Overview
Request 935972 accepted
- Non-existent patches must be listed twice to appear as added in a
unified diff against a version that had them. Only that can make
factory-auto happy. Here we go:
* makedumpfile-Retrieve-MAX_PHYSMEM_BITS-from-vmcoreinfo.patch
* makedumpfile-arm64-Add-support-for-ARMv8.2-LPA-52-bit-PA-su.patch
Request History
ptesarik created request
- Non-existent patches must be listed twice to appear as added in a
unified diff against a version that had them. Only that can make
factory-auto happy. Here we go:
* makedumpfile-Retrieve-MAX_PHYSMEM_BITS-from-vmcoreinfo.patch
* makedumpfile-arm64-Add-support-for-ARMv8.2-LPA-52-bit-PA-su.patch
factory-auto added opensuse-review-team as a reviewer
Please review sources
factory-auto accepted review
skipping the staging process since only .changes modifications
factory-auto accepted review
Check script succeeded
licensedigger accepted review
ok
dimstar accepted review
dimstar approved review
dimstar_suse accepted request
Accept to openSUSE:Factory
say what???
This is probably the weirdest thing I've ever read and makes very little sense.
@sweiberg does that refer to a decline in SLE which was odd? e.g. patches added in a SP that are dropped in SP4 with the package update?
You bet. Makedumpfile must be upgraded for SLE15 SP4 (because the kernel was updated). These two patches had to be backported into SLE15 SP3 makedumpfile:
However, by that time, Factory already contained a newer version, which included the changes from these two patches. That's how these patches never made it into Tumbleweed. So far so good…
Now comes the funny part. The changelog for SLE goes like this:
To sync that with Factory, I added these lines (as you can see above):
As it happens, the two lines with the patch names happen to match exactly, so they are seen as unchanged by factory-auto. Technically, they are not listed on a line that starts with a
+
when factory-auto does a unified diff. Consequently, the script concludes that the patches were removed without being mentioned:And the submit request was declined automatically before anyone even had a chance to have a look. See comments to the corresponding IBS request: https://build.suse.de/request/show/259746
For reference, the genius code that forced me to add my funny changelog entry lives here:
https://github.com/openSUSE/openSUSE-release-tools/blob/master/check_source.pl
The decision to enforce Factory-first policy for SLE with a script was made by project managers long ago against my objections. I'm between a rock and a hard place, you see.
sadly, https://build.suse.de/request/show/259746 is in a state that shows a totally bogus diff; making it very hard to debug. I'd have loved to see the original case
But well - let's take it like this so at least you won't have to suffer with future submissions to SP4
The original file for the diff was this one:
https://build.opensuse.org/package/view_file/SUSE:SLE-15-SP3:Update/makedumpfile/makedumpfile.changes?expand=1
The new one for the request was like in Factory before accepting this submit request.
Good luck!