* [PULL 01/14] hw/vfio/listener.c: remove CONFIG_KVM
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 02/14] hw/vfio/helpers.c: extract kvm helpers in kvm-helpers.c Cédric Le Goater
` (13 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Code concerned is under a kvm_enabled() guard.
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-2-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/listener.c | 4 ----
1 file changed, 4 deletions(-)
diff --git a/hw/vfio/listener.c b/hw/vfio/listener.c
index 960da9e0a9341e8123baed7839a32bc3f3acca57..31c3113f8fbe305081ffdd16cb28cf8379926971 100644
--- a/hw/vfio/listener.c
+++ b/hw/vfio/listener.c
@@ -20,9 +20,7 @@
#include "qemu/osdep.h"
#include <sys/ioctl.h>
-#ifdef CONFIG_KVM
#include <linux/kvm.h>
-#endif
#include <linux/vfio.h>
#include "exec/target_page.h"
@@ -303,11 +301,9 @@ static bool vfio_ram_discard_register_listener(VFIOContainer *bcontainer,
if (bcontainer->dma_max_mappings) {
unsigned int vrdl_count = 0, vrdl_mappings = 0, max_memslots = 512;
-#ifdef CONFIG_KVM
if (kvm_enabled()) {
max_memslots = kvm_get_max_memslots();
}
-#endif
QLIST_FOREACH(vrdl, &bcontainer->vrdl_list, next) {
hwaddr start, end;
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 02/14] hw/vfio/helpers.c: extract kvm helpers in kvm-helpers.c
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
2026-04-21 16:34 ` [PULL 01/14] hw/vfio/listener.c: remove CONFIG_KVM Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 03/14] hw/vfio/pci-quirks.c: remove CONFIG_VFIO_IGD Cédric Le Goater
` (12 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Because those functions use kvm specific types, they need to be isolated
in another source file.
This allows us to link kvm-helpers only in configurations with
CONFIG_KVM.
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-3-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/helpers.c | 172 -------------------------------------
hw/vfio/kvm-helpers.c | 192 ++++++++++++++++++++++++++++++++++++++++++
hw/vfio/kvm-stubs.c | 26 ++++++
hw/vfio/meson.build | 2 +
4 files changed, 220 insertions(+), 172 deletions(-)
create mode 100644 hw/vfio/kvm-helpers.c
create mode 100644 hw/vfio/kvm-stubs.c
diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c
index 00d42d3b98ef98368bb30ff11098615c881cf0a8..65c6dba0428954170ce8a889edffa34ce9b24b13 100644
--- a/hw/vfio/helpers.c
+++ b/hw/vfio/helpers.c
@@ -22,7 +22,6 @@
#include "qemu/osdep.h"
#include <sys/ioctl.h>
-#include "system/kvm.h"
#include "exec/cpu-common.h"
#include "hw/vfio/vfio-device.h"
#include "hw/core/hw-error.h"
@@ -107,177 +106,6 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info,
return true;
}
-#ifdef CONFIG_KVM
-/*
- * We have a single VFIO pseudo device per KVM VM. Once created it lives
- * for the life of the VM. Closing the file descriptor only drops our
- * reference to it and the device's reference to kvm. Therefore once
- * initialized, this file descriptor is only released on QEMU exit and
- * we'll re-use it should another vfio device be attached before then.
- */
-int vfio_kvm_device_fd = -1;
-
-/*
- * Confidential virtual machines:
- * During reset of confidential vms, the kvm vm file descriptor changes.
- * In this case, the old vfio kvm file descriptor is
- * closed and a new descriptor is created against the new kvm vm file
- * descriptor.
- */
-
-typedef struct VFIODeviceFd {
- int fd;
- QLIST_ENTRY(VFIODeviceFd) node;
-} VFIODeviceFd;
-
-static QLIST_HEAD(, VFIODeviceFd) vfio_device_fds =
- QLIST_HEAD_INITIALIZER(vfio_device_fds);
-
-static void vfio_device_fd_list_add(int fd)
-{
- VFIODeviceFd *file_fd;
- file_fd = g_malloc0(sizeof(*file_fd));
- file_fd->fd = fd;
- QLIST_INSERT_HEAD(&vfio_device_fds, file_fd, node);
-}
-
-static void vfio_device_fd_list_remove(int fd)
-{
- VFIODeviceFd *file_fd, *next;
-
- QLIST_FOREACH_SAFE(file_fd, &vfio_device_fds, node, next) {
- if (file_fd->fd == fd) {
- QLIST_REMOVE(file_fd, node);
- g_free(file_fd);
- break;
- }
- }
-}
-
-static int vfio_device_fd_rebind(NotifierWithReturn *notifier, void *data,
- Error **errp)
-{
- VFIODeviceFd *file_fd;
- struct kvm_device_attr attr = {
- .group = KVM_DEV_VFIO_FILE,
- .attr = KVM_DEV_VFIO_FILE_ADD,
- };
- struct kvm_create_device cd = {
- .type = KVM_DEV_TYPE_VFIO,
- };
-
- /* we are not interested in pre vmfd change notification */
- if (((VmfdChangeNotifier *)data)->pre) {
- return 0;
- }
-
- if (kvm_vm_ioctl(kvm_state, KVM_CREATE_DEVICE, &cd)) {
- error_setg_errno(errp, errno, "Failed to create KVM VFIO device");
- return -errno;
- }
-
- if (vfio_kvm_device_fd != -1) {
- close(vfio_kvm_device_fd);
- }
-
- vfio_kvm_device_fd = cd.fd;
-
- QLIST_FOREACH(file_fd, &vfio_device_fds, node) {
- attr.addr = (uint64_t)(unsigned long)&file_fd->fd;
- if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
- error_setg_errno(errp, errno,
- "Failed to add fd %d to KVM VFIO device",
- file_fd->fd);
- return -errno;
- }
- }
- return 0;
-}
-
-static struct NotifierWithReturn vfio_vmfd_change_notifier = {
- .notify = vfio_device_fd_rebind,
-};
-
-#endif
-
-void vfio_kvm_device_close(void)
-{
-#ifdef CONFIG_KVM
- kvm_close();
- if (vfio_kvm_device_fd != -1) {
- close(vfio_kvm_device_fd);
- vfio_kvm_device_fd = -1;
- }
-#endif
-}
-
-int vfio_kvm_device_add_fd(int fd, Error **errp)
-{
-#ifdef CONFIG_KVM
- struct kvm_device_attr attr = {
- .group = KVM_DEV_VFIO_FILE,
- .attr = KVM_DEV_VFIO_FILE_ADD,
- .addr = (uint64_t)(unsigned long)&fd,
- };
-
- if (!kvm_enabled()) {
- return 0;
- }
-
- if (vfio_kvm_device_fd < 0) {
- struct kvm_create_device cd = {
- .type = KVM_DEV_TYPE_VFIO,
- };
-
- if (kvm_vm_ioctl(kvm_state, KVM_CREATE_DEVICE, &cd)) {
- error_setg_errno(errp, errno, "Failed to create KVM VFIO device");
- return -errno;
- }
-
- vfio_kvm_device_fd = cd.fd;
- /*
- * If the vm file descriptor changes, add a notifier so that we can
- * re-create the vfio_kvm_device_fd.
- */
- kvm_vmfd_add_change_notifier(&vfio_vmfd_change_notifier);
- }
-
- if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
- error_setg_errno(errp, errno, "Failed to add fd %d to KVM VFIO device",
- fd);
- return -errno;
- }
-
- vfio_device_fd_list_add(fd);
-#endif
- return 0;
-}
-
-int vfio_kvm_device_del_fd(int fd, Error **errp)
-{
-#ifdef CONFIG_KVM
- struct kvm_device_attr attr = {
- .group = KVM_DEV_VFIO_FILE,
- .attr = KVM_DEV_VFIO_FILE_DEL,
- .addr = (uint64_t)(unsigned long)&fd,
- };
-
- if (vfio_kvm_device_fd < 0) {
- error_setg(errp, "KVM VFIO device isn't created yet");
- return -EINVAL;
- }
-
- if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
- error_setg_errno(errp, errno,
- "Failed to remove fd %d from KVM VFIO device", fd);
- return -errno;
- }
-
- vfio_device_fd_list_remove(fd);
-#endif
- return 0;
-}
-
struct vfio_device_info *vfio_get_device_info(int fd)
{
struct vfio_device_info *info;
diff --git a/hw/vfio/kvm-helpers.c b/hw/vfio/kvm-helpers.c
new file mode 100644
index 0000000000000000000000000000000000000000..d71c9590aaa53dc4e436987a34575f57a6397c6b
--- /dev/null
+++ b/hw/vfio/kvm-helpers.c
@@ -0,0 +1,192 @@
+/*
+ * low level and IOMMU backend agnostic helpers used by VFIO devices,
+ * related to regions, interrupts, capabilities
+ *
+ * Copyright Red Hat, Inc. 2012
+ *
+ * Authors:
+ * Alex Williamson <alex.williamson@redhat.com>
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ *
+ * Based on qemu-kvm device-assignment:
+ * Adapted for KVM by Qumranet.
+ * Copyright (c) 2007, Neocleus, Alex Novik (alex@neocleus.com)
+ * Copyright (c) 2007, Neocleus, Guy Zana (guy@neocleus.com)
+ * Copyright (C) 2008, Qumranet, Amit Shah (amit.shah@qumranet.com)
+ * Copyright (C) 2008, Red Hat, Amit Shah (amit.shah@redhat.com)
+ * Copyright (C) 2008, IBM, Muli Ben-Yehuda (muli@il.ibm.com)
+ */
+
+#include "qemu/osdep.h"
+#include <sys/ioctl.h>
+
+#include <linux/kvm.h>
+#include "system/kvm.h"
+#include "exec/cpu-common.h"
+#include "hw/vfio/vfio-device.h"
+#include "hw/core/hw-error.h"
+#include "qapi/error.h"
+#include "vfio-helpers.h"
+
+/*
+ * We have a single VFIO pseudo device per KVM VM. Once created it lives
+ * for the life of the VM. Closing the file descriptor only drops our
+ * reference to it and the device's reference to kvm. Therefore once
+ * initialized, this file descriptor is only released on QEMU exit and
+ * we'll re-use it should another vfio device be attached before then.
+ */
+int vfio_kvm_device_fd = -1;
+
+/*
+ * Confidential virtual machines:
+ * During reset of confidential vms, the kvm vm file descriptor changes.
+ * In this case, the old vfio kvm file descriptor is
+ * closed and a new descriptor is created against the new kvm vm file
+ * descriptor.
+ */
+
+typedef struct VFIODeviceFd {
+ int fd;
+ QLIST_ENTRY(VFIODeviceFd) node;
+} VFIODeviceFd;
+
+static QLIST_HEAD(, VFIODeviceFd) vfio_device_fds =
+ QLIST_HEAD_INITIALIZER(vfio_device_fds);
+
+static void vfio_device_fd_list_add(int fd)
+{
+ VFIODeviceFd *file_fd;
+ file_fd = g_malloc0(sizeof(*file_fd));
+ file_fd->fd = fd;
+ QLIST_INSERT_HEAD(&vfio_device_fds, file_fd, node);
+}
+
+static void vfio_device_fd_list_remove(int fd)
+{
+ VFIODeviceFd *file_fd, *next;
+
+ QLIST_FOREACH_SAFE(file_fd, &vfio_device_fds, node, next) {
+ if (file_fd->fd == fd) {
+ QLIST_REMOVE(file_fd, node);
+ g_free(file_fd);
+ break;
+ }
+ }
+}
+
+static int vfio_device_fd_rebind(NotifierWithReturn *notifier, void *data,
+ Error **errp)
+{
+ VFIODeviceFd *file_fd;
+ struct kvm_device_attr attr = {
+ .group = KVM_DEV_VFIO_FILE,
+ .attr = KVM_DEV_VFIO_FILE_ADD,
+ };
+ struct kvm_create_device cd = {
+ .type = KVM_DEV_TYPE_VFIO,
+ };
+
+ /* we are not interested in pre vmfd change notification */
+ if (((VmfdChangeNotifier *)data)->pre) {
+ return 0;
+ }
+
+ if (kvm_vm_ioctl(kvm_state, KVM_CREATE_DEVICE, &cd)) {
+ error_setg_errno(errp, errno, "Failed to create KVM VFIO device");
+ return -errno;
+ }
+
+ if (vfio_kvm_device_fd != -1) {
+ close(vfio_kvm_device_fd);
+ }
+
+ vfio_kvm_device_fd = cd.fd;
+
+ QLIST_FOREACH(file_fd, &vfio_device_fds, node) {
+ attr.addr = (uint64_t)(unsigned long)&file_fd->fd;
+ if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
+ error_setg_errno(errp, errno,
+ "Failed to add fd %d to KVM VFIO device",
+ file_fd->fd);
+ return -errno;
+ }
+ }
+ return 0;
+}
+
+static struct NotifierWithReturn vfio_vmfd_change_notifier = {
+ .notify = vfio_device_fd_rebind,
+};
+
+void vfio_kvm_device_close(void)
+{
+ kvm_close();
+ if (vfio_kvm_device_fd != -1) {
+ close(vfio_kvm_device_fd);
+ vfio_kvm_device_fd = -1;
+ }
+}
+
+int vfio_kvm_device_add_fd(int fd, Error **errp)
+{
+ struct kvm_device_attr attr = {
+ .group = KVM_DEV_VFIO_FILE,
+ .attr = KVM_DEV_VFIO_FILE_ADD,
+ .addr = (uint64_t)(unsigned long)&fd,
+ };
+
+ if (!kvm_enabled()) {
+ return 0;
+ }
+
+ if (vfio_kvm_device_fd < 0) {
+ struct kvm_create_device cd = {
+ .type = KVM_DEV_TYPE_VFIO,
+ };
+
+ if (kvm_vm_ioctl(kvm_state, KVM_CREATE_DEVICE, &cd)) {
+ error_setg_errno(errp, errno, "Failed to create KVM VFIO device");
+ return -errno;
+ }
+
+ vfio_kvm_device_fd = cd.fd;
+ /*
+ * If the vm file descriptor changes, add a notifier so that we can
+ * re-create the vfio_kvm_device_fd.
+ */
+ kvm_vmfd_add_change_notifier(&vfio_vmfd_change_notifier);
+ }
+
+ if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
+ error_setg_errno(errp, errno, "Failed to add fd %d to KVM VFIO device",
+ fd);
+ return -errno;
+ }
+
+ vfio_device_fd_list_add(fd);
+ return 0;
+}
+
+int vfio_kvm_device_del_fd(int fd, Error **errp)
+{
+ struct kvm_device_attr attr = {
+ .group = KVM_DEV_VFIO_FILE,
+ .attr = KVM_DEV_VFIO_FILE_DEL,
+ .addr = (uint64_t)(unsigned long)&fd,
+ };
+
+ if (vfio_kvm_device_fd < 0) {
+ error_setg(errp, "KVM VFIO device isn't created yet");
+ return -EINVAL;
+ }
+
+ if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
+ error_setg_errno(errp, errno,
+ "Failed to remove fd %d from KVM VFIO device", fd);
+ return -errno;
+ }
+
+ vfio_device_fd_list_remove(fd);
+ return 0;
+}
diff --git a/hw/vfio/kvm-stubs.c b/hw/vfio/kvm-stubs.c
new file mode 100644
index 0000000000000000000000000000000000000000..5a489d1b7110a5564054398cc354bb806ac39334
--- /dev/null
+++ b/hw/vfio/kvm-stubs.c
@@ -0,0 +1,26 @@
+/*
+ * Stubs for kvm helpers
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+
+#include "hw/vfio/vfio-device.h"
+#include "qapi/error.h"
+#include "vfio-helpers.h"
+
+void vfio_kvm_device_close(void)
+{
+ return;
+}
+
+int vfio_kvm_device_add_fd(int fd, Error **errp)
+{
+ return 0;
+}
+
+int vfio_kvm_device_del_fd(int fd, Error **errp)
+{
+ return 0;
+}
diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build
index 82f68698fb8483196d5ced096f79d1024bf9c4bc..f2a7728d3d00f70b6a6f6e3ea24749e6246b5c43 100644
--- a/hw/vfio/meson.build
+++ b/hw/vfio/meson.build
@@ -7,6 +7,8 @@ vfio_ss.add(files(
'container-legacy.c',
'helpers.c',
))
+vfio_ss.add(when: 'CONFIG_KVM', if_true: files('kvm-helpers.c'))
+stub_ss.add(files('kvm-stubs.c'))
vfio_ss.add(when: 'CONFIG_PSERIES', if_true: files('spapr.c'))
vfio_ss.add(when: 'CONFIG_VFIO_PCI', if_true: files(
'pci-quirks.c',
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 03/14] hw/vfio/pci-quirks.c: remove CONFIG_VFIO_IGD
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
2026-04-21 16:34 ` [PULL 01/14] hw/vfio/listener.c: remove CONFIG_KVM Cédric Le Goater
2026-04-21 16:34 ` [PULL 02/14] hw/vfio/helpers.c: extract kvm helpers in kvm-helpers.c Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 04/14] hw/vfio: eradicate CONFIG_IOMMU from sources Cédric Le Goater
` (11 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Add stubs for needed functions.
vfio_pci_hot_reset_info must be forwarded declared on non linux platforms:
In file included from ../hw/vfio/igd-stubs.c:9:
../hw/vfio/pci.h:265:44: error: 'struct vfio_pci_hot_reset_info' declared inside parameter list will not be visible outside of this definition or declaration [-Werror]
265 | struct vfio_pci_hot_reset_info **info_p);
|
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-4-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/pci.h | 2 ++
hw/vfio/igd-stubs.c | 20 ++++++++++++++++++++
hw/vfio/pci-quirks.c | 5 -----
hw/vfio/meson.build | 1 +
4 files changed, 23 insertions(+), 5 deletions(-)
create mode 100644 hw/vfio/igd-stubs.c
diff --git a/hw/vfio/pci.h b/hw/vfio/pci.h
index d6495d7f297513605db363dcdb3dd9b8dfb29a8d..c3a1f53d35004c914854c97d6e3608e4a20b121a 100644
--- a/hw/vfio/pci.h
+++ b/hw/vfio/pci.h
@@ -258,6 +258,8 @@ bool vfio_probe_igd_config_quirk(VFIOPCIDevice *vdev, Error **errp);
extern const PropertyInfo qdev_prop_nv_gpudirect_clique;
+struct vfio_pci_hot_reset_info;
+
void vfio_pci_pre_reset(VFIOPCIDevice *vdev);
void vfio_pci_post_reset(VFIOPCIDevice *vdev);
bool vfio_pci_host_match(PCIHostDeviceAddress *addr, const char *name);
diff --git a/hw/vfio/igd-stubs.c b/hw/vfio/igd-stubs.c
new file mode 100644
index 0000000000000000000000000000000000000000..f7687d909125c2233e76f7e6d03aa90f38c3b89f
--- /dev/null
+++ b/hw/vfio/igd-stubs.c
@@ -0,0 +1,20 @@
+/*
+ * IGD device quirks stubs
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "qapi/qmp/qerror.h"
+#include "pci.h"
+#include "pci-quirks.h"
+
+void vfio_probe_igd_bar0_quirk(VFIOPCIDevice *vdev, int nr)
+{
+ return;
+}
+
+bool vfio_probe_igd_config_quirk(VFIOPCIDevice *vdev, Error **errp)
+{
+ return true;
+}
diff --git a/hw/vfio/pci-quirks.c b/hw/vfio/pci-quirks.c
index 66e02b15a481aa810cca6fac465e64639673e970..bccf31751fd47387fd89110e27b695e5c822d5c4 100644
--- a/hw/vfio/pci-quirks.c
+++ b/hw/vfio/pci-quirks.c
@@ -11,7 +11,6 @@
*/
#include "qemu/osdep.h"
-#include CONFIG_DEVICES
#include "exec/memop.h"
#include "qemu/units.h"
#include "qemu/log.h"
@@ -1128,11 +1127,9 @@ static void vfio_probe_rtl8168_bar2_quirk(VFIOPCIDevice *vdev, int nr)
*/
bool vfio_config_quirk_setup(VFIOPCIDevice *vdev, Error **errp)
{
-#ifdef CONFIG_VFIO_IGD
if (!vfio_probe_igd_config_quirk(vdev, errp)) {
return false;
}
-#endif
return true;
}
@@ -1179,9 +1176,7 @@ void vfio_bar_quirk_setup(VFIOPCIDevice *vdev, int nr)
vfio_probe_nvidia_bar5_quirk(vdev, nr);
vfio_probe_nvidia_bar0_quirk(vdev, nr);
vfio_probe_rtl8168_bar2_quirk(vdev, nr);
-#ifdef CONFIG_VFIO_IGD
vfio_probe_igd_bar0_quirk(vdev, nr);
-#endif
}
void vfio_bar_quirk_exit(VFIOPCIDevice *vdev, int nr)
diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build
index f2a7728d3d00f70b6a6f6e3ea24749e6246b5c43..6c00a7f51bbe79d8e45d828a11d4deecf21512be 100644
--- a/hw/vfio/meson.build
+++ b/hw/vfio/meson.build
@@ -17,6 +17,7 @@ vfio_ss.add(when: 'CONFIG_VFIO_PCI', if_true: files(
vfio_ss.add(when: 'CONFIG_VFIO_CCW', if_true: files('ccw.c'))
vfio_ss.add(when: 'CONFIG_VFIO_AP', if_true: files('ap.c'))
vfio_ss.add(when: 'CONFIG_VFIO_IGD', if_true: files('igd.c'))
+stub_ss.add(files('igd-stubs.c'))
specific_ss.add_all(when: 'CONFIG_VFIO', if_true: vfio_ss)
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 04/14] hw/vfio: eradicate CONFIG_IOMMU from sources
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (2 preceding siblings ...)
2026-04-21 16:34 ` [PULL 03/14] hw/vfio/pci-quirks.c: remove CONFIG_VFIO_IGD Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 05/14] hw/vfio/pci.c: eradicate CONFIG_KVM Cédric Le Goater
` (10 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
This commit removes usage of CONFIG_IOMMU in hw/vfio sources, exposing
inconditionally iommufd related properties, which are declared
statically (in const arrays). The alternative to expose them dynamically
is more complex and requires boilerplate to set properties at runtime,
with set_* callbacks and added logic to check if iommufd backend is
available, with no obvious benefit.
One possible difference is that user may see a different error message
when trying to attach a vfio device with a QEMU not supporting iommufd,
without declaring iommufd object associated.
Instead of:
```
$ qemu-system-* -device vfio-pci,host=0000:01:00.0,iommufd=iommufd0
qemu-system-*: -device vfio-pci,host=0000:01:00.0,iommufd=iommufd0: Property 'vfio-pci.iommufd' not found
```
User will now see:
```
qemu-system-aarch64: -device vfio-pci,host=0000:01:00.0,iommufd=iommufd0: Device 'iommufd0' not found
```
However, since declaring the iommufd object is needed, error reported
before and after is still the same:
```
$ qemu-system-* -object iommufd,id=iommufd0 -device vfio-pci,host=0000:01:00.0,iommufd=iommufd0
qemu-system-*: invalid object type: iommufd
```
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-5-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/ap.c | 9 ---------
hw/vfio/ccw.c | 9 ---------
hw/vfio/pci.c | 11 -----------
3 files changed, 29 deletions(-)
diff --git a/hw/vfio/ap.c b/hw/vfio/ap.c
index 5c8f3056530c7027670f042ac920bd9e4a3da6e9..be20d7e7dc67f21363863f8c6d8b2a7cdedb8762 100644
--- a/hw/vfio/ap.c
+++ b/hw/vfio/ap.c
@@ -11,7 +11,6 @@
*/
#include "qemu/osdep.h"
-#include CONFIG_DEVICES /* CONFIG_IOMMUFD */
#include <linux/vfio.h>
#include <sys/ioctl.h>
#include "qapi/error.h"
@@ -279,10 +278,8 @@ static void vfio_ap_unrealize(DeviceState *dev)
static const Property vfio_ap_properties[] = {
DEFINE_PROP_STRING("sysfsdev", VFIOAPDevice, vdev.sysfsdev),
-#ifdef CONFIG_IOMMUFD
DEFINE_PROP_LINK("iommufd", VFIOAPDevice, vdev.iommufd,
TYPE_IOMMUFD_BACKEND, IOMMUFDBackend *),
-#endif
};
static void vfio_ap_reset(DeviceState *dev)
@@ -320,21 +317,17 @@ static void vfio_ap_instance_init(Object *obj)
vbasedev->mdev = true;
}
-#ifdef CONFIG_IOMMUFD
static void vfio_ap_set_fd(Object *obj, const char *str, Error **errp)
{
vfio_device_set_fd(&VFIO_AP_DEVICE(obj)->vdev, str, errp);
}
-#endif
static void vfio_ap_class_init(ObjectClass *klass, const void *data)
{
DeviceClass *dc = DEVICE_CLASS(klass);
device_class_set_props(dc, vfio_ap_properties);
-#ifdef CONFIG_IOMMUFD
object_class_property_add_str(klass, "fd", NULL, vfio_ap_set_fd);
-#endif
dc->vmsd = &vfio_ap_vmstate;
dc->desc = "VFIO-based AP device assignment";
set_bit(DEVICE_CATEGORY_MISC, dc->categories);
@@ -347,11 +340,9 @@ static void vfio_ap_class_init(ObjectClass *klass, const void *data)
object_class_property_set_description(klass, /* 3.1 */
"sysfsdev",
"Host sysfs path of assigned device");
-#ifdef CONFIG_IOMMUFD
object_class_property_set_description(klass, /* 9.0 */
"iommufd",
"Set host IOMMUFD backend device");
-#endif
}
static const TypeInfo vfio_ap_info = {
diff --git a/hw/vfio/ccw.c b/hw/vfio/ccw.c
index ce9c014e6a7f23b2eceecc05bc60509ece270e4b..ff6ab9a7408f01b9e64ba06f2c8da4a69fa1f9ff 100644
--- a/hw/vfio/ccw.c
+++ b/hw/vfio/ccw.c
@@ -15,7 +15,6 @@
*/
#include "qemu/osdep.h"
-#include CONFIG_DEVICES /* CONFIG_IOMMUFD */
#include <linux/vfio.h>
#include <linux/vfio_ccw.h>
#include <sys/ioctl.h>
@@ -647,10 +646,8 @@ static void vfio_ccw_unrealize(DeviceState *dev)
static const Property vfio_ccw_properties[] = {
DEFINE_PROP_STRING("sysfsdev", VFIOCCWDevice, vdev.sysfsdev),
DEFINE_PROP_BOOL("force-orb-pfch", VFIOCCWDevice, force_orb_pfch, false),
-#ifdef CONFIG_IOMMUFD
DEFINE_PROP_LINK("iommufd", VFIOCCWDevice, vdev.iommufd,
