qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Steven Sistare <steven.sistare@oracle.com>
To: "Duan, Zhenzhong" <zhenzhong.duan@intel.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: Alex Williamson <alex.williamson@redhat.com>,
	Cedric Le Goater <clg@redhat.com>,
	"Liu, Yi L" <yi.l.liu@intel.com>,
	Eric Auger <eric.auger@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Marcel Apfelbaum <marcel.apfelbaum@gmail.com>,
	Peter Xu <peterx@redhat.com>, Fabiano Rosas <farosas@suse.de>
Subject: Re: [PATCH V3 36/42] vfio/iommufd: preserve descriptors
Date: Mon, 19 May 2025 11:53:04 -0400	[thread overview]
Message-ID: <fee4df02-876e-420b-8afa-ca8745c81c1f@oracle.com> (raw)
In-Reply-To: <SJ0PR11MB67441B4BA69BF42F30261EB19293A@SJ0PR11MB6744.namprd11.prod.outlook.com>

On 5/16/2025 6:06 AM, Duan, Zhenzhong wrote:
>> -----Original Message-----
>> From: Steve Sistare <steven.sistare@oracle.com>
>> Subject: [PATCH V3 36/42] vfio/iommufd: preserve descriptors
>>
>> Save the iommu and vfio device fd in CPR state when it is created.
>> After CPR, the fd number is found in CPR state and reused.  Remember
>> the reused status for subsequent patches.  The reused status is cleared
>> when vmstate load finishes.
>>
>> Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
>> ---
>> backends/iommufd.c       | 19 ++++++++++---------
>> hw/vfio/cpr-iommufd.c    | 16 ++++++++++++++++
>> hw/vfio/device.c         | 10 ++--------
>> hw/vfio/iommufd.c        | 13 +++++++++++--
>> include/system/iommufd.h |  1 +
>> 5 files changed, 40 insertions(+), 19 deletions(-)
>>
>> diff --git a/backends/iommufd.c b/backends/iommufd.c
>> index 6fed1c1..492747c 100644
>> --- a/backends/iommufd.c
>> +++ b/backends/iommufd.c
>> @@ -16,12 +16,18 @@
>> #include "qemu/module.h"
>> #include "qom/object_interfaces.h"
>> #include "qemu/error-report.h"
>> +#include "migration/cpr.h"
>> #include "monitor/monitor.h"
>> #include "trace.h"
>> #include "hw/vfio/vfio-device.h"
>> #include <sys/ioctl.h>
>> #include <linux/iommufd.h>
>>
>> +static const char *iommufd_fd_name(IOMMUFDBackend *be)
>> +{
>> +    return object_get_canonical_path_component(OBJECT(be));
>> +}
>> +
>> static void iommufd_backend_init(Object *obj)
>> {
>>      IOMMUFDBackend *be = IOMMUFD_BACKEND(obj);
>> @@ -47,9 +53,8 @@ static void iommufd_backend_set_fd(Object *obj, const
>> char *str, Error **errp)
>>      IOMMUFDBackend *be = IOMMUFD_BACKEND(obj);
>>      int fd = -1;
>>
>> -    fd = monitor_fd_param(monitor_cur(), str, errp);
>> +    fd = cpr_get_fd_param(iommufd_fd_name(be), str, 0, &be->cpr_reused, errp);
>>      if (fd == -1) {
>> -        error_prepend(errp, "Could not parse remote object fd %s:", str);
>>          return;
>>      }
>>      be->fd = fd;
>> @@ -95,14 +100,9 @@ bool iommufd_change_process(IOMMUFDBackend *be,
>> Error **errp)
>>
>> bool iommufd_backend_connect(IOMMUFDBackend *be, Error **errp)
>> {
>> -    int fd;
>> -
>>      if (be->owned && !be->users) {
>> -        fd = qemu_open("/dev/iommu", O_RDWR, errp);
>> -        if (fd < 0) {
>> -            return false;
>> -        }
>> -        be->fd = fd;
>> +        be->fd = cpr_open_fd("/dev/iommu", O_RDWR, iommufd_fd_name(be), 0,
>> +                             &be->cpr_reused, errp);
> 
> Need to check error before assign to be->fd.

will do.

>>      }
>>      be->users++;
>>
>> @@ -121,6 +121,7 @@ void iommufd_backend_disconnect(IOMMUFDBackend
>> *be)
>>          be->fd = -1;
>>      }
>> out:
>> +    cpr_delete_fd(iommufd_fd_name(be), 0);
>>      trace_iommufd_backend_disconnect(be->fd, be->users);
>> }
>>
>> diff --git a/hw/vfio/cpr-iommufd.c b/hw/vfio/cpr-iommufd.c
>> index 46f2006..b760bd3 100644
>> --- a/hw/vfio/cpr-iommufd.c
>> +++ b/hw/vfio/cpr-iommufd.c
>> @@ -8,6 +8,7 @@
>> #include "qemu/osdep.h"
>> #include "qapi/error.h"
>> #include "hw/vfio/vfio-cpr.h"
>> +#include "hw/vfio/vfio-device.h"
>> #include "migration/blocker.h"
>> #include "migration/cpr.h"
>> #include "migration/migration.h"
>> @@ -25,10 +26,25 @@ static bool vfio_cpr_supported(VFIOIOMMUFDContainer
>> *container, Error **errp)
>>      return true;
>> }
>>
>> +static int vfio_container_post_load(void *opaque, int version_id)
>> +{
>> +    VFIOIOMMUFDContainer *container = opaque;
>> +    VFIOContainerBase *bcontainer = &container->bcontainer;
>> +    VFIODevice *vbasedev;
>> +
>> +    QLIST_FOREACH(vbasedev, &bcontainer->device_list, container_next) {
>> +        vbasedev->cpr.reused = false;
>> +    }
>> +    container->be->cpr_reused = false;
> 
> It's strange to set iommufd and vfio device's reused in container's post load,
> Maybe better to do it in their own post load handler?

vfio_container_post_load has MIG_PRI_LOW so it is called last, which guarantees
that be->cpr_reused remains true while all devices are loaded.  This is required
so that we supress dma_map calls during device load processing:

   iommufd_backend_map_file_dma()
     if (be->cpr_reused)
       return 0;

"vbasedev->cpr.reused = false" could be moved to vfio_device_post_load.
I put it here to be future proof -- al reused flags are cleared together,
at the end of post_load, and to be consistent with cpr-legacy.c:vfio_container_post_load

>> +
>> +    return 0;
>> +}
>> +
>> static const VMStateDescription vfio_container_vmstate = {
>>      .name = "vfio-iommufd-container",
>>      .version_id = 0,
>>      .minimum_version_id = 0,
>> +    .post_load = vfio_container_post_load,
>>      .needed = cpr_needed_for_reuse,
>>      .fields = (VMStateField[]) {
>>          VMSTATE_END_OF_LIST()
>> diff --git a/hw/vfio/device.c b/hw/vfio/device.c
>> index 8e9de68..02f384e 100644
>> --- a/hw/vfio/device.c
>> +++ b/hw/vfio/device.c
>> @@ -312,14 +312,8 @@ bool vfio_device_get_name(VFIODevice *vbasedev,
>> Error **errp)
>>
>> void vfio_device_set_fd(VFIODevice *vbasedev, const char *str, Error **errp)
>> {
>> -    ERRP_GUARD();
>> -    int fd = monitor_fd_param(monitor_cur(), str, errp);
>> -
>> -    if (fd < 0) {
>> -        error_prepend(errp, "Could not parse remote object fd %s:", str);
>> -        return;
>> -    }
>> -    vbasedev->fd = fd;
>> +    vbasedev->fd = cpr_get_fd_param(vbasedev->dev->id, str, 0,
>> +                                    &vbasedev->cpr.reused, errp);
> 
> Same here.

Do you mean, "need to check error"?
If so, no need.  The new function definition is:

void vfio_device_set_fd(VFIODevice *vbasedev, const char *str, Error **errp)
{
     vbasedev->fd = cpr_get_fd_param(vbasedev->dev->id, str, 0,
                                     &vbasedev->cpr.reused, errp);
}

cpr_get_fd_param() returns -1 on error and sets errp.

