 |
| T42 | <b100dian> thaodan: I mean I use the separate miniaudiopolicy not the env.mk change in droidmedia. But this is a good question, it is probably an old build. The version on the phone is 202107062252 | 14:08 |
| T42 | <b100dian> Also, I think it is this one https://github.com/sailfishos-sony-tama/miniaudiopolicy | 14:13 |
| T42 | <b100dian> I am not close to the computer to check the ref though.. | 14:17 |
| T42 | <elros34> Is it normal that I have "gecko-camera[31471]: droid-codec init" when trying to play video in browser? | 14:50 |
| Thaodan | b100dian: there's a newer one that is just a fake miniaudiopolicy, it should work beter | 14:56 |
| T42 | <b100dian> This one, right? https://github.com/sailfishos/droidmedia/pull/92 | 15:06 |
| Thaodan | ye | 15:06 |
| Thaodan | s | 15:06 |
| T42 | <elros34> hm so after removing gecko-camera-droid-plugin gmp-droid start and videos in browser also works | 15:17 |
| piggz_ | argh, hit again with OBS problem https://build.sailfishos.org/package/show/nemo:devel:hw:pine:dontbeevil/kernel-adaptation-pine64# | 15:25 |
| piggz_ | unable to resolve github.com | 15:25 |
| piggz_ | lbt: ^^ | 15:25 |
| T42 | <elros34> I think I hit the same but for droid-config submodule | 15:33 |
| piggz_ | yeah, is happening a lot | 15:35 |
| piggz_ | mal: Thaodan: what is the error here? https://paste.mozilla.org/b00BT30F | 19:58 |
| piggz_ | is it the line with "relocation R_AARCH64_ABS32 against" | 19:58 |
| mal | piggz_: hmm, strange, where is that built? | 20:13 |
| piggz_ | mal: im building that locally in the sdk for now as OBS isnt wrking | 20:16 |
| Mister_Magister | obs is working tho | 20:17 |
| Mister_Magister | lbt just fixed it | 20:17 |
| piggz_ | ah cool, ill try there too | 20:18 |
| piggz_ | mal: quite frustrating to get to the end of a looooooong build and get such weird failure | 20:18 |
| Umeaboy | Thoadan: You here? | 20:18 |
| Mister_Magister | piggz_: if jolla didn't screw up obs, you could just do osc build and chroot into the build system after it fails to figure stuff out or continue build without clearing build system but here we are | 20:19 |
| mal | piggz_: yeah, there is that one warning earlier in log also "WARNING: EXPORT symbol "gsi_write_channel_scratch" [vmlinux] version generation failed, symbol will not be versioned." | 20:19 |
| Umeaboy | You guys know how to fix the SHA1 issue when syncing? | 20:20 |
| Umeaboy | I can't get passed it. | 20:20 |
| Umeaboy | Some commits ids are wrong. | 20:20 |
| mal | what issue? | 20:21 |
| Umeaboy | mal: repo sync --fetch-submodules fail to finish due to wrong SHA1. | 20:22 |
| Umeaboy | On 4 repos. | 20:23 |
| Umeaboy | I can recreate it all the time. | 20:23 |
| Umeaboy | No matter what OS I use. | 20:23 |
| piggz_ | i guess those sha;s dont exist in the repos? | 20:23 |
| Umeaboy | I'll give you the output soon. | 20:25 |
| Umeaboy | https://pst.klgrth.io/paste/kz3rn | 20:29 |
| Umeaboy | Your thoughts? | 20:32 |
| Umeaboy | I've done nothing wrong from my side. | 20:33 |
| mal | can you show the manifets you use? | 20:43 |
| Umeaboy | mal: Content of $DEVICE.xml is here https://pst.klgrth.io/paste/a7vap | 20:48 |
| Umeaboy | I haven't touched any other file. | 20:49 |
| mal | you are using normal hybris-18.1 branch? | 20:56 |
| Umeaboy | Yes. I tried both. | 21:05 |
| Umeaboy | Same result. | 21:05 |
| T42 | <elros34> Does it cause any issues when you build hybris-hal? I think these wrong commits were reported long time ago | 21:08 |
| Umeaboy | What more repos do I need to add to my manifest? The exynos9820-common right? | 21:10 |
| T42 | <elros34> check lineage.dependencies in your device repo | 21:11 |