TYPE_IOMMUFD_BACKEND, IOMMUFDBackend *),
-#endif
DEFINE_PROP_CCW_LOADPARM("loadparm", CcwDevice, loadparm),
};
@@ -679,12 +676,10 @@ static void vfio_ccw_instance_init(Object *obj)
DEVICE(vcdev), true);
}
-#ifdef CONFIG_IOMMUFD
static void vfio_ccw_set_fd(Object *obj, const char *str, Error **errp)
{
vfio_device_set_fd(&VFIO_CCW(obj)->vdev, str, errp);
}
-#endif
static void vfio_ccw_class_init(ObjectClass *klass, const void *data)
{
@@ -692,9 +687,7 @@ static void vfio_ccw_class_init(ObjectClass *klass, const void *data)
S390CCWDeviceClass *cdc = S390_CCW_DEVICE_CLASS(klass);
device_class_set_props(dc, vfio_ccw_properties);
-#ifdef CONFIG_IOMMUFD
object_class_property_add_str(klass, "fd", NULL, vfio_ccw_set_fd);
-#endif
dc->vmsd = &vfio_ccw_vmstate;
dc->desc = "VFIO-based subchannel assignment";
set_bit(DEVICE_CATEGORY_MISC, dc->categories);
@@ -713,11 +706,9 @@ static void vfio_ccw_class_init(ObjectClass *klass, const void *data)
object_class_property_set_description(klass, /* 3.0 */
"force-orb-pfch",
"Force unlimited prefetch");
-#ifdef CONFIG_IOMMUFD
object_class_property_set_description(klass, /* 9.0 */
"iommufd",
"Set host IOMMUFD backend device");
-#endif
object_class_property_set_description(klass, /* 9.2 */
"loadparm",
"Define which devices that can be used for booting");
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 1945751ffd44dd129768ac7788f0375b216f1747..1f1f3aaad9a901a087ecad1b5c46e8756b3a6339 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -19,7 +19,6 @@
*/
#include "qemu/osdep.h"
-#include CONFIG_DEVICES /* CONFIG_IOMMUFD */
#include <linux/vfio.h>
#include <sys/ioctl.h>
@@ -3471,9 +3470,7 @@ static void vfio_pci_realize(PCIDevice *pdev, Error **errp)
~vdev->host.slot || ~vdev->host.function)) {
error_setg(errp, "No provided host device");
error_append_hint(errp, "Use -device vfio-pci,host=DDDD:BB:DD.F "
-#ifdef CONFIG_IOMMUFD
"or -device vfio-pci,fd=DEVICE_FD "
-#endif
"or -device vfio-pci,sysfsdev=PATH_TO_DEVICE\n");
return;
}
@@ -3816,22 +3813,18 @@ static const Property vfio_pci_properties[] = {
qdev_prop_nv_gpudirect_clique, uint8_t),
DEFINE_PROP_OFF_AUTO_PCIBAR("x-msix-relocation", VFIOPCIDevice, msix_relo,
OFF_AUTO_PCIBAR_OFF),
-#ifdef CONFIG_IOMMUFD
DEFINE_PROP_LINK("iommufd", VFIOPCIDevice, vbasedev.iommufd,
TYPE_IOMMUFD_BACKEND, IOMMUFDBackend *),
-#endif
DEFINE_PROP_BOOL("skip-vsc-check", VFIOPCIDevice, skip_vsc_check, true),
DEFINE_PROP_UINT16("x-vpasid-cap-offset", VFIOPCIDevice,
vpasid_cap_offset, 0),
};
-#ifdef CONFIG_IOMMUFD
static void vfio_pci_set_fd(Object *obj, const char *str, Error **errp)
{
VFIOPCIDevice *vdev = VFIO_PCI_DEVICE(obj);
vfio_device_set_fd(&vdev->vbasedev, str, errp);
}
-#endif
static void vfio_pci_class_init(ObjectClass *klass, const void *data)
{
@@ -3840,9 +3833,7 @@ static void vfio_pci_class_init(ObjectClass *klass, const void *data)
device_class_set_legacy_reset(dc, vfio_pci_reset);
device_class_set_props(dc, vfio_pci_properties);
-#ifdef CONFIG_IOMMUFD
object_class_property_add_str(klass, "fd", NULL, vfio_pci_set_fd);
-#endif
dc->vmsd = &vfio_cpr_pci_vmstate;
dc->desc = "VFIO-based PCI device assignment";
pdc->realize = vfio_pci_realize;
@@ -3944,11 +3935,9 @@ static void vfio_pci_class_init(ObjectClass *klass, const void *data)
"vf-token",
"Specify UUID VF token. Required for VF when PF is owned "
"by another VFIO driver");
-#ifdef CONFIG_IOMMUFD
object_class_property_set_description(klass, /* 9.0 */
"iommufd",
"Set host IOMMUFD backend device");
-#endif
object_class_property_set_description(klass, /* 9.1 */
"x-device-dirty-page-tracking",
"Disable device dirty page tracking and use "
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 05/14] hw/vfio/pci.c: eradicate CONFIG_KVM
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (3 preceding siblings ...)
2026-04-21 16:34 ` [PULL 04/14] hw/vfio: eradicate CONFIG_IOMMU from sources Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 06/14] hw/vfio/ap.c: use full path for target specific header Cédric Le Goater
` (9 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Philippe Mathieu-Daudé,
Cédric Le Goater
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
We just need to add kvm_enabled() guard when calling concerned
functions, but no need to extract those kvm functions since they are not
using any kvm specific types that would not be visible at compilation
time.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-6-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/pci.c | 28 ++++++++++------------------
1 file changed, 10 insertions(+), 18 deletions(-)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 1f1f3aaad9a901a087ecad1b5c46e8756b3a6339..b2a07f6bb4217bcacead194ce75b359b1896ef3a 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -152,7 +152,6 @@ void vfio_pci_intx_eoi(VFIODevice *vbasedev)
static bool vfio_intx_enable_kvm(VFIOPCIDevice *vdev, Error **errp)
{
-#ifdef CONFIG_KVM
PCIDevice *pdev = PCI_DEVICE(vdev);
int irq_fd = event_notifier_get_fd(&vdev->intx.interrupt);
@@ -206,14 +205,10 @@ fail:
qemu_set_fd_handler(irq_fd, vfio_intx_interrupt, NULL, vdev);
vfio_device_irq_unmask(&vdev->vbasedev, VFIO_PCI_INTX_IRQ_INDEX);
return false;
-#else
- return true;
-#endif
}
static bool vfio_cpr_intx_enable_kvm(VFIOPCIDevice *vdev, Error **errp)
{
-#ifdef CONFIG_KVM
if (vdev->no_kvm_intx || !kvm_irqfds_enabled() ||
vdev->intx.route.mode != PCI_INTX_ENABLED ||
!kvm_resamplefds_enabled()) {
@@ -236,14 +231,10 @@ static bool vfio_cpr_intx_enable_kvm(VFIOPCIDevice *vdev, Error **errp)
vdev->intx.kvm_accel = true;
trace_vfio_intx_enable_kvm(vdev->vbasedev.name);
return true;
-#else
- return true;
-#endif
}
static void vfio_intx_disable_kvm(VFIOPCIDevice *vdev)
{
-#ifdef CONFIG_KVM
PCIDevice *pdev = PCI_DEVICE(vdev);
if (!vdev->intx.kvm_accel) {
@@ -277,7 +268,6 @@ static void vfio_intx_disable_kvm(VFIOPCIDevice *vdev)
vfio_device_irq_unmask(&vdev->vbasedev, VFIO_PCI_INTX_IRQ_INDEX);
trace_vfio_intx_disable_kvm(vdev->vbasedev.name);
-#endif
}
static void vfio_intx_update(VFIOPCIDevice *vdev, PCIINTxRoute *route)
@@ -287,7 +277,9 @@ static void vfio_intx_update(VFIOPCIDevice *vdev, PCIINTxRoute *route)
trace_vfio_intx_update(vdev->vbasedev.name,
vdev->intx.route.irq, route->irq);
- vfio_intx_disable_kvm(vdev);
+ if (kvm_enabled()) {
+ vfio_intx_disable_kvm(vdev);
+ }
vdev->intx.route = *route;
@@ -295,7 +287,7 @@ static void vfio_intx_update(VFIOPCIDevice *vdev, PCIINTxRoute *route)
return;
}
- if (!vfio_intx_enable_kvm(vdev, &err)) {
+ if (kvm_enabled() && !vfio_intx_enable_kvm(vdev, &err)) {
warn_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
}
@@ -350,16 +342,14 @@ static bool vfio_intx_enable(VFIOPCIDevice *vdev, Error **errp)
vdev->intx.pin = pin - 1; /* Pin A (1) -> irq[0] */
pci_config_set_interrupt_pin(pdev->config, pin);
-#ifdef CONFIG_KVM
/*
* Only conditional to avoid generating error messages on platforms
* where we won't actually use the result anyway.
*/
- if (kvm_irqfds_enabled() && kvm_resamplefds_enabled()) {
+ if (kvm_enabled() && kvm_irqfds_enabled() && kvm_resamplefds_enabled()) {
vdev->intx.route = pci_device_route_intx_to_irq(pdev,
vdev->intx.pin);
}
-#endif
if (!vfio_notifier_init(vdev, &vdev->intx.interrupt, "intx-interrupt", 0,
errp)) {
@@ -370,7 +360,7 @@ static bool vfio_intx_enable(VFIOPCIDevice *vdev, Error **errp)
if (cpr_is_incoming()) {
- if (!vfio_cpr_intx_enable_kvm(vdev, &err)) {
+ if (kvm_enabled() && !vfio_cpr_intx_enable_kvm(vdev, &err)) {
warn_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
}
goto skip_signaling;
@@ -383,7 +373,7 @@ static bool vfio_intx_enable(VFIOPCIDevice *vdev, Error **errp)
return false;
}
- if (!vfio_intx_enable_kvm(vdev, &err)) {
+ if (kvm_enabled() && !vfio_intx_enable_kvm(vdev, &err)) {
warn_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
}
@@ -400,7 +390,9 @@ static void vfio_intx_disable(VFIOPCIDevice *vdev)
int fd;
timer_del(vdev->intx.mmap_timer);
- vfio_intx_disable_kvm(vdev);
+ if (kvm_enabled()) {
+ vfio_intx_disable_kvm(vdev);
+ }
vfio_device_irq_disable(&vdev->vbasedev, VFIO_PCI_INTX_IRQ_INDEX);
vdev->intx.pending = false;
pci_irq_deassert(pdev);
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 06/14] hw/vfio/ap.c: use full path for target specific header
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (4 preceding siblings ...)
2026-04-21 16:34 ` [PULL 05/14] hw/vfio/pci.c: eradicate CONFIG_KVM Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 07/14] hw/vfio/spapr.c: extract vfio_spapr_kvm_attach_tce to hw/vfio/kvm-spapr.c Cédric Le Goater
` (8 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
In addition, we fix target/s390x/kvm/kvm_s390x.h cpu-qom include also.
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-7-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
target/s390x/kvm/kvm_s390x.h | 2 +-
hw/vfio/ap.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/s390x/kvm/kvm_s390x.h b/target/s390x/kvm/kvm_s390x.h
index 649dae5948a6e4069beb526d6be1df667d68e292..7b1cce3e60d5a407239d4055bbf0196edbc8c686 100644
--- a/target/s390x/kvm/kvm_s390x.h
+++ b/target/s390x/kvm/kvm_s390x.h
@@ -10,7 +10,7 @@
#ifndef KVM_S390X_H
#define KVM_S390X_H
-#include "cpu-qom.h"
+#include "target/s390x/cpu-qom.h"
struct kvm_s390_irq;
diff --git a/hw/vfio/ap.c b/hw/vfio/ap.c
index be20d7e7dc67f21363863f8c6d8b2a7cdedb8762..276ae8c207bf2a1d7a0d5c0fc7f2ef98cec695d1 100644
--- a/hw/vfio/ap.c
+++ b/hw/vfio/ap.c
@@ -25,7 +25,7 @@
#include "qemu/module.h"
#include "qemu/option.h"
#include "qemu/config-file.h"
-#include "kvm/kvm_s390x.h"
+#include "target/s390x/kvm/kvm_s390x.h"
#include "migration/vmstate.h"
#include "hw/core/qdev-properties.h"
#include "hw/s390x/ap-bridge.h"
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 07/14] hw/vfio/spapr.c: extract vfio_spapr_kvm_attach_tce to hw/vfio/kvm-spapr.c
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (5 preceding siblings ...)
