vfio: Convert to ram_block_discard_disable()

VFIO is (except devices without a physical IOMMU or some mediated devices)
incompatible with discarding of RAM. The kernel will pin basically all VM
memory. Let's convert to ram_block_discard_disable(), which can now
fail, in contrast to qemu_balloon_inhibit().

Leave "x-balloon-allowed" named as it is for now.

Reviewed-by: Tony Krowiak <akrowiak@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Cc: Cornelia Huck <cohuck@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Tony Krowiak <akrowiak@linux.ibm.com>
Cc: Halil Pasic <pasic@linux.ibm.com>
Cc: Pierre Morel <pmorel@linux.ibm.com>
Cc: Eric Farman <farman@linux.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200626072248.78761-4-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This commit is contained in:
David Hildenbrand 2020-06-26 09:22:30 +02:00 committed by Michael S. Tsirkin
parent d24f31db3b
commit aff92b8286
5 changed files with 44 additions and 38 deletions

View File

@ -105,12 +105,12 @@ static void vfio_ap_realize(DeviceState *dev, Error **errp)
vapdev->vdev.dev = dev;
/*
* vfio-ap devices operate in a way compatible with
* memory ballooning, as no pages are pinned in the host.
* vfio-ap devices operate in a way compatible with discarding of
* memory in RAM blocks, as no pages are pinned in the host.
* This needs to be set before vfio_get_device() for vfio common to
* handle the balloon inhibitor.
* handle ram_block_discard_disable().
*/
vapdev->vdev.balloon_allowed = true;
vapdev->vdev.ram_block_discard_allowed = true;
ret = vfio_get_device(vfio_group, mdevid, &vapdev->vdev, errp);
if (ret) {

View File

@ -574,12 +574,13 @@ static void vfio_ccw_get_device(VFIOGroup *group, VFIOCCWDevice *vcdev,
/*
* All vfio-ccw devices are believed to operate in a way compatible with
* memory ballooning, ie. pages pinned in the host are in the current
* working set of the guest driver and therefore never overlap with pages
* available to the guest balloon driver. This needs to be set before
* vfio_get_device() for vfio common to handle the balloon inhibitor.
* discarding of memory in RAM blocks, ie. pages pinned in the host are
* in the current working set of the guest driver and therefore never
* overlap e.g., with pages available to the guest balloon driver. This
* needs to be set before vfio_get_device() for vfio common to handle
* ram_block_discard_disable().
*/
vcdev->vdev.balloon_allowed = true;
vcdev->vdev.ram_block_discard_allowed = true;
if (vfio_get_device(group, vcdev->cdev.mdevid, &vcdev->vdev, errp)) {
goto out_err;

