From: "Theo de Raadt" <deraadt@openbsd.org>
Subject: Re: xenocara release hashes
To: Theo Buehler <tb@theobuehler.org>
Cc: tech@openbsd.org
Date: Mon, 10 Nov 2025 10:05:07 -0700

Uhm, the idea is that you use a seperate RELEASEDIR for X.

I'm very surprised someone is using the same directory.

if you do this >> then building X twice in a row will collect
old hashes followed by new (different) hashes in the same file.
How is that desireable?


Theo Buehler <tb@theobuehler.org> wrote:

> If you build a full snapshot, the xenocara release target will wipe out
> the hashes computed for the base release.
> 
> Now claudio ran into this same issue I've had this M in my tree for a
> long time since my workflow involves experimenting with libcrypto
> changes and seeing how the ports tree copes with my abuse.
> 
> Would a diff like this be acceptable or would that break someone else?
> 
> Index: Makefile
> ===================================================================
> RCS file: /cvs/xenocara/Makefile,v
> diff -u -p -r1.78 Makefile
> --- Makefile	28 Apr 2020 07:04:49 -0000	1.78
> +++ Makefile	3 Jun 2024 21:46:12 -0000
> @@ -96,7 +96,7 @@ sha: release-clean release-install dist 
>  
>  hash: dist
>  	-cd ${RELEASEDIR}; \
> -		cksum -a sha256 x*tgz > SHA256
> +		cksum -a sha256 x*tgz >> SHA256
>  
>  .ORDER: release-clean release-install dist hash
>  .endif
>