2026-04-21 16:34 ` [PULL 06/14] hw/vfio/ap.c: use full path for target specific header Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 08/14] hw/vfio: all vfio files can now be common files Cédric Le Goater
` (7 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Philippe Mathieu-Daudé,
Cédric Le Goater
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Since this function needs kvm specific types, we need to extract in
another file and link it only for KVM builds.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-8-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/kvm-spapr.h | 12 ++++++++++++
hw/vfio/kvm-spapr.c | 47 +++++++++++++++++++++++++++++++++++++++++++++
hw/vfio/kvm-stubs.c | 8 ++++++++
hw/vfio/spapr.c | 30 ++++-------------------------
hw/vfio/meson.build | 1 +
5 files changed, 72 insertions(+), 26 deletions(-)
create mode 100644 hw/vfio/kvm-spapr.h
create mode 100644 hw/vfio/kvm-spapr.c
diff --git a/hw/vfio/kvm-spapr.h b/hw/vfio/kvm-spapr.h
new file mode 100644
index 0000000000000000000000000000000000000000..b1f68c686a7defbed94cc4c0d56b786bc436ed89
--- /dev/null
+++ b/hw/vfio/kvm-spapr.h
@@ -0,0 +1,12 @@
+/*
+ * VFIO sPAPR KVM specific functions
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "hw/vfio/vfio-container.h"
+#include "qapi/error.h"
+
+bool vfio_spapr_kvm_attach_tce(VFIOContainer *bcontainer,
+ MemoryRegionSection *section,
+ Error **errp);
diff --git a/hw/vfio/kvm-spapr.c b/hw/vfio/kvm-spapr.c
new file mode 100644
index 0000000000000000000000000000000000000000..ad71c5a85e273fbe8df9fda71ba21c7df171914d
--- /dev/null
+++ b/hw/vfio/kvm-spapr.c
@@ -0,0 +1,47 @@
+/*
+ * VFIO sPAPR KVM specific functions
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include <sys/ioctl.h>
+#include <linux/vfio.h>
+#include <linux/kvm.h>
+
+#include "hw/vfio/vfio-container-legacy.h"
+#include "hw/vfio/kvm-spapr.h"
+#include "qapi/error.h"
+#include "trace.h"
+#include "vfio-helpers.h"
+
+bool vfio_spapr_kvm_attach_tce(VFIOContainer *bcontainer,
+ MemoryRegionSection *section,
+ Error **errp)
+{
+ VFIOLegacyContainer *container = VFIO_IOMMU_LEGACY(bcontainer);
+ VFIOGroup *group;
+ IOMMUMemoryRegion *iommu_mr = IOMMU_MEMORY_REGION(section->mr);
+ struct kvm_vfio_spapr_tce param;
+ struct kvm_device_attr attr = {
+ .group = KVM_DEV_VFIO_GROUP,
+ .attr = KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE,
+ .addr = (uint64_t)(unsigned long)¶m,
+ };
+
+ if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_FD,
+ ¶m.tablefd)) {
+ QLIST_FOREACH(group, &container->group_list, container_next) {
+ param.groupfd = group->fd;
+ if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
+ error_setg_errno(errp, errno,
+ "vfio: failed GROUP_SET_SPAPR_TCE for "
+ "KVM VFIO device %d and group fd %d",
+ param.tablefd, param.groupfd);
+ return false;
+ }
+ trace_vfio_spapr_group_attach(param.groupfd, param.tablefd);
+ }
+ }
+ return true;
+}
diff --git a/hw/vfio/kvm-stubs.c b/hw/vfio/kvm-stubs.c
index 5a489d1b7110a5564054398cc354bb806ac39334..78c51b9915559ba3c328cb3bcc0d9e2896269f9f 100644
--- a/hw/vfio/kvm-stubs.c
+++ b/hw/vfio/kvm-stubs.c
@@ -6,6 +6,7 @@
#include "qemu/osdep.h"
+#include "hw/vfio/kvm-spapr.h"
#include "hw/vfio/vfio-device.h"
#include "qapi/error.h"
#include "vfio-helpers.h"
@@ -24,3 +25,10 @@ int vfio_kvm_device_del_fd(int fd, Error **errp)
{
return 0;
}
+
+bool vfio_spapr_kvm_attach_tce(VFIOContainer *bcontainer,
+ MemoryRegionSection *section,
+ Error **errp)
+{
+ g_assert_not_reached();
+}
diff --git a/hw/vfio/spapr.c b/hw/vfio/spapr.c
index a9f093c357057eaa07378c77c86da096fe9f6297..42690e4323d20915c55591129d9fce0ae46b0cf3 100644
--- a/hw/vfio/spapr.c
+++ b/hw/vfio/spapr.c
@@ -16,6 +16,7 @@
#include "system/address-spaces.h"
#include "hw/vfio/vfio-container-legacy.h"
+#include "hw/vfio/kvm-spapr.h"
#include "hw/core/hw-error.h"
#include "qemu/error-report.h"
#include "qapi/error.h"
@@ -406,33 +407,10 @@ vfio_spapr_container_add_section_window(VFIOContainer *bcontainer,
vfio_host_win_add(scontainer, section->offset_within_address_space,
section->offset_within_address_space +
int128_get64(section->size) - 1, pgsize);
-#ifdef CONFIG_KVM
- if (kvm_enabled()) {
- VFIOGroup *group;
- IOMMUMemoryRegion *iommu_mr = IOMMU_MEMORY_REGION(section->mr);
- struct kvm_vfio_spapr_tce param;
- struct kvm_device_attr attr = {
- .group = KVM_DEV_VFIO_GROUP,
- .attr = KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE,
- .addr = (uint64_t)(unsigned long)¶m,
- };
-
- if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_FD,
- ¶m.tablefd)) {
- QLIST_FOREACH(group, &container->group_list, container_next) {
- param.groupfd = group->fd;
- if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) {
- error_setg_errno(errp, errno,
- "vfio: failed GROUP_SET_SPAPR_TCE for "
- "KVM VFIO device %d and group fd %d",
- param.tablefd, param.groupfd);
- return false;
- }
- trace_vfio_spapr_group_attach(param.groupfd, param.tablefd);
- }
- }
+ if (kvm_enabled() && !vfio_spapr_kvm_attach_tce(bcontainer, section, errp)) {
+ return false;
}
-#endif
+
return true;
}
diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build
index 6c00a7f51bbe79d8e45d828a11d4deecf21512be..bab5f2b7f1555eed452e7a91b8a8984bb1f262c6 100644
--- a/hw/vfio/meson.build
+++ b/hw/vfio/meson.build
@@ -10,6 +10,7 @@ vfio_ss.add(files(
vfio_ss.add(when: 'CONFIG_KVM', if_true: files('kvm-helpers.c'))
stub_ss.add(files('kvm-stubs.c'))
vfio_ss.add(when: 'CONFIG_PSERIES', if_true: files('spapr.c'))
+vfio_ss.add(when: ['CONFIG_KVM', 'CONFIG_PSERIES'], if_true: files('kvm-spapr.c'))
vfio_ss.add(when: 'CONFIG_VFIO_PCI', if_true: files(
'pci-quirks.c',
'pci.c',
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 08/14] hw/vfio: all vfio files can now be common files
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (6 preceding siblings ...)
2026-04-21 16:34 ` [PULL 07/14] hw/vfio/spapr.c: extract vfio_spapr_kvm_attach_tce to hw/vfio/kvm-spapr.c Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 09/14] iommufd: Rename all the idev and idevc variables to hiod and hiodc Cédric Le Goater
` (6 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Pierrick Bouvier, Philippe Mathieu-Daudé,
Cédric Le Goater
From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Cédric Le Goater <clg@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260318174733.1717643-9-pierrick.bouvier@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio/meson.build | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build
index bab5f2b7f1555eed452e7a91b8a8984bb1f262c6..fa0ea6ecf0e1478e88769af73da1badc9daca921 100644
--- a/hw/vfio/meson.build
+++ b/hw/vfio/meson.build
@@ -20,7 +20,7 @@ vfio_ss.add(when: 'CONFIG_VFIO_AP', if_true: files('ap.c'))
vfio_ss.add(when: 'CONFIG_VFIO_IGD', if_true: files('igd.c'))
stub_ss.add(files('igd-stubs.c'))
-specific_ss.add_all(when: 'CONFIG_VFIO', if_true: vfio_ss)
+system_ss.add_all(when: 'CONFIG_VFIO', if_true: vfio_ss)
system_ss.add(when: 'CONFIG_VFIO', if_true: files(
'cpr.c',
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 09/14] iommufd: Rename all the idev and idevc variables to hiod and hiodc
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (7 preceding siblings ...)
2026-04-21 16:34 ` [PULL 08/14] hw/vfio: all vfio files can now be common files Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 10/14] hw/vfio/iommufd: Control dirty tracking for nesting parent HWPT Cédric Le Goater
` (5 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Zhenzhong Duan, Cédric Le Goater,
Philippe Mathieu-Daudé
From: Zhenzhong Duan <zhenzhong.duan@intel.com>
We used idev and idevc naming for HostIOMMUDeviceIOMMUFD and corresponding
class variables which followed the iommufd_device naming in linux kernel.
This is mixed with the hiod naming for base type HostIOMMUDevice. Rename
HostIOMMUDeviceIOMMUFD* to hiodi* for consistency in QEMU.
No functional change intended.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260401080354.1347212-1-zhenzhong.duan@intel.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/arm/smmuv3-accel.h | 2 +-
include/system/iommufd.h | 12 ++---
backends/iommufd.c | 26 +++++------
hw/arm/smmuv3-accel.c | 93 +++++++++++++++++++------------------
hw/i386/intel_iommu_accel.c | 44 +++++++++---------
hw/vfio/container-legacy.c | 10 ++--
hw/vfio/iommufd.c | 24 +++++-----
7 files changed, 107 insertions(+), 104 deletions(-)
diff --git a/hw/arm/smmuv3-accel.h b/hw/arm/smmuv3-accel.h
index dba6c71de5269a2223a9048e703ae2f829513f8e..908a13cbcc6b44c1d524221c489695176bc07ed1 100644
--- a/hw/arm/smmuv3-accel.h
+++ b/hw/arm/smmuv3-accel.h
@@ -34,7 +34,7 @@ typedef struct SMMUS1Hwpt {
typedef struct SMMUv3AccelDevice {
SMMUDevice sdev;
- HostIOMMUDeviceIOMMUFD *idev;
+ HostIOMMUDeviceIOMMUFD *hiodi;
SMMUS1Hwpt *s1_hwpt;
IOMMUFDVdev *vdev;
QLIST_ENTRY(SMMUv3AccelDevice) next;
diff --git a/include/system/iommufd.h b/include/system/iommufd.h
index 7062944fe61d161fa6ca1c0cff1be879e6449955..2925d116acdf285ae3ff9968a018a924be6138db 100644
--- a/include/system/iommufd.h
+++ b/include/system/iommufd.h
@@ -136,7 +136,7 @@ struct HostIOMMUDeviceIOMMUFDClass {
*
* Mandatory callback.
*
- * @idev: host IOMMU device backed by IOMMUFD backend.
+ * @hiodi: host IOMMU device backed by IOMMUFD backend.
*
* @hwpt_id: ID of IOMMUFD hardware page table.
*
@@ -144,7 +144,7 @@ struct HostIOMMUDeviceIOMMUFDClass {
*
* Returns: true on success, false on failure.
*/
- bool (*attach_hwpt)(HostIOMMUDeviceIOMMUFD *idev, uint32_t hwpt_id,
+ bool (*attach_hwpt)(HostIOMMUDeviceIOMMUFD *hiodi, uint32_t hwpt_id,
Error **errp);
/**
* @detach_hwpt: detach host IOMMU device from IOMMUFD hardware page table.
@@ -152,17 +152,17 @@ struct HostIOMMUDeviceIOMMUFDClass {
*
* Mandatory callback.
*
- * @idev: host IOMMU device backed by IOMMUFD backend.
+ * @hiodi: host IOMMU device backed by IOMMUFD backend.
*
* @errp: pass an Error out when attachment fails.
*
* Returns: true on success, false on failure.