>> }
>>
>> static VFIODeviceIOOps vfio_device_io_ops_ioctl;
>> diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c
>> index dabb948..046f601 100644
>> --- a/hw/vfio/iommufd.c
>> +++ b/hw/vfio/iommufd.c
>> @@ -26,6 +26,7 @@
>> #include "qemu/cutils.h"
>> #include "qemu/chardev_open.h"
>> #include "migration/blocker.h"
>> +#include "migration/cpr.h"
>> #include "pci.h"
>> #include "vfio-iommufd.h"
>> #include "vfio-helpers.h"
>> @@ -530,13 +531,18 @@ static bool iommufd_cdev_attach(const char *name,
>> VFIODevice *vbasedev,
>>
>> VFIO_IOMMU_CLASS(object_class_by_name(TYPE_VFIO_IOMMU_IOMMUFD));
>>
>>      if (vbasedev->fd < 0) {
>> -        devfd = iommufd_cdev_getfd(vbasedev->sysfsdev, errp);
>> +        devfd = cpr_find_fd(vbasedev->name, 0);
>> +        vbasedev->cpr.reused = (devfd >= 0);
>> +        if (!vbasedev->cpr.reused) {
>> +            devfd = iommufd_cdev_getfd(vbasedev->sysfsdev, errp);
>> +        }
>>          if (devfd < 0) {
>>              return false;
>>          }
>>          vbasedev->fd = devfd;
>>      } else {
>>          devfd = vbasedev->fd;
>> +        /* reused was set in iommufd_backend_set_fd */
> 
> Should be vfio_device_set_fd

yes.

- Steve

>>      }
>>
>>      if (!iommufd_cdev_connect_and_bind(vbasedev, errp)) {
>> @@ -634,7 +640,9 @@ found_container:
>>
>>      vfio_device_prepare(vbasedev, bcontainer, &dev_info);
>>      vfio_iommufd_cpr_register_device(vbasedev);
>> -
>> +    if (!vbasedev->cpr.reused) {
>> +        cpr_save_fd(vbasedev->name, 0, vbasedev->fd);
>> +    }
>>      trace_iommufd_cdev_device_info(vbasedev->name, devfd, vbasedev-
>>> num_irqs,
>>                                     vbasedev->num_regions, vbasedev->flags);
>>      return true;
>> @@ -673,6 +681,7 @@ static void iommufd_cdev_detach(VFIODevice *vbasedev)
>>
>>      migrate_del_blocker(&vbasedev->cpr.id_blocker);
>>      vfio_iommufd_cpr_unregister_device(vbasedev);
>> +    cpr_delete_fd(vbasedev->name, 0);
>>      iommufd_cdev_unbind_and_disconnect(vbasedev);
>>      close(vbasedev->fd);
>> }
>> diff --git a/include/system/iommufd.h b/include/system/iommufd.h
>> index db9ed53..5c17abd 100644
>> --- a/include/system/iommufd.h
>> +++ b/include/system/iommufd.h
>> @@ -32,6 +32,7 @@ struct IOMMUFDBackend {
>>      /*< protected >*/
>>      int fd;            /* /dev/iommu file descriptor */
>>      bool owned;        /* is the /dev/iommu opened internally */
>> +    bool cpr_reused;   /* fd is reused after CPR */
>>      uint32_t users;
>>
>>      /*< public >*/
>> --
>> 1.8.3.1
> 



  reply	other threads:[~2025-05-19 15:53 UTC|newest]

