Re: [PATCH v10 00/22] kasan: add software tag-based mode for arm64

[Andrey Konovalov <andreyknvl@google.com>]

On Wed, Nov 7, 2018 at 4:34 PM, Will Deacon <will.deacon@arm.com> wrote:
> On Wed, Nov 07, 2018 at 04:11:35PM +0100, Andrey Konovalov wrote:
>> On Wed, Nov 7, 2018 at 3:59 PM, Will Deacon <will.deacon@arm.com> wrote:
>> > On Wed, Nov 07, 2018 at 03:56:03PM +0100, Andrey Konovalov wrote:
>> >> On Tue, Nov 6, 2018 at 6:30 PM, Andrey Konovalov <andreyknvl@google.com> wrote:
>> >> > This patchset adds a new software tag-based mode to KASAN [1].
>> >> > (Initially this mode was called KHWASAN, but it got renamed,
>> >> >  see the naming rationale at the end of this section).
>> >>
>> >> [...]
>> >>
>> >> > Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
>> >> > Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
>> >> > Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
>> >>
>> >> Hi Andrew,
>> >>
>> >> This patchset has now been reviewed-by KASAN maintainers. Could you
>> >> take a look and consider taking this into the -mm tree?
>> >
>> > I would much prefer to take the arm64 parts (which still need to be reviewed
>> > by Catalin afaict) via the arm64 tree, so please can you split those out
>> > separately?
>>
>> Which parts do you mean exactly, which patches? I don't think it makes
>> sense to split this patchset, as one part won't function without the
>> other.
>
> I would like the patches that touch code under arch/arm64/ to be reviewed by
> somebody from the arm64 community. Since the core parts have already been
> reviewed, I was suggesting that you could split them out so that they are
> not blocked by the architecture code. Is it not possible to preserve the
> existing KASAN behaviour for arm64 with the core parts merged? I figured it
> must be, since you're not touching any other architectures here and they
> assumedly continue to function correctly.
>
> However, if you'd rather keep everything together, please can we give it a
> couple of weeks so we can at least get the architecture bits reviewed? Most
> people are out at LPC next week (and I'm at another conference this week).
>
> Will