Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20211019134204.3382645-1-agruenba@redhat.com> <CAHk-=wh0_3y5s7-G74U0Pcjm7Y_yHB608NYrQSvgogVNBxsWSQ@mail.gmail.com>
 <YXBFqD9WVuU8awIv@arm.com> <CAHk-=wgv=KPZBJGnx_O5-7hhST8CL9BN4wJwtVuycjhv_1MmvQ@mail.gmail.com>
 <YXCbv5gdfEEtAYo8@arm.com> <CAHk-=wgP058PNY8eoWW=5uRMox-PuesDMrLsrCWPS+xXhzbQxQ@mail.gmail.com>
 <YXL9tRher7QVmq6N@arm.com> <CAHk-=wg4t2t1AaBDyMfOVhCCOiLLjCB5TFVgZcV4Pr8X2qptJw@mail.gmail.com>
In-Reply-To: <CAHk-=wg4t2t1AaBDyMfOVhCCOiLLjCB5TFVgZcV4Pr8X2qptJw@mail.gmail.com>
From:   Andreas Gruenbacher <agruenba@redhat.com>
Date:   Mon, 25 Oct 2021 21:00:43 +0200
Message-ID: <CAHc6FU7BEfBJCpm8wC3P+8GTBcXxzDWcp6wAcgzQtuaJLHrqZA@mail.gmail.com>
Subject: Re: [PATCH v8 00/17] gfs2: Fix mmap + page fault deadlocks
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Catalin Marinas <catalin.marinas@arm.com>,
        Paul Mackerras <paulus@ozlabs.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        Christoph Hellwig <hch@infradead.org>,
        "Darrick J. Wong" <djwong@kernel.org>, Jan Kara <jack@suse.cz>,
        Matthew Wilcox <willy@infradead.org>,
        cluster-devel <cluster-devel@redhat.com>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        ocfs2-devel@oss.oracle.com, kvm-ppc@vger.kernel.org,
        linux-btrfs <linux-btrfs@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Oct 22, 2021 at 9:23 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> On Fri, Oct 22, 2021 at 8:06 AM Catalin Marinas <catalin.marinas@arm.com> wrote:
> > Probing only the first byte(s) in fault_in() would be ideal, no need to
> > go through all filesystems and try to change the uaccess/probing order.
>
> Let's try that. Or rather: probing just the first page - since there
> are users like that btrfs ioctl, and the direct-io path.

For direct I/O, we actually only want to trigger page fault-in so that
we can grab page references with bio_iov_iter_get_pages. Probing for
sub-page error domains will only slow things down. If we hit -EFAULT
during the actual copy-in or copy-out, we know that the error can't be
page fault related. Similarly, in the buffered I/O case, we only
really care about the next byte, so any probing beyond that is
unnecessary.

So maybe we should split the sub-page error domain probing off from
the fault-in functions. Or at least add an argument to the fault-in
functions that specifies the amount of memory to probe.

Thanks,
Andreas