From: Jason Gunthorpe <jgg@nvidia.com>
To: "Liu, Yi L" <yi.l.liu@intel.com>
Cc: "mjrosato@linux.ibm.com" <mjrosato@linux.ibm.com>,
"jasowang@redhat.com" <jasowang@redhat.com>,
"Hao, Xudong" <xudong.hao@intel.com>,
"Duan, Zhenzhong" <zhenzhong.duan@intel.com>,
"peterx@redhat.com" <peterx@redhat.com>,
"Xu, Terrence" <terrence.xu@intel.com>,
"chao.p.peng@linux.intel.com" <chao.p.peng@linux.intel.com>,
"linux-s390@vger.kernel.org" <linux-s390@vger.kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"lulu@redhat.com" <lulu@redhat.com>,
"Jiang, Yanting" <yanting.jiang@intel.com>,
"joro@8bytes.org" <joro@8bytes.org>,
"nicolinc@nvidia.com" <nicolinc@nvidia.com>,
"Tian, Kevin" <kevin.tian@intel.com>,
"Zhao, Yan Y" <yan.y.zhao@intel.com>,
"intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>,
"eric.auger@redhat.com" <eric.auger@redhat.com>,
"intel-gvt-dev@lists.freedesktop.org"
<intel-gvt-dev@lists.freedesktop.org>,
"yi.y.sun@linux.intel.com" <yi.y.sun@linux.intel.com>,
"clegoate@redhat.com" <clegoate@redhat.com>,
"cohuck@redhat.com" <cohuck@redhat.com>,
"shameerali.kolothum.thodi@huawei.com"
<shameerali.kolothum.thodi@huawei.com>,
"suravee.suthikulpanit@amd.com" <suravee.suthikulpanit@amd.com>,
"robin.murphy@arm.com" <robin.murphy@arm.com>
Subject: Re: [Intel-gfx] [PATCH v11 20/23] vfio: Add VFIO_DEVICE_[AT|DE]TACH_IOMMUFD_PT
Date: Tue, 6 Jun 2023 11:42:59 -0300 [thread overview]
Message-ID: <ZH9F8zFtD7pPeUE8@nvidia.com> (raw)
In-Reply-To: <DS0PR11MB7529407F01EE55AE4A0A9F1FC3469@DS0PR11MB7529.namprd11.prod.outlook.com>
On Thu, May 25, 2023 at 03:03:54AM +0000, Liu, Yi L wrote:
> vIOMMU may introduce some performance deduction if there
> are frequent map/unmap.
DPDK doesn't do that.
And once you turn on the HW IOMMU you negate alot of the micro
performance wins of bypassing. Maybe there is still some argument
about giant/huge pages or something.
> without vIOMMU is supposed to be more robust. But I'm not
> sure if the noiommu userspace will adapt to cdev noiommu.
> Perhaps yes if group may be deprecated in future.
I think that is more a distro question, and we don't have to answer it
fully now.
Jason
WARNING: multiple messages have this Message-ID (diff)
From: Jason Gunthorpe <jgg@nvidia.com>
To: "Liu, Yi L" <yi.l.liu@intel.com>
Cc: Alex Williamson <alex.williamson@redhat.com>,
"Tian, Kevin" <kevin.tian@intel.com>,
"joro@8bytes.org" <joro@8bytes.org>,
"robin.murphy@arm.com" <robin.murphy@arm.com>,
"cohuck@redhat.com" <cohuck@redhat.com>,
"eric.auger@redhat.com" <eric.auger@redhat.com>,
"nicolinc@nvidia.com" <nicolinc@nvidia.com>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"mjrosato@linux.ibm.com" <mjrosato@linux.ibm.com>,
"chao.p.peng@linux.intel.com" <chao.p.peng@linux.intel.com>,
"yi.y.sun@linux.intel.com" <yi.y.sun@linux.intel.com>,
"peterx@redhat.com" <peterx@redhat.com>,
"jasowang@redhat.com" <jasowang@redhat.com>,
"shameerali.kolothum.thodi@huawei.com"
<shameerali.kolothum.thodi@huawei.com>,
"lulu@redhat.com" <lulu@redhat.com>,
"suravee.suthikulpanit@amd.com" <suravee.suthikulpanit@amd.com>,
"intel-gvt-dev@lists.freedesktop.org"
<intel-gvt-dev@lists.freedesktop.org>,
"intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>,
"linux-s390@vger.kernel.org" <linux-s390@vger.kernel.org>,
"Hao, Xudong" <xudong.hao@intel.com>,
"Zhao, Yan Y" <yan.y.zhao@intel.com>,
"Xu, Terrence" <terrence.xu@intel.com>,
"Jiang, Yanting" <yanting.jiang@intel.com>,
"Duan, Zhenzhong" <zhenzhong.duan@intel.com>,
"clegoate@redhat.com" <clegoate@redhat.com>
Subject: Re: [PATCH v11 20/23] vfio: Add VFIO_DEVICE_[AT|DE]TACH_IOMMUFD_PT
Date: Tue, 6 Jun 2023 11:42:59 -0300 [thread overview]
Message-ID: <ZH9F8zFtD7pPeUE8@nvidia.com> (raw)
In-Reply-To: <DS0PR11MB7529407F01EE55AE4A0A9F1FC3469@DS0PR11MB7529.namprd11.prod.outlook.com>
On Thu, May 25, 2023 at 03:03:54AM +0000, Liu, Yi L wrote:
> vIOMMU may introduce some performance deduction if there
> are frequent map/unmap.
