MIME-Version: 1.0
References: <20230202112915.867409-1-usama.anjum@collabora.com>
 <20230202112915.867409-4-usama.anjum@collabora.com> <CABb0KFEgsk+YidSXBYQ9mM8nVV6PuEOQf=bbNn7hsoG1hUeLZg@mail.gmail.com>
 <36ddfd75-5c58-197b-16c9-9f819099ea6d@collabora.com> <CABb0KFGWi0dtgXZ-AeUuHb55EgnwTu3JfJ9cW3ftCqezKi8dAQ@mail.gmail.com>
 <CANaxB-x2OrTPziL_hgwgQ1xe-ypVrvEJZK5i4ZvmUwsLqfTcvA@mail.gmail.com>
In-Reply-To: <CANaxB-x2OrTPziL_hgwgQ1xe-ypVrvEJZK5i4ZvmUwsLqfTcvA@mail.gmail.com>
From:   =?UTF-8?B?TWljaGHFgiBNaXJvc8WCYXc=?= <emmir@google.com>
Date:   Sat, 25 Feb 2023 10:38:54 +0100
Message-ID: <CABb0KFHYr-_o9bQAwqaXkKC9Bipo18b95FawhkUG-vOQwGNpxQ@mail.gmail.com>
Subject: Re: [PATCH v10 3/6] fs/proc/task_mmu: Implement IOCTL to get and/or
 the clear info about PTEs
To:     Andrei Vagin <avagin@gmail.com>
Cc:     Muhammad Usama Anjum <usama.anjum@collabora.com>,
        Mike Rapoport <rppt@kernel.org>, Nadav Amit <namit@vmware.com>,
        David Hildenbrand <david@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Paul Gofman <pgofman@codeweavers.com>,
        Cyrill Gorcunov <gorcunov@gmail.com>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        Shuah Khan <shuah@kernel.org>,
        Christian Brauner <brauner@kernel.org>,
        Yang Shi <shy828301@gmail.com>,
        Vlastimil Babka <vbabka@suse.cz>,
        "Liam R . Howlett" <Liam.Howlett@oracle.com>,
        Yun Zhou <yun.zhou@windriver.com>,
        Suren Baghdasaryan <surenb@google.com>,
        Alex Sierra <alex.sierra@amd.com>,
        Peter Xu <peterx@redhat.com>,
        Matthew Wilcox <willy@infradead.org>,
        Pasha Tatashin <pasha.tatashin@soleen.com>,
        Axel Rasmussen <axelrasmussen@google.com>,
        "Gustavo A . R . Silva" <gustavoars@kernel.org>,
        Dan Williams <dan.j.williams@intel.com>,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        linux-mm@kvack.org, linux-kselftest@vger.kernel.org,
        Greg KH <gregkh@linuxfoundation.org>, kernel@collabora.com,
        Danylo Mocherniuk <mdanylo@google.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, 24 Feb 2023 at 03:20, Andrei Vagin <avagin@gmail.com> wrote:
>
> On Tue, Feb 21, 2023 at 4:42=E2=80=AFAM Micha=C5=82 Miros=C5=82aw <emmir@=
google.com> wrote:
> >
> > On Tue, 21 Feb 2023 at 11:28, Muhammad Usama Anjum
> > <usama.anjum@collabora.com> wrote:
> > >
> > > Hi Micha=C5=82,
> > >
> > > Thank you so much for comment!
> > >
> > > On 2/17/23 8:18=E2=80=AFPM, Micha=C5=82 Miros=C5=82aw wrote:
> > [...]
> > > > For the page-selection mechanism, currently required_mask and
> > > > excluded_mask have conflicting
> > > They are opposite of each other:
> > > All the set bits in required_mask must be set for the page to be sele=
cted.
> > > All the set bits in excluded_mask must _not_ be set for the page to b=
e
> > > selected.
> > >
> > > > responsibilities. I suggest to rework that to:
> > > > 1. negated_flags: page flags which are to be negated before applyin=
g
> > > > the page selection using following masks;
> > > Sorry I'm unable to understand the negation (which is XOR?). Lets loo=
k at
> > > the truth table:
> > > Page Flag       negated_flags
> > > 0               0                       0
> > > 0               1                       1
> > > 1               0                       1
> > > 1               1                       0
> > >
> > > If a page flag is 0 and negated_flag is 1, the result would be 1 whic=
h has
> > > changed the page flag. It isn't making sense to me. Why the page flag=
 bit
> > > is being fliped?
> > >
> > > When Anrdei had proposed these masks, they seemed like a fancy way of
> > > filtering inside kernel and it was straight forward to understand. Th=
ese
> > > masks would help his use cases for CRIU. So I'd included it. Please c=
an you
> > > elaborate what is the purpose of negation?
> >
> > The XOR is a way to invert the tested value of a flag (from positive
> > to negative and the other way) without having the API with invalid
> > values (with required_flags and excluded_flags you need to define a
> > rule about what happens if a flag is present in both of the masks -
> > either prioritise one mask over the other or reject the call).
> > (Note: the XOR is applied only to the value of the flags for the
> > purpose of testing page-selection criteria.)
>
> Micha=C5=82,
>
> Your API isn't much different from the current one, but it requires
> a bit more brain activity for understanding.
>
> The current set of masks can be easy translated to the new one:
> negated_flags =3D excluded_flags
> required_flags_new =3D excluded_flags | required_flags
>
> As for invalid values, I think it is an advantage of the current API.
> I mean we can easily detect invalid values and return EINVAL. With your
> API, such mistakes will be undetectable.
>
> As for priorities, I don't see this problem here If I don't miss somethin=
g.
>
> We can rewrite the code this way:
> ```
> if (required_mask && ((page_flags & required_mask) !=3D required_mask)
>   skip page;
> if (anyof_mask && !(page_flags & anyof_mask))
>   skip page;
> if (page_flags & excluded_mask)
>   skip page;
> ```
>
> I think the result is always the same no matter in what order each
> mask is applied.

Hi,

I would not want the discussion to wander into easier/harder territory
as that highty depends on experience one has. What I'm arguing about
is the consistency of the API. Let me expand a bit on that.

We have two ways to look at the page_flags:
 A. the field represents a *set of elements* (tags, attributes)
present on the page;
 B. the field represents a bitfield (structure; a fixed set of boolean
fields having a value of 0 or 1)

From A follows the include/exclude way of API design for matching the
flags, and from B the matched mask (which flags to check) + value set
(what values to require).

My argument is that B is consistent with how the flags are used in the
kernel: we don't have operations that add or remove flags, but we have
operations that set or change their value.

Best Regards
Micha=C5=82 Miros=C5=82aw