In addition to memfd, a blob resource can also have its backing
storage in a VFIO device region. Since, there is no effective way
to determine where the backing storage is located, we first try to
create a dmabuf assuming it is in memfd. If that fails, we try to
create a dmabuf assuming it is in VFIO device region.
So, we first call virtio_gpu_create_udmabuf() to check if the blob's
backing storage is located in a memfd or not. If it is not, we invoke
the vfio_device_create_dmabuf_fd() API which identifies the right
VFIO device and eventually creates a dmabuf fd.
Note that in virtio_gpu_remap_dmabuf(), we first try to test if
the VFIO dmabuf exporter supports mmap or not. If it doesn't, we
fall back to the vfio_device_mmap_dmabuf() API to get a mapping
created for the dmabuf.
Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
Cc: Alex Bennée <alex.bennee@linaro.org>
Cc: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Cc: Alex Williamson <alex@shazbot.org>
Cc: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
---
hw/display/virtio-gpu-dmabuf.c | 35 +++++++++++++++++++++++++++-------
1 file changed, 28 insertions(+), 7 deletions(-)
diff --git a/hw/display/virtio-gpu-dmabuf.c b/hw/display/virtio-gpu-dmabuf.c
index b2badfc1bd..533d3d601c 100644
--- a/hw/display/virtio-gpu-dmabuf.c
+++ b/hw/display/virtio-gpu-dmabuf.c
@@ -74,13 +74,21 @@ static int virtio_gpu_create_udmabuf(struct virtio_gpu_simple_resource *res,
static bool virtio_gpu_remap_dmabuf(struct virtio_gpu_simple_resource *res,
Error **errp)
{
+ Error *mmap_err = NULL;
void *map;
map = mmap(NULL, res->blob_size, PROT_READ, MAP_SHARED, res->dmabuf_fd, 0);
if (map == MAP_FAILED) {
- error_setg_errno(errp, errno, "dmabuf mmap failed");
+ error_setg_errno(&mmap_err, errno, "dmabuf mmap failed: ");
res->remapped = NULL;
- return false;
+
+ if (!vfio_device_mmap_dmabuf(res->iov, res->iov_cnt, &map,
+ res->blob_size, errp)) {
+ error_report_err(mmap_err);
+ return false;
+ }
+ error_free(mmap_err);
+ mmap_err = NULL;
}
res->remapped = map;
return true;
@@ -137,7 +145,7 @@ bool virtio_gpu_have_udmabuf(void)
void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
{
- Error *local_err = NULL;
+ Error *local_err = NULL, *vfio_err = NULL;
void *pdata = NULL;
int ret;
@@ -151,15 +159,28 @@ void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
error_prepend(&local_err, "Cannot create dmabuf: ");
if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
- qemu_log_mask(LOG_GUEST_ERROR,
- "Cannot create dmabuf: incompatible memory\n");
- error_free(local_err);
- return;
+ ret = vfio_device_create_dmabuf_fd(res->iov, res->iov_cnt,
+ &vfio_err);
+ if (ret > 0) {
+ error_free(local_err);
+ local_err = NULL;
+ goto out;
+ }
+
+ if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
+ qemu_log_mask(LOG_GUEST_ERROR,
+ "Cannot create dmabuf: incompatible memory\n");
+ error_free(vfio_err);
+ error_free(local_err);
+ return;
+ }
+ error_report_err(vfio_err);
}
error_report_err(local_err);
return;
}
+out:
res->dmabuf_fd = ret;
if (!virtio_gpu_remap_dmabuf(res, &local_err)) {
error_report_err(local_err);
--
2.53.0
On 2026/02/28 7:13, Vivek Kasireddy wrote:
> In addition to memfd, a blob resource can also have its backing
> storage in a VFIO device region. Since, there is no effective way
> to determine where the backing storage is located, we first try to
> create a dmabuf assuming it is in memfd. If that fails, we try to
> create a dmabuf assuming it is in VFIO device region.
