From: Alex Williamson <alex.williamson@redhat.com>
To: qemu-devel@nongnu.org
Cc: Kirti Wankhede <kwankhede@nvidia.com>, Neo Jia <cjia@nvidia.com>
Subject: [PULL 15/32] vfio: Add ioctl to get dirty pages bitmap during dma unmap
Date: Mon, 26 Oct 2020 13:34:35 -0600 [thread overview]
Message-ID: <160374087514.22414.3514803987689298850.stgit@gimli.home> (raw)
In-Reply-To: <160374054442.22414.10832953989449611268.stgit@gimli.home>
From: Kirti Wankhede <kwankhede@nvidia.com>
With vIOMMU, IO virtual address range can get unmapped while in pre-copy
phase of migration. In that case, unmap ioctl should return pages pinned
in that range and QEMU should find its correcponding guest physical
addresses and report those dirty.
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
---
hw/vfio/common.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 92 insertions(+), 4 deletions(-)
diff --git a/hw/vfio/common.c b/hw/vfio/common.c
index c0b5b6245a47..49c68a5253ae 100644
--- a/hw/vfio/common.c
+++ b/hw/vfio/common.c
@@ -321,11 +321,94 @@ static bool vfio_devices_all_stopped_and_saving(VFIOContainer *container)
return true;
}
+static bool vfio_devices_all_running_and_saving(VFIOContainer *container)
+{
+ VFIOGroup *group;
+ VFIODevice *vbasedev;
+ MigrationState *ms = migrate_get_current();
+
+ if (!migration_is_setup_or_active(ms->state)) {
+ return false;
+ }
+
+ QLIST_FOREACH(group, &container->group_list, container_next) {
+ QLIST_FOREACH(vbasedev, &group->device_list, next) {
+ VFIOMigration *migration = vbasedev->migration;
+
+ if (!migration) {
+ return false;
+ }
+
+ if ((migration->device_state & VFIO_DEVICE_STATE_SAVING) &&
+ (migration->device_state & VFIO_DEVICE_STATE_RUNNING)) {
+ continue;
+ } else {
+ return false;
+ }
+ }
+ }
+ return true;
+}
+
+static int vfio_dma_unmap_bitmap(VFIOContainer *container,
+ hwaddr iova, ram_addr_t size,
+ IOMMUTLBEntry *iotlb)
+{
+ struct vfio_iommu_type1_dma_unmap *unmap;
+ struct vfio_bitmap *bitmap;
+ uint64_t pages = TARGET_PAGE_ALIGN(size) >> TARGET_PAGE_BITS;
+ int ret;
+
+ unmap = g_malloc0(sizeof(*unmap) + sizeof(*bitmap));
+
+ unmap->argsz = sizeof(*unmap) + sizeof(*bitmap);
+ unmap->iova = iova;
+ unmap->size = size;
+ unmap->flags |= VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP;
+ bitmap = (struct vfio_bitmap *)&unmap->data;
+
+ /*
+ * cpu_physical_memory_set_dirty_lebitmap() expects pages in bitmap of
+ * TARGET_PAGE_SIZE to mark those dirty. Hence set bitmap_pgsize to
+ * TARGET_PAGE_SIZE.
