Excerpts from Paul Mackerras's message of June 15, 2021 9:05 am:
> This ensures that we don't get a decrementer interrupt arriving before
> we have set up a handler for it.
Would this be better off merged in the previous patch (maybe with
comment)? Why don't other platform_init()s seem to require this?
Thanks,
Nick
>
> Signed-off-by: Paul Mackerras <pau...@ozlabs.org>
> ---
> arch/powerpc/boot/microwatt.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/powerpc/boot/microwatt.c b/arch/powerpc/boot/microwatt.c
> index ac922dd0aa4d..86a07bceaadf 100644
> --- a/arch/powerpc/boot/microwatt.c
> +++ b/arch/powerpc/boot/microwatt.c
> @@ -12,6 +12,7 @@ void platform_init(unsigned long r3, unsigned long r4,
> unsigned long r5)
> {
> unsigned long heapsize = 16*1024*1024 - (unsigned long)_end;
>
> + __asm__ volatile("mtmsrd %0,1" : : "r" (0));
> simple_alloc_init(_end, heapsize, 32, 64);
> fdt_init(_dtb_start);
> serial_console_init();
> --
> 2.31.1
>
>
