Hi Akihiko,
> Subject: Re: [PATCH v7 06/10] virtio-gpu-dmabuf: Improve error handling
> with 'Error **' and err enum
>
> >>
> >> On 2026/02/23 17:00, Vivek Kasireddy wrote:
> >>> Make the error handling more robust in virtio_gpu_init_udmabuf()
> >>> by introducing 'Error **' parameter to capture errors and an enum
> >>> to categorize Guest and Host errors. This allows for better error
> >>> reporting and handling of errors from virtio_gpu_create_udmabuf()
> >>> and virtio_gpu_remap_dmabuf().
> >>>
> >>> Cc: Marc-André Lureau <[email protected]>
> >>> Cc: Alex Bennée <[email protected]>
> >>> Cc: Akihiko Odaki <[email protected]>
> >>> Cc: Dmitry Osipenko <[email protected]>
> >>> Cc: Alex Williamson <[email protected]>
> >>> Cc: Cédric Le Goater <[email protected]>
> >>> Signed-off-by: Vivek Kasireddy <[email protected]>
> >>> ---
> >>> hw/display/virtio-gpu-dmabuf.c | 65 +++++++++++++++++++++++++-
> --
> >> ------
> >>> 1 file changed, 48 insertions(+), 17 deletions(-)
> >>>
> >>> diff --git a/hw/display/virtio-gpu-dmabuf.c b/hw/display/virtio-gpu-
> >> dmabuf.c
> >>> index 8d67ef7c2a..fdcf8e53a2 100644
> >>> --- a/hw/display/virtio-gpu-dmabuf.c
> >>> +++ b/hw/display/virtio-gpu-dmabuf.c
> >>> @@ -27,7 +27,17 @@
> >>> #include "standard-headers/linux/udmabuf.h"
> >>> #include "standard-headers/drm/drm_fourcc.h"
> >>>
> >>> -static void virtio_gpu_create_udmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>> +typedef enum VirtIOGPUErrorType {
> >>> + VIRTIO_GPU_NO_ERROR = 0,
> >>> + /* Guest is responsible for this error */
> >>> + VIRTIO_GPU_GUEST_ERROR = 1,
> >>> + /* Host is at fault for this error */
> >>> + VIRTIO_GPU_HOST_ERROR = 2,
> >>
> >> include/qapi/error.h says:
> >> > - Whenever practical, also return a value that indicates success /
> >> > failure. This can make the error checking more concise, and can
> >> > avoid useless error object creation and destruction. Note that
> >> > we still have many functions returning void. We recommend
> >> > . bool-valued functions return true on success / false on failure,
> >> > . pointer-valued functions return non-null / null pointer, and
> >> > . integer-valued functions return non-negative / negative.
> >>
> >> Do what it says. Non-negative if *errp is set. Negative otherwise.
> > Ok, I'll change the values in the following way:
> > typedef enum VirtIOGPUErrorType {
> > VIRTIO_GPU_NO_ERROR = 1,
> > /* Guest is responsible for this error */
> > VIRTIO_GPU_GUEST_ERROR = -1,
> > /* Host is at fault for this error */
> > VIRTIO_GPU_HOST_ERROR = -2,
> > } VirtIOGPUErrorType;
> >
> >>
> >>> +} VirtIOGPUErrorType;
> >>> +
> >>> +static VirtIOGPUErrorType
> >>> +virtio_gpu_create_udmabuf(struct virtio_gpu_simple_resource *res,
> >> int *fd,
> >>> + Error **errp)
> >>> {
> >>> g_autofree struct udmabuf_create_list *list = NULL;
> >>> RAMBlock *rb;
> >>> @@ -36,7 +46,8 @@ static void virtio_gpu_create_udmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>>
> >>> udmabuf = udmabuf_fd();
> >>> if (udmabuf < 0) {
> >>> - return;
> >>> + error_setg(errp, "udmabuf device not available");
> >>> + return VIRTIO_GPU_HOST_ERROR;
> >>> }
> >>>
> >>> list = g_malloc0(sizeof(struct udmabuf_create_list) +
> >>> @@ -45,7 +56,8 @@ static void virtio_gpu_create_udmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>> for (i = 0; i < res->iov_cnt; i++) {
> >>> rb = qemu_ram_block_from_host(res->iov[i].iov_base, false,
> >> &offset);
> >>> if (!rb || rb->fd < 0) {
> >>> - return;
> >>> + error_setg(errp, "Could not find valid ramblock");
> >>
> >> It is a valid ramblock even if rb->fd < 0. It is just "incompatible"
> >> with udmabuf.
> > Does "IOV memory address incompatible with udmabuf " sound better?
