Re: Bug report - ssh signing causes git tag -l malloc failed

From: Junio C Hamano <gitster@pobox.com>
Date: May 2, 2022, 20:24 UTC
Message-ID: <xmqq7d73y97i.fsf@gitster.g>
In-Reply-To: <16668F1B-7670-4136-8AAF-ABAAA802C7D4@b-rad.info>

Brad Beam <brad.beam@b-rad.info> writes:

Thanks for a report.

> What did you do before the bug happened? (Steps to reproduce your issue)
>
> When using ssh signing `git tag -l --format='%(contents:body)' <tag>` returns `fatal: Out of memory, malloc failed (tried to allocate 18446744073709551323 bytes)`

An obvious first follow-up question is if there is any difference in behaviour if another kind of signing (like PGP) is used.

Show 16 quoted lines

> What did you expect to happen? (Expected behavior)
>
> The tag to be returned without crashing
>
> What happened instead? (Actual behavior)
>
> git crashed with the above error (`fatal: Out of memory, malloc failed (tried to allocate 18446744073709551323 bytes)`)
>
> What's different between what you expected and what actually happened?
>
> Anything else you want to add:
>
>
> ```
> [14:23:54] (0):~/src/github.com/my/repo
> % git tag -sam v0.0.9 v0.0.9

Here, or before this step, there would probably have been something to say "No, I do not use the default PGP sign, but I want SSH sign"?

For those who are reading this bug report and mistakenly thought it is a bug to show a self-recursive tag (like I did during my first reading of the report), the first v0.0.9 is merely an argument to the "-m" option, and the second v0.0.9 is the name of the newly created tag. We are creating a signed tag with v0.0.9 in the message that points at HEAD and has "tag v0.0.9" in the header, and storing the resulting tag at refs/tags/v0.0.9