[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CA+fCnZeD4CJheGP6D+x5dzUc4ABRZz1Db0h7hNVmwH3gUC2zyg@mail.gmail.com>
Date: Wed, 14 Aug 2024 18:10:57 +0200
From: Andrey Konovalov <andreyknvl@...il.com>
To: Miguel Ojeda <miguel.ojeda.sandonis@...il.com>
Cc: andrey.konovalov@...ux.dev, Marco Elver <elver@...gle.com>,
Matthew Maurer <mmaurer@...gle.com>, Alexander Potapenko <glider@...gle.com>,
Dmitry Vyukov <dvyukov@...gle.com>, Andrey Ryabinin <ryabinin.a.a@...il.com>, kasan-dev@...glegroups.com,
Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
Miguel Ojeda <ojeda@...nel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kasan: simplify and clarify Makefile
On Wed, Aug 14, 2024 at 5:37 PM Miguel Ojeda
<miguel.ojeda.sandonis@...il.com> wrote:
>
> On Wed, Aug 14, 2024 at 12:40 AM <andrey.konovalov@...ux.dev> wrote:
> >
> > Signed-off-by: Andrey Konovalov <andreyknvl@...il.com>
>
> It is easier to read now, and indeed GCC 5.1+ and LLVM 13+ both
> support the flags, so `CFLAGS_KASAN_SHADOW` can't be empty.
>
> > +# First, enable -fsanitize=kernel-address together with providing the shadow
> > +# mapping offset, as for GCC, -fasan-shadow-offset fails without -fsanitize
> > +# (GCC accepts the shadow mapping offset via -fasan-shadow-offset instead of
> > +# a normal --param). Instead of ifdef-checking the compiler, rely on cc-option.
>
> I guess "a normal --param" means here that it is the usual way to
> tweak the rest of the KASAN parameters, right?
Yes, clarified in v2.
> > +# Now, add other parameters enabled in a similar way with GCC and Clang.
>
> I think the "with" sounds strange, but I am not a native speaker.
> Perhaps "in a similar way with" -> "similarly in both"?
Sure, done in v2.
Thank you!
Powered by blists - more mailing lists