Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Paul Burton <paul.burton@mips.com>
To:     Steven Price <steven.price@arm.com>
CC:     Mark Rutland <Mark.Rutland@arm.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Will Deacon <will.deacon@arm.com>,
        "linux-mips@vger.kernel.org" <linux-mips@vger.kernel.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        "Liang, Kan" <kan.liang@linux.intel.com>,
        "x86@kernel.org" <x86@kernel.org>, Ingo Molnar <mingo@redhat.com>,
        James Hogan <jhogan@kernel.org>, Arnd Bergmann <arnd@arndb.de>,
        =?iso-8859-1?Q?J=E9r=F4me_Glisse?= <jglisse@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        Andy Lutomirski <luto@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        "linux-arm-kernel@lists.infradead.org" 
        <linux-arm-kernel@lists.infradead.org>,
        Ard Biesheuvel <ard.biesheuvel@linaro.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Ralf Baechle <ralf@linux-mips.org>,
        James Morse <james.morse@arm.com>
Subject: Re: [PATCH v3 11/34] mips: mm: Add p?d_large() definitions
Thread-Topic: [PATCH v3 11/34] mips: mm: Add p?d_large() definitions
Thread-Index: AQHUzr7i/d0uU6ROnE+g9kAjsVHJ0KX0eXKAgACmggCAAHDjAA==
Date:   Thu, 28 Feb 2019 18:55:28 +0000
Message-ID: <20190228185526.hdryn2zsfign7vht@pburton-laptop>
References: <20190227170608.27963-1-steven.price@arm.com>
 <20190227170608.27963-12-steven.price@arm.com>
 <20190228021526.bb6m3my46ohb4o6h@pburton-laptop>
 <74944d83-f3c0-ff02-590e-b7e5abcea485@arm.com>
In-Reply-To: <74944d83-f3c0-ff02-590e-b7e5abcea485@arm.com>
Accept-Language: en-US
Content-Language: en-US
user-agent: NeoMutt/20180716
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [67.207.99.198]
x-ms-publictraffictype: Email
x-ms-office365-filtering-correlation-id: 550c1196-5533-4a3f-2915-08d69dae4e26
x-microsoft-antispam: BCL:0;PCL:0;RULEID:(2390118)(7020095)(4652040)(8989299)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(5600127)(711020)(4605104)(2017052603328)(7153060)(7193020);SRVR:MWHPR2201MB1359;
x-ms-traffictypediagnostic: MWHPR2201MB1359:
x-microsoft-exchange-diagnostics: =?iso-8859-1?Q?1;MWHPR2201MB1359;23:APQz/DE9p4wWq6b0dZpgpoDdUz3tcNtWFG4R3?=
 =?iso-8859-1?Q?c0SV6754VZ1bgIRlfhf/JpFe0B9xDmVvJ980SCtXoIn6GsCluDx4ng/vga?=
 =?iso-8859-1?Q?Z6MBuXYzMolQlFMMjV4s8HWnG/isT/TcSlu1ATdYhqUB6Lhehb/EQfc11b?=
 =?iso-8859-1?Q?KppF3LNFiXAvVEXwcK1SkCfE5aXXcU2Eox6OzxgEGmFDp8rCcla5Bbti0Z?=
 =?iso-8859-1?Q?eUSKhtccHf50S4STr25a/mvb/Ec24qt437bV9E7jrDqDJzLkPGWpyuaUxq?=
 =?iso-8859-1?Q?LlG1DvGqxIdUt3hqaAy8wUsLB2H6zOBaqfN1qdiI4wDfHjs0HtDxf73v0u?=
 =?iso-8859-1?Q?0+8CtFQHSf/nUW3GuUz/q3Pf/BvfjxSSQWMvQcg1Hm/Z26oJDWdiojK2G/?=
 =?iso-8859-1?Q?5PUozKEd+mJiNVmxeu6qyUi2eOFabVK0Wu1C+XI9KFf2dz01RhhU4QyZpJ?=
 =?iso-8859-1?Q?0QLyXgkgfQocjWq419QWEB3kvYrBquiFweQVgTRNKKYNIFQtJnS3b32KK+?=
 =?iso-8859-1?Q?EUVBa5X0YQJwLmz74Q+IcDDfs+zkrhaOKpxgF6v8WDrjJsHwgxMT6fFOv1?=
 =?iso-8859-1?Q?m1jb83oDEn3+RuWvfmUH3it2kpOaqsqzRff3yfIRYKYLinkv+mqfnOsj9f?=
 =?iso-8859-1?Q?Ci3okklujtmg0fjSvs8pVzdnmH/yjVOjhY90gH4KUvFlZXOQ07Lqr1qDKi?=
 =?iso-8859-1?Q?UczI/luYgvGcFvySRa/35dWfe4JaSkMz74M5xq1S2yMudANxIG4ekmGLex?=
 =?iso-8859-1?Q?rvBoYMIGN3UDHbEC+6cwdklNER7AoS2M8GHJTxZ6/wYqyaoYZ3q+iiqBGp?=
 =?iso-8859-1?Q?JDTtv3SSnhQ3131Y7tHvdL5CDAgf2EjSX9GFDuu2tnVHWW1a2Xbs6gxKvH?=
 =?iso-8859-1?Q?CFA9t2GTka3a6xkYIEgdZEO7snoaTJr46x8Q2QYjWpQvaJVtvZ8q4pYzE3?=
 =?iso-8859-1?Q?66VBzV+0d+ONOOI98vtBY2KflW4E+vbBZ1pRphFAZ6fMD0nDl+6Y4ikWrz?=
 =?iso-8859-1?Q?HSvE9s/rHwDmZi5qkTdDPHFiTxoZsT2j0wU2bIHDSfNmupgg1w9/RgQhe9?=
 =?iso-8859-1?Q?eHTJPAbWR7SSLbbDx1+rWnv+XQ0xSQWsBkhPNfrXSax/n0fZSWAUN4Oz04?=
 =?iso-8859-1?Q?wwtzWYWA+P0V/X5UWKVBNbFdc0+K0/HUY94/UeRae+7S+c5MzjPquc5/RZ?=
 =?iso-8859-1?Q?Y1s4qdv/RuJHkcGKGWVwti841wULQV/1f1b3EwjTq27dNswehDoKLpyltj?=
 =?iso-8859-1?Q?NuliAZXR3L8V3iHUoDTWhKP5EAeynilGhHdrVvp3ifuKBtHasnjEu5S3r9?=
 =?iso-8859-1?Q?wvQSPI01GQn2YHziU1Qj0dCaa0gEWf1cPDv6Xz2JV0xjCI6A2hHvtfdLH9?=
 =?iso-8859-1?Q?zfICz2vkZSEV4jxvkuJWxvt0BeHyn1mf4rRFoCrZObypXQLPJxE4H9XylU?=
 =?iso-8859-1?Q?Cc1VrDz9NxASPKqti2h0ZI/zQQCUiY5JmnOkN?=
x-microsoft-antispam-prvs: <MWHPR2201MB1359F18426AD3A01B32E4F36C1750@MWHPR2201MB1359.namprd22.prod.outlook.com>
x-forefront-prvs: 0962D394D2
x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(7916004)(346002)(136003)(376002)(366004)(396003)(39850400004)(51914003)(199004)(189003)(186003)(6246003)(6486002)(486006)(11346002)(106356001)(446003)(476003)(42882007)(52116002)(53546011)(305945005)(6506007)(386003)(66066001)(93886005)(7416002)(6916009)(256004)(229853002)(478600001)(6436002)(102836004)(33716001)(76176011)(53936002)(9686003)(14444005)(6512007)(4326008)(81156014)(81166006)(25786009)(71190400001)(6116002)(8676002)(71200400001)(5660300002)(97736004)(3846002)(2906002)(1076003)(44832011)(68736007)(316002)(26005)(14454004)(8936002)(54906003)(99286004)(7736002)(105586002)(58126008);DIR:OUT;SFP:1102;SCL:1;SRVR:MWHPR2201MB1359;H:MWHPR2201MB1277.namprd22.prod.outlook.com;FPR:;SPF:None;LANG:en;PTR:InfoNoRecords;A:1;MX:1;
received-spf: None (protection.outlook.com: wavecomp.com does not designate
 permitted sender hosts)
x-ms-exchange-senderadcheck: 1
x-microsoft-antispam-message-info: sQdOpLduEp5xt+4qfJPbXOED9r3yioy1Wm4u843ee1EHOpDndmuGOWtHeeEtfv7fmvwKXnrqYQ/WdZPeolaf87flYadfVSP0Q02VGgzMHyl0kRAOREUwzDb7DHDhSE2MJ+atyIHKdNIRJ984narfxtSwpjcYIrzJ6J0lMFwHJNAwnTn3Idkyz5nDJP/hzUBnuV3sLFcaGf9QVMr9H4ql6tzAhAkRY+x2XgCCbVYVmDHBAy2B7By8rTX5T+JkwV2mSn8ZIeepb/MPFnunBlvSZ91Q6DsXQBgpl39gRSNh4h3oPrpxRaie9ioPH/gaMTRYwvro7ji8pnMi9Wm/+F9GVMo6yOyTJcisaR0x3TD8dpWuaErqJ+/ZEilf5J+MFevYHeLLbG+C4ldpsP80HzkUOwSKiJiUYPD3YpLV6RmHfNE=
Content-Type: text/plain; charset="iso-8859-1"
Content-ID: <FFBA1DC130D1404CBBCECA4240085204@namprd22.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: mips.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 550c1196-5533-4a3f-2915-08d69dae4e26
X-MS-Exchange-CrossTenant-originalarrivaltime: 28 Feb 2019 18:55:27.9917
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-id: 463607d3-1db3-40a0-8a29-970c56230104
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MWHPR2201MB1359
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Hi Steven,

