Skip to contentSkip to conversation

arch/nios2 for v3.19

GIT PULL George Spelvin 2 participants 3 messages Dec 10, 2014

lore
Share & tools
lore
#0 lore raw
Quoted 2 lines · Thomas Gleixner

It's a bit sad that all the reviewed/acked-by information which has been sent throughout the review process is completely lost.

Well, in the case here where the ack comes in after the pull request, obviously only the merger (Linus) can add it, but for older ones, it seems like an obvious git feature request to have comments in the annotated tag automatically copied to the merge commit message draft.

That way, the subsystem maintainer could write a branch summary, like the [PATCH 0/n] summaries that are popular, and have it automatically included in the history.

That just seems a more sensible place to put things than the pull request e-mail.

(Preserving the signed annotated tag object itself is another intriguing possibility, but not only is it a huge format change to create a way for a commit to point to a tag, but I also don't see how to avoid tag name collision problems.)

#1 lore raw
Quoted 5 lines · George Spelvin

(Preserving the signed annotated tag object itself is another intriguing possibility, but not only is it a huge format change to create a way for a commit to point to a tag, but I also don't see how to avoid tag name collision problems.)

Actually, git already preserves the signed annotated tag. That's the only way git show --show-signature can verify the tag's gpg signature.

You can see original signed annotated tag via a command like this:

git cat-file -p 18b34d9a7a085ba8f9cafa6a0d002e2cbac87c1f

- Ted

#2 lore raw
Quoted 6 lines

Actually, git already preserves the signed annotated tag. That's the only way git show --show-signature can verify the tag's gpg signature.

You can see original signed annotated tag via a command like this:

git cat-file -p 18b34d9a7a085ba8f9cafa6a0d002e2cbac87c1f

TIL, thank you!

For the others watching in, as you know all commits have "headers" which are normally not printed. The above commit looks like

Quoted 10 lines

tree 939755a58a68aae5e1241b2a905ea612aeb2e589
parent 502fde1a0a2990ec54eab5241d3135c545da7372
parent bf40c92635d63fcc574c52649f7cda13e0418ac1
author Linus Torvalds <[email protected]> 1405282495 -0700 committer Linus Torvalds <[email protected]> 1405282495 -0700 mergetag object bf40c92635d63fcc574c52649f7cda13e0418ac1
type commit
tag ext4_for_linus_stable
tagger Theodore Ts'o <[email protected]> 1405281748 -0400

Quoted 6 lines

More bug fixes for ext4 -- most importantly, a fix for a bug (introduced in 3.15) that can end up triggering a file system corruption error after a journal replay. (It shouldn't lead to any actual data corruption, but it is scary and can force file systems to be remounted read-only, etc.)
-----BEGIN PGP SIGNATURE-----

The "mergetag" header contains the original annotated tag, indented by one space.

In this case, Ted included the same message (formatted slightly differently) in the pull e-mail and Linus copied it into the merge commit, so it worked out in the end.