T42 | <b100dian> mal: thanks for looking into this. Gstreamer is very opaque to me. But I suppose timestamp come from the droidmedia side? | 07:18 |
---|---|---|
T42 | <adampigg> in what crazy world do you want video frames out of order :D | 07:40 |
T42 | <edp_17> @elros34 : I ran that make * defconfig-bluetooth command: https://paste.ubuntu.com/p/t2xXMkBPyD/ | 07:46 |
*** rdr_ is now known as rdr | 10:10 | |
steadfasterx | good old IRC times.. 😄 oh my.. hi everyone.. will look into porting sailfishOS for at least the fairphone6 and/or shiftphone8.. while having years of experience with Android building, zero experience with SFOS yet, so this will be fun I guess.. | 10:20 |
rinigus | mal: any tips on how can we check our video streams? | 10:52 |
mal | rinigus: I saw the timestamps in "GST_DEBUG=videodecoder:7 jolla-gallery | grep finish\ frame.*sync" and there checking how the PTS value look like | 10:55 |
rinigus | mal: thanks! will try to check tonight | 10:56 |
T42 | <edp_17> steadfasterx, Have you seen/read HADK, HADK-faq and HADK-hot documents? They are useful when you start porting. | 11:29 |
steadfasterx | T42: bookmarked them already, will review them once starting with this... and likely come back with questions then 😄 | 11:31 |
mal | steadfasterx: I already have some initial things done for fp6, waiting for lineage for it to become buildable | 11:33 |
steadfasterx | mal: great! I saw you ported already FP4 and 5 so I hoped you will go with FP6 as well.. will be a good starting point for me as I could base on your work to learn how this all is tight together.. | 11:40 |
T42 | <elros34> @edp_17 that was with .config removed? | 14:22 |
T42 | <elros34> are you sure you have applied (.config) all required kernel configs from faq? | 14:24 |
T42 | <elros34> like CONFIG_CRYPTO_CMAC? | 14:29 |
T42 | <edp_17> Yes. Before I ran the command, I deleted the .config. | 15:15 |
T42 | <edp_17> Yes, I did. All of them are still in kernel/ xyanogenmod_hammerhead_defconfig. However, this particular one is not in out/.../obj/KERNEL_OBJ/.config | 15:18 |
T42 | <edp_17> The other four are there. Only this CONFIG_CRYPTO_CMAC is missing. | 15:20 |
T42 | <elros34> get used to checking .config, defconfig is "useless" as you already know | 15:24 |
T42 | <edp_17> Yes, thanks. So, back to the drawing board :) | 15:47 |
T42 | <elros34> didn't you forget to apply cmac patch? | 15:47 |
T42 | <edp_17> No, I haven't. It's there: https://github.com/edp17/android_kernel_lge_hammerhead/commits/hybris-12.1-edp17/ | 15:49 |
T42 | <edp_17> Hmm, wait, in my local the kernel is on hybris-12.1-edp17-bluez5 branch. | 15:50 |
T42 | <edp_17> Hmm, I've rebuil kernel & backports then local rpms and installed them on devices. The uname -a gives me the same name that I have in /lib/modules. However, when I try to modprobe a module from there, it says: "modprobe: ERROR: could not insert xxxx: Exec format error" | 16:18 |
T42 | <edp_17> Just like when the module is not for this kernel. | 16:18 |
T42 | <edp_17> I am burning a new image as could be the multirom interferes with this. | 16:19 |
T42 | <edp_17> I doubt it, but you never know. | 16:19 |
T42 | <elros34> file <kernel module.so>, and check dmesg | 16:22 |
T42 | <edp_17> :1 | 16:26 |
rinigus | mal: as I am not sure how to read it, here is GSTREAMER log: https://dpaste.org/J8Zxm/raw | 18:09 |
rinigus | that's in nagara - same issue with playback. works fine on PC | 18:09 |
*** xmn_ is now known as xmn | 18:29 | |
mal | rinigus: that specific part didn't have timestamp issue but that was very small part, and you didn't grep the output to get just the relevant parts | 18:58 |
rinigus | mal: sorry - was not paying attention. here it is grepped for finish frame: https://dpaste.org/gerzT/raw | 19:02 |
rinigus | but that's output from gallery where the playback just stops in this occasion | 19:04 |
rinigus | I wonder where ubuntu devs - in particular ofono ones - are hanging out. would like to discuss ims and hangup of the calls. maybe they know how to do it | 19:09 |
rinigus | or debug it. | 19:09 |
rinigus | setting persist.vendor.radio.ril_extra_debug 1 didn't make the ims stack suddenly more chatty either | 19:11 |
mal | I don't see that issue in your log | 19:27 |
rinigus | mal: in my case, for this video, playback just stops and never recovers. | 19:28 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!