qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Steve Sistare <steven.sistare@oracle.com>
To: qemu-devel@nongnu.org
Cc: Alex Williamson <alex.williamson@redhat.com>,
	Cedric Le Goater <clg@redhat.com>, Yi Liu <yi.l.liu@intel.com>,
	Eric Auger <eric.auger@redhat.com>,
	Zhenzhong Duan <zhenzhong.duan@intel.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Marcel Apfelbaum <marcel.apfelbaum@gmail.com>,
	Peter Xu <peterx@redhat.com>, Fabiano Rosas <farosas@suse.de>,
	Steve Sistare <steven.sistare@oracle.com>
Subject: [PATCH V4 42/43] iommufd: preserve DMA mappings
Date: Thu, 29 May 2025 12:24:38 -0700	[thread overview]
Message-ID: <1748546679-154091-43-git-send-email-steven.sistare@oracle.com> (raw)
In-Reply-To: <1748546679-154091-1-git-send-email-steven.sistare@oracle.com>

During cpr-transfer load in new QEMU, the vfio_memory_listener causes
spurious calls to map and unmap DMA regions, as devices are created and
the address space is built.  This memory was already already mapped by the
device in old QEMU, so suppress the map and unmap callbacks during incoming
CPR.

Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
---
 backends/iommufd.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/backends/iommufd.c b/backends/iommufd.c
index 98d83aa..62d1f71 100644
--- a/backends/iommufd.c
+++ b/backends/iommufd.c
@@ -245,6 +245,10 @@ int iommufd_backend_map_file_dma(IOMMUFDBackend *be, uint32_t ioas_id,
         .length = size,
     };
 
+    if (cpr_is_incoming()) {
+        return 0;
+    }
+
     if (!readonly) {
         map.flags |= IOMMU_IOAS_MAP_WRITEABLE;
     }
@@ -274,6 +278,10 @@ int iommufd_backend_unmap_dma(IOMMUFDBackend *be, uint32_t ioas_id,
         .length = size,
     };
 
+    if (cpr_is_incoming()) {
+        return 0;
+    }
+
     ret = ioctl(fd, IOMMU_IOAS_UNMAP, &unmap);
     /*
      * IOMMUFD takes mapping as some kind of object, unmapping
-- 
1.8.3.1



  parent reply	other threads:[~2025-05-29 19:29 UTC|newest]