>
> So, we first call virtio_gpu_create_udmabuf() to check if the blob's
> backing storage is located in a memfd or not. If it is not, we invoke
> the vfio_device_create_dmabuf_fd() API which identifies the right
> VFIO device and eventually creates a dmabuf fd.
>
> Note that in virtio_gpu_remap_dmabuf(), we first try to test if
> the VFIO dmabuf exporter supports mmap or not. If it doesn't, we
> fall back to the vfio_device_mmap_dmabuf() API to get a mapping
> created for the dmabuf.
>
> Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
> Cc: Alex Bennée <alex.bennee@linaro.org>
> Cc: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
> Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
> Cc: Alex Williamson <alex@shazbot.org>
> Cc: Cédric Le Goater <clg@redhat.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> hw/display/virtio-gpu-dmabuf.c | 35 +++++++++++++++++++++++++++-------
> 1 file changed, 28 insertions(+), 7 deletions(-)
>
> diff --git a/hw/display/virtio-gpu-dmabuf.c b/hw/display/virtio-gpu-dmabuf.c
> index b2badfc1bd..533d3d601c 100644
> --- a/hw/display/virtio-gpu-dmabuf.c
> +++ b/hw/display/virtio-gpu-dmabuf.c
> @@ -74,13 +74,21 @@ static int virtio_gpu_create_udmabuf(struct virtio_gpu_simple_resource *res,
> static bool virtio_gpu_remap_dmabuf(struct virtio_gpu_simple_resource *res,
> Error **errp)
> {
> + Error *mmap_err = NULL;
> void *map;
>
> map = mmap(NULL, res->blob_size, PROT_READ, MAP_SHARED, res->dmabuf_fd, 0);
> if (map == MAP_FAILED) {
> - error_setg_errno(errp, errno, "dmabuf mmap failed");
> + error_setg_errno(&mmap_err, errno, "dmabuf mmap failed: ");
> res->remapped = NULL;
> - return false;
> +
> + if (!vfio_device_mmap_dmabuf(res->iov, res->iov_cnt, &map,
> + res->blob_size, errp)) {
Let's call vfio_device_mmap_dmabuf() only when
vfio_device_create_dmabuf_fd() is called. Hopefully it will simplify the
logic structure.
Regards,
Akihiko Odaki
> + error_report_err(mmap_err);
> + return false;
> + }
> + error_free(mmap_err);
> + mmap_err = NULL;
> }
> res->remapped = map;
> return true;
> @@ -137,7 +145,7 @@ bool virtio_gpu_have_udmabuf(void)
>
> void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
> {
> - Error *local_err = NULL;
> + Error *local_err = NULL, *vfio_err = NULL;
> void *pdata = NULL;
> int ret;
>
> @@ -151,15 +159,28 @@ void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
> error_prepend(&local_err, "Cannot create dmabuf: ");
>
> if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
> - qemu_log_mask(LOG_GUEST_ERROR,
> - "Cannot create dmabuf: incompatible memory\n");
> - error_free(local_err);
> - return;
> + ret = vfio_device_create_dmabuf_fd(res->iov, res->iov_cnt,
> + &vfio_err);
> + if (ret > 0) {
> + error_free(local_err);
> + local_err = NULL;
> + goto out;
> + }
> +
> + if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
> + qemu_log_mask(LOG_GUEST_ERROR,
> + "Cannot create dmabuf: incompatible memory\n");
> + error_free(vfio_err);
> + error_free(local_err);
> + return;
> + }
> + error_report_err(vfio_err);
> }
> error_report_err(local_err);
> return;
> }
>
> +out:
> res->dmabuf_fd = ret;
> if (!virtio_gpu_remap_dmabuf(res, &local_err)) {
> error_report_err(local_err);
Hi Akihiko,
> Subject: Re: [PATCH v8 10/10] virtio-gpu-dmabuf: Create dmabuf for blobs
> associated with VFIO devices
>
> On 2026/02/28 7:13, Vivek Kasireddy wrote:
> > In addition to memfd, a blob resource can also have its backing
> > storage in a VFIO device region. Since, there is no effective way
> > to determine where the backing storage is located, we first try to
> > create a dmabuf assuming it is in memfd. If that fails, we try to
> > create a dmabuf assuming it is in VFIO device region.
