Hello,
As there is no next tree since a few days and this will continue to be true for
a few days, I tried to create my own next.
I did
git show next/master:Next/Trees | sed 1,4d > trees
git checkout linus/master
And then repeatedly run:
while read rname protocol repohashbranch; do
repo="${repohashbranch%#*}"
branch="${repohashbranch#*#}"
# git:// seems to be ratelimited at github
repo="${repo/#git:\/\/github.com\//https:\/\/github.com\/}"
case "$repo" in
*git.libc.org*)
# if repo it on git.libc.org, skip it "Name or service not known"
continue
;;
esac
echo "fetch $repo $branch"
rev="$(git ls-remote "$repo" "$branch" | awk '{ print $1 }')" || continue
echo $rev
if git merge-base --is-ancestor "$rev" HEAD 2>/dev/null; then
continue
fi
pgit pull "$repo" "$branch" || break
done < trees
and fixed merge conflicts in between.
I was so bold to tag the result as next-20221230, it's available from
https://git.pengutronix.de/git/ukl/linux tags/next-20221230
It doesn't contain the usual meta-data in the top commit (as I didn't
find the scripts next is usually created with) and I didn't do any build
tests.
There were a few conflicts:
- In drivers/dma-buf/dma-buf.c between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
f728a5ea27c9 ("dma-buf: fix dma_buf_export init order v2")
and
git://anongit.freedesktop.org/drm/drm-misc for-linux-next-fixes
28743e25fa1c ("dma-buf: Remove obsoleted internal lock")
- In arch/arm/boot/dts/nuvoton-wpcm450-supermicro-x9sci-ln4f.dts between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
62d127eeac27 ("ARM: dts: nuvoton,wpcm450-supermicro-x9sci-ln4f: Add GPIO line names")
and
https://git.kernel.org/pub/scm/linux/kernel/git/joel/bmc for-next
220a041d4cca ("ARM: dts: nuvoton,wpcm450-supermicro-x9sci-ln4f: Add GPIO line names")
which is a problem because both commits are identical but the second tree has
some more commits touching the file.
- In drivers/gpu/drm/amd/amdgpu/amdgpu_device.c between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
afa6646b1c5d ("drm/amdgpu: skip MES for S0ix as well since it's part of GFX")
and
https://gitlab.freedesktop.org/agd5f/linux drm-next
5620a1889e4c ("drm/amdgpu: skip MES for S0ix as well since it's part of GFX")
which is a problem because both commits are identical but the second tree has
some more commits touching the file.
- In drivers/gpu/drm/i915/i915_drv.h between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
8f70f1ec587d ("drm/i915/mtl: Add Wa_14017073508 for SAMedia")
and
git://anongit.freedesktop.org/drm-intel for-linux-next
2357f2b271ad ("drm/i915/mtl: Initial display workarounds")
- In arch/arm64/kernel/stacktrace.c between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
0fbcd8abf337 ("arm64: Prohibit instrumentation on arch_stack_walk()")
and
git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi.git next
c2530a04a73e ("arm64: efi: Account for the EFI runtime stack in stack unwinder")
- In various files below mm/ between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
???
and
git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab.git for-next
???
I didn't feel confident to fix this conflict correctly and skipped merging the slab tree.
- In various files belwo mm/ between
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
???
and
git://git.infradead.org/users/willy/pagecache.git for-next
???
As above I didn't even try to fix these up and skipped merging the folio tree
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | https://www.pengutronix.de/ |
On 12/30/22 11:30, Uwe Kleine-König wrote:
> - In various files below mm/ between
>
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> ???
>
> and
>
> git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab.git for-next
> ???
That's weird, because slab.git for-next is still exactly what was merged to
6.2-rc1, so when I try to merge it to master I just get "Already up to date."
> I didn't feel confident to fix this conflict correctly and skipped merging the slab tree.
Nothing is lost then, but maybe it indicates some issue with your approach?
(didn't try to replicate it in full)
> - In various files belwo mm/ between
>
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> ???
>
> and
>
> git://git.infradead.org/users/willy/pagecache.git for-next
> ???
>
> As above I didn't even try to fix these up and skipped merging the folio tree
>
> Best regards
> Uwe
>