Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.100.0.2.22\))
Subject: Re: Do we need to unrevert "fs: do not prefault sys_write() user
 buffer pages"?
From:   Nadav Amit <nadav.amit@gmail.com>
In-Reply-To: <YNItqqZA9Y1wOnZY@casper.infradead.org>
Date:   Tue, 22 Jun 2021 11:51:51 -0700
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        David Howells <dhowells@redhat.com>,
        Al Viro <viro@zeniv.linux.org.uk>, Ted Ts'o <tytso@mit.edu>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Linux-MM <linux-mm@kvack.org>,
        Ext4 Developers List <linux-ext4@vger.kernel.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <F78E1A78-DB7E-4F3A-8C7C-842AA757E4FE@gmail.com>
References: <CAHk-=wh=YxjEtTpYyhgypKmPJQ8eVLJ4qowmwbnG1bOU06_4Bg@mail.gmail.com>
 <3221175.1624375240@warthog.procyon.org.uk>
 <YNIBb5WPrk8nnKKn@zeniv-ca.linux.org.uk>
 <YNIDdgn0m8d2a0P3@zeniv-ca.linux.org.uk>
 <YNIdJaKrNj5GoT7w@casper.infradead.org>
 <3231150.1624384533@warthog.procyon.org.uk>
 <YNImEkqizzuStW72@casper.infradead.org>
 <CAHk-=wicC9ZTNNH1E-oHebcT3+r4Q4Wf1tXBindXrCdotj20Gg@mail.gmail.com>
 <YNIqjhsvEms6+vk9@casper.infradead.org>
 <CAHk-=wiRumzeOn1Fk-m-FiGf+sA0dSS3YPu--KAkT8-5W5yEHA@mail.gmail.com>
 <YNItqqZA9Y1wOnZY@casper.infradead.org>
To:     Matthew Wilcox <willy@infradead.org>
Precedence: bulk


> On Jun 22, 2021, at 11:36 AM, Matthew Wilcox <willy@infradead.org> =
wrote:
>=20
> On Tue, Jun 22, 2021 at 11:28:30AM -0700, Linus Torvalds wrote:
>> On Tue, Jun 22, 2021 at 11:23 AM Matthew Wilcox <willy@infradead.org> =
wrote:
>>>=20
>>> It wouldn't be _that_ bad necessarily.  filemap_fault:
>>=20
>> It's not actually the mm code that is the biggest problem. We
>> obviously already have readahead support.
>>=20
>> It's the *fault* side.
>>=20
>> In particular, since the fault would return without actually filling
>> in the page table entry (because the page isn't ready yet, and you
>> cannot expose it to other threads!), you also have to jump over the
>> instruction that caused this all.
>=20
> Oh, I was assuming that it'd be a function call like
> get_user_pages_fast(), not an instruction that was specially marked to
> be jumped over.  Gag reflex diminishing now?

Just reminding the alternative (in the RFC that I mentioned before):
a vDSO exception table entry for a memory accessing function in the
vDSO. It then behaves as a sort of MADV_WILLNEED for the faulting
page if an exception is triggered. Unlike MADV_WILLNEED it maps the
page if no IO is needed. It can return through a register whether
the page was present or not.

I once implemented (another) alternative, in which the ELF had a section
with an exception-table (holding all the =E2=80=9CAsync-#PF=E2=80=9D =
instructions),
which described where to skip to if a #PF occurs, but this solution
seemed too heavy-weight/intrusive.