> >
> >>
> >>> + return VIRTIO_GPU_GUEST_ERROR;
> >>> }
> >>>
> >>> list->list[i].memfd = rb->fd;
> >>> @@ -56,22 +68,30 @@ static void virtio_gpu_create_udmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>> list->count = res->iov_cnt;
> >>> list->flags = UDMABUF_FLAGS_CLOEXEC;
> >>>
> >>> - res->dmabuf_fd = ioctl(udmabuf, UDMABUF_CREATE_LIST, list);
> >>> - if (res->dmabuf_fd < 0) {
> >>> - warn_report("%s: UDMABUF_CREATE_LIST: %s", __func__,
> >>> - strerror(errno));
> >>> + *fd = ioctl(udmabuf, UDMABUF_CREATE_LIST, list);
> >>> + if (*fd < 0) {
> >>> + error_setg_errno(errp, errno, "UDMABUF_CREATE_LIST: ioctl
> >> failed");
> >>> + if (errno == EINVAL || errno == EBADFD) {
> >>> + return VIRTIO_GPU_GUEST_ERROR;
> >>> + }
> >>> + return VIRTIO_GPU_HOST_ERROR;
> >>> }
> >>> + return VIRTIO_GPU_NO_ERROR;
> >>> }
> >>>
> >>> -static void virtio_gpu_remap_dmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>> +static bool virtio_gpu_remap_dmabuf(struct
> >> virtio_gpu_simple_resource *res,
> >>> + Error **errp)
> >>> {
> >>> - res->remapped = mmap(NULL, res->blob_size, PROT_READ,
> >>> - MAP_SHARED, res->dmabuf_fd, 0);
> >>> - if (res->remapped == MAP_FAILED) {
> >>> - warn_report("%s: dmabuf mmap failed: %s", __func__,
> >>> - strerror(errno));
> >>> + 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");
> >>> res->remapped = NULL;
> >>> + return false;
> >>> }
> >>> + res->remapped = map;
> >>> + return true;
> >>> }
> >>>
> >>> static void virtio_gpu_destroy_dmabuf(struct
> >> virtio_gpu_simple_resource *res)
> >>> @@ -125,19 +145,30 @@ bool virtio_gpu_have_udmabuf(void)
> >>>
> >>> void virtio_gpu_init_dmabuf(struct virtio_gpu_simple_resource
> *res)
> >>> {
> >>> + Error *local_err = NULL;
> >>> + VirtIOGPUErrorType err;
> >>> void *pdata = NULL;
> >>> + int fd;
> >>>
> >>> res->dmabuf_fd = -1;
> >>> if (res->iov_cnt == 1 &&
> >>> res->iov[0].iov_len < 4096) {
> >>> pdata = res->iov[0].iov_base;
> >>> } else {
> >>> - virtio_gpu_create_udmabuf(res);
> >>> - if (res->dmabuf_fd < 0) {
> >>> + err = virtio_gpu_create_udmabuf(res, &fd, &local_err);
> >>> + if (err != VIRTIO_GPU_NO_ERROR) {
> >>> + error_prepend(&local_err, "Cannot create dmabuf: ");
> >>> + if (err == VIRTIO_GPU_GUEST_ERROR) {
> >>> + qemu_log_mask(LOG_GUEST_ERROR,
> >>> + "Cannot create dmabuf: incompatible
> >>> mem\n");
> >>
> >> s/mem/memory/
> >>
> >> The log is not that long so we can have a full correct spelling.
> > It was going over the 80 char limit and I was reluctant to split it
> > across lines.
>
> scripts/checkpatch.pl explicitly allows to have a longer line for string
> literals. Priotize the user-visible behavior than keeping the code tidy.
Sounds good. I'll replace mem with memory.
>
> >
> >>
> >>> + }
> >>> + error_report_err(local_err);
> >>
> >> This prints a message even for a guest error.
> > This is intentional. My thinking is that if the user did not enable logging
> > then it will fail silently in the case of Guest error. So, I thought it
> > makes
> > sense to print an error regardless of the type of error.
>
> Do not locally override the behavior of the codebase-wide
> infrastructure. If you think a guest error needs to be logged by
> default, you should make a change to the logging infrastructure.
>
> I also think the decision of the logging infrastructure not logging
> guest errors by default makes sense since, if you log guest errors by
> default, it will allow the guest to spam the output and can cause
> security and reliability issues.
Ok, I'll change it such that error_report_err(local_err) is only done in the
case of Host error. And, in the case of Guest error, I plan to just have
error_free(local_err) in addition to invoking qemu_log_mask().
Thanks,
Vivek
>
> Regards,
> Akihiko Odaki
>
> >
> > Thanks,
> > Vivek
> >
> