Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.33 as permitted sender) client-ip=23.128.96.33;
Date:   Thu, 12 Oct 2023 16:09:31 -0300
From:   Jason Gunthorpe <jgg@nvidia.com>
To:     Yi Liu <yi.l.liu@intel.com>
Cc:     joro@8bytes.org, alex.williamson@redhat.com, kevin.tian@intel.com,
        robin.murphy@arm.com, baolu.lu@linux.intel.com, cohuck@redhat.com,
        eric.auger@redhat.com, nicolinc@nvidia.com, kvm@vger.kernel.org,
        mjrosato@linux.ibm.com, chao.p.peng@linux.intel.com,
        yi.y.sun@linux.intel.com, peterx@redhat.com, jasowang@redhat.com,
        shameerali.kolothum.thodi@huawei.com, lulu@redhat.com,
        suravee.suthikulpanit@amd.com, iommu@lists.linux.dev,
        linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org,
        zhenzhong.duan@intel.com, joao.m.martins@oracle.com
Subject: Re: [PATCH v4 05/17] iommufd: Separate kernel-managed HWPT
 alloc/destroy/abort functions
Message-ID: <20231012190931.GO3952@nvidia.com>
References: <20230921075138.124099-1-yi.l.liu@intel.com>
 <20230921075138.124099-6-yi.l.liu@intel.com>
 <20231010184932.GT3952@nvidia.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20231010184932.GT3952@nvidia.com>
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?NT9eamY/mDhphzD1vehN45NtM+i4v0T3CVyTc9NrDeMPMEy9RbWSmyL8k4ZK?=
 =?us-ascii?Q?IwPC5peMxHMR2FbHHhrxMUhbm+ecMr69g7IQeVVLuRGqa82p6bE4qH6uSNPS?=
 =?us-ascii?Q?/NbbP76sfRmF6mTUukiSoySBhrDd8lhBsKPeRMqcN3ybbxd+pNgYS05eRBvt?=
 =?us-ascii?Q?GQNJTfZCr0tLVJleqkDDjoujq5Edsyyh5YPitoOxiyphpY3Meh0lamARSL0z?=
 =?us-ascii?Q?ylqOwu1LAayg4YDUXXmo6exM5i8SZ7IJuyYlA7UMIXsEexCzW28LXIUOu0ox?=
 =?us-ascii?Q?xdrhuEq7t1cB7eoPYR2DW0/ZkONKQ2822pThNyVrtNaRP0YtqB3zrUX1iyw8?=
 =?us-ascii?Q?ijnDCW3x5H0eNk+E6NU42kB+gPNBF4+b8zlf8hZ15nTHIWePzVFID56r8BBj?=
 =?us-ascii?Q?V/1Kp0SOUQ8Y1Dnxv9I9eg0egEpiyos4IndfREGXSeLasU8HiePxdRkvLQS6?=
 =?us-ascii?Q?Wzj9ZPPau5zj27HWtcnDxa1ngYShMCE39f8HWXgjLELb68qJaP05enTNh/L1?=
 =?us-ascii?Q?0MXmKKl2irRyimP50FBFOREoeUVROiAAaHJRNT5mL4NxGza2anHC0AQ+dWkC?=
 =?us-ascii?Q?witq0tsBt570penzJfdejuGMCn79x0yBqMohZ0gs9sSexUtKiU76B8vqwxMA?=
 =?us-ascii?Q?znAfuoJIqjz6j0lpJNvt/twBYZDV13LZrsXmk1p95jBY20kRwNLErqryd+S7?=
 =?us-ascii?Q?YlQ88z6x+BPk+oVJ82d4XgsqGyLWsrq8CEThC/EmVul6oYI7C57VYHctWccT?=
 =?us-ascii?Q?ru8rE/qSiztIyyqulq9L+irJX2TzqAMha/yUFtTgto6Jnn3AFSN4LSjs88mf?=
 =?us-ascii?Q?kMDqe7SRm47eO0o54pSv4/oz2lxWOterhoWsKrgdc/eviSeSSMdNbrj1U0H3?=
 =?us-ascii?Q?iKJ7kV+qjNrX+bBf7VacbK5zZY34yK3zjZMAqF0px/Wx1gZeFFOZBLubqXNX?=
 =?us-ascii?Q?n2VHd1gB26bSKLOmIRgqACV7OKl6tMX+zc6WVkX0lP/a3q8KpXczWTsgK/+2?=
 =?us-ascii?Q?c1MmmRlApAquROrmwQQxMxkV/JvYOxIHAbWzoeHQ3/F++zNhSiTvvWYwVusE?=
 =?us-ascii?Q?+AUzDH3tD7bw09YFwRu9f60HZFNBisFP9xhfS/gwcQgGsyWeKz6sXwATurdo?=
 =?us-ascii?Q?k/qajq3phkuU/3D4xHgBj4savOaAOppozMgLfYyWvBsZBdpUYuv8O/Sf5dXA?=
 =?us-ascii?Q?7YX6OM2flpCL6tSTuAS87Ggtx6onUjTFbmmIz4ndR9JOyjD7lUPbsxViXVQV?=
 =?us-ascii?Q?h65/LfeFKkuj6dewqtSY1TYD8q3iL4rWMaqdfRhk+mok6UTCJEy6fiMsjVOU?=
 =?us-ascii?Q?bC0laCKgzJPnF68/wsr5vRgHvCFtJcLCo7HtVasN+H/g7vZfnx74bK0yhoLG?=
 =?us-ascii?Q?DiBAq/PpaTgSf6Kniwhp6O6vsS4lKXSIu1wVerLf83FekZFfMqp909Niiuv3?=
 =?us-ascii?Q?CYI1lK3ysE/VIBfJZttt6MhLKFY75wdXsniE2G1PRyL1DzYCHvDm3F7kqDxC?=
 =?us-ascii?Q?QkXwwVqzRLMpEX54DzR9V7S/gLQqXpWACo2W1563J92iLM9YVONeSVgjatWS?=
 =?us-ascii?Q?AbSEX0OWfDzdddB1064IRjFUExsWMZEL52I1yXxu?=
X-OriginatorOrg: Nvidia.com
X-MS-Exchange-CrossTenant-Network-Message-Id: c5ea6eb3-69a5-4c94-3c7d-08dbcb56c417
X-MS-Exchange-CrossTenant-AuthSource: LV2PR12MB5869.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 12 Oct 2023 19:09:32.2381
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: jPGbKjVhfdjcPtr+JrGICoIYSaPHvZC0mgxYp5KiVVDuyizn3akpyZbnoDaz7vI8
X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH7PR12MB9176
X-Spam-Status: No, score=-0.9 required=5.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED,
	DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,
	MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS autolearn=unavailable
	autolearn_force=no version=3.4.6
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lipwig.vger.email
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
X-Greylist: Sender passed SPF test, not delayed by milter-greylist-4.6.4 (lipwig.vger.email [0.0.0.0]); Thu, 12 Oct 2023 12:09:58 -0700 (PDT)