*/
- bool (*detach_hwpt)(HostIOMMUDeviceIOMMUFD *idev, Error **errp);
+ bool (*detach_hwpt)(HostIOMMUDeviceIOMMUFD *hiodi, Error **errp);
};
-bool host_iommu_device_iommufd_attach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+bool host_iommu_device_iommufd_attach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
uint32_t hwpt_id, Error **errp);
-bool host_iommu_device_iommufd_detach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+bool host_iommu_device_iommufd_detach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
Error **errp);
#endif
diff --git a/backends/iommufd.c b/backends/iommufd.c
index e1fee16acf2e15ad1e9066aa32a270bb3945b52d..410b0443707a2baed91ddf221bae4df0f2376c8a 100644
--- a/backends/iommufd.c
+++ b/backends/iommufd.c
@@ -538,24 +538,24 @@ bool iommufd_backend_alloc_veventq(IOMMUFDBackend *be, uint32_t viommu_id,
return true;
}
-bool host_iommu_device_iommufd_attach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+bool host_iommu_device_iommufd_attach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
uint32_t hwpt_id, Error **errp)
{
- HostIOMMUDeviceIOMMUFDClass *idevc =
- HOST_IOMMU_DEVICE_IOMMUFD_GET_CLASS(idev);
+ HostIOMMUDeviceIOMMUFDClass *hiodic =
+ HOST_IOMMU_DEVICE_IOMMUFD_GET_CLASS(hiodi);
- g_assert(idevc->attach_hwpt);
- return idevc->attach_hwpt(idev, hwpt_id, errp);
+ g_assert(hiodic->attach_hwpt);
+ return hiodic->attach_hwpt(hiodi, hwpt_id, errp);
}
-bool host_iommu_device_iommufd_detach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+bool host_iommu_device_iommufd_detach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
Error **errp)
{
- HostIOMMUDeviceIOMMUFDClass *idevc =
- HOST_IOMMU_DEVICE_IOMMUFD_GET_CLASS(idev);
+ HostIOMMUDeviceIOMMUFDClass *hiodic =
+ HOST_IOMMU_DEVICE_IOMMUFD_GET_CLASS(hiodi);
- g_assert(idevc->detach_hwpt);
- return idevc->detach_hwpt(idev, errp);
+ g_assert(hiodic->detach_hwpt);
+ return hiodic->detach_hwpt(hiodi, errp);
}
static int hiod_iommufd_get_cap(HostIOMMUDevice *hiod, int cap, Error **errp)
@@ -591,10 +591,10 @@ static bool hiod_iommufd_get_pasid_info(HostIOMMUDevice *hiod,
static void hiod_iommufd_class_init(ObjectClass *oc, const void *data)
{
- HostIOMMUDeviceClass *hioc = HOST_IOMMU_DEVICE_CLASS(oc);
+ HostIOMMUDeviceClass *hiodc = HOST_IOMMU_DEVICE_CLASS(oc);
- hioc->get_cap = hiod_iommufd_get_cap;
- hioc->get_pasid_info = hiod_iommufd_get_pasid_info;
+ hiodc->get_cap = hiod_iommufd_get_cap;
+ hiodc->get_pasid_info = hiod_iommufd_get_pasid_info;
};
static const TypeInfo types[] = {
diff --git a/hw/arm/smmuv3-accel.c b/hw/arm/smmuv3-accel.c
index 65c2f44880a06402049ec9ba885c7eb082419f79..3630078751ca544d023398ed088047c6c561c1ca 100644
--- a/hw/arm/smmuv3-accel.c
+++ b/hw/arm/smmuv3-accel.c
@@ -129,16 +129,16 @@ smmuv3_accel_check_hw_compatible(SMMUv3State *s,
}
static bool
-smmuv3_accel_hw_compatible(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
+smmuv3_accel_hw_compatible(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *hiodi,
Error **errp)
{
struct iommu_hw_info_arm_smmuv3 info;
uint32_t data_type;
uint64_t caps;
- if (!iommufd_backend_get_device_info(idev->iommufd, idev->devid, &data_type,
- &info, sizeof(info), &caps, NULL,
- errp)) {
+ if (!iommufd_backend_get_device_info(hiodi->iommufd, hiodi->devid,
+ &data_type, &info, sizeof(info), &caps,
+ NULL, errp)) {
return false;
}
@@ -182,15 +182,15 @@ static bool
smmuv3_accel_alloc_vdev(SMMUv3AccelDevice *accel_dev, int sid, Error **errp)
{
SMMUv3AccelState *accel = accel_dev->s_accel;
- HostIOMMUDeviceIOMMUFD *idev = accel_dev->idev;
+ HostIOMMUDeviceIOMMUFD *hiodi = accel_dev->hiodi;
IOMMUFDVdev *vdev = accel_dev->vdev;
uint32_t vdevice_id;
- if (!idev || vdev) {
+ if (!hiodi || vdev) {
return true;
}
- if (!iommufd_backend_alloc_vdev(idev->iommufd, idev->devid,
+ if (!iommufd_backend_alloc_vdev(hiodi->iommufd, hiodi->devid,
accel->viommu->viommu_id, sid,
&vdevice_id, errp)) {
return false;
@@ -209,7 +209,7 @@ smmuv3_accel_dev_alloc_translate(SMMUv3AccelDevice *accel_dev, STE *ste,
{
uint64_t ste_0 = (uint64_t)ste->word[0] | (uint64_t)ste->word[1] << 32;
uint64_t ste_1 = (uint64_t)ste->word[2] | (uint64_t)ste->word[3] << 32;
- HostIOMMUDeviceIOMMUFD *idev = accel_dev->idev;
+ HostIOMMUDeviceIOMMUFD *hiodi = accel_dev->hiodi;
SMMUv3AccelState *accel = accel_dev->s_accel;
struct iommu_hwpt_arm_smmuv3 nested_data = {
.ste = {
@@ -220,7 +220,7 @@ smmuv3_accel_dev_alloc_translate(SMMUv3AccelDevice *accel_dev, STE *ste,
uint32_t hwpt_id = 0, flags = 0;
SMMUS1Hwpt *s1_hwpt;
- if (!iommufd_backend_alloc_hwpt(idev->iommufd, idev->devid,
+ if (!iommufd_backend_alloc_hwpt(hiodi->iommufd, hiodi->devid,
accel->viommu->viommu_id, flags,
IOMMU_HWPT_DATA_ARM_SMMUV3,
sizeof(nested_data), &nested_data,
@@ -242,7 +242,7 @@ bool smmuv3_accel_install_ste(SMMUv3State *s, SMMUDevice *sdev, int sid,
.inval_ste_allowed = true};
SMMUv3AccelState *accel = s->s_accel;
SMMUv3AccelDevice *accel_dev;
- HostIOMMUDeviceIOMMUFD *idev;
+ HostIOMMUDeviceIOMMUFD *hiodi;
uint32_t config, hwpt_id = 0;
SMMUS1Hwpt *s1_hwpt = NULL;
const char *type;
@@ -257,7 +257,7 @@ bool smmuv3_accel_install_ste(SMMUv3State *s, SMMUDevice *sdev, int sid,
return true;
}
- idev = accel_dev->idev;
+ hiodi = accel_dev->hiodi;
if (!smmuv3_accel_alloc_vdev(accel_dev, sid, errp)) {
return false;
}
@@ -300,9 +300,9 @@ bool smmuv3_accel_install_ste(SMMUv3State *s, SMMUDevice *sdev, int sid,
return false;
}
- if (!host_iommu_device_iommufd_attach_hwpt(idev, hwpt_id, errp)) {
+ if (!host_iommu_device_iommufd_attach_hwpt(hiodi, hwpt_id, errp)) {
if (s1_hwpt) {
- iommufd_backend_free_id(idev->iommufd, s1_hwpt->hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, s1_hwpt->hwpt_id);
g_free(s1_hwpt);
}
return false;
@@ -310,7 +310,7 @@ bool smmuv3_accel_install_ste(SMMUv3State *s, SMMUDevice *sdev, int sid,
/* Free the previous s1_hwpt */
if (accel_dev->s1_hwpt) {
- iommufd_backend_free_id(idev->iommufd, accel_dev->s1_hwpt->hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, accel_dev->s1_hwpt->hwpt_id);
g_free(accel_dev->s1_hwpt);
}
@@ -524,7 +524,7 @@ free_veventq:
}
static bool
-smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
+smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *hiodi,
Error **errp)
{
SMMUv3AccelState *accel = s->s_accel;
@@ -534,11 +534,11 @@ smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
struct iommu_hwpt_arm_smmuv3 abort_data = {
.ste = { SMMU_STE_VALID, 0x0ULL },
};
- uint32_t s2_hwpt_id = idev->hwpt_id;
+ uint32_t s2_hwpt_id = hiodi->hwpt_id;
uint32_t viommu_id, hwpt_id;
IOMMUFDViommu *viommu;
- if (!iommufd_backend_alloc_viommu(idev->iommufd, idev->devid,
+ if (!iommufd_backend_alloc_viommu(hiodi->iommufd, hiodi->devid,
IOMMU_VIOMMU_TYPE_ARM_SMMUV3,
s2_hwpt_id, &viommu_id, errp)) {
return false;
@@ -547,21 +547,21 @@ smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
viommu = g_new0(IOMMUFDViommu, 1);
viommu->viommu_id = viommu_id;
viommu->s2_hwpt_id = s2_hwpt_id;
- viommu->iommufd = idev->iommufd;
+ viommu->iommufd = hiodi->iommufd;
accel->viommu = viommu;
/*
* Pre-allocate HWPTs for S1 bypass and abort cases. These will be attached
* later for guest STEs or GBPAs that require bypass or abort configuration.
*/
- if (!iommufd_backend_alloc_hwpt(idev->iommufd, idev->devid, viommu_id,
+ if (!iommufd_backend_alloc_hwpt(hiodi->iommufd, hiodi->devid, viommu_id,
0, IOMMU_HWPT_DATA_ARM_SMMUV3,
sizeof(abort_data), &abort_data,
&accel->abort_hwpt_id, errp)) {
goto free_viommu;
}
- if (!iommufd_backend_alloc_hwpt(idev->iommufd, idev->devid, viommu_id,
+ if (!iommufd_backend_alloc_hwpt(hiodi->iommufd, hiodi->devid, viommu_id,
0, IOMMU_HWPT_DATA_ARM_SMMUV3,
sizeof(bypass_data), &bypass_data,
&accel->bypass_hwpt_id, errp)) {
@@ -575,7 +575,7 @@ smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
/* Attach a HWPT based on SMMUv3 GBPA.ABORT value */
hwpt_id = smmuv3_accel_gbpa_hwpt(s, accel);
- if (!host_iommu_device_iommufd_attach_hwpt(idev, hwpt_id, errp)) {
+ if (!host_iommu_device_iommufd_attach_hwpt(hiodi, hwpt_id, errp)) {
goto free_veventq;
}
return true;
@@ -583,11 +583,11 @@ smmuv3_accel_alloc_viommu(SMMUv3State *s, HostIOMMUDeviceIOMMUFD *idev,
free_veventq:
smmuv3_accel_free_veventq(accel);
free_bypass_hwpt:
- iommufd_backend_free_id(idev->iommufd, accel->bypass_hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, accel->bypass_hwpt_id);
free_abort_hwpt:
- iommufd_backend_free_id(idev->iommufd, accel->abort_hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, accel->abort_hwpt_id);
free_viommu:
- iommufd_backend_free_id(idev->iommufd, viommu->viommu_id);
+ iommufd_backend_free_id(hiodi->iommufd, viommu->viommu_id);
g_free(viommu);
accel->viommu = NULL;
return false;
@@ -596,20 +596,20 @@ free_viommu:
static bool smmuv3_accel_set_iommu_device(PCIBus *bus, void *opaque, int devfn,
HostIOMMUDevice *hiod, Error **errp)
{
- HostIOMMUDeviceIOMMUFD *idev = HOST_IOMMU_DEVICE_IOMMUFD(hiod);
+ HostIOMMUDeviceIOMMUFD *hiodi = HOST_IOMMU_DEVICE_IOMMUFD(hiod);
SMMUState *bs = opaque;
SMMUv3State *s = ARM_SMMUV3(bs);
SMMUPciBus *sbus = smmu_get_sbus(bs, bus);
SMMUv3AccelDevice *accel_dev = smmuv3_accel_get_dev(bs, sbus, bus, devfn);
- if (!idev) {
+ if (!hiodi) {
return true;
}
- if (accel_dev->idev) {
- if (accel_dev->idev != idev) {
- error_setg(errp, "Device already has an associated idev 0x%x",
- idev->devid);
+ if (accel_dev->hiodi) {
+ if (accel_dev->hiodi != hiodi) {
+ error_setg(errp, "Device already has an associated hiodi 0x%x",
+ hiodi->devid);
return false;
}
return true;
@@ -619,7 +619,7 @@ static bool smmuv3_accel_set_iommu_device(PCIBus *bus, void *opaque, int devfn,
* Check the host SMMUv3 associated with the dev is compatible with the
* QEMU SMMUv3 accel.
