Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 2 Jun 2023 17:08:08 -0700
From:   Mike Kravetz <mike.kravetz@oracle.com>
To:     Matthew Wilcox <willy@infradead.org>
Cc:     Tarun Sahu <tsahu@linux.ibm.com>, linux-mm@kvack.org,
        akpm@linux-foundation.org, muchun.song@linux.dev,
        aneesh.kumar@linux.ibm.com, sidhartha.kumar@oracle.com,
        gerald.schaefer@linux.ibm.com, linux-kernel@vger.kernel.org,
        jaypatel@linux.ibm.com
Subject: Re: [PATCH v2] mm/folio: Avoid special handling for order value 0 in
 folio_set_order
Message-ID: <20230603000808.GA29961@monkey>
References: <20230515170809.284680-1-tsahu@linux.ibm.com>
 <ZGJo4UhdyakGFwP7@casper.infradead.org>
 <20230515174513.GB3848@monkey>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230515174513.GB3848@monkey>
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?WAMDpWyt+rQWpYIJ1r1DmFRVJCdGzJ6fY9OMDE6Dt3Dk/wIMMDI+DiEj3XWa?=
 =?us-ascii?Q?76aYNCJg0qL2enHqVlGQDEZSj94hVD03fITQouJ9XDEdOIFDpoiU3EWQx6bv?=
 =?us-ascii?Q?uF+Hi4ur9wceltDZ7eVVicJw+Mi/vfkZ1o0NUlojk7Xu/MQJsTCQ3A+7eq9u?=
 =?us-ascii?Q?JfHusPKauZ9yKTNb35kCWkxtfWNv6xwnTeuGSn9bP0l4hzCItHtVEVkstSvI?=
 =?us-ascii?Q?oD/5rNneQIB/yUIz860Hx3D0obb4BJ3VDEslBvg5DiRAb83H5YvQGPqlNkim?=
 =?us-ascii?Q?2c1FAmvcx7NyIEarVChnpLd1IytGZxPjdpFlvWLGXCrCVBOhd1deAzytdlEM?=
 =?us-ascii?Q?O+Yzow6kZblTC9MiAF4VUzm29PZW9EnZKKGUP96+T19Bppdho3MYQXEZIZ9L?=
 =?us-ascii?Q?J3FCktY15YZYW7M5hNwWmvNC/iNa+6jGyo0yYvVf3v3VNd0hEuQ7/bjwuxUd?=
 =?us-ascii?Q?QDv/DNs/DrNgC/ZZ4mCimXj+2cBELLCUbA4lR1QHAHQln2wJUTeOn4Gs3sJ7?=
 =?us-ascii?Q?haFaqigvcrXk/jHS73eZRuyCoAPIhiuecNIAHF0nIcGKAQrog5f/IkMr7pFv?=
 =?us-ascii?Q?N2sd1CmI6skq8iGZ3QjbkYIgIpBZXH4UV0PE+onJas870usmKR+RykylC2jQ?=
 =?us-ascii?Q?EvVNFVn40LBex5N4aI6MQsSKJs73l6iDW1PUlF7aYNTnyaz67uSWvGCOytRv?=
 =?us-ascii?Q?gBoIe4mkwPg/EKa7+8HYGSl37MMB+g0xQ1qSoydf2X8d6tJYiPU1cTsEza7Q?=
 =?us-ascii?Q?nTD9+hGT7pPyIFijL5gjFPQ6s/XftSD16ShbcHLZFcbkk51nbzFinGPhTQg7?=
 =?us-ascii?Q?DxM0U2gUyYAwZ1yNvxieEDvg+rYmrzCwSP5WXHbdufZguy9zAjQEAmX6pBDZ?=
 =?us-ascii?Q?DNDic9ohfDESyjrcIIZow7suQOumqlVAVPVVcgwqwggSeGco75eZMmZCHNEo?=
 =?us-ascii?Q?i/POfcpSLCfH2qY0nHD8Q0PDq7/9OMdvW6dDxZGsQAGt4qgyXKKV53G02NCN?=
 =?us-ascii?Q?llDRdOZhtKt74EzM+SjMOPJ/1npuPq95GJP4/4/JuF67n0tyTLyRdaXiDVo0?=
 =?us-ascii?Q?jd4T4zm8sQCS0DmLlkRM30bmkhJDilJlUc+ApsWQGpPv6ByruzHH+PojEYmp?=
 =?us-ascii?Q?lx0eflr0TfFzb0vCynYyNm1+FSCkUFo376j5kEcxkg9fRmp1Pckv7PpFw5Vg?=
 =?us-ascii?Q?DV809/dZ3fAIBcvKEeJYdlVmvICjjiG7Xk/syIhrMTWdYLMorTDKXdqayLP5?=
 =?us-ascii?Q?MZ+Xz7oPh4Zkm0+HJP9EM+RauKE3WNSa9+EhuPI4sffCVDBg/5QTDiicF/JY?=
 =?us-ascii?Q?FgSrHSjFMGoitgun8JexQ5CCdAC3XlSKiktlBCePjtIl5X8kK3flG0n2uo86?=
 =?us-ascii?Q?d0P3QQ1hGP6eFjlhOntInJsGcagZGn8lniRFx5NUeafm+bcK1qOFwnOjgXXx?=
 =?us-ascii?Q?1Fj65G5fub+FL3xpEoJv0VlQmH7Vy2OKtU720nB8w9oAE5nB3/GoZcsakg5+?=
