Skip to content

ppsh compatible Qt5 configuration

Yann E. MORIN buildroot 1 version New

Get series with b4
b4 am 20190310160630.GD25009@scaer

Versions newest first

  1. v1 Latest0 of 2 patches 0 repliesNew

Patches in v1 No applies recorded yet

None of the 2 patches the cover letter announces are in this archive.

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [Buildroot] [PATCH 0/2] ppsh compatible Qt5 configuration

Thomas, All,

On 2019-03-09 15:22 +0100, Thomas Petazzoni spake thusly:

Quoted 31 lines
Quoted 11 lines · On Sat, 9 Mar 2019 15:10:20 +0100 Andreas Naumann <[email protected]>…

you were asking for ideas on "Making the Qt5 packaging compatible with per-package folder". Lately I had some time to look into it and came up with the following patches. Basically they use a custom qt.conf to manipulate qmakes builtin pathes. Details are in the commit-messages.

I did a compile test with almost all qt5 modules available, as well as other buildroot provided qt5 applications. At the moment I dont have equipment for runtime tests. As far as I could see the install procedure now works as expected, no more failures.
Unfortunately webengine and webkit-example failed sometime during the build but it seems that these were unrelated issues.

Thanks a lot for looking into this!

However, there is one thing that this doesn't seem to address: the installation part.

Currently, the staging installation is just:

make install

with no DESTDIR/INSTALL_ROOT, and therefore for the target installation, we have to manually install libraries, QML files and other stuff, which is really annoying to maintain. Ideally, we would like Qt5 to behave like most other packages, and do:

	make INSTALL_ROOT=$(STAGING_DIR) install
Quoted 8 lines

for the staging installation, and:

	make INSTALL_ROOT=$(TARGET_DIR) install

for the target installation. This change is not strictly needed for per-package host/target directories, but I really would like to take the opportunity of changing all this Qt5 logic to also address this related problem.

I still believe a long-term goal would be to actually only ever install everything in staging, and generate target right at the end as a copy from staging as a starting point for all the cleaning we already do in target-finalize.

This would have nicely worked-around this issue.

Regards,
Yann E. MORIN.

Quoted 15 lines

I had prepared something like this:
https://github.com/tpetazzoni/buildroot/commit/05e776abf531319a749f5a7d65f265979ea6c4ab,
but I don't remember if it was working completely or not.

Thanks,

Thomas

Signature · 7 lines
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'