Re: [RFC PATCH 09/10] powerpc: snapshot thread flags

[Michael Ellerman]

Mark Rutland <mark.rutland@xxxxxxx> writes:
> Some thread flags can be set remotely, and so even when IRQs are
> disabled, the flags can change under our feet. Generally this is
> unlikely to cause a problem in practice, but it is somewhat unsound, and
> KCSAN will legitimately warn that there is a data race.
>
> To avoid such issues, we should snapshot the flags prior to using them.
> Let's use the new helpers to do so on powerpc.
>
> Signed-off-by: Mark Rutland <mark.rutland@xxxxxxx>
> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx>
> Cc: Paul Mackerras <paulus@xxxxxxxxx>
> ---
>  arch/powerpc/kernel/interrupt.c     | 16 ++++++++--------
>  arch/powerpc/kernel/ptrace/ptrace.c |  3 +--
>  2 files changed, 9 insertions(+), 10 deletions(-)

This looks good.

Acked-by: Michael Ellerman <mpe@xxxxxxxxxxxxxx> (powerpc)


But, it clashes terribly with a series I'm planning to take for v5.14
that reworks our interrupt return code.

https://lore.kernel.org/linuxppc-dev/20210610130921.706938-1-npiggin@xxxxxxxxx/T/#t

So if you're also targeting v5.14 then it might be best to drop this
patch from the series, and we can do the conversion later.

cheers