+ */
+
+ bitmap->pgsize = TARGET_PAGE_SIZE;
+ bitmap->size = ROUND_UP(pages, sizeof(__u64) * BITS_PER_BYTE) /
+ BITS_PER_BYTE;
+
+ if (bitmap->size > container->max_dirty_bitmap_size) {
+ error_report("UNMAP: Size of bitmap too big 0x%llx", bitmap->size);
+ ret = -E2BIG;
+ goto unmap_exit;
+ }
+
+ bitmap->data = g_try_malloc0(bitmap->size);
+ if (!bitmap->data) {
+ ret = -ENOMEM;
+ goto unmap_exit;
+ }
+
+ ret = ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, unmap);
+ if (!ret) {
+ cpu_physical_memory_set_dirty_lebitmap((uint64_t *)bitmap->data,
+ iotlb->translated_addr, pages);
+ } else {
+ error_report("VFIO_UNMAP_DMA with DIRTY_BITMAP : %m");
+ }
+
+ g_free(bitmap->data);
+unmap_exit:
+ g_free(unmap);
+ return ret;
+}
+
/*
* DMA - Mapping and unmapping for the "type1" IOMMU interface used on x86
*/
static int vfio_dma_unmap(VFIOContainer *container,
- hwaddr iova, ram_addr_t size)
+ hwaddr iova, ram_addr_t size,
+ IOMMUTLBEntry *iotlb)
{
struct vfio_iommu_type1_dma_unmap unmap = {
.argsz = sizeof(unmap),
@@ -334,6 +417,11 @@ static int vfio_dma_unmap(VFIOContainer *container,
.size = size,
};
+ if (iotlb && container->dirty_pages_supported &&
+ vfio_devices_all_running_and_saving(container)) {
+ return vfio_dma_unmap_bitmap(container, iova, size, iotlb);
+ }
+
while (ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, &unmap)) {
/*
* The type1 backend has an off-by-one bug in the kernel (71a7d3d78e3c
@@ -381,7 +469,7 @@ static int vfio_dma_map(VFIOContainer *container, hwaddr iova,
* the VGA ROM space.
*/
if (ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) == 0 ||
- (errno == EBUSY && vfio_dma_unmap(container, iova, size) == 0 &&
+ (errno == EBUSY && vfio_dma_unmap(container, iova, size, NULL) == 0 &&
ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) == 0)) {
return 0;
}
@@ -531,7 +619,7 @@ static void vfio_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
iotlb->addr_mask + 1, vaddr, ret);
}
} else {
- ret = vfio_dma_unmap(container, iova, iotlb->addr_mask + 1);
+ ret = vfio_dma_unmap(container, iova, iotlb->addr_mask + 1, iotlb);
if (ret) {
error_report("vfio_dma_unmap(%p, 0x%"HWADDR_PRIx", "
"0x%"HWADDR_PRIx") = %d (%m)",
@@ -834,7 +922,7 @@ static void vfio_listener_region_del(MemoryListener *listener,
}
if (try_unmap) {
- ret = vfio_dma_unmap(container, iova, int128_get64(llsize));
+ ret = vfio_dma_unmap(container, iova, int128_get64(llsize), NULL);
if (ret) {
error_report("vfio_dma_unmap(%p, 0x%"HWADDR_PRIx", "
"0x%"HWADDR_PRIx") = %d (%m)",
next prev parent reply other threads:[~2020-10-26 19:57 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-26 19:32 [PULL 00/32] VFIO updates 2020-10-26 (for QEMU 5.2 soft-freeze) Alex Williamson
2020-10-26 19:32 ` [PULL 01/32] vfio: Add function to unmap VFIO region Alex Williamson
2020-10-26 19:32 ` [PULL 02/32] vfio: Add vfio_get_object callback to VFIODeviceOps Alex Williamson
2020-10-26 19:32 ` [PULL 03/32] vfio: Add save and load functions for VFIO PCI devices Alex Williamson
2020-10-26 19:32 ` [PULL 04/32] vfio: Add migration region initialization and finalize function Alex Williamson
2020-10-26 19:33 ` [PULL 05/32] vfio: Add VM state change handler to know state of VM Alex Williamson
