| T42 | <edp_17> I cannot work out why I get that error. I outputted value of some vars with $(warning) : https://paste.ubuntu.com/p/NfYQqt3gPx/ | 00:13 |
|---|---|---|
| T42 | <edp_17> If I change the text backports: at line 48 to backports-edp17, the error message also changes to "writing to readonly directory: "backports-edp17"". So, I guess this error message is a red-herring as that folder in message does not exist. | 00:16 |
| T42 | <edp_17> Also, I have no idea where from it builds the .config file. | 00:17 |
| *** ChanServ changes topic to "Sailfish OS Porters | checks in-order: https://sailfishos.org/hadk http://bit.ly/faq-hadk http://bit.ly/port-channel-log http://bit.ly/always-grep-irc-logs | http://bit.ly/port-status | https://forum.sailfishos.org/c/hw-adaptation/10 | old http://bit.ly/port-requests | http://bit.ly/irc-house-rules | http://bit.ly/jolla-oss-contributors-policy" | 00:41 | |
| mal | @edp_17 it doesn't get far enough to create the .config file because it somehow things the .mk file is not valid | 09:20 |
| T42 | <edp_17> mal : Yeah, I've tried to understand the syntax but I couldn't. I understand line 50 and 51 are executing the relevant functions defined at line 33 and 38. However, I don't know what line 45 and 48 are doing. | 09:40 |
| mal | 48 just defines build target names backports which has dependency to the config generation target | 09:43 |
| mal | and 45 defines the config generation target | 09:44 |
| T42 | <edp_17> So, should the build target name "bakcports" be a directory somewhere or just an internal command label within the .mk file? | 09:51 |
| mal | no | 09:57 |
| mal | it's just a label, not sure what is wrong in the syntax | 09:57 |
| T42 | <edp_17> I see. so line 45 is also a label? | 09:58 |
| T42 | <edp_17> I think understand now. So, at line 48 it jumps back to line 45 (label) and executes line 46, then that fails and the script reports as a problem with "backports" at line 48. Am I right or am on a wrong track? | 10:01 |
| mal | I thought it fails at line 49? | 10:15 |
| T42 | <edp_17> mal : Yes, you're right. It fails at line 49. However, in that case I don't understand why didn't generate the .config in external/backports-bluetooth folder. Or if it couldn't do that, why there was no error about it. | 10:32 |
| T42 | <edp_17> Interesting. I've commented out lines 46, 49, 50 and 51. Then added "$(warning "hello") right after line 45. Then, I got the same error but mentioned line 48 in there. | 10:57 |
| T42 | <edp_17> Therefore, I don't think anything is wrong with those line but something wrong before them. That's how I meant was that error message looks like a red-herring. | 10:58 |
| T42 | <edp_17> mal : In "backports-bluetooth/Makefile.kernel" there is "export BACKPORT_DIR = backports/". However, in "backports-bluetooth/Android.mk" the BACKPORT_DIR declared as $(call my-dir) whic is "external/backports-bluetooth". Is this correct? | 11:18 |
| mal | I need to check those makefiles later | 12:44 |
| T42 | <edp_17> mal : Thanks! | 12:44 |
| T42 | <edp_17> Hi all! Another question about a different device. Since 4.4.0.72 vibra and key tones do not work. | 15:06 |
| T42 | <edp_17> Permission of /sys/class/timed_output/vibrator/enable is system/input. | 15:06 |
| T42 | <edp_17> journal: https://paste.ubuntu.com/p/nRg4tgPcJJ/ | 15:07 |
| T42 | <edp_17> debug of ngfd: ngfd -vvv: https://paste.ubuntu.com/p/7DG9xW9sCc/ | 15:07 |
| T42 | <edp_17> Nothing happened when I echoed 1 (or 200) into /sys/class/timed_output/vibrator/enable is system/input . | 15:08 |
| T42 | <edp_17> Although, when I echoed as root, the device vibrated. | 15:09 |
| piggz | @edp_17 permissions are set in android .rc files, patch them and rebuild hybris-hal | 15:36 |
| T42 | <edp_17> Permissions are correct. | 15:40 |
| T42 | <edp_17> Since, as root, I echoed 256 into /sys/class/timed_output/vibrator/enable, I also can echo into it as user and vibra reacts. Even after reboot, so permissions are correct. | 15:41 |
| T42 | <edp_17> When I close an app, the device vibrate slightly and I hear the closure sound although, very-very quiet. | 15:41 |
| T42 | <edp_17> However, neither in CSD nor in Fingerterm, vibra doesn't work. Hm, odd. | 15:43 |
| T42 | <b100dian> does keyboard vibrate? | 21:28 |
| T42 | <edp_17> No. | 21:44 |
| T42 | <b100dian> I think you are missing qt haptics. This might be fixable through making qt talk to ngfd like https://github.com/sailfishos-on-tucana/droid-config-tucana/commit/15a8a30881caaf6b614d7af3bc641de0b6383883 | 22:04 |
| T42 | <b100dian> But there may be other solutions I am not aware of, for timed_output compatible devices. | 22:05 |
| mal | those go through the same things | 22:44 |
| T42 | <b100dian> do you mean qt (e.g. keyboard) vibration goes through the same thing as ngfd? I did not find that true by default on a custom effects device like mine. That's why I am not sure that the advice applies to a device that works with timed_output, to use libngf-qt5-qtfeedback may indeed not be necessary | 22:55 |
| mal | I meant the different types of backends all go through same code | 23:01 |
| mal | official devices also have Requires: libngf-qt5-qtfeedback in patterns | 23:02 |
| T42 | <b100dian> I see, this is the way:) | 23:04 |
| mal | here is an example of the change https://github.com/mer-hybris/droid-config-sony-ganges-pie/commit/9598012a6f4f739090dd46b1782955d76411f507 | 23:04 |
| T42 | <b100dian> Thanks! Let's see how that works for @edp_17 | 23:09 |
| T42 | <b100dian> btw, creating libngf-qt5-qtfeedback was an excellent idea and I admit I was suprised in a good way that the two mechanisms were merged, thanks! | 23:11 |
| mal | yeah, makes maintenance easier when there is no duplication of some of the code | 23:16 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!