Re: [PATCH] arm64: ptrace: user_regset_copyin_ignore() always returns 0

[Catalin Marinas <catalin.marinas@arm.com>]

On Tue, Sep 27, 2022 at 09:46:11PM +0300, Sergey Shtylyov wrote:
> On 9/26/22 5:01 PM, Catalin Marinas wrote:
> >>>> user_regset_copyin_ignore() always return 0, so checking its result seems
> >>>> pointless -- don't do this...
[...]
> >>> I think it would be better to have user_regset_copyin_ignore() return void
> >>> so that we don't run the risk of missing an error code if it starts
> >>> returning one in future.
> >>
> >>    That's the plan! But I need to convert the users 1st, right?
> > 
> > Right, though normally we'd like to see the full series that does the
> > arch clean-up followed by the user_regset_copyin_ignore() return changed
> > to void. If for some reason the last patch is rejected by the maintainer
> > because there are plans to actually return some non-zero value, we'd
> > have to revert the above.
> 
>    Hm... I usually try to avoid a cross-tree patch series but here I should
> be able to use the linux-next repo as I series' base. Is that OK?

It depends on how you plan to merge it, though basing the series off
linux-next is not ideal. It's fine to merge individual patches in one
release though specific trees and the final patch in another but when
you posted this single patch we didn't have the context. So ideally I'd
like to see an ack on the patch converting user_regset_copyin_ignore()
to void. It wasn't obvious there's such patch.

You can also have a cross-tree series and convince one of the maintainer
to pick it up (usually cc'ing akpm does the trick ;)).

-- 
Catalin

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel