Hi All!
On 7/6/22 01:24, Miro Hrončok wrote:
On 06. 07. 22 1:17, Miro Hrončok wrote:
On 06. 07. 22 0:14, Kevin Kofler via devel wrote:
Stephen Smoogen wrote:
Hyperbole aside, it isn't a joke. Looking at the chain we see a common
If mesa's i686 support should be removed, then this proposal would need to be reverted.
I had filled bugs against all native direct dependencies of java (eg subversion) and against all packages, which transitively depends on java, and are themsleves important (being dependence of 78+ other pkgs[that is where the non linear
curve really started to grow]). All direct no-arch deps got injected ExclusiveArch: %{java_arches}. See https://fedoraproject.org/wiki/Changes/Drop_i686_JDKs#Workflow; there are also exact listings of what was filled bug agasint.
problem where subversion relies on java-11-openjdk and without it is going
to cause a lot of packages to be removed. Either subversion needs to lose
that requirement or a lot of packages are going to get removed as failure
to build in i686.
libglvnd-glx<-mesa-libGL<-libxcb<-doxygen<-git<-subversion<-java-11-
openjdk-devel
This email isn't a comment about it being a good or bad thing.. it is just
what is being presented.
I do not see why bogus "bug" reports are filed against a bazillion packages
when subversion is the only package that needs to be fixed.
I cant agree. Of course subversion is main to act, but afaik all its dependence should be warned. Otherwise they would suddenly got very weird FTBFS on i686.
My investigations shown, probably nothing will be affected by removal of jdk on i686 and the fix in subversion and automake/autotools is fluid and, again, will damage nothing. However I'm not sure. I can not possibly see into all details of
all affected packages.
Where casual pkg which is being depndent by nothing, may simply get weird FTBFS, but not a package, on which half of the system depend.
File a bug against subversion, put it on the release blocker tracker, and do
not waste everyone else's time.
Thus saying, is it really waste of time? I really doubt. It is making people aware of quite major happening, and thus about fact, that they may be affected, even if they did not know. Maybe "the script that generated this data and filed
bugs for
packages affected by the removal of Java packages on i686 was a bit over-zealous." ( to quote :) ) but I still somehow finds it correct.
Jiri, could you please close all the bugzillas that were only opened due to the subversion<-java dependency now when https://bugzilla.redhat.com/show_bug.cgi?id=2103909 was fixed?
I've closed some bugs for very important components manualy, but there are simply too many.
Argh:( You should have not. I'm really reluctant to do so. How can you proof the package is really not affected by the change?
I have actually second set of dependent packages (with 50-77 dependencies) prepared to fill bug aganst, if the mass rebuild next week goes bad.
If this thread asks me to close the transitive depndencies of subversion to be closed, I will do so. But I relly think it is bad idea. Owner should check the impact, and close on his own. And actually it is no tso much. It is less then 50
which depends *only* on subversion.
Sorry for delayed reply, Miro, thanx for ping.
J.
--
Jiri Vanek Mgr.
Principal QA Software Engineer
Red Hat Inc.
+420 775 39 01 09
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