On Thu, Feb 28, 2019 at 12:11:24PM +0000, Steven Price wrote:
> On 28/02/2019 02:15, Paul Burton wrote:
> > On Wed, Feb 27, 2019 at 05:05:45PM +0000, Steven Price wrote:
> >> For mips, we don't support large pages on 32 bit so add stubs returnin=
g 0.
> >=20
> > So far so good :)
> >=20
> >> For 64 bit look for _PAGE_HUGE flag being set. This means exposing the
> >> flag when !CONFIG_MIPS_HUGE_TLB_SUPPORT.
> >=20
> > Here I have to ask why? We could just return 0 like the mips32 case whe=
n
> > CONFIG_MIPS_HUGE_TLB_SUPPORT=3Dn, let the compiler optimize the whole
> > thing out and avoid redundant work at runtime.
> >=20
> > This could be unified too in asm/pgtable.h - checking for
> > CONFIG_MIPS_HUGE_TLB_SUPPORT should be sufficient to cover the mips32
> > case along with the subset of mips64 configurations without huge pages.
>=20
> The intention here is to define a new set of macros/functions which will
> always tell us whether we're at the leaf of a page table walk, whether
> or not huge pages are compiled into the kernel. Basically this allows
> the page walking code to be used on page tables other than user space,
> for instance the kernel page tables (which e.g. might use a large
> mapping for linear memory even if huge pages are not compiled in) or
> page tables from firmware (e.g. EFI on arm64).
>=20
> I'm not familiar enough with mips to know how it handles things like the
> linear map so I don't know how relevant that is, but I'm trying to
> introduce a new set of functions which differ from the existing
> p?d_huge() macros by not depending on whether these mappings could exist
> for a user space VMA (i.e. not depending on HUGETLB support and existing
> for all levels that architecturally they can occur at).