> >
> > So, we first call virtio_gpu_create_udmabuf() to check if the blob's
> > backing storage is located in a memfd or not. If it is not, we invoke
> > the vfio_device_create_dmabuf_fd() API which identifies the right
> > VFIO device and eventually creates a dmabuf fd.
> >
> > Note that in virtio_gpu_remap_dmabuf(), we first try to test if
> > the VFIO dmabuf exporter supports mmap or not. If it doesn't, we
> > fall back to the vfio_device_mmap_dmabuf() API to get a mapping
> > created for the dmabuf.
> >
> > Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
> > Cc: Alex Bennée <alex.bennee@linaro.org>
> > Cc: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
> > Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
> > Cc: Alex Williamson <alex@shazbot.org>
> > Cc: Cédric Le Goater <clg@redhat.com>
> > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > ---
> > hw/display/virtio-gpu-dmabuf.c | 35 +++++++++++++++++++++++++++-
> ------
> > 1 file changed, 28 insertions(+), 7 deletions(-)
> >
> > diff --git a/hw/display/virtio-gpu-dmabuf.c b/hw/display/virtio-gpu-
> dmabuf.c
> > index b2badfc1bd..533d3d601c 100644
> > --- a/hw/display/virtio-gpu-dmabuf.c
> > +++ b/hw/display/virtio-gpu-dmabuf.c
> > @@ -74,13 +74,21 @@ static int virtio_gpu_create_udmabuf(struct
> virtio_gpu_simple_resource *res,
> > static bool virtio_gpu_remap_dmabuf(struct
> virtio_gpu_simple_resource *res,
> > Error **errp)
> > {
> > + Error *mmap_err = NULL;
> > void *map;
> >
> > map = mmap(NULL, res->blob_size, PROT_READ, MAP_SHARED, res-
> >dmabuf_fd, 0);
> > if (map == MAP_FAILED) {
> > - error_setg_errno(errp, errno, "dmabuf mmap failed");
> > + error_setg_errno(&mmap_err, errno, "dmabuf mmap failed: ");
> > res->remapped = NULL;
> > - return false;
> > +
> > + if (!vfio_device_mmap_dmabuf(res->iov, res->iov_cnt, &map,
> > + res->blob_size, errp)) {
>
> Let's call vfio_device_mmap_dmabuf() only when
> vfio_device_create_dmabuf_fd() is called. Hopefully it will simplify the
> logic structure.
The logic looks OK to me but are you suggesting that it would be simpler to
remove the call to vfio_device_mmap_dmabuf() from virtio_gpu_remap_dmabuf()
and call it where vfio_device_create_dmabuf_fd() gets called?
Or, if we keep the call to vfio_device_mmap_dmabuf() in virtio_gpu_remap_dmabuf(),
then to implement your suggestion, I think we need a flag to identify dmabufs
that are created via VFIO so that we would call vfio_device_mmap_dmabuf()
only for these ones.
Also, for VFIO based dmabufs, should we directly call vfio_device_mmap_dmabuf()
or is it ok if we first call mmap() and check if it works or not?
