On Sunday, 13 March 2022 05:51:39 PDT Alexey Rochev wrote:
> You mean second component of version by minor, right? So supported range
> for 6.3 is 6.0-6.7 (or 6.3-6.7)?

It would if we had made that decision earlier. Starting with 6.3, it will be 
N-4 to N+4, so for 6.3 it is 6.3 to 6.7.

> I think it would be a good idea to test this in CI builds. If support of
> different host Qt versions is not guaranteed then users can't rely on
> binary Qt builds provided by their distributions (there are at least two
> Linux distros that have Qt 6 packages now, Fedora and openSUSE Tumbleweed),
> forcing them build host Qt from source if they want to cross-compile it.

Yes, but we also need to limit the amount of testing. Maybe test at release 
time that you can still cross-compile the past full releases and cross-compile 
this release with those past releases as host? That way, it's not an N^2 
problem, but only N*2..

We don't also need to *cross* compile. We can set QT_HOST_TOOLS and make 
another host build. This is the type of build that interests me. It's 
currently working, but with patches not yet integrated into dev and my host 
build is not THAT far from my other host build -- only by a few weeks at most.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel DPG Cloud Engineering



_______________________________________________
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest

Reply via email to