Re: [PATCH v2] tools/nolibc: add support for Alpha

Willy Tarreau <[email protected]> Sat, 1 Aug 2026 21:20:35 +0200
Newsgroups org.kernel.vger.linux-alpha,org.kernel.vger.linux-kernel,org.kernel.vger.linux-kselftest
Message-ID <[email protected]>
Hi Thomas,

On Wed, Jul 22, 2026 at 12:07:05AM +0200, Thomas Weißschuh wrote:
> A straightforward new architecture.
> 
> Signed-off-by: Thomas Weißschuh <[email protected]>
> ---
> Only tested on QEMU so far.
> Testing on real hardware would be very welcome.

I'd love to, but last year I could fix my serial port (replaced the chip
on the motherboard) and the new one died a few minutes later :-(  I'm
starting to seriously suspect a flaky PSU that doesn't regulate its 12V
output well and that is progressively killing the mobo.

(...)
> diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c
> index 996e8d13508e..66accf99f560 100644
> --- a/tools/testing/selftests/nolibc/nolibc-test.c
> +++ b/tools/testing/selftests/nolibc/nolibc-test.c
> @@ -756,6 +756,10 @@ int run_startup(int min, int max)
>  	/* checking NULL for argv/argv0, environ and _auxv is not enough, let's compare with sbrk(0) or &end */
>  	extern char end;
>  	char *brk = sbrk(0) != (void *)-1 ? sbrk(0) : &end;
> +#if defined(__alpha__)
> +	/* the ordering above does not work on an alpha kernel due to STACK_TOP != TASK_SIZEi */
                                                                                   ^^^^^^^^^^
I suspect this trailing "i" is a leftover from an attempt to switch vi to
insertion mode :-)

Otherwise, while I'm not competent on the alpha side, it looks clean from
the nolibc perspective, thank you!

Acked-by: Willy Tarreau <[email protected]>

Willy