| piggz_ | Umeaboy: are there errors in the manfest, possibly invalid characters at the end of the sha1 in the file? | 21:15 |
| T42 | <elros34> more likely git force push | 21:16 |
| Umeaboy | piggz_: Yeah...... there is an invalid character, but I didn't add it. | 21:19 |
| Umeaboy | The commit in that repo did. | 21:19 |
| mal | which repo? | 21:31 |
| Umeaboy | mal: 4 of them. Read https://pst.klgrth.io/paste/kz3rn | 21:50 |
| T42 | <elros34> why dont you either ignore these wrong commits or fix them and move forward? | 21:53 |
| Umeaboy | I'd fix them if I knew where to edit. | 21:56 |
| mal | I meant you said "there is an invalid character, but I didn't add it" but I have no idea what that is referring to | 22:03 |
| Umeaboy | mal: Row 238. | 22:12 |
| Umeaboy | The ^ in the end shouldn't be there right? | 22:13 |
| Umeaboy | Even row 236. | 22:13 |
| T42 | <elros34> if you do not know where then ignore it. If build fails then you will worry about it, didn't you port this beyond2lte for a long time already? you need to speed up little | 22:15 |
| Umeaboy | elros34: I am trying. | 22:19 |
| Umeaboy | I have only ported Lineage before and I got help to port Sailfish OS to the Sony Xperia 10. | 22:20 |
| T42 | <elros34> trying? so do you run make hybris-hal finally? | 22:22 |
| T42 | <elros34> Did you follow steps from hadk-faq about hybris-18 first (hybris-patches)? | 22:23 |
| mal | show the actual full manifest, not just the device manifest | 22:27 |
| Umeaboy | elros34: There is no section named hybris-18.1 in the hadk-faq. Newest is for hybris-17.1 | 22:30 |
| Umeaboy | mal: Coming up....... | 22:31 |
| Umeaboy | mal: You to be clear...... You mean from ~/hadk/.repo/manifests/default.xml ? | 22:33 |
| Umeaboy | Right? | 22:33 |
| T42 | <elros34> yeah, but applying patches step is same, you can find few more 18 related advices here (some are not valid anymore): collabedit.com/kr9xx | 22:34 |
| T42 | <elros34> about commit, you can cllearly see it's wrong when you open submodule: https://github.com/Etar-Group/android_frameworks_opt_colorpicker/tree/ca69f1e26e84490284abab46dd7929613d8cb325 | 22:34 |
| mal | Umeaboy: well what ever the .repo/manifest.xml points to, afaik that is some symlink | 22:39 |
| Umeaboy | I'm not trying to be rude, but could you perhaps seperate the ones for 18.1? | 22:42 |
| mal | ? | 22:44 |
| Umeaboy | elros34 said that some aren't valid anymore. How to tell if they are or not? | 22:45 |
| Umeaboy | mal: Content of the sym linked default.xml: https://pst.klgrth.io/paste/symrk | 22:47 |
| mal | Umeaboy: what happens if you run repo sync without "--fetch-submodules" | 23:05 |
| Umeaboy | mal: Then it syncs just fine. | 23:05 |
| mal | so then just continue | 23:06 |
| mal | I remember now that the Etar repo is broken | 23:06 |
| Umeaboy | That doesn't explain the other ones though. | 23:06 |
| Umeaboy | For instance: error: Cannot fetch platform/external/python/pybind11/tools/clang from https://android.googlesource.com/platform/external/wjakob/clang-cindex-python3 | 23:07 |
| Umeaboy | That is NOT related to Etar. | 23:07 |
| mal | well no idea about those, if you can build hybris-hal then everything is fine | 23:11 |
| Umeaboy | I'll check. | 23:12 |
| Umeaboy | Syncing finished successfully now. | 23:12 |
| Umeaboy | NO errors. | 23:12 |
| Umeaboy | mal: It's a tad confusing that the HADK tells you to make your own manifest and when you run breakfast $DEVICE you see an error about duplicate path in roomservice.xml. I guess it's because the device already has been ported to Lineage and has it's own manifest already. | 23:35 |
| Umeaboy | I'll pull my manifest and resync. | 23:35 |
| Umeaboy | Nope. That didn't help. | 23:49 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!