2020-10-26 19:33 ` [PULL 06/32] vfio: Add migration state change notifier Alex Williamson
2020-10-26 19:33 ` [PULL 07/32] vfio: Register SaveVMHandlers for VFIO device Alex Williamson
2020-10-26 19:33 ` [PULL 08/32] vfio: Add save state functions to SaveVMHandlers Alex Williamson
2020-10-26 19:33 ` [PULL 09/32] vfio: Add load " Alex Williamson
2020-10-26 19:33 ` [PULL 10/32] memory: Set DIRTY_MEMORY_MIGRATION when IOMMU is enabled Alex Williamson
2020-10-26 19:34 ` [PULL 11/32] vfio: Get migration capability flags for container Alex Williamson
2020-10-26 19:34 ` [PULL 12/32] vfio: Add function to start and stop dirty pages tracking Alex Williamson
2020-10-26 19:34 ` [PULL 13/32] vfio: Add vfio_listener_log_sync to mark dirty pages Alex Williamson
2020-10-26 19:34 ` [PULL 14/32] vfio: Dirty page tracking when vIOMMU is enabled Alex Williamson
2020-10-26 19:34 ` Alex Williamson [this message]
2020-10-26 19:34 ` [PULL 16/32] vfio: Make vfio-pci device migration capable Alex Williamson
2020-10-26 19:34 ` [PULL 17/32] qapi: Add VFIO devices migration stats in Migration stats Alex Williamson
2020-10-26 19:35 ` [PULL 18/32] update-linux-headers: Add vfio_zdev.h Alex Williamson
2020-10-26 19:35 ` [PULL 19/32] linux-headers: update against 5.10-rc1 Alex Williamson
2020-10-26 19:35 ` [PULL 20/32] s390x/pci: Move header files to include/hw/s390x Alex Williamson
2020-10-26 19:35 ` [PULL 21/32] vfio: Create shared routine for scanning info capabilities Alex Williamson
2020-10-26 19:35 ` [PULL 22/32] vfio: Find DMA available capability Alex Williamson
2020-10-26 19:35 ` [PULL 23/32] s390x/pci: Add routine to get the vfio dma available count Alex Williamson
2020-10-26 19:35 ` [PULL 24/32] s390x/pci: Honor DMA limits set by vfio Alex Williamson
2020-10-26 19:35 ` [PULL 25/32] s390x/pci: create a header dedicated to PCI CLP Alex Williamson
2020-10-26 19:36 ` [PULL 26/32] s390x/pci: use a PCI Group structure Alex Williamson
2020-10-26 19:36 ` [PULL 27/32] s390x/pci: clean up s390 PCI groups Alex Williamson
2020-10-26 19:36 ` [PULL 28/32] s390x/pci: use a PCI Function structure Alex Williamson
2020-10-26 19:36 ` [PULL 29/32] vfio: Add routine for finding VFIO_DEVICE_GET_INFO capabilities Alex Williamson
2020-10-26 19:36 ` [PULL 30/32] s390x/pci: get zPCI function info from host Alex Williamson
2020-10-26 19:36 ` [PULL 31/32] hw/vfio: Use lock guard macros Alex Williamson
2020-10-26 19:36 ` [PULL 32/32] vfio: fix incorrect print type Alex Williamson
2020-10-27 23:42 ` [PULL 00/32] VFIO updates 2020-10-26 (for QEMU 5.2 soft-freeze) Peter Maydell
2020-10-28 2:00 ` Alex Williamson
2020-10-28 9:18 ` Dr. David Alan Gilbert
2020-10-28 15:23 ` Miklos Szeredi
2020-10-28 9:21 ` Max Reitz
2020-10-28 8:11 ` Cornelia Huck
2020-10-28 9:28 ` Max Reitz
2020-10-28 9:41 ` Max Reitz
2020-10-28 10:12 ` Cornelia Huck
2020-10-28 15:07 ` Peter Maydell
2020-10-28 15:20 ` Matthew Rosato
2020-10-28 15:41 ` Alex Williamson
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=160374087514.22414.3514803987689298850.stgit@gimli.home \
--to=alex.williamson@redhat.com \
--cc=cjia@nvidia.com \
--cc=kwankhede@nvidia.com \
--cc=qemu-devel@nongnu.org \
/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).