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;
From:   Nadav Amit <namit@vmware.com>
To:     Matthew Wilcox <willy@infradead.org>
CC:     Linux-MM <linux-mm@kvack.org>, LKML <linux-kernel@vger.kernel.org>,
        Hugh Dickins <hughd@google.com>,
        Andy Lutomirski <luto@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        Sean Christopherson <seanjc@google.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        "x86@kernel.org" <x86@kernel.org>
Subject: Re: [RFC 0/6] x86: prefetch_page() vDSO call
Thread-Topic: [RFC 0/6] x86: prefetch_page() vDSO call
Thread-Index: AQHXC3A3AKdXDDevAkSbJ7drCRkvXKppF4sAgAAKEoCAAAWsgA==
Date:   Thu, 25 Feb 2021 17:53:12 +0000
Message-ID: <C47CFA1A-2183-4EC8-B201-C95BB1C7F882@vmware.com>
References: <20210225072910.2811795-1-namit@vmware.com>
 <20210225121645.GZ2858050@casper.infradead.org>
 <0EFCDB0B-DB73-4866-9C0B-7192737CA372@vmware.com>
 <20210225173253.GB2858050@casper.infradead.org>
In-Reply-To: <20210225173253.GB2858050@casper.infradead.org>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <AFDF8ECAA403AD40AC358C22B7D2F7B8@namprd05.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 944e52f7-cc78-483b-8e14-08d8d9b63830
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Feb 2021 17:53:12.2237
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: b39138ca-3cee-4b4a-a4d6-cd83d9dd62f0
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: HWWw1zKQVADCtNFopUOq2umYPIvazN0o8EGb1X4LkR0CgVQLEy0w8LtsD68uIUmYgw4JwTcrh6k+5nkqGAn9DQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SJ0PR05MB7644
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


> On Feb 25, 2021, at 9:32 AM, Matthew Wilcox <willy@infradead.org> wrote:
>=20
> On Thu, Feb 25, 2021 at 04:56:50PM +0000, Nadav Amit wrote:
>>=20
>>> On Feb 25, 2021, at 4:16 AM, Matthew Wilcox <willy@infradead.org> wrote=
:
>>>=20
>>> On Wed, Feb 24, 2021 at 11:29:04PM -0800, Nadav Amit wrote:
>>>> Just as applications can use prefetch instructions to overlap
>>>> computations and memory accesses, applications may want to overlap the
>>>> page-faults and compute or overlap the I/O accesses that are required
>>>> for page-faults of different pages.
>>>=20
>>> Isn't this madvise(MADV_WILLNEED)?
>>=20
>> Good point that I should have mentioned. In a way prefetch_page() a
>> combination of mincore() and MADV_WILLNEED.
>>=20
>> There are 4 main differences from MADV_WILLNEED:
>>=20
>> 1. Much lower invocation cost if the readahead is not needed: this allow=
s
>> to prefetch pages more abundantly.
>=20
> That seems like something that could be fixed in libc -- if we add a
> page prefetch vdso call, an application calling posix_madvise() could
> be implemented by calling this fast path.  Assuming the performance
> increase justifies this extra complexity.
>=20
>> 2. Return value: return value tells you whether the page is accessible.
>> This makes it usable for coroutines, for instance. In this regard the
>> call is more similar to mincore() than MADV_WILLNEED.
>=20
> I don't quite understand the programming model you're describing here.
>=20
>> 3. The PTEs are mapped if the pages are already present in the
>> swap/page-cache, preventing an additional page-fault just to map them.
>=20
> We could enhance madvise() to do this, no?
>=20
>> 4. Avoiding heavy-weight reclamation on low memory (this may need to
>> be selective, and can be integrated with MADV_WILLNEED).
>=20
> Likewise.
>=20
> I don't want to add a new Linux-specific call when there's already a
> POSIX interface that communicates the exact same thing.  The return
> value seems like the only problem.

I agree that this call does not have to be exposed to the application.

I am not sure there is a lot of extra complexity now, but obviously
some evaluations are needed.