*/
- if (!smmuv3_accel_hw_compatible(s, idev, errp)) {
+ if (!smmuv3_accel_hw_compatible(s, hiodi, errp)) {
return false;
}
@@ -627,17 +627,17 @@ static bool smmuv3_accel_set_iommu_device(PCIBus *bus, void *opaque, int devfn,
goto done;
}
- if (!smmuv3_accel_alloc_viommu(s, idev, errp)) {
- error_append_hint(errp, "Unable to alloc vIOMMU: idev devid 0x%x: ",
- idev->devid);
+ if (!smmuv3_accel_alloc_viommu(s, hiodi, errp)) {
+ error_append_hint(errp, "Unable to alloc vIOMMU: hiodi devid 0x%x: ",
+ hiodi->devid);
return false;
}
done:
- accel_dev->idev = idev;
+ accel_dev->hiodi = hiodi;
accel_dev->s_accel = s->s_accel;
QLIST_INSERT_HEAD(&s->s_accel->device_list, accel_dev, next);
- trace_smmuv3_accel_set_iommu_device(devfn, idev->devid);
+ trace_smmuv3_accel_set_iommu_device(devfn, hiodi->devid);
return true;
}
@@ -646,7 +646,7 @@ static void smmuv3_accel_unset_iommu_device(PCIBus *bus, void *opaque,
{
SMMUState *bs = opaque;
SMMUPciBus *sbus = g_hash_table_lookup(bs->smmu_pcibus_by_busptr, bus);
- HostIOMMUDeviceIOMMUFD *idev;
+ HostIOMMUDeviceIOMMUFD *hiodi;
SMMUv3AccelDevice *accel_dev;
SMMUv3AccelState *accel;
IOMMUFDVdev *vdev;
@@ -662,16 +662,16 @@ static void smmuv3_accel_unset_iommu_device(PCIBus *bus, void *opaque,
}
accel_dev = container_of(sdev, SMMUv3AccelDevice, sdev);
- idev = accel_dev->idev;
+ hiodi = accel_dev->hiodi;
accel = accel_dev->s_accel;
/* Re-attach the default s2 hwpt id */
- if (!host_iommu_device_iommufd_attach_hwpt(idev, idev->hwpt_id, NULL)) {
- error_report("Unable to attach the default HW pagetable: idev devid "
- "0x%x", idev->devid);
+ if (!host_iommu_device_iommufd_attach_hwpt(hiodi, hiodi->hwpt_id, NULL)) {
+ error_report("Unable to attach the default HW pagetable: hiodi devid "
+ "0x%x", hiodi->devid);
}
if (accel_dev->s1_hwpt) {
- iommufd_backend_free_id(accel_dev->idev->iommufd,
+ iommufd_backend_free_id(accel_dev->hiodi->iommufd,
accel_dev->s1_hwpt->hwpt_id);
g_free(accel_dev->s1_hwpt);
accel_dev->s1_hwpt = NULL;
@@ -684,10 +684,10 @@ static void smmuv3_accel_unset_iommu_device(PCIBus *bus, void *opaque,
accel_dev->vdev = NULL;
}
- accel_dev->idev = NULL;
+ accel_dev->hiodi = NULL;
accel_dev->s_accel = NULL;
QLIST_REMOVE(accel_dev, next);
- trace_smmuv3_accel_unset_iommu_device(devfn, idev->devid);
+ trace_smmuv3_accel_unset_iommu_device(devfn, hiodi->devid);
if (QLIST_EMPTY(&accel->device_list)) {
smmuv3_accel_free_viommu(accel);
@@ -879,10 +879,11 @@ bool smmuv3_accel_attach_gbpa_hwpt(SMMUv3State *s, Error **errp)
hwpt_id = smmuv3_accel_gbpa_hwpt(s, accel);
QLIST_FOREACH(accel_dev, &accel->device_list, next) {
- if (!host_iommu_device_iommufd_attach_hwpt(accel_dev->idev, hwpt_id,
+ if (!host_iommu_device_iommufd_attach_hwpt(accel_dev->hiodi, hwpt_id,
&local_err)) {
error_append_hint(&local_err, "Failed to attach GBPA hwpt %u for "
- "idev devid %u", hwpt_id, accel_dev->idev->devid);
+ "hiodi devid %u", hwpt_id,
+ accel_dev->hiodi->devid);
error_report_err(local_err);
local_err = NULL;
all_ok = false;
diff --git a/hw/i386/intel_iommu_accel.c b/hw/i386/intel_iommu_accel.c
index 67d54849f2f08187629fca5e2c081f8dd003b6b2..ed3793602bc51bbce0f4428bfabf6ad1772f3d13 100644
--- a/hw/i386/intel_iommu_accel.c
+++ b/hw/i386/intel_iommu_accel.c
@@ -69,7 +69,7 @@ VTDHostIOMMUDevice *vtd_find_hiod_iommufd(VTDAddressSpace *as)
return NULL;
}
-static bool vtd_create_fs_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+static bool vtd_create_fs_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
VTDPASIDEntry *pe, uint32_t *fs_hwpt_id,
Error **errp)
{
@@ -81,27 +81,27 @@ static bool vtd_create_fs_hwpt(HostIOMMUDeviceIOMMUFD *idev,
vtd.addr_width = vtd_pe_get_fs_aw(pe);
vtd.pgtbl_addr = (uint64_t)vtd_pe_get_fspt_base(pe);
- return iommufd_backend_alloc_hwpt(idev->iommufd, idev->devid, idev->hwpt_id,
- 0, IOMMU_HWPT_DATA_VTD_S1, sizeof(vtd),
- &vtd, fs_hwpt_id, errp);
+ return iommufd_backend_alloc_hwpt(hiodi->iommufd, hiodi->devid,
+ hiodi->hwpt_id, 0, IOMMU_HWPT_DATA_VTD_S1,
+ sizeof(vtd), &vtd, fs_hwpt_id, errp);
}
-static void vtd_destroy_old_fs_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+static void vtd_destroy_old_fs_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
VTDAddressSpace *vtd_as)
{
if (!vtd_as->fs_hwpt_id) {
return;
}
- iommufd_backend_free_id(idev->iommufd, vtd_as->fs_hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, vtd_as->fs_hwpt_id);
vtd_as->fs_hwpt_id = 0;
}
static bool vtd_device_attach_iommufd(VTDHostIOMMUDevice *vtd_hiod,
VTDAddressSpace *vtd_as, Error **errp)
{
- HostIOMMUDeviceIOMMUFD *idev = HOST_IOMMU_DEVICE_IOMMUFD(vtd_hiod->hiod);
+ HostIOMMUDeviceIOMMUFD *hiodi = HOST_IOMMU_DEVICE_IOMMUFD(vtd_hiod->hiod);
VTDPASIDEntry *pe = &vtd_as->pasid_cache_entry.pasid_entry;
- uint32_t hwpt_id = idev->hwpt_id;
+ uint32_t hwpt_id = hiodi->hwpt_id;
bool ret;
/*
@@ -116,21 +116,21 @@ static bool vtd_device_attach_iommufd(VTDHostIOMMUDevice *vtd_hiod,
}
if (vtd_pe_pgtt_is_fst(pe)) {
- if (!vtd_create_fs_hwpt(idev, pe, &hwpt_id, errp)) {
+ if (!vtd_create_fs_hwpt(hiodi, pe, &hwpt_id, errp)) {
return false;
}
}
- ret = host_iommu_device_iommufd_attach_hwpt(idev, hwpt_id, errp);
- trace_vtd_device_attach_hwpt(idev->devid, vtd_as->pasid, hwpt_id, ret);
+ ret = host_iommu_device_iommufd_attach_hwpt(hiodi, hwpt_id, errp);
+ trace_vtd_device_attach_hwpt(hiodi->devid, vtd_as->pasid, hwpt_id, ret);
if (ret) {
/* Destroy old fs_hwpt if it's a replacement */
- vtd_destroy_old_fs_hwpt(idev, vtd_as);
+ vtd_destroy_old_fs_hwpt(hiodi, vtd_as);
if (vtd_pe_pgtt_is_fst(pe)) {
vtd_as->fs_hwpt_id = hwpt_id;
}
} else if (vtd_pe_pgtt_is_fst(pe)) {
- iommufd_backend_free_id(idev->iommufd, hwpt_id);
+ iommufd_backend_free_id(hiodi->iommufd, hwpt_id);
}
return ret;
@@ -139,27 +139,28 @@ static bool vtd_device_attach_iommufd(VTDHostIOMMUDevice *vtd_hiod,
static bool vtd_device_detach_iommufd(VTDHostIOMMUDevice *vtd_hiod,
VTDAddressSpace *vtd_as, Error **errp)
{
- HostIOMMUDeviceIOMMUFD *idev = HOST_IOMMU_DEVICE_IOMMUFD(vtd_hiod->hiod);
+ HostIOMMUDeviceIOMMUFD *hiodi = HOST_IOMMU_DEVICE_IOMMUFD(vtd_hiod->hiod);
IntelIOMMUState *s = vtd_as->iommu_state;
uint32_t pasid = vtd_as->pasid;
bool ret;
if (s->dmar_enabled && s->root_scalable) {
- ret = host_iommu_device_iommufd_detach_hwpt(idev, errp);
- trace_vtd_device_detach_hwpt(idev->devid, pasid, ret);
+ ret = host_iommu_device_iommufd_detach_hwpt(hiodi, errp);
+ trace_vtd_device_detach_hwpt(hiodi->devid, pasid, ret);
} else {
/*
* If DMAR remapping is disabled or guest switches to legacy mode,
* we fallback to the default HWPT which contains shadow page table.
* So guest DMA could still work.
*/
- ret = host_iommu_device_iommufd_attach_hwpt(idev, idev->hwpt_id, errp);
- trace_vtd_device_reattach_def_hwpt(idev->devid, pasid, idev->hwpt_id,
+ ret = host_iommu_device_iommufd_attach_hwpt(hiodi, hiodi->hwpt_id,
+ errp);
+ trace_vtd_device_reattach_def_hwpt(hiodi->devid, pasid, hiodi->hwpt_id,
ret);
}
if (ret) {
- vtd_destroy_old_fs_hwpt(idev, vtd_as);
+ vtd_destroy_old_fs_hwpt(hiodi, vtd_as);
}
return ret;
@@ -211,13 +212,14 @@ static void vtd_flush_host_piotlb_locked(gpointer key, gpointer value,
did = VTD_SM_PASID_ENTRY_DID(&pc_entry->pasid_entry);
if (piotlb_info->domain_id == did && piotlb_info->pasid == PASID_0) {
- HostIOMMUDeviceIOMMUFD *idev =
+ HostIOMMUDeviceIOMMUFD *hiodi =
HOST_IOMMU_DEVICE_IOMMUFD(vtd_hiod->hiod);
uint32_t entry_num = 1; /* Only implement one request for simplicity */
Error *local_err = NULL;
struct iommu_hwpt_vtd_s1_invalidate *cache = piotlb_info->inv_data;
- if (!iommufd_backend_invalidate_cache(idev->iommufd, vtd_as->fs_hwpt_id,
+ if (!iommufd_backend_invalidate_cache(hiodi->iommufd,
+ vtd_as->fs_hwpt_id,
IOMMU_HWPT_INVALIDATE_DATA_VTD_S1,
sizeof(*cache), &entry_num, cache,
&local_err)) {
diff --git a/hw/vfio/container-legacy.c b/hw/vfio/container-legacy.c
index 625f151364c060135076cd19537810bbe61ab998..d301b27aa652964d4f5f2cb5bd22de69048eb337 100644
--- a/hw/vfio/container-legacy.c
+++ b/hw/vfio/container-legacy.c
@@ -1244,12 +1244,12 @@ static void vfio_iommu_legacy_instance_init(Object *obj)
static void hiod_legacy_vfio_class_init(ObjectClass *oc, const void *data)
{
- HostIOMMUDeviceClass *hioc = HOST_IOMMU_DEVICE_CLASS(oc);
+ HostIOMMUDeviceClass *hiodc = HOST_IOMMU_DEVICE_CLASS(oc);
- hioc->realize = hiod_legacy_vfio_realize;
- hioc->get_cap = hiod_legacy_vfio_get_cap;
- hioc->get_iova_ranges = hiod_legacy_vfio_get_iova_ranges;
- hioc->get_page_size_mask = hiod_legacy_vfio_get_page_size_mask;
+ hiodc->realize = hiod_legacy_vfio_realize;
+ hiodc->get_cap = hiod_legacy_vfio_get_cap;
+ hiodc->get_iova_ranges = hiod_legacy_vfio_get_iova_ranges;
+ hiodc->get_page_size_mask = hiod_legacy_vfio_get_page_size_mask;
};
static const TypeInfo types[] = {
diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c
index 3e33dfbb35652376d1e285485221f27fcea84d16..399b36aa752ebe24d5ea7eb4d0a991785c51e642 100644
--- a/hw/vfio/iommufd.c
+++ b/hw/vfio/iommufd.c
@@ -917,19 +917,19 @@ static void vfio_iommu_iommufd_class_init(ObjectClass *klass, const void *data)
};
static bool
-host_iommu_device_iommufd_vfio_attach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+host_iommu_device_iommufd_vfio_attach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
uint32_t hwpt_id, Error **errp)
{
- VFIODevice *vbasedev = HOST_IOMMU_DEVICE(idev)->agent;
+ VFIODevice *vbasedev = HOST_IOMMU_DEVICE(hiodi)->agent;
return !iommufd_cdev_attach_ioas_hwpt(vbasedev, hwpt_id, errp);
}
static bool
-host_iommu_device_iommufd_vfio_detach_hwpt(HostIOMMUDeviceIOMMUFD *idev,
+host_iommu_device_iommufd_vfio_detach_hwpt(HostIOMMUDeviceIOMMUFD *hiodi,
Error **errp)
{
- VFIODevice *vbasedev = HOST_IOMMU_DEVICE(idev)->agent;
+ VFIODevice *vbasedev = HOST_IOMMU_DEVICE(hiodi)->agent;
return iommufd_cdev_detach_ioas_hwpt(vbasedev, errp);
}
@@ -938,7 +938,7 @@ static bool hiod_iommufd_vfio_realize(HostIOMMUDevice *hiod, void *opaque,
Error **errp)
{
VFIODevice *vdev = opaque;
- HostIOMMUDeviceIOMMUFD *idev;
+ HostIOMMUDeviceIOMMUFD *hiodi;
HostIOMMUDeviceCaps *caps = &hiod->caps;
VendorCaps *vendor_caps = &caps->vendor_caps;
enum iommu_hw_info_type type;
@@ -958,10 +958,10 @@ static bool hiod_iommufd_vfio_realize(HostIOMMUDevice *hiod, void *opaque,
caps->hw_caps = hw_caps;
caps->max_pasid_log2 = max_pasid_log2;
- idev = HOST_IOMMU_DEVICE_IOMMUFD(hiod);
- idev->iommufd = vdev->iommufd;
- idev->devid = vdev->devid;
- idev->hwpt_id = vdev->hwpt->hwpt_id;
+ hiodi = HOST_IOMMU_DEVICE_IOMMUFD(hiod);
+ hiodi->iommufd = vdev->iommufd;
+ hiodi->devid = vdev->devid;
+ hiodi->hwpt_id = vdev->hwpt->hwpt_id;
return true;
}
@@ -988,14 +988,14 @@ hiod_iommufd_vfio_get_page_size_mask(HostIOMMUDevice *hiod)
static void hiod_iommufd_vfio_class_init(ObjectClass *oc, const void *data)
{
HostIOMMUDeviceClass *hiodc = HOST_IOMMU_DEVICE_CLASS(oc);
- HostIOMMUDeviceIOMMUFDClass *idevc = HOST_IOMMU_DEVICE_IOMMUFD_CLASS(oc);
+ HostIOMMUDeviceIOMMUFDClass *hiodic = HOST_IOMMU_DEVICE_IOMMUFD_CLASS(oc);
hiodc->realize = hiod_iommufd_vfio_realize;
hiodc->get_iova_ranges = hiod_iommufd_vfio_get_iova_ranges;
hiodc->get_page_size_mask = hiod_iommufd_vfio_get_page_size_mask;
- idevc->attach_hwpt = host_iommu_device_iommufd_vfio_attach_hwpt;
- idevc->detach_hwpt = host_iommu_device_iommufd_vfio_detach_hwpt;
+ hiodic->attach_hwpt = host_iommu_device_iommufd_vfio_attach_hwpt;
+ hiodic->detach_hwpt = host_iommu_device_iommufd_vfio_detach_hwpt;
};
static const TypeInfo types[] = {
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 10/14] hw/vfio/iommufd: Control dirty tracking for nesting parent HWPT
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (8 preceding siblings ...)
2026-04-21 16:34 ` [PULL 09/14] iommufd: Rename all the idev and idevc variables to hiod and hiodc Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 11/14] vfio: Add VFIO I/O backend capability flags for feature support Cédric Le Goater
` (4 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, Shameer Kolothum, Zhenzhong Duan,
Cédric Le Goater
From: Shameer Kolothum <skolothumtho@nvidia.com>
QEMU smmuv3 accel does not support live migration yet, so dirty
tracking for the nesting parent HWPT is not useful.