On Tue, Oct 10, 2023 at 03:49:32PM -0300, Jason Gunthorpe wrote:
> > diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h
> > index 1d3b1a74e854..3e89c3d530f3 100644
> > --- a/drivers/iommu/iommufd/iommufd_private.h
> > +++ b/drivers/iommu/iommufd/iommufd_private.h
> > @@ -234,6 +234,9 @@ struct iommufd_hw_pagetable {
> >  	struct iommufd_object obj;
> >  	struct iommu_domain *domain;
> >  
> > +	void (*abort)(struct iommufd_object *obj);
> > +	void (*destroy)(struct iommufd_object *obj);
> > +
> >  	union {
> >  		struct { /* kernel-managed */
> >  			struct iommufd_ioas *ioas;
> 
> I think if you are doing this you are trying too hard to share the
> struct.. Defaintely want to avoid function pointers in general, and
> function pointers in a writable struct in particular.

I looked at this for a while and I do still have the feeling that
probably two structs and even two type IDs is probably a cleaner
design.

Like this:

// Or maybe use obj.type ?
enum iommufd_hw_pagetable_type {
	IOMMUFD_HWPT_PAGING,
	IOMMUFD_HWPT_NESTED,
};

struct iommufd_hw_pagetable_common {
	struct iommufd_object obj;
	struct iommu_domain *domain;
	enum iommufd_hw_pagetable_type type;
};

struct iommufd_hw_pagetable {
	struct iommufd_hw_pagetable_common common;
	struct iommufd_ioas *ioas;
	bool auto_domain : 1;
	bool enforce_cache_coherency : 1;
	bool msi_cookie : 1;
	/* Head at iommufd_ioas::hwpt_list */
	struct list_head hwpt_item;
};

struct iommufd_hw_pagetable_nested {
	struct iommufd_hw_pagetable_common common;
	// ??
};

I poked at it in an editor for a bit and it was looking OK but
requires breaking up a bunch of functions then I ran out of time

Also, we probably should feed enforce_cache_coherency through the
alloc_hwpt uapi and not try to autodetect it..

Jason