 =?us-ascii?Q?YPRzi0ATjXhCoLL/BVlq2Ng+EhzBropoLaE3iuCgRuTdWwEUPjAgfiavX+f4?=
 =?us-ascii?Q?jrir9W6CVCHqpzyXsX1jVaYOppSVVaQCrLzTVQMt?=
X-MS-Exchange-AntiSpam-ExternalHop-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-ExternalHop-MessageData-0: =?us-ascii?Q?UiY1IZtvjHrdSMqdaDZyFYKv62+RikznxdbRk4itBKlBQRpxB09ySWhWRmK5?=
 =?us-ascii?Q?KspoDWxMCeIBSyoBA/hiiOnfvlJkkfPQh3IwhYEGuMShXFW8FaLYuHRqfF2R?=
 =?us-ascii?Q?i4R9alVD2n+OpCI+7DhTnVaHaTkKNZ6r4Nwtqhz5GuTY254SMWIxzgmyTEEz?=
 =?us-ascii?Q?EeUAkxhikg3V69xicacQsL+yhHRKHnrOE5MKDQRj3H0wuGHlU1rr8xw3RnBF?=
 =?us-ascii?Q?lKOzRbBjGaismM/D75yFj02a6Nvc8PDwiGo5sgiKIzoszsGF12E6dtLJmzKe?=
 =?us-ascii?Q?I0h+l9XkWjJzQH11TQ74aasdR8QGbvxDZYWCObTaMCPcDxy2I7EDxCMhT2Qi?=
 =?us-ascii?Q?YrSTGccCMoPEFNrtTIHUxlPELSgTyeFGaBlCpXllQFVKb3vRYIJbyxA/c8V7?=
 =?us-ascii?Q?asJWJoKauGwGL+5laMFxHCk7+6wUK4X+NTto+TUtZg1Q2cX5kkKWOH/hUY2P?=
 =?us-ascii?Q?GfddJndtzzkiVDtmMrFdj3jdkjWxQQg+A+dUFnS746HemJmYpVbQ/ojJAsLI?=
 =?us-ascii?Q?T5TxItj1vnxVOS2ERyftOUVU7XqG6ilSbpfDFw7QIbaZBFb7ngyz/WaYK62S?=
 =?us-ascii?Q?qKVjsv50uB6xraWRBagnH+FW/2pJQQ2fPX73+kOv1a0O05rN7q8EMoow6FWe?=
 =?us-ascii?Q?ogdRCx0op7O1ilet7P9vh4fXYQHtGxb5ptqBWKGVci0X8YezXOmTeQuaIDQG?=
 =?us-ascii?Q?b7WCDPzBAsQq3CA0cPGR7PUQ7PIZc96lHzeIY4/jIlnrFfdKf7+8704Jgn4Y?=
 =?us-ascii?Q?DPMQ6SgUq11wGMZv4QVeHFK3ao3MW3y6F7dRbcb/QlrIw0QHNMFtsASk+bgX?=
 =?us-ascii?Q?y1tZBjhBaQw+d+51eyTSPPrpr3QMUC+qPDAqz2K8Jki/6vRCzrIXONCUNcX3?=
 =?us-ascii?Q?F5m20SXRIblnv2QWUoFzmmwRFvOw2TclV7t2IXL3DNFgpa31nf82H514m/hv?=
X-OriginatorOrg: oracle.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 5e017dd7-b553-4fbe-18e5-08db63c69e71
X-MS-Exchange-CrossTenant-AuthSource: BY5PR10MB4196.namprd10.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 Jun 2023 00:08:11.9277
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 4e2c6054-71cb-48f1-bd6c-3a9705aca71b
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: G4tSo2ROcXdOBjar9ZPdwgxVQ2ZA3wHH5G4DjjxfRVp/KC+QJtWrTpIQ/EbLZSudr0DPTu9CAUE+ElzjYStypQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MW4PR10MB5774
X-Proofpoint-Virus-Version: vendor=baseguard
 engine=ICAP:2.0.254,Aquarius:18.0.957,Hydra:6.0.573,FMLib:17.11.176.26
 definitions=2023-06-02_18,2023-06-02_02,2023-05-22_02
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 suspectscore=0 adultscore=0 spamscore=0
 phishscore=0 malwarescore=0 mlxlogscore=999 mlxscore=0 bulkscore=0
 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2304280000
 definitions=main-2306020191
X-Proofpoint-ORIG-GUID: YfwqAshmsEkX2BYZbrvyxzjjuaT89bvc
X-Proofpoint-GUID: YfwqAshmsEkX2BYZbrvyxzjjuaT89bvc
X-Spam-Status: No, score=-2.8 required=5.0 tests=BAYES_00,DKIM_SIGNED,
        DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,RCVD_IN_DNSWL_LOW,
        RCVD_IN_MSPIKE_H5,RCVD_IN_MSPIKE_WL,SPF_HELO_NONE,SPF_NONE,
