Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <YzN+ZYLjK6HI1P1C@ZenIV> <YzSSl1ItVlARDvG3@ZenIV>
 <YzpcXU2WO8e22Cmi@iweiny-desk3> <7714.1664794108@jrobl> <Yzs4mL3zrrC0/vN+@iweiny-mobl>
 <YztfvaAFOe2kGvDz@ZenIV> <4011.1664837894@jrobl> <YztyLFZJKKTWcMdO@ZenIV>
In-Reply-To: <YztyLFZJKKTWcMdO@ZenIV>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Mon, 3 Oct 2022 17:20:03 -0700
Message-ID: <CAHk-=whsOyuRhjmUQ5c1dBQYt1E4ANhObAbEspWtUyt+Pq=Kmw@mail.gmail.com>
Subject: Re: [PATCH][CFT] [coredump] don't use __kernel_write() on kmap_local_page()
To:     Al Viro <viro@zeniv.linux.org.uk>
Cc:     "J. R. Okajima" <hooanon05g@gmail.com>,
        Ira Weiny <ira.weiny@intel.com>, linux-kernel@vger.kernel.org,
        linux-fsdevel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Mon, Oct 3, 2022 at 4:37 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> One variant would be to revert the original patch, put its
> (hopefully) fixed variant into -next and let it sit there for
> a while.  Another is to put this incremental into -next and
> merge it into mainline once it gets a sane amount of testing.

Just do the incremental fix. It looks obvious enough ("oops, we need
to get the pos _after_ we've done any skip-lseeks on the core file")
that I think it would be just harder to follow a "revert and follow up
with a fix".

I don't think it needs a ton of extra testing, with Okajima having
already confirmed it fixes his problem case..

                Linus