pkgsrc-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkgsrc on OpenBSD



On Sat, Aug 29, 2020 at 9:42 AM Ottavio Caruso
<ottavio2006-usenet2012%yahoo.com@localhost> wrote:
> On 29/08/2020 01:28, Greg Troxel wrote:


>> I think it means that pkgsrc is confusing the values of arch and/or
>> machine arch.
>
> A quick look at all "arch" strings in my bootstrap log on OpenbBSD
> 6.6:
>
> ===> running: (cd /home/oc/cvs/pkgsrc/bootstrap/work/bmake && /bin/sh
> configure  --prefix=/home/oc/cvs/pkgsrc/bootstrap/work
> --with-default-sys-path=/home/oc/cvs/pkgsrc/bootstrap/work/share/mk
> --with-machine-arch=x86_64 )
>
> [..]
>
> checking for MACHINE & MACHINE_ARCH...
> defaults: MACHINE=amd64, MACHINE_ARCH=amd64
> Using: MACHINE=amd64, MACHINE_ARCH=x86_64
>
> [...]
>
> Is this what it should be?
>
> Then it appears bmake is broken:
>
> oc@OpenBSD:~/cvs/pkgsrc/bootstrap$ /usr/pkg/bin/bmake show-var
> VARNAME=MACHINE
> malloc() warning: unknown char in MALLOC_OPTIONS
> bmake: don't know how to make show-var. Stop
>
> bmake: stopped in /home/oc/cvs/pkgsrc/bootstrap
>
> Maybe just me? This was reported here:
> https://mail-index.netbsd.org/pkgsrc-users/2017/04/12/msg024700.html

The MALLOC problem's a different issue (that I pointed out in my initial email)


> Then, in the pkgsrc guide:
> http://cdn.netbsd.org/pub/pkgsrc/current/pkgsrc/doc/pkgsrc.html#plist.misc
>
> "PLIST will be preprocessed before actually used, and the symbol
> ???${MACHINE_ARCH}??? will be replaced by what uname -p gives."
>
> NetBSD:
>
> oc@NetBSD:/home/oc$ uname -p
> x86_64
>
> OpenBSD:
>
> oc@OpenBSD:~$ uname -p
> amd64
>
> Is this what causes the problem, or am I looking in the wrong place?

This is the problem. But the scripts run "arch" rather than "uname". Same thing.


Home | Main Index | Thread Index | Old Index