Thread overview: 157+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-12 15:32 [PATCH V3 00/42] Live update: vfio and iommufd Steve Sistare
2025-05-12 15:32 ` [PATCH V3 01/42] MAINTAINERS: Add reviewer for CPR Steve Sistare
2025-05-15  7:36   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 02/42] migration: cpr helpers Steve Sistare
2025-05-15  7:43   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 03/42] migration: lower handler priority Steve Sistare
2025-05-12 15:32 ` [PATCH V3 04/42] vfio: vfio_find_ram_discard_listener Steve Sistare
2025-05-12 15:32 ` [PATCH V3 05/42] vfio: move vfio-cpr.h Steve Sistare
2025-05-15  7:46   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 06/42] vfio/container: register container for cpr Steve Sistare
2025-05-15  7:54   ` Cédric Le Goater
2025-05-15 19:06     ` Steven Sistare
2025-05-16 16:20       ` Cédric Le Goater
2025-05-16 17:21         ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 07/42] vfio/container: preserve descriptors Steve Sistare
2025-05-15 12:59   ` Cédric Le Goater
2025-05-15 19:08     ` Steven Sistare
2025-05-19 13:20       ` Cédric Le Goater
2025-05-19 16:21         ` Steven Sistare
2025-05-22 13:51   ` Cédric Le Goater
2025-05-22 13:56     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 08/42] vfio/container: export vfio_legacy_dma_map Steve Sistare
2025-05-15 13:42   ` Cédric Le Goater
2025-05-15 19:08     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 09/42] vfio/container: discard old DMA vaddr Steve Sistare
2025-05-15 13:30   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 10/42] vfio/container: restore " Steve Sistare
2025-05-15 13:42   ` Cédric Le Goater
2025-05-15 19:08     ` Steven Sistare
2025-05-19 13:32       ` Cédric Le Goater
2025-05-19 16:33         ` Steven Sistare
2025-05-22  6:37   ` Cédric Le Goater
2025-05-22 14:00     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 11/42] vfio/container: mdev cpr blocker Steve Sistare
2025-05-16  8:16   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 12/42] vfio/container: recover from unmap-all-vaddr failure Steve Sistare
2025-05-20  6:29   ` Cédric Le Goater
2025-05-20 13:39     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 13/42] pci: export msix_is_pending Steve Sistare
2025-05-12 15:32 ` [PATCH V3 14/42] pci: skip reset during cpr Steve Sistare
2025-05-16  8:19   ` Cédric Le Goater
2025-05-16 17:58     ` Steven Sistare
2025-05-24  9:34     ` Michael S. Tsirkin
2025-05-27 20:42       ` Steven Sistare
2025-05-27 21:03         ` Michael S. Tsirkin
2025-05-28 16:11           ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 15/42] vfio-pci: " Steve Sistare
2025-05-20  6:48   ` Cédric Le Goater
2025-05-20 13:44     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 16/42] vfio/pci: vfio_vector_init Steve Sistare
2025-05-16  8:32   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 17/42] vfio/pci: vfio_notifier_init Steve Sistare
2025-05-16  8:29   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 18/42] vfio/pci: pass vector to virq functions Steve Sistare
2025-05-16  8:28   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 19/42] vfio/pci: vfio_notifier_init cpr parameters Steve Sistare
2025-05-16  8:29   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 20/42] vfio/pci: vfio_notifier_cleanup Steve Sistare
2025-05-16  8:30   ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 21/42] vfio/pci: export MSI functions Steve Sistare
2025-05-16  8:31   ` Cédric Le Goater
2025-05-16 17:58     ` Steven Sistare
2025-05-20  5:52       ` Cédric Le Goater
2025-05-20 14:56         ` Steven Sistare
2025-05-20 15:10           ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 22/42] vfio-pci: preserve MSI Steve Sistare
2025-05-28 17:44   ` Steven Sistare
2025-06-01 17:28     ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 23/42] vfio-pci: preserve INTx Steve Sistare
2025-05-12 15:32 ` [PATCH V3 24/42] migration: close kvm after cpr Steve Sistare
2025-05-16  8:35   ` Cédric Le Goater
2025-05-16 17:14     ` Peter Xu
2025-05-16 19:17       ` Steven Sistare
2025-05-16 18:18     ` Steven Sistare
2025-05-19  8:51       ` Cédric Le Goater
2025-05-19 19:07         ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 25/42] migration: cpr_get_fd_param helper Steve Sistare
2025-05-19 21:22   ` Fabiano Rosas
2025-05-12 15:32 ` [PATCH V3 26/42] vfio: return mr from vfio_get_xlat_addr Steve Sistare