Thanks for the explanation - the background helps.

Right now for MIPS, with one exception, there'll be no difference
between a page being huge or large. So for the vast majority of kernels
with CONFIG_MIPS_HUGE_TLB_SUPPORT=3Dn we should just return 0.

The one exception I mentioned is old SGI IP27 support, which allows the
kernel to be mapped through the TLB & does that using 2x 16MB pages when
CONFIG_MAPPED_KERNEL=3Dy. However even there your patch as-is won't pick
up on that for 2 reasons:

  1) The pages in question don't appear to actually be recorded in the
     page tables - they're just written straight into the TLB as wired
     entries (ie. entries that will never be evicted).

  2) Even if they were in the page tables the _PAGE_HUGE bit isn't set.

Since those pages aren't recorded in the page tables anyway we'd either
need to:

  a) Add them to the page tables, and set the _PAGE_HUGE bit.

  b) Ignore them if the code you're working on won't be operating on the
     memory mapping the kernel.

For other platforms the kernel is run from unmapped memory, and for all
cases including IP27 the kernel will use unmapped memory to access
lowmem or peripherals when possible. That is, MIPS has virtual address
regions ((c)kseg[01] or xkphys) which are architecturally defined as
linear maps to physical memory & so VA->PA translation doesn't use the
TLB at all.

So my thought would be that for almost everything we could just do:

  #define pmd_large(pmd)	pmd_huge(pmd)
  #define pud_large(pmd)	pud_huge(pmd)

And whether we need to do anything about IP27 depends on whether a) or
b) is chosen above.

Or alternatively you could do something like:

  #ifdef _PAGE_HUGE

  static inline int pmd_large(pmd_t pmd)
  {
  	return (pmd_val(pmd) & _PAGE_HUGE) !=3D 0;
  }

  static inline int pud_large(pud_t pud)
  {
  	return (pud_val(pud) & _PAGE_HUGE) !=3D 0;
  }

  #else
  # define pmd_large(pmd)	0
  # define pud_large(pud)	0
  #endif

That would cover everything except for the IP27, but would make it pick
up the IP27 kernel pages automatically if someone later defines
_PAGE_HUGE for IP27 CONFIG_MAPPED_KERNEL=3Dy & makes use of it for those
pages.

Thanks,
    Paul