Claire, le lun. 05 oct. 2026 04:00:06 +0200, a ecrit:
> The report targets a different, older set of headers: the ones
> shipped INSIDE the official preinstalled image
> debian-hurd-amd64-20260314.img, which is still the "latest"
> image on cdimage.debian.org/ports today.  On that image:
> 
>     $ dpkg -l gnumach-image-1.8-amd64-up | tail -1
>     ii  gnumach-image-1.8-amd64-up 2:1.8+git20260224-4 hurd-amd64
> 
> and the typedef is absent - the grep you ran returns nothing:
> 
>     $ grep -n "processor_name_array_t" \
>           /usr/include/x86_64-gnu/mach/mach_types.h
>     (no match)

Yes.

>     $ grep -rn "processor_name_array_t" /usr/include/mach/
>     /usr/include/mach/mach_host.h:538:  processor_name_array_t
>         *processors_list,
>     /usr/include/mach/mach_host.h:1092: processor_name_array_t
>         *processors_list,

No. There is no such occurrence on the debian-hurd-amd64-20260314.img.

I don't know what you did with the image, but there is really no such
bug.

> So the minimal reproducer fails on the stock image:
> 
>     $ printf '#include <mach.h>\nint main(void){return 0;}\n' \
>           > /tmp/r.c && gcc /tmp/r.c -o /tmp/r

It does build fine here with the stock debian-hurd-amd64-20260314.img
image.

Samuel

Reply via email to