bpo-45711: Do not push/pop traceback on stack as part of exc_info

[iritkatriel]

https://bugs.python.org/issue45711

[iritkatriel]

Marked as draft because I still need to update the opcodes documentation

[iritkatriel]

The next step is to remove exc_traceback from exc_info and then this will not be here.

[bedevere-bot]

🤖 New build scheduled with the buildbot fleet by @iritkatriel for commit 98085fe 🤖

If you want to schedule another build, you need to add the ":hammer: test-with-buildbots" label again.

[iritkatriel]

This clashes with the except* branch, and I'd rather merge that one first and update this one.

[markshannon]

This clashes with the except* branch, and I'd rather merge that one first and update this one.

I would rather merge this first.
Apart from the trivial conflict in dis.rst this seems ready to merge. The except* branch is not.

[iritkatriel]

This clashes with the except* branch, and I'd rather merge that one first and update this one.

I would rather merge this first. Apart from the trivial conflict in dis.rst this seems ready to merge. The except* branch is not.

We won't see a speedup for this because of the get_exc_traceback calls. We should remove them before measuring, and probably merge the whole thing together.

[iritkatriel]

We won't see a speedup for this because of the get_exc_traceback calls. We should remove them before measuring, and probably merge the whole thing together.

And that requires PR 29922.

[arhadthedev]

Spacing between tokens is strange.

[iritkatriel]

Superseded by #30122