View File

@ -33,7 +33,6 @@
#include "qemu/error-report.h"
#include "qemu/main-loop.h"
#include "qemu/range.h"
#include "sysemu/balloon.h"
#include "sysemu/kvm.h"
#include "sysemu/reset.h"
#include "trace.h"
@ -1215,31 +1214,36 @@ static int vfio_connect_container(VFIOGroup *group, AddressSpace *as,
space = vfio_get_address_space(as);
/*
* VFIO is currently incompatible with memory ballooning insofar as the
* VFIO is currently incompatible with discarding of RAM insofar as the
* madvise to purge (zap) the page from QEMU's address space does not
* interact with the memory API and therefore leaves stale virtual to
* physical mappings in the IOMMU if the page was previously pinned. We
* therefore add a balloon inhibit for each group added to a container,
* therefore set discarding broken for each group added to a container,
* whether the container is used individually or shared. This provides
* us with options to allow devices within a group to opt-in and allow
* ballooning, so long as it is done consistently for a group (for instance
* discarding, so long as it is done consistently for a group (for instance
* if the device is an mdev device where it is known that the host vendor
* driver will never pin pages outside of the working set of the guest
* driver, which would thus not be ballooning candidates).
* driver, which would thus not be discarding candidates).
*
* The first opportunity to induce pinning occurs here where we attempt to
* attach the group to existing containers within the AddressSpace. If any
* pages are already zapped from the virtual address space, such as from a
* previous ballooning opt-in, new pinning will cause valid mappings to be
* pages are already zapped from the virtual address space, such as from
* previous discards, new pinning will cause valid mappings to be
* re-established. Likewise, when the overall MemoryListener for a new
* container is registered, a replay of mappings within the AddressSpace
* will occur, re-establishing any previously zapped pages as well.
*
* NB. Balloon inhibiting does not currently block operation of the
* balloon driver or revoke previously pinned pages, it only prevents
* calling madvise to modify the virtual mapping of ballooned pages.
* Especially virtio-balloon is currently only prevented from discarding
* new memory, it will not yet set ram_block_discard_set_required() and
* therefore, neither stops us here or deals with the sudden memory
* consumption of inflated memory.
*/
qemu_balloon_inhibit(true);
ret = ram_block_discard_disable(true);
if (ret) {
error_setg_errno(errp, -ret, "Cannot set discarding of RAM broken");
return ret;
}
QLIST_FOREACH(container, &space->containers, next) {
if (!ioctl(group->fd, VFIO_GROUP_SET_CONTAINER, &container->fd)) {
@ -1405,7 +1409,7 @@ close_fd_exit:
close(fd);
put_space_exit:
qemu_balloon_inhibit(false);
ram_block_discard_disable(false);
vfio_put_address_space(space);
return ret;
@ -1526,8 +1530,8 @@ void vfio_put_group(VFIOGroup *group)
return;
}
if (!group->balloon_allowed) {
qemu_balloon_inhibit(false);
if (!group->ram_block_discard_allowed) {
ram_block_discard_disable(false);
}
vfio_kvm_device_del_group(group);
vfio_disconnect_container(group);
@ -1565,22 +1569,23 @@ int vfio_get_device(VFIOGroup *group, const char *name,
}
/*
* Clear the balloon inhibitor for this group if the driver knows the
* device operates compatibly with ballooning. Setting must be consistent
* per group, but since compatibility is really only possible with mdev
* currently, we expect singleton groups.
* Set discarding of RAM as not broken for this group if the driver knows
* the device operates compatibly with discarding. Setting must be
* consistent per group, but since compatibility is really only possible
* with mdev currently, we expect singleton groups.
*/
if (vbasedev->balloon_allowed != group->balloon_allowed) {
if (vbasedev->ram_block_discard_allowed !=
group->ram_block_discard_allowed) {
if (!QLIST_EMPTY(&group->device_list)) {
error_setg(errp,
"Inconsistent device balloon setting within group");
error_setg(errp, "Inconsistent setting of support for discarding "
"RAM (e.g., balloon) within group");
close(fd);
return -1;
}
if (!group->balloon_allowed) {
group->balloon_allowed = true;
qemu_balloon_inhibit(false);
if (!group->ram_block_discard_allowed) {
group->ram_block_discard_allowed = true;
ram_block_discard_disable(false);
}
}

View File

@ -2789,7 +2789,7 @@ static void vfio_realize(PCIDevice *pdev, Error **errp)
}
/*
* Mediated devices *might* operate compatibly with memory ballooning, but
* Mediated devices *might* operate compatibly with discarding of RAM, but
* we cannot know for certain, it depends on whether the mdev vendor driver
* stays in sync with the active working set of the guest driver. Prevent
* the x-balloon-allowed option unless this is minimally an mdev device.
@ -2802,7 +2802,7 @@ static void vfio_realize(PCIDevice *pdev, Error **errp)
trace_vfio_mdev(vdev->vbasedev.name, is_mdev);
if (vdev->vbasedev.balloon_allowed && !is_mdev) {
if (vdev->vbasedev.ram_block_discard_allowed && !is_mdev) {
error_setg(errp, "x-balloon-allowed only potentially compatible "
"with mdev devices");
vfio_put_group(group);
@ -3156,7 +3156,7 @@ static Property vfio_pci_dev_properties[] = {
VFIO_FEATURE_ENABLE_IGD_OPREGION_BIT, false),
DEFINE_PROP_BOOL("x-no-mmap", VFIOPCIDevice, vbasedev.no_mmap, false),
DEFINE_PROP_BOOL("x-balloon-allowed", VFIOPCIDevice,
vbasedev.balloon_allowed, false),
vbasedev.ram_block_discard_allowed, false),
DEFINE_PROP_BOOL("x-no-kvm-intx", VFIOPCIDevice, no_kvm_intx, false),
DEFINE_PROP_BOOL("x-no-kvm-msi", VFIOPCIDevice, no_kvm_msi, false),
DEFINE_PROP_BOOL("x-no-kvm-msix", VFIOPCIDevice, no_kvm_msix, false),

View File

@ -108,7 +108,7 @@ typedef struct VFIODevice {
bool reset_works;
bool needs_reset;
bool no_mmap;
bool balloon_allowed;
bool ram_block_discard_allowed;
VFIODeviceOps *ops;
unsigned int num_irqs;
unsigned int num_regions;
@ -128,7 +128,7 @@ typedef struct VFIOGroup {
QLIST_HEAD(, VFIODevice) device_list;
QLIST_ENTRY(VFIOGroup) next;
QLIST_ENTRY(VFIOGroup) container_next;
bool balloon_allowed;
bool ram_block_discard_allowed;
} VFIOGroup;
typedef struct VFIODMABuf {