DPDK doesn't do that.
And once you turn on the HW IOMMU you negate alot of the micro
performance wins of bypassing. Maybe there is still some argument
about giant/huge pages or something.
> without vIOMMU is supposed to be more robust. But I'm not
> sure if the noiommu userspace will adapt to cdev noiommu.
> Perhaps yes if group may be deprecated in future.
I think that is more a distro question, and we don't have to answer it
fully now.
Jason
next prev parent reply other threads:[~2023-06-06 14:43 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-13 13:28 [Intel-gfx] [PATCH v11 00/23] Add vfio_device cdev for iommufd support Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 01/23] vfio: Allocate per device file structure Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 02/23] vfio: Refine vfio file kAPIs for KVM Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 03/23] vfio: Accept vfio device file in the KVM facing kAPI Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 19:42 ` [Intel-gfx] " Alex Williamson
2023-05-22 19:42 ` Alex Williamson
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 04/23] kvm/vfio: Rename kvm_vfio_group to prepare for accepting vfio device fd Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 05/23] kvm/vfio: Accept vfio device file from userspace Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 06/23] vfio: Pass struct vfio_device_file * to vfio_device_open/close() Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 07/23] vfio: Block device access via device fd until device is opened Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 08/23] vfio: Add cdev_device_open_cnt to vfio_group Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 09/23] vfio: Make vfio_device_open() single open for device cdev path Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 10/23] vfio-iommufd: Move noiommu compat probe out of vfio_iommufd_bind() Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 20:24 ` [Intel-gfx] " Alex Williamson
2023-05-22 20:24 ` Alex Williamson
2023-05-23 0:45 ` [Intel-gfx] " Liu, Yi L
2023-05-23 0:45 ` Liu, Yi L
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 11/23] vfio-iommufd: Split bind/attach into two steps Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 12/23] vfio: Record devid in vfio_device_file Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 13/23] vfio-iommufd: Add detach_ioas support for physical VFIO devices Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 20:46 ` [Intel-gfx] " Alex Williamson
2023-05-22 20:46 ` Alex Williamson
2023-05-22 20:59 ` [Intel-gfx] " Alex Williamson
2023-05-22 20:59 ` Alex Williamson
2023-05-23 1:34 ` [Intel-gfx] " Liu, Yi L
2023-05-23 1:34 ` Liu, Yi L
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 14/23] iommufd/device: Add iommufd_access_detach() API Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 15/23] vfio-iommufd: Add detach_ioas support for emulated VFIO devices Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 16/23] vfio: Name noiommu vfio_device with "noiommu-" prefix Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 17/23] vfio: Move vfio_device_group_unregister() to be the first operation in unregister Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 18/23] vfio: Add cdev for vfio_device Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 19/23] vfio: Add VFIO_DEVICE_BIND_IOMMUFD Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 22:01 ` [Intel-gfx] " Alex Williamson