Thanks,
Vivek
>
> Regards,
> Akihiko Odaki
>
> > + error_report_err(mmap_err);
> > + return false;
> > + }
> > + error_free(mmap_err);
> > + mmap_err = NULL;
> > }
> > res->remapped = map;
> > return true;
> > @@ -137,7 +145,7 @@ bool virtio_gpu_have_udmabuf(void)
> >
> > void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
> > {
> > - Error *local_err = NULL;
> > + Error *local_err = NULL, *vfio_err = NULL;
> > void *pdata = NULL;
> > int ret;
> >
> > @@ -151,15 +159,28 @@ void virtio_gpu_init_dmabuf(struct
> virtio_gpu_simple_resource *res)
> > error_prepend(&local_err, "Cannot create dmabuf: ");
> >
> > if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
> > - qemu_log_mask(LOG_GUEST_ERROR,
> > - "Cannot create dmabuf: incompatible memory\n");
> > - error_free(local_err);
> > - return;
> > + ret = vfio_device_create_dmabuf_fd(res->iov, res->iov_cnt,
> > + &vfio_err);
> > + if (ret > 0) {
> > + error_free(local_err);
> > + local_err = NULL;
> > + goto out;
> > + }
> > +
> > + if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
> > + qemu_log_mask(LOG_GUEST_ERROR,
> > + "Cannot create dmabuf: incompatible memory\n");
> > + error_free(vfio_err);
> > + error_free(local_err);
> > + return;
> > + }
> > + error_report_err(vfio_err);
> > }
> > error_report_err(local_err);
> > return;
> > }
> >
> > +out:
> > res->dmabuf_fd = ret;
> > if (!virtio_gpu_remap_dmabuf(res, &local_err)) {
> > error_report_err(local_err);
On 2026/03/02 15:21, Kasireddy, Vivek wrote:
> Hi Akihiko,
>
>> Subject: Re: [PATCH v8 10/10] virtio-gpu-dmabuf: Create dmabuf for blobs
>> associated with VFIO devices
>>
>> On 2026/02/28 7:13, Vivek Kasireddy wrote:
>>> In addition to memfd, a blob resource can also have its backing
>>> storage in a VFIO device region. Since, there is no effective way
>>> to determine where the backing storage is located, we first try to
>>> create a dmabuf assuming it is in memfd. If that fails, we try to
>>> create a dmabuf assuming it is in VFIO device region.
>>>
>>> So, we first call virtio_gpu_create_udmabuf() to check if the blob's
>>> backing storage is located in a memfd or not. If it is not, we invoke
>>> the vfio_device_create_dmabuf_fd() API which identifies the right
>>> VFIO device and eventually creates a dmabuf fd.
>>>
>>> Note that in virtio_gpu_remap_dmabuf(), we first try to test if
>>> the VFIO dmabuf exporter supports mmap or not. If it doesn't, we
>>> fall back to the vfio_device_mmap_dmabuf() API to get a mapping
>>> created for the dmabuf.
>>>
>>> Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
>>> Cc: Alex Bennée <alex.bennee@linaro.org>
>>> Cc: Akihiko Odaki <odaki@rsg.ci.i.u-tokyo.ac.jp>
>>> Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
>>> Cc: Alex Williamson <alex@shazbot.org>
>>> Cc: Cédric Le Goater <clg@redhat.com>
>>> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
>>> ---
>>> hw/display/virtio-gpu-dmabuf.c | 35 +++++++++++++++++++++++++++-
>> ------
>>> 1 file changed, 28 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/hw/display/virtio-gpu-dmabuf.c b/hw/display/virtio-gpu-
>> dmabuf.c
>>> index b2badfc1bd..533d3d601c 100644
>>> --- a/hw/display/virtio-gpu-dmabuf.c
>>> +++ b/hw/display/virtio-gpu-dmabuf.c
>>> @@ -74,13 +74,21 @@ static int virtio_gpu_create_udmabuf(struct
>> virtio_gpu_simple_resource *res,
>>> static bool virtio_gpu_remap_dmabuf(struct
>> virtio_gpu_simple_resource *res,
>>> Error **errp)
>>> {
>>> + Error *mmap_err = NULL;
>>> void *map;
>>>
>>> map = mmap(NULL, res->blob_size, PROT_READ, MAP_SHARED, res-
>>> dmabuf_fd, 0);
>>> if (map == MAP_FAILED) {
>>> - error_setg_errno(errp, errno, "dmabuf mmap failed");
>>> + error_setg_errno(&mmap_err, errno, "dmabuf mmap failed: ");
>>> res->remapped = NULL;
>>> - return false;
>>> +
>>> + if (!vfio_device_mmap_dmabuf(res->iov, res->iov_cnt, &map,
>>> + res->blob_size, errp)) {
>>
>> Let's call vfio_device_mmap_dmabuf() only when
>> vfio_device_create_dmabuf_fd() is called. Hopefully it will simplify the
>> logic structure.