Also, nested vIOMMU use cases can break on some platforms. For
example, SMMUv3 with HTTU may advertise dirty tracking capability,
but the kernel supports it only for stage-1. Requesting dirty
tracking for a nesting parent HWPT (stage-2) can fail.
Add a vIOMMU flag to explicitly request dirty tracking for the
nesting parent HWPT. For nested cases, dirty tracking is enabled
only when requested by the vIOMMU.
Non-nested cases and Intel vIOMMU keep the existing behavior.
Fixes: fc6dafb98cec ("hw/arm/smmuv3: Implement get_viommu_cap() callback")
Signed-off-by: Shameer Kolothum <skolothumtho@nvidia.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Link: https://lore.kernel.org/qemu-devel/20260401084133.56266-1-skolothumtho@nvidia.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/hw/core/iommu.h | 2 ++
include/hw/vfio/vfio-device.h | 1 +
hw/i386/intel_iommu.c | 7 +++++--
hw/vfio/device.c | 11 +++++++++++
hw/vfio/iommufd.c | 11 +++++++++--
5 files changed, 28 insertions(+), 4 deletions(-)
diff --git a/include/hw/core/iommu.h b/include/hw/core/iommu.h
index 86af315c153763d4d217a2e255c3f7aced228e91..cd59a367ceb9517885a5a8c14a40af3404e99832 100644
--- a/include/hw/core/iommu.h
+++ b/include/hw/core/iommu.h
@@ -21,6 +21,8 @@ enum viommu_flags {
/* vIOMMU needs nesting parent HWPT to create nested HWPT */
VIOMMU_FLAG_WANT_NESTING_PARENT = BIT_ULL(0),
VIOMMU_FLAG_PASID_SUPPORTED = BIT_ULL(1),
+ /* vIOMMU needs dirty tracking on the nesting parent HWPT for nested use */
+ VIOMMU_FLAG_WANT_NESTING_DIRTY_TRACKING = BIT_ULL(2),
};
/* Host IOMMU quirks. Extracted from host IOMMU capabilities */
diff --git a/include/hw/vfio/vfio-device.h b/include/hw/vfio/vfio-device.h
index 828a31c006ec9fc372760fee023da31b0e7acc4b..a95c5bf50302a7394582dcc493d96aef00b47dc8 100644
--- a/include/hw/vfio/vfio-device.h
+++ b/include/hw/vfio/vfio-device.h
@@ -268,6 +268,7 @@ void vfio_device_prepare(VFIODevice *vbasedev, VFIOContainer *bcontainer,
void vfio_device_unprepare(VFIODevice *vbasedev);
bool vfio_device_get_viommu_flags_want_nesting(VFIODevice *vbasedev);
+bool vfio_device_get_viommu_flags_want_nesting_dirty(VFIODevice *vbasedev);
bool vfio_device_get_host_iommu_quirk_bypass_ro(VFIODevice *vbasedev,
uint32_t type, void *caps,
uint32_t size);
diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
index f395fa248c009c663be281774c8455ee3b85d8a2..b784c5f10aebbd3fe1f1c332c4ecc5aecc2b9136 100644
--- a/hw/i386/intel_iommu.c
+++ b/hw/i386/intel_iommu.c
@@ -4825,9 +4825,12 @@ static void vtd_dev_unset_iommu_device(PCIBus *bus, void *opaque, int devfn)
static uint64_t vtd_get_viommu_flags(void *opaque)
{
IntelIOMMUState *s = opaque;
- uint64_t flags;
+ uint64_t flags = 0;
- flags = s->fsts ? VIOMMU_FLAG_WANT_NESTING_PARENT : 0;
+ if (s->fsts) {
+ flags = VIOMMU_FLAG_WANT_NESTING_PARENT |
+ VIOMMU_FLAG_WANT_NESTING_DIRTY_TRACKING;
+ }
return flags;
}
diff --git a/hw/vfio/device.c b/hw/vfio/device.c
index 973fc35b59d4bdaa34c0f2773ec78df0f151dfa2..8f7ae919a55dd9e08bdd072eee33f5b46a91aa2a 100644
--- a/hw/vfio/device.c
+++ b/hw/vfio/device.c
@@ -522,6 +522,17 @@ void vfio_device_unprepare(VFIODevice *vbasedev)
vbasedev->bcontainer = NULL;
}
+bool vfio_device_get_viommu_flags_want_nesting_dirty(VFIODevice *vbasedev)
+{
+ VFIOPCIDevice *vdev = vfio_pci_from_vfio_device(vbasedev);
+
+ if (vdev) {
+ return !!(pci_device_get_viommu_flags(PCI_DEVICE(vdev)) &
+ VIOMMU_FLAG_WANT_NESTING_DIRTY_TRACKING);
+ }
+ return false;
+}
+
bool vfio_device_get_viommu_flags_want_nesting(VFIODevice *vbasedev)
{
VFIOPCIDevice *vdev = vfio_pci_from_vfio_device(vbasedev);
diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c
index 399b36aa752ebe24d5ea7eb4d0a991785c51e642..df148a49a762b1f6b05d548cb9f9b0413b743439 100644
--- a/hw/vfio/iommufd.c
+++ b/hw/vfio/iommufd.c
@@ -352,6 +352,7 @@ static bool iommufd_cdev_autodomains_get(VFIODevice *vbasedev,
ERRP_GUARD();
IOMMUFDBackend *iommufd = vbasedev->iommufd;
VFIOContainer *bcontainer = VFIO_IOMMU(container);
+ bool viommu_nesting, viommu_nesting_dirty;
uint32_t type, flags = 0;
uint64_t hw_caps;
VendorCaps caps;
@@ -405,8 +406,14 @@ static bool iommufd_cdev_autodomains_get(VFIODevice *vbasedev,
return false;
}
+ viommu_nesting = vfio_device_get_viommu_flags_want_nesting(vbasedev);
+ viommu_nesting_dirty =
+ vfio_device_get_viommu_flags_want_nesting_dirty(vbasedev);
+
if (hw_caps & IOMMU_HW_CAP_DIRTY_TRACKING) {
- flags = IOMMU_HWPT_ALLOC_DIRTY_TRACKING;
+ if (!viommu_nesting || viommu_nesting_dirty) {
+ flags |= IOMMU_HWPT_ALLOC_DIRTY_TRACKING;
+ }
}
/*
@@ -414,7 +421,7 @@ static bool iommufd_cdev_autodomains_get(VFIODevice *vbasedev,
* force to create it so that it could be reused by vIOMMU to create
* nested HWPT.
*/
- if (vfio_device_get_viommu_flags_want_nesting(vbasedev)) {
+ if (viommu_nesting) {
flags |= IOMMU_HWPT_ALLOC_NEST_PARENT;
if (vfio_device_get_host_iommu_quirk_bypass_ro(vbasedev, type,
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 11/14] vfio: Add VFIO I/O backend capability flags for feature support
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (9 preceding siblings ...)
2026-04-21 16:34 ` [PULL 10/14] hw/vfio/iommufd: Control dirty tracking for nesting parent HWPT Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 12/14] vfio-user: support VFIO_USER_DEVICE_FEATURE Cédric Le Goater
` (3 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel; +Cc: Alex Williamson, Cédric Le Goater, John Levon
Different VFIO I/O backends support different features. For example,
the kernel VFIO backend supports DMA-BUF creation, while vfio-user
does not. Currently, this is handled by attempting the operation and
checking for -ENOTTY, which can lead to misleading warnings when a
feature is simply not supported by a particular backend.
Introduce a capability flags mechanism in VFIODeviceIOOps that allows
backends to explicitly advertise which features they support. Callers
can check these capabilities before attempting operations, avoiding
spurious errors and warnings.
Cc: John Levon <john.levon@nutanix.com>
Reviewed-by: John Levon <john.levon@nutanix.com>
Link: https://lore.kernel.org/qemu-devel/20260409114312.1704062-1-clg@redhat.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/hw/vfio/vfio-device.h | 13 +++++++++++++
hw/vfio/device.c | 2 ++
hw/vfio/region.c | 5 +++++
3 files changed, 20 insertions(+)
diff --git a/include/hw/vfio/vfio-device.h b/include/hw/vfio/vfio-device.h
index a95c5bf50302a7394582dcc493d96aef00b47dc8..380a55d6e5ea280c8f44016c7a8a1dcb8f4c00fe 100644
--- a/include/hw/vfio/vfio-device.h
+++ b/include/hw/vfio/vfio-device.h
@@ -172,12 +172,25 @@ typedef QLIST_HEAD(VFIODeviceList, VFIODevice) VFIODeviceList;
extern VFIODeviceList vfio_device_list;
#ifdef CONFIG_LINUX
+/*
+ * VFIO backend I/O operation capabilities
+ */
+#define VFIO_IO_CAP_DMA_BUF (1ULL << 0)
+
/*
* How devices communicate with the server. The default option is through
* ioctl() to the kernel VFIO driver, but vfio-user can use a socket to a remote
* process.
*/
struct VFIODeviceIOOps {
+ /**
+ * @capabilities
+ *
+ * Bitmask of VFIO_IO_CAP_* flags indicating which features this
+ * backend supports.
+ */
+ uint64_t capabilities;
+
/**
* @device_feature
*
diff --git a/hw/vfio/device.c b/hw/vfio/device.c
index 8f7ae919a55dd9e08bdd072eee33f5b46a91aa2a..3ffd69a579b8ac135ef729d3d836c8545e36786b 100644
--- a/hw/vfio/device.c
+++ b/hw/vfio/device.c
@@ -648,6 +648,8 @@ static int vfio_device_io_region_write(VFIODevice *vbasedev, uint8_t index,
}
static VFIODeviceIOOps vfio_device_io_ops_ioctl = {
+ .capabilities = VFIO_IO_CAP_DMA_BUF,
+
.device_feature = vfio_device_io_device_feature,
.get_region_info = vfio_device_io_get_region_info,
.get_irq_info = vfio_device_io_get_irq_info,
diff --git a/hw/vfio/region.c b/hw/vfio/region.c
index 47fdc2df349b65c6be6c9605b7a38a4e367f0475..0342ca712acd0ad846685115c29432c15aa966a0 100644
--- a/hw/vfio/region.c
+++ b/hw/vfio/region.c
@@ -293,6 +293,11 @@ static bool vfio_region_create_dma_buf(VFIORegion *region, Error **errp)
size_t total_size;
int i, ret;
+ /* Check if backend supports DMA-BUF creation */
+ if (!(vbasedev->io_ops->capabilities & VFIO_IO_CAP_DMA_BUF)) {
+ return true;
+ }
+
total_size = sizeof(*feature) + sizeof(*dma_buf) +
sizeof(struct vfio_region_dma_range) * region->nr_mmaps;
feature = g_malloc0(total_size);
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 12/14] vfio-user: support VFIO_USER_DEVICE_FEATURE
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (10 preceding siblings ...)
2026-04-21 16:34 ` [PULL 11/14] vfio: Add VFIO I/O backend capability flags for feature support Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 13/14] vfio-user: correct protocol for DMA reads/writes Cédric Le Goater
` (2 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel; +Cc: Alex Williamson, John Levon, Cédric Le Goater
From: John Levon <john.levon@nutanix.com>
Plumb through vfio_device_get_feature to the vfio-user server. Note that
we translate EINVAL into ENOTTY, as the existing generic vfio code is
expecting the latter to mean "unsupported".
As part of adding a trace point, clean up the trace file.
Signed-off-by: John Levon <john.levon@nutanix.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260410085716.877185-2-john.levon@nutanix.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/vfio-user/protocol.h | 12 +++++++++++
hw/vfio-user/device.c | 42 +++++++++++++++++++++++++++++++++++++++
hw/vfio-user/trace-events | 23 +++++++++++----------
3 files changed, 67 insertions(+), 10 deletions(-)
diff --git a/hw/vfio-user/protocol.h b/hw/vfio-user/protocol.h
index 3249a4a6b62af95e3718ed5053bbf310cd08c024..2a0c31e7c57f8ed86681ebd96b20df3e32792dec 100644
--- a/hw/vfio-user/protocol.h
+++ b/hw/vfio-user/protocol.h
@@ -40,6 +40,7 @@ enum vfio_user_command {
VFIO_USER_DEVICE_RESET = 13,
VFIO_USER_DIRTY_PAGES = 14,
VFIO_USER_REGION_WRITE_MULTI = 15,
+ VFIO_USER_DEVICE_FEATURE = 16,
VFIO_USER_MAX,
};
@@ -239,4 +240,15 @@ typedef struct {
VFIOUserWROne wrs[VFIO_USER_MULTI_MAX];
} VFIOUserWRMulti;
+/*
+ * VFIO_USER_DEVICE_FEATURE
+ * imported from struct vfio_device_feature
+ */
+typedef struct {
+ VFIOUserHdr hdr;
+ uint32_t argsz;
+ uint32_t flags;
+ char data[];
+} VFIOUserDeviceFeature;
+
#endif /* VFIO_USER_PROTOCOL_H */
diff --git a/hw/vfio-user/device.c b/hw/vfio-user/device.c
index 64ef35b3209429f7158d4ea79f095a5f16950d77..b8d2b7c1a84bb1914a71fdccbbdff675373cef3b 100644
--- a/hw/vfio-user/device.c
+++ b/hw/vfio-user/device.c
@@ -74,6 +74,47 @@ void vfio_user_device_reset(VFIOUserProxy *proxy)
}
}
+static int
+vfio_user_device_io_device_feature(VFIODevice *vbasedev,
+ struct vfio_device_feature *feature)
+{
+ g_autofree VFIOUserDeviceFeature *msgp = NULL;
+ int size = sizeof(VFIOUserHdr) + feature->argsz;
+ VFIOUserProxy *proxy = vbasedev->proxy;
+ Error *local_err = NULL;
+
+ msgp = g_malloc0(size);
+
+ vfio_user_request_msg(&msgp->hdr, VFIO_USER_DEVICE_FEATURE, size, 0);
+
+ memcpy(&msgp->argsz, &feature->argsz, feature->argsz);
+
+ if (!vfio_user_send_wait(proxy, &msgp->hdr, NULL, size, &local_err)) {
+ error_prepend(&local_err, "%s: ", __func__);
+ error_report_err(local_err);
+ return -EFAULT;
+ }
+
+ if (msgp->hdr.flags & VFIO_USER_ERROR) {
+ /*
+ * Client expects ENOTTY for "not supported", but the protocol may
+ * return EINVAL (which should only occur in the case the feature isn't
+ * actually supported on the server).