2023-05-22 22:01 ` Alex Williamson
2023-05-23 1:41 ` [Intel-gfx] " Liu, Yi L
2023-05-23 1:41 ` Liu, Yi L
2023-05-23 15:51 ` [Intel-gfx] " Alex Williamson
2023-05-23 15:51 ` Alex Williamson
2023-05-24 2:20 ` [Intel-gfx] " Liu, Yi L
2023-05-24 2:20 ` Liu, Yi L
2023-05-24 2:39 ` [Intel-gfx] " Tian, Kevin
2023-05-24 2:39 ` Tian, Kevin
2023-05-24 2:40 ` [Intel-gfx] " Liu, Yi L
2023-05-24 2:40 ` Liu, Yi L
2023-05-24 8:31 ` [Intel-gfx] " Liu, Yi L
2023-05-24 8:31 ` Liu, Yi L
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 20/23] vfio: Add VFIO_DEVICE_[AT|DE]TACH_IOMMUFD_PT Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 22:15 ` [Intel-gfx] " Alex Williamson
2023-05-22 22:15 ` Alex Williamson
2023-05-23 1:20 ` [Intel-gfx] " Liu, Yi L
2023-05-23 1:20 ` Liu, Yi L
2023-05-23 15:50 ` [Intel-gfx] " Alex Williamson
2023-05-23 15:50 ` Alex Williamson
2023-05-24 2:12 ` [Intel-gfx] " Liu, Yi L
2023-05-24 2:12 ` Liu, Yi L
2023-05-24 15:31 ` [Intel-gfx] " Alex Williamson
2023-05-24 15:31 ` Alex Williamson
2023-05-25 3:03 ` [Intel-gfx] " Liu, Yi L
2023-05-25 3:03 ` Liu, Yi L
2023-05-25 15:59 ` [Intel-gfx] " Alex Williamson
2023-05-25 15:59 ` Alex Williamson
2023-05-26 8:38 ` [Intel-gfx] " Liu, Yi L
2023-05-26 8:38 ` Liu, Yi L
2023-06-06 14:42 ` Jason Gunthorpe [this message]
2023-06-06 14:42 ` Jason Gunthorpe
2023-06-06 14:40 ` [Intel-gfx] " Jason Gunthorpe
2023-06-06 14:40 ` Jason Gunthorpe
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 21/23] vfio: Determine noiommu device in __vfio_register_dev() Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-22 23:04 ` [Intel-gfx] " Alex Williamson
2023-05-22 23:04 ` Alex Williamson
2023-05-23 2:13 ` [Intel-gfx] " Liu, Yi L
2023-05-23 2:13 ` Liu, Yi L
2023-05-24 8:14 ` [Intel-gfx] " Liu, Yi L
2023-05-24 8:14 ` Liu, Yi L
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 22/23] vfio: Compile vfio_group infrastructure optionally Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 13:28 ` [Intel-gfx] [PATCH v11 23/23] docs: vfio: Add vfio device cdev description Yi Liu
2023-05-13 13:28 ` Yi Liu
2023-05-13 14:13 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for Add vfio_device cdev for iommufd support (rev13) Patchwork
2023-05-18 5:39 ` [Intel-gfx] [PATCH v11 00/23] Add vfio_device cdev for iommufd support Xu, Terrence
2023-05-18 5:39 ` Xu, Terrence
2023-05-23 7:42 ` [Intel-gfx] " Duan, Zhenzhong
2023-05-23 7:42 ` Duan, Zhenzhong
2023-05-24 8:33 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for Add vfio_device cdev for iommufd support (rev14) Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZH9F8zFtD7pPeUE8@nvidia.com \
--to=jgg@nvidia.com \
--cc=chao.p.peng@linux.intel.com \
--cc=clegoate@redhat.com \
--cc=cohuck@redhat.com \
--cc=eric.auger@redhat.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-gvt-dev@lists.freedesktop.org \
--cc=jasowang@redhat.com \
--cc=joro@8bytes.org \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=lulu@redhat.com \
--cc=mjrosato@linux.ibm.com \
--cc=nicolinc@nvidia.com \
--cc=peterx@redhat.com \
--cc=robin.murphy@arm.com \
--cc=shameerali.kolothum.thodi@huawei.com \
--cc=suravee.suthikulpanit@amd.com \
--cc=terrence.xu@intel.com \
--cc=xudong.hao@intel.com \
--cc=yan.y.zhao@intel.com \
--cc=yanting.jiang@intel.com \
--cc=yi.l.liu@intel.com \
--cc=yi.y.sun@linux.intel.com \
--cc=zhenzhong.duan@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.