> The logic looks OK to me but are you suggesting that it would be simpler to
> remove the call to vfio_device_mmap_dmabuf() from virtio_gpu_remap_dmabuf()
> and call it where vfio_device_create_dmabuf_fd() gets called?
>
> Or, if we keep the call to vfio_device_mmap_dmabuf() in virtio_gpu_remap_dmabuf(),
> then to implement your suggestion, I think we need a flag to identify dmabufs
> that are created via VFIO so that we would call vfio_device_mmap_dmabuf()
> only for these ones.
>
> Also, for VFIO based dmabufs, should we directly call vfio_device_mmap_dmabuf()
> or is it ok if we first call mmap() and check if it works or not?
The current logic structure looks like as follows:
Initialize DMA-BUF
Create DMA-BUF
Create DMA-BUF, assuming it is memfd
If it turned out it is not memfd,
Create DMA-BUF, assuming it is VFIO
Map memory
Map memory, assuming it is memfd
If it turned out it is not memfd,
Map memory, assuming it is VFIO
The suggested logic structure looks like as follows:
Initialize DMA-BUF
Initialize DMA-BUF, assuming it is memfd
Create DMA-BUF
Map memory
If it turned out it is not memfd
Initialize DMA-BUF, assuming it is VFIO
Create DMA-BUF
Map memory
See we will have less "if".
Regards,
Akihiko Odaki
>
> Thanks,
> Vivek
>
>>
>> Regards,
>> Akihiko Odaki
>>
>>> + error_report_err(mmap_err);
>>> + return false;
>>> + }
>>> + error_free(mmap_err);
>>> + mmap_err = NULL;
>>> }
>>> res->remapped = map;
>>> return true;
>>> @@ -137,7 +145,7 @@ bool virtio_gpu_have_udmabuf(void)
>>>
>>> void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource *res)
>>> {
>>> - Error *local_err = NULL;
>>> + Error *local_err = NULL, *vfio_err = NULL;
>>> void *pdata = NULL;
>>> int ret;
>>>
>>> @@ -151,15 +159,28 @@ void virtio_gpu_init_dmabuf(struct
>> virtio_gpu_simple_resource *res)
>>> error_prepend(&local_err, "Cannot create dmabuf: ");
>>>
>>> if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
>>> - qemu_log_mask(LOG_GUEST_ERROR,
>>> - "Cannot create dmabuf: incompatible memory\n");
>>> - error_free(local_err);
>>> - return;
>>> + ret = vfio_device_create_dmabuf_fd(res->iov, res->iov_cnt,
>>> + &vfio_err);
>>> + if (ret > 0) {
>>> + error_free(local_err);
>>> + local_err = NULL;
>>> + goto out;
>>> + }
>>> +
>>> + if (ret == VFIO_DMABUF_CREATE_GUEST_ERROR) {
>>> + qemu_log_mask(LOG_GUEST_ERROR,
>>> + "Cannot create dmabuf: incompatible memory\n");
>>> + error_free(vfio_err);
>>> + error_free(local_err);
>>> + return;
>>> + }
>>> + error_report_err(vfio_err);
>>> }
>>> error_report_err(local_err);
>>> return;
>>> }
>>>
>>> +out:
>>> res->dmabuf_fd = ret;
>>> if (!virtio_gpu_remap_dmabuf(res, &local_err)) {
>>> error_report_err(local_err);
>
© 2016 - 2026 Red Hat, Inc.