        T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on
        lindbergh.monkeyblade.net
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 05/15/23 10:45, Mike Kravetz wrote:
> On 05/15/23 18:16, Matthew Wilcox wrote:
> > On Mon, May 15, 2023 at 10:38:09PM +0530, Tarun Sahu wrote:
> > > @@ -1951,9 +1950,6 @@ static bool __prep_compound_gigantic_folio(struct folio *folio,
> > >  	struct page *p;
> > >  
> > >  	__folio_clear_reserved(folio);
> > > -	__folio_set_head(folio);
> > > -	/* we rely on prep_new_hugetlb_folio to set the destructor */
> > > -	folio_set_order(folio, order);
> > >  	for (i = 0; i < nr_pages; i++) {
> > >  		p = folio_page(folio, i);
> > >  
> > > @@ -1999,6 +1995,9 @@ static bool __prep_compound_gigantic_folio(struct folio *folio,
> > >  		if (i != 0)
> > >  			set_compound_head(p, &folio->page);
> > >  	}
> > > +	__folio_set_head(folio);
> > > +	/* we rely on prep_new_hugetlb_folio to set the destructor */
> > > +	folio_set_order(folio, order);
> > 
> > This makes me nervous, as I said before.  This means that
> > compound_head(tail) can temporarily point to a page which is not marked
> > as a head page.  That's different from prep_compound_page().  You need to
> > come up with some good argumentation for why this is safe, and no amount
> > of testing you do can replace it -- any race in this area will be subtle.

We could continue to set up the head page first as in the current code,
but we need to move the freezing of that page outside the loop.  That is
better then the existing code, however I am not sure if it is any better
than what is proposed here.  I still believe my reasoning below as to
why this proposal is better then the existing code is correct.

Also, that 'folio_set_order(folio, 0)' only exists in the error path of
the current code.  I am not sure if it is actually needed.  Why?  Right
after returning an error, the pages associated with the gigantic page
will be freed.  This is similar to the reason why it can be removed in
__destroy_compound_gigantic_folio.

> I added comments supporting this approach in the first version of the patch.
> My argument was that this is actually safer than the existing code.  That is
> because we freeze the page (ref count 0) before setting compound_head(tail).
> So, nobody should be taking any speculative refs on those tail pages.
> 
> In the existing code, we set the compound page order in the head before
> freezing the head or any tail pages.  Therefore, speculative refs can be
> taken on any of the pages while in this state.
> 
> If we want prep_compound_gigantic_folio to work like prep_compound_page
> we would need to take two passes through the pages.  In the first pass,
> freeze all the pages and in the second set up the compound page.

-- 
Mike Kravetz