+ */
+ if (msgp->hdr.error_reply == EINVAL) {
+ return -ENOTTY;
+ }
+
+ return -msgp->hdr.error_reply;
+ }
+
+ memcpy(feature, &msgp->argsz, feature->argsz);
+
+ trace_vfio_user_device_io_device_feature(msgp->argsz, msgp->flags);
+
+ return 0;
+}
+
static int vfio_user_get_region_info(VFIOUserProxy *proxy,
struct vfio_region_info *info,
VFIOUserFDs *fds)
@@ -432,6 +473,7 @@ static int vfio_user_device_io_region_write(VFIODevice *vbasedev, uint8_t index,
* Socket-based io_ops
*/
VFIODeviceIOOps vfio_user_device_io_ops_sock = {
+ .device_feature = vfio_user_device_io_device_feature,
.get_region_info = vfio_user_device_io_get_region_info,
.get_irq_info = vfio_user_device_io_get_irq_info,
.set_irqs = vfio_user_device_io_set_irqs,
diff --git a/hw/vfio-user/trace-events b/hw/vfio-user/trace-events
index abb67f4c119af533a77ad04e383fa027e5aaa684..503e1c82d43a7d9361b1e5d487929ecc5a8a83da 100644
--- a/hw/vfio-user/trace-events
+++ b/hw/vfio-user/trace-events
@@ -2,19 +2,22 @@
#
# SPDX-License-Identifier: GPL-2.0-or-later
-# common.c
+# container.c
+vfio_user_dma_map(uint64_t iova, uint64_t size, uint64_t off, uint32_t flags, bool async_ops) " iova 0x%"PRIx64" size 0x%"PRIx64" off 0x%"PRIx64" flags 0x%x async_ops %d"
+vfio_user_dma_unmap(uint64_t iova, uint64_t size, uint32_t flags, bool async_ops) " iova 0x%"PRIx64" size 0x%"PRIx64" flags 0x%x async_ops %d"
+
+# device.c
+vfio_user_device_io_device_feature(uint32_t argsz, uint32_t flags) " argsz 0x%x flags 0x%x"
+vfio_user_get_info(uint32_t nregions, uint32_t nirqs) " #regions %d #irqs %d"
+vfio_user_get_irq_info(uint32_t index, uint32_t flags, uint32_t count) " index %d flags 0x%x count %d"
+vfio_user_set_irqs(uint32_t index, uint32_t start, uint32_t count, uint32_t flags) " index %d start %d count %d flags 0x%x"
+vfio_user_get_region_info(uint32_t index, uint32_t flags, uint64_t size) " index %d flags 0x%x size 0x%"PRIx64
+vfio_user_region_rw(uint32_t region, uint64_t off, uint32_t count) " region %d offset 0x%"PRIx64" count %d"
+
+# proxy.c
vfio_user_recv_hdr(const char *name, uint16_t id, uint16_t cmd, uint32_t size, uint32_t flags) " (%s) id 0x%x cmd 0x%x size 0x%x flags 0x%x"
vfio_user_recv_read(uint16_t id, int read) " id 0x%x read 0x%x"
vfio_user_recv_request(uint16_t cmd) " command 0x%x"
vfio_user_send_write(uint16_t id, int wrote) " id 0x%x wrote 0x%x"
vfio_user_version(uint16_t major, uint16_t minor, const char *caps) " major %d minor %d caps: %s"
-vfio_user_get_info(uint32_t nregions, uint32_t nirqs) " #regions %d #irqs %d"
-vfio_user_get_region_info(uint32_t index, uint32_t flags, uint64_t size) " index %d flags 0x%x size 0x%"PRIx64
-vfio_user_region_rw(uint32_t region, uint64_t off, uint32_t count) " region %d offset 0x%"PRIx64" count %d"
-vfio_user_get_irq_info(uint32_t index, uint32_t flags, uint32_t count) " index %d flags 0x%x count %d"
-vfio_user_set_irqs(uint32_t index, uint32_t start, uint32_t count, uint32_t flags) " index %d start %d count %d flags 0x%x"
vfio_user_wrmulti(const char *s, uint64_t wr_cnt) " %s count 0x%"PRIx64
-
-# container.c
-vfio_user_dma_map(uint64_t iova, uint64_t size, uint64_t off, uint32_t flags, bool async_ops) " iova 0x%"PRIx64" size 0x%"PRIx64" off 0x%"PRIx64" flags 0x%x async_ops %d"
-vfio_user_dma_unmap(uint64_t iova, uint64_t size, uint32_t flags, bool async_ops) " iova 0x%"PRIx64" size 0x%"PRIx64" flags 0x%x async_ops %d"
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 13/14] vfio-user: correct protocol for DMA reads/writes
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (11 preceding siblings ...)
2026-04-21 16:34 ` [PULL 12/14] vfio-user: support VFIO_USER_DEVICE_FEATURE Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-21 16:34 ` [PULL 14/14] vfio-user: fix DMA write reply Cédric Le Goater
2026-04-23 1:03 ` [PULL 00/14] vfio queue Stefan Hajnoczi
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, John Levon, Patrick Mooney,
Cédric Le Goater
From: John Levon <john.levon@nutanix.com>
The "count" parameter in the DMA write reply should be 8 bytes in size.
Reported-by: Patrick Mooney <patrick@matx.com>
Signed-off-by: John Levon <john.levon@nutanix.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260410085716.877185-3-john.levon@nutanix.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
docs/interop/vfio-user.rst | 2 +-
hw/vfio-user/protocol.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/docs/interop/vfio-user.rst b/docs/interop/vfio-user.rst
index 303fc13d6ec4c32c32bd95604c75239342d85062..d4766487ea6d0ee021316890d443814957cf6284 100644
--- a/docs/interop/vfio-user.rst
+++ b/docs/interop/vfio-user.rst
@@ -1463,7 +1463,7 @@ Reply
+=========+========+==========+
| address | 0 | 8 |
+---------+--------+----------+
-| count | 8 | 4 |
+| count | 8 | 8 |
+---------+--------+----------+
* *address* is the client DMA memory address being accessed.
diff --git a/hw/vfio-user/protocol.h b/hw/vfio-user/protocol.h
index 2a0c31e7c57f8ed86681ebd96b20df3e32792dec..c478d1353f9d618ef8c202cd28cd638aeed3bd57 100644
--- a/hw/vfio-user/protocol.h
+++ b/hw/vfio-user/protocol.h
@@ -210,7 +210,7 @@ typedef struct {
typedef struct {
VFIOUserHdr hdr;
uint64_t offset;
- uint32_t count;
+ uint64_t count;
char data[];
} VFIOUserDMARW;
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* [PULL 14/14] vfio-user: fix DMA write reply
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (12 preceding siblings ...)
2026-04-21 16:34 ` [PULL 13/14] vfio-user: correct protocol for DMA reads/writes Cédric Le Goater
@ 2026-04-21 16:34 ` Cédric Le Goater
2026-04-23 1:03 ` [PULL 00/14] vfio queue Stefan Hajnoczi
14 siblings, 0 replies; 16+ messages in thread
From: Cédric Le Goater @ 2026-04-21 16:34 UTC (permalink / raw)
To: qemu-devel
Cc: Alex Williamson, John Levon, Patrick Mooney,
Cédric Le Goater
From: John Levon <john.levon@nutanix.com>
The protocol specifies that DMA write replies should include
address+count, but the client code was only doing so for read. Fix that
up.
In addition, add a protocol clarification over how short writes may be
reported in that reply. QEMU never reports a short write via the
count field.
Reported-by: Patrick Mooney <patrick@matx.com>
Signed-off-by: John Levon <john.levon@nutanix.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260410085716.877185-4-john.levon@nutanix.com
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
docs/interop/vfio-user.rst | 6 ++++++
hw/vfio-user/pci.c | 30 ++++++++++++++++++++++--------
2 files changed, 28 insertions(+), 8 deletions(-)
diff --git a/docs/interop/vfio-user.rst b/docs/interop/vfio-user.rst
index d4766487ea6d0ee021316890d443814957cf6284..12deb25102af217f7f292b4cb655dc4a6a92439a 100644
--- a/docs/interop/vfio-user.rst
+++ b/docs/interop/vfio-user.rst
@@ -1429,6 +1429,9 @@ Reply
* *count* is the size of the data transferred.
* *data* is the data read.
+Note that whether short reads return an error or just set count appropriately is
+a client-side choice; servers should be prepared to handle both cases.
+
``VFIO_USER_DMA_WRITE``
-----------------------
@@ -1469,6 +1472,9 @@ Reply
* *address* is the client DMA memory address being accessed.
* *count* is the size of the data transferred.
+Note that whether short writes return an error or just set count appropriately
+is a client-side choice; servers should be prepared to handle both cases.
+
``VFIO_USER_DEVICE_RESET``
--------------------------
diff --git a/hw/vfio-user/pci.c b/hw/vfio-user/pci.c
index 64b8b3cb8cf5d66879af1703c4f2b5ae0c7418ca..facc79727acb82c35fec5b4bef79e78b3878531a 100644
--- a/hw/vfio-user/pci.c
+++ b/hw/vfio-user/pci.c
@@ -109,6 +109,10 @@ static void vfio_user_dma_read(VFIOPCIDevice *vdev, VFIOUserDMARW *msg)
r = pci_dma_read(pdev, res->offset, &res->data, res->count);
+ /*
+ * pci_dma_read() doesn't support reporting short reads via the reply's
+ * count parameter; in this case, we'll reply with an error instead.
+ */
switch (r) {
case MEMTX_OK:
if (res->hdr.flags & VFIO_USER_NO_REPLY) {
@@ -136,6 +140,7 @@ static void vfio_user_dma_write(VFIOPCIDevice *vdev, VFIOUserDMARW *msg)
{
PCIDevice *pdev = PCI_DEVICE(vdev);
VFIOUserProxy *proxy = vdev->vbasedev.proxy;
+ VFIOUserDMARW *res;
MemTxResult r;
if (msg->hdr.size < sizeof(*msg)) {
@@ -150,26 +155,35 @@ static void vfio_user_dma_write(VFIOPCIDevice *vdev, VFIOUserDMARW *msg)
r = pci_dma_write(pdev, msg->offset, &msg->data, msg->count);
+ res = g_malloc0(sizeof(*res));
+ memcpy(res, msg, sizeof(*res));
+ g_free(msg);
+
+ /*
+ * pci_dma_write() doesn't support reporting short writes via the reply's
+ * count parameter; in this case, we'll reply with an error instead.
+ */
switch (r) {
case MEMTX_OK:
- if ((msg->hdr.flags & VFIO_USER_NO_REPLY) == 0) {
- vfio_user_send_reply(proxy, &msg->hdr, sizeof(msg->hdr));
- } else {
- g_free(msg);
+ if (res->hdr.flags & VFIO_USER_NO_REPLY) {
+ g_free(res);
+ return;
}
+
+ vfio_user_send_reply(proxy, &res->hdr, sizeof(*res));
break;
case MEMTX_ERROR:
- vfio_user_send_error(proxy, &msg->hdr, EFAULT);
+ vfio_user_send_error(proxy, &res->hdr, EFAULT);
break;
case MEMTX_DECODE_ERROR:
- vfio_user_send_error(proxy, &msg->hdr, ENODEV);
+ vfio_user_send_error(proxy, &res->hdr, ENODEV);
break;
case MEMTX_ACCESS_ERROR:
- vfio_user_send_error(proxy, &msg->hdr, EPERM);
+ vfio_user_send_error(proxy, &res->hdr, EPERM);
break;
default:
error_printf("vfio_user_dma_write unknown error %d\n", r);
- vfio_user_send_error(vdev->vbasedev.proxy, &msg->hdr, EINVAL);
+ vfio_user_send_error(vdev->vbasedev.proxy, &res->hdr, EINVAL);
}
}
--
2.53.0
^ permalink raw reply related [flat|nested] 16+ messages in thread* Re: [PULL 00/14] vfio queue
2026-04-21 16:34 [PULL 00/14] vfio queue Cédric Le Goater
` (13 preceding siblings ...)
2026-04-21 16:34 ` [PULL 14/14] vfio-user: fix DMA write reply Cédric Le Goater
@ 2026-04-23 1:03 ` Stefan Hajnoczi
14 siblings, 0 replies; 16+ messages in thread
From: Stefan Hajnoczi @ 2026-04-23 1:03 UTC (permalink / raw)
To: Cédric Le Goater; +Cc: qemu-devel, Alex Williamson, Cédric Le Goater
[-- Attachment #1: Type: text/plain, Size: 17 bytes --]
Applied, thanks.
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply [flat|nested] 16+ messages in thread