Re: kern/60814: Kernel build using existing /usr/tools fails with: "src/sys/dev/fdt/files.fdt:205: `fdt' is not a plain attribute"

[Alexander Schreiber <als%thangorodrim.ch@localhost>]

On Mon, Sep 28, 2026 at 08:30:02AM +0000, Martin Husemann via gnats wrote:
> The following reply was made to PR kern/60814; it has been noted by GNATS.
> 
> From: Martin Husemann <martin%duskware.de@localhost>
> To: gnats-bugs%netbsd.org@localhost
> Cc: 
> Subject: Re: kern/60814: Kernel build using existing /usr/tools fails with:
>  "src/sys/dev/fdt/files.fdt:205: `fdt' is not a plain attribute"
> Date: Mon, 28 Sep 2026 10:28:36 +0200
> 
>  On Sun, Sep 27, 2026 at 07:45:01PM +0000, als%thangorodrim.ch@localhost via gnats wrote:
>  > Build GENERIC kernel from HEAD/20260927052045Z snapshot fails with:
>  
>  I have seen simmilar fallout when trying to build with tools from the
>  wrong branch.
>  
>  Just to make sure: you used a -current tools build to configure the
>  current (HEAD) kernel? Using a 11.0 tools build would not work (and
>  is not supposed to).

Not knowing any better, I used 11.0 RELEASE tools, essentially following 
 https://www.netbsd.org/docs/guide/en/chap-updating.html
 "35.1.2. Building a new kernel":

$ cd /usr/src
$ ./build.sh -O ../obj -T ../tools -U distribution

but modified to
$ cd /usr/build/HEAD/${TIMESTAMP}/usr/src
$ ./build.sh -O ../obj -T /usr/tools -U distribution

Dug around the docs a little more, found 
 https://www.netbsd.org/docs/guide/en/chap-kernel.html
 "34.5. Building the kernel using build.sh"

$ cd /usr/src
$ ./build.sh kernel=MYKERNEL

modified to
$ cd /usr/build/HEAD/${TIMESTAMP}/usr/src
$ ./build.sh kernel=MYKERNEL

and that pretty quickly points out that it does indeed need new tools and
that I should just run "./build.sh tools" to get this done - after which the
kernel build works fine. 

So I guess I looked at the wrong corner of the docs, aka "operator error".

Maybe some more ... foolproofing of the docs would be helpful ;-)

Although I suspect my specific use case of "I dug up this machine, noticed
NetBSD now somewhat works on it, and want to find any remaining breakages
so they can be fixed ... all while not being a NetBSD developer (and thus
lacking some helpful experience" is not exactly common.

Kind regards,
           Alex.
-- 
"Opportunity is missed by most people because it is dressed in overalls and
 looks like work."                                      -- Thomas A. Edison