On Thu, Nov 5, 2020 at 3:14 PM Vincenzo Frascino
[off-list ref] wrote:

On 11/5/20 12:14 PM, Andrey Konovalov wrote:

quoted

On Thu, Nov 5, 2020 at 12:39 PM Vincenzo Frascino
[off-list ref] wrote:

quoted

On 11/5/20 11:35 AM, Andrey Konovalov wrote:

quoted

This will work. Any preference on the name of this function?

I called it in my current iteration mte_enable(), and calling it from
cpu_enable_mte().

quoted

Alternatively we can rename mte_init_tags() to something else and let
it handle both RRND and sync/async.

This is an option but then you need to change the name of kasan_init_tags and
the init_tags indirection name as well. I would go for the simpler and just
splitting the function as per above.

What do you think?

OK, let's split. mte_enable() as a name sounds good to me. Both
functions will still be called one right after another from
kasan_init_hw_tags (as it's now called) though. I think the name
works, as it means initializing the hw_tags mode, not just the tags.

I agree. When you finish with v9, could you please provide a tree with both the
sets on top similar to [1]? I would like to repeat the tests (ltp + kselftests)
and even to rebase my async code on top of it since we are aligning with the
development.

[1] https://github.com/xairy/linux/tree/up-boot-mte-v1

Sure, will share the trees, probably later today.

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