2025-05-12 20:51   ` John Levon
2025-05-14 17:03     ` Cédric Le Goater
2025-05-15  8:22       ` David Hildenbrand
2025-05-15 19:13         ` Steven Sistare
2025-05-15 17:24     ` Steven Sistare
2025-05-13 11:12   ` Mark Cave-Ayland
2025-05-15 19:40     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 27/42] vfio: pass ramblock to vfio_container_dma_map Steve Sistare
2025-05-16  8:26   ` Duan, Zhenzhong
2025-05-12 15:32 ` [PATCH V3 28/42] backends/iommufd: iommufd_backend_map_file_dma Steve Sistare
2025-05-16  8:26   ` Duan, Zhenzhong
2025-05-19 15:51     ` Steven Sistare
2025-05-20 19:32       ` Steven Sistare
2025-05-21  2:48         ` Duan, Zhenzhong
2025-05-12 15:32 ` [PATCH V3 29/42] backends/iommufd: change process ioctl Steve Sistare
2025-05-16  8:42   ` Duan, Zhenzhong
2025-05-19 15:51     ` Steven Sistare
2025-05-20 19:34       ` Steven Sistare
2025-05-21  3:11         ` Duan, Zhenzhong
2025-05-21 13:01           ` Steven Sistare
2025-05-22  3:19             ` Duan, Zhenzhong
2025-05-22 21:11               ` Steven Sistare
2025-05-23  8:56                 ` Duan, Zhenzhong
2025-05-23 14:56                   ` Steven Sistare
2025-05-23 19:19                     ` Steven Sistare
2025-05-26  2:31                       ` Duan, Zhenzhong
2025-05-28 13:31                         ` Steven Sistare
2025-05-30  9:56                           ` Duan, Zhenzhong
2025-05-12 15:32 ` [PATCH V3 30/42] physmem: qemu_ram_get_fd_offset Steve Sistare
2025-05-16  8:40   ` Duan, Zhenzhong
2025-05-12 15:32 ` [PATCH V3 31/42] vfio/iommufd: use IOMMU_IOAS_MAP_FILE Steve Sistare
2025-05-16  8:48   ` Duan, Zhenzhong
2025-05-19 15:52     ` Steven Sistare
2025-05-20 19:39       ` Steven Sistare
2025-05-21  3:13         ` Duan, Zhenzhong
2025-05-20 12:27   ` Cédric Le Goater
2025-05-20 13:58     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 32/42] vfio/iommufd: export iommufd_cdev_get_info_iova_range Steve Sistare
2025-05-21 18:35   ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 33/42] vfio/iommufd: define hwpt constructors Steve Sistare
2025-05-16  8:55   ` Duan, Zhenzhong
2025-05-19 15:55     ` Steven Sistare
2025-05-23 17:47       ` Steven Sistare
2025-05-20 12:34     ` Cédric Le Goater
2025-05-21  2:48       ` Duan, Zhenzhong
2025-05-21  8:19         ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 34/42] vfio/iommufd: invariant device name Steve Sistare
2025-05-16  9:29   ` Duan, Zhenzhong
2025-05-19 15:52     ` Steven Sistare
2025-05-20 13:55   ` Cédric Le Goater
2025-05-20 21:00     ` Steven Sistare
2025-05-21  8:20       ` Cédric Le Goater
2025-05-12 15:32 ` [PATCH V3 35/42] vfio/iommufd: register container for cpr Steve Sistare
2025-05-16 10:23   ` Duan, Zhenzhong
2025-05-19 15:52     ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 36/42] vfio/iommufd: preserve descriptors Steve Sistare
2025-05-16 10:06   ` Duan, Zhenzhong
2025-05-19 15:53     ` Steven Sistare [this message]
2025-05-20  9:15       ` Duan, Zhenzhong
2025-05-12 15:32 ` [PATCH V3 37/42] vfio/iommufd: reconstruct device Steve Sistare
2025-05-16 10:22   ` Duan, Zhenzhong
2025-05-19 15:53     ` Steven Sistare
2025-05-20  9:14       ` Duan, Zhenzhong
2025-05-21 18:38   ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 38/42] vfio/iommufd: reconstruct hw_caps Steve Sistare
2025-05-21 19:59   ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 39/42] vfio/iommufd: reconstruct hwpt Steve Sistare
2025-05-19  3:25   ` Duan, Zhenzhong
2025-05-19 15:53     ` Steven Sistare
2025-05-20  9:16       ` Duan, Zhenzhong
2025-05-21 17:40         ` Steven Sistare
2025-05-12 15:32 ` [PATCH V3 40/42] vfio/iommufd: change process Steve Sistare
2025-05-12 15:32 ` [PATCH V3 41/42] iommufd: preserve DMA mappings Steve Sistare
2025-05-12 15:32 ` [PATCH V3 42/42] vfio/container: delete old cpr register Steve Sistare
2025-05-16 16:37 ` [PATCH V3 00/42] Live update: vfio and iommufd Cédric Le Goater
2025-05-16 17:17   ` Steven Sistare
2025-05-16 19:48     ` Steven Sistare
2025-05-19  8:54       ` Cédric Le Goater

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=fee4df02-876e-420b-8afa-ca8745c81c1f@oracle.com \
    --to=steven.sistare@oracle.com \
    --cc=alex.williamson@redhat.com \
    --cc=clg@redhat.com \
    --cc=eric.auger@redhat.com \
    --cc=farosas@suse.de \
    --cc=marcel.apfelbaum@gmail.com \
    --cc=mst@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=yi.l.liu@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).