[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CACzwLxjpH=n5RDWnQCPZjH8i1GZAAbG-6BzuWCcVf99=qTO7HQ@mail.gmail.com>
Date: Fri, 27 Sep 2024 20:18:46 +0500
From: Sabyrzhan Tasbolatov <snovitoll@...il.com>
To: andreyknvl@...il.com
Cc: akpm@...ux-foundation.org, bp@...en8.de, brauner@...nel.org,
dave.hansen@...ux.intel.com, dhowells@...hat.com, dvyukov@...gle.com,
glider@...gle.com, hpa@...or.com, kasan-dev@...glegroups.com,
linux-kernel@...r.kernel.org, linux-mm@...ck.org, mingo@...hat.com,
ryabinin.a.a@...il.com, tglx@...utronix.de, vincenzo.frascino@....com,
x86@...nel.org
Subject: Re: [PATCH v5] mm: x86: instrument __get/__put_kernel_nofault
On Mon, Sep 23, 2024 at 11:09 AM Sabyrzhan Tasbolatov
<snovitoll@...il.com> wrote:
>
> Instead of adding KASAN, KCSAN checks per arch macro,
> here is the alternative, generic way with a wrapper.
> I've tested it on x86_64 only, going to test on arm64
> with KASAN_SW_TAGS, KASAN_HW_TAGS if I can do it in qemu,
> and form a new patch for all arch
> and this PATCH v5 for x86 only can be abandoned.
>
> Please let me know if this wrapper is good enough,
> I will see in kasan_test.c how I should use SW/HW_TAG, probably,
> they should be a separate test with
> KASAN_TEST_NEEDS_CONFIG_ON(test, CONFIG_KASAN_SW_TAGS);
> ---
Hello,
I've sent a different patch [1] to support all arch checks,
tested on x86_64 and arm64 with SW/HW_TAGS.
This [PATCH v5] for the x86 only can be ignored.
[1] https://lore.kernel.org/linux-mm/20240927151438.2143936-1-snovitoll@gmail.com/T/#u
Powered by blists - more mailing lists