Thread overview: 90+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-29 19:23 [PATCH V4 00/43] Live update: vfio and iommufd Steve Sistare
2025-05-29 19:23 ` [PATCH V4 01/43] MAINTAINERS: Add reviewer for CPR Steve Sistare
2025-05-29 19:23 ` [PATCH V4 02/43] vfio: return mr from vfio_get_xlat_addr Steve Sistare
2025-06-03 10:39   ` Duan, Zhenzhong
2025-05-29 19:23 ` [PATCH V4 03/43] vfio/container: pass MemoryRegion to DMA operations Steve Sistare
2025-06-03 10:39   ` Duan, Zhenzhong
2025-05-29 19:24 ` [PATCH V4 04/43] vfio/pci: vfio_pci_put_device on failure Steve Sistare
2025-06-03 10:40   ` Duan, Zhenzhong
2025-06-03 14:09     ` Steven Sistare
2025-06-04  3:55       ` Duan, Zhenzhong
2025-06-04 13:33         ` Steven Sistare
2025-06-05  3:02           ` Duan, Zhenzhong
2025-06-05 15:16             ` Steven Sistare
2025-06-05 21:14               ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 05/43] migration: cpr helpers Steve Sistare
2025-05-29 19:24 ` [PATCH V4 06/43] migration: lower handler priority Steve Sistare
2025-05-29 19:24 ` [PATCH V4 07/43] vfio: vfio_find_ram_discard_listener Steve Sistare
2025-06-03 10:59   ` Duan, Zhenzhong
2025-05-29 19:24 ` [PATCH V4 08/43] vfio: move vfio-cpr.h Steve Sistare
2025-06-03 11:01   ` Duan, Zhenzhong
2025-05-29 19:24 ` [PATCH V4 09/43] vfio/container: register container for cpr Steve Sistare
2025-06-01 15:21   ` Cédric Le Goater
2025-06-03 11:57   ` Duan, Zhenzhong
2025-06-03 14:09     ` Steven Sistare
2025-06-03 14:17       ` Steven Sistare
2025-06-03 15:27         ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 10/43] vfio/container: preserve descriptors Steve Sistare
2025-06-01 16:57   ` Cédric Le Goater
2025-06-03 11:57   ` Duan, Zhenzhong
2025-05-29 19:24 ` [PATCH V4 11/43] vfio/container: discard old DMA vaddr Steve Sistare
2025-05-29 19:24 ` [PATCH V4 12/43] vfio/container: restore " Steve Sistare
2025-06-01 16:48   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 13/43] vfio/container: mdev cpr blocker Steve Sistare
2025-05-29 19:24 ` [PATCH V4 14/43] vfio/container: recover from unmap-all-vaddr failure Steve Sistare
2025-05-29 19:24 ` [PATCH V4 15/43] pci: export msix_is_pending Steve Sistare
2025-05-29 19:24 ` [PATCH V4 16/43] pci: skip reset during cpr Steve Sistare
2025-06-01 16:38   ` Cédric Le Goater
2025-06-01 19:07     ` Michael S. Tsirkin
2025-06-02 12:36       ` Steven Sistare
2025-06-04  7:09         ` Cédric Le Goater
2025-06-04 11:59           ` Cédric Le Goater
2025-06-04 13:15             ` Steven Sistare
2025-06-04 13:48               ` Cédric Le Goater
2025-06-10 16:31                 ` Michael S. Tsirkin
2025-06-10 17:05                   ` Steven Sistare
2025-06-10 17:11                     ` Cédric Le Goater
2025-06-10 17:14                       ` Steven Sistare
2025-06-10 17:19                         ` Cédric Le Goater
2025-06-10 17:09                   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 17/43] vfio-pci: " Steve Sistare
2025-06-01 16:39   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 18/43] vfio/pci: vfio_pci_vector_init Steve Sistare
2025-06-01 15:25   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 19/43] vfio/pci: vfio_notifier_init Steve Sistare
2025-05-29 19:24 ` [PATCH V4 20/43] vfio/pci: pass vector to virq functions Steve Sistare
2025-05-29 19:24 ` [PATCH V4 21/43] vfio/pci: vfio_notifier_init cpr parameters Steve Sistare
2025-05-29 19:24 ` [PATCH V4 22/43] vfio/pci: vfio_notifier_cleanup Steve Sistare
2025-05-29 19:24 ` [PATCH V4 23/43] vfio/pci: export MSI functions Steve Sistare
2025-06-01 15:27   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 24/43] vfio-pci: preserve MSI Steve Sistare
2025-05-29 19:24 ` [PATCH V4 25/43] vfio-pci: preserve INTx Steve Sistare
2025-05-29 19:24 ` [PATCH V4 26/43] migration: close kvm after cpr Steve Sistare
2025-05-29 19:24 ` [PATCH V4 27/43] migration: cpr_get_fd_param helper Steve Sistare
2025-05-29 19:24 ` [PATCH V4 28/43] backends/iommufd: iommufd_backend_map_file_dma Steve Sistare
2025-05-29 19:24 ` [PATCH V4 29/43] backends/iommufd: change process ioctl Steve Sistare
2025-05-29 19:24 ` [PATCH V4 30/43] physmem: qemu_ram_get_fd_offset Steve Sistare
2025-05-29 19:24 ` [PATCH V4 31/43] vfio/iommufd: use IOMMU_IOAS_MAP_FILE Steve Sistare
2025-05-29 19:24 ` [PATCH V4 32/43] vfio/iommufd: invariant device name Steve Sistare
2025-06-10  6:10   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 33/43] vfio/iommufd: add vfio_device_free_name Steve Sistare
2025-06-10  6:12   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 34/43] vfio/iommufd: device name blocker Steve Sistare
2025-05-29 19:24 ` [PATCH V4 35/43] vfio/iommufd: register container for cpr Steve Sistare
2025-06-09 20:30   ` Cédric Le Goater
2025-06-09 20:47     ` Steven Sistare
2025-06-10  6:11       ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 36/43] migration: vfio cpr state hook Steve Sistare
2025-06-10  6:14   ` Cédric Le Goater
2025-05-29 19:24 ` [PATCH V4 37/43] vfio/iommufd: cpr state Steve Sistare
2025-05-29 19:24 ` [PATCH V4 38/43] vfio/iommufd: preserve descriptors Steve Sistare
2025-05-29 19:24 ` [PATCH V4 39/43] vfio/iommufd: reconstruct device Steve Sistare
2025-05-29 19:24 ` [PATCH V4 40/43] vfio/iommufd: reconstruct hwpt Steve Sistare
2025-05-29 19:24 ` [PATCH V4 41/43] vfio/iommufd: change process Steve Sistare
2025-05-29 19:24 ` Steve Sistare [this message]
2025-05-29 19:24 ` [PATCH V4 43/43] vfio/container: delete old cpr register Steve Sistare
2025-06-10  6:14   ` Cédric Le Goater
2025-06-01 17:26 ` [PATCH V4 00/43] Live update: vfio and iommufd Cédric Le Goater
2025-06-02 12:42   ` Steven Sistare
2025-06-03 12:09 ` Duan, Zhenzhong
2025-06-03 14:09   ` Steven Sistare

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=1748546679-154091-43-git-send-email-steven.sistare@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).