Igor Posted January 3, 2017 Posted January 3, 2017 Shall we start to finalise and prepare for images rebuild? Some new features were already documented, but there are some ongoing things and we are also preparing new download organisation, which could be included. - it looks like I manage to disable / merge little too much images, which needs to be built. @tkaiser was right I made a mistake, so please check .conf if I forgot something. - beta builds. we can't check all images, but some should be enough to assume everything is o.k. ... Do we have a minimum set yet? - I will try to enable beta testing protocol / call for testing within this build. What else should be included / finished and pushed into this build? Shall we put some H3 mainline builds out? At least for: OPi PC+, Zero, ... ? Is released due for 17.1. O.K.?
zador.blood.stained Posted January 3, 2017 Posted January 3, 2017 Shall we start to finalise and prepare for images rebuild? Some new features were already documented, but there are some ongoing things and we are also preparing new download organisation, which could be included. My concerns: 1. Wait for the mainline 4.9.1 to drop swconfig b53 patch for Lamobo-R1. If at least basic network configuration works on R1 with DSA, then the issue #511 will be resolved. 2. Is the firmware issue #484 resolved? 3. What is the status of M2+ legacy and mainline (issue #564)? If it works for you with and without network and/or HDMI, then it is solved, otherwise maybe it should be excluded from stable images for now? 4. It would be good to improve download pages as much as possible before the release: add kernel comparison tabs for mvebu, i.mx and all sunxi families; add "Known issues" tab (at least per $LINUXFAMILY, with a list of generic issues and board-specific issues like embedded wireless modules problems), add more text on nightly tabs to state that these are automated untested builds and no end user support is provided for them), ... Shall we put some H3 mainline builds out? At least for: OPi PC+, Zero, ... ? I would prefer to leave these as nightlies for now, but we can add nightlies for more boards that have good enough configurations, such as One, Zero, PC, ...
lvmc Posted January 3, 2017 Posted January 3, 2017 For 5.25, I suggest solving (issue #579) since it affects all H3 boards and offers a "bad first experience to Armbian" on boot.
zador.blood.stained Posted January 3, 2017 Posted January 3, 2017 For 5.25, I suggest solving (issue #579) since it affects all H3 boards and offers a "bad first experience to Armbian" on boot. This is quite simple - disable video support in u-boot for legacy releases for now, and deal with it properly later.
lvmc Posted January 3, 2017 Posted January 3, 2017 @zador.blood.stained, how to disable video on u-boot?
zador.blood.stained Posted January 3, 2017 Posted January 3, 2017 @lvmc After looking at GH issue #579 I found and applied @jernej's patch and it looks fine. Answering your question - to disable video in u-boot you have to remove CONFIG_VIDEO from respective board defconfig, it should be enough.
Christos Posted January 3, 2017 Posted January 3, 2017 @lvmc After looking at GH issue #579 I found and applied @jernej's patch and it looks fine. Answering your question - to disable video in u-boot you have to remove CONFIG_VIDEO from respective board defconfig, it should be enough. @zador.blood.stained, so, if we build a new H3 image, the boot artifacts will be gone? Its the 1d38c5b patch right? Good..
zador.blood.stained Posted January 3, 2017 Posted January 3, 2017 @zador.blood.stained, so, if we build a new H3 image, the boot artifacts will be gone? Its the 1d38c5b patch right? Good.. I ran a CLI Xenial image on OrangePI+2E without any artifacts with this commit.
tkaiser Posted January 3, 2017 Posted January 3, 2017 2. Is the firmware issue #484 resolved? IMO related: https://github.com/igorpecovnik/lib/commit/6df11c8889cac3c105bb6dea6e20310cd1a850df#commitcomment-20310982 Apart from that I also would prefer to fix download pages first.
Igor Posted January 5, 2017 Author Posted January 5, 2017 1. OK 2. Yes it works now, while cleaning the mess, as Thomas suggested, remain open. 3. Problems are only when something is on USB in all other cases I could not reproduce - when using a proper PSU. I suspect it has something to do with power stability. If some not so perfect source is used, board might not boot. There is nothing more to be done here. 4. I'll try to squeeze this into this release.
zador.blood.stained Posted January 5, 2017 Posted January 5, 2017 3. Problems are only when something is on USB in all other cases I could not reproduce - when using a proper PSU. I suspect it has something to do with power stability. If some not so perfect source is used, board might not boot. There is nothing more to be done here. Then it's a hardware problem, GH issue should be closed and this mess should be present in the "Known issues" tab on the download page.
Christos Posted January 5, 2017 Posted January 5, 2017 In my recent builds got this Processing file /home/christos/armbian/lib/patch/kernel/sun8i-default/ov5640-enhanced-driver-with-vfe-drc-table-fix.patch patch unexpectedly ends in middle of line I build always with RT enabled (option 4) so don't know if this is the cause, even though I'm not using the ov5640, just mentioning that FYI if you plan for a new release.
vincele Posted January 5, 2017 Posted January 5, 2017 Hello, Some new features were already documented Looks like 5.24 was forgotten...
zador.blood.stained Posted January 5, 2017 Posted January 5, 2017 Looks like 5.24 was forgotten... 5.24 is the current version for the nightly and user-built images, so nightly and manually built packages have higher version than stable release.
vincele Posted January 6, 2017 Posted January 6, 2017 5.24 is the current version for the nightly and user-built images, so nightly and manually built packages have higher version than stable release. So 5.24 will never be a release number ? I was only speaking of the Release_Changelog.md file content, maybe something like that could be added in this file to avoid surprising the readers: v5.24 - not released, used for the nightly or user-built images just my 2 cts...
Igor Posted January 6, 2017 Author Posted January 6, 2017 @vincele Since you have a good eye in spotting such things - I just gave you commit rights for lib.docs ... quick fixing can be more convenient way. We are already filled with various tasks. Thanks.
vincele Posted January 6, 2017 Posted January 6, 2017 Since you have a good eye in spotting such things - I just gave you commit rights for lib.docs ... quick fixing can be more convenient way. We are already filled with various tasks. Thanks. OK thanks. Is my proposition above the right thing to put in the release doc ? EDIT: DONE: https://github.com/igorpecovnik/lib.docs/commit/ed96d5ce1662ae2962770c04b85a9771e6e1e876 1
lvmc Posted January 6, 2017 Posted January 6, 2017 @lex, can you take a look at it? In my recent builds got this Processing file /home/christos/armbian/lib/patch/kernel/sun8i-default/ov5640-enhanced-driver-with-vfe-drc-table-fix.patch patch unexpectedly ends in middle of line I build always with RT enabled (option 4) so don't know if this is the cause, even though I'm not using the ov5640, just mentioning that FYI if you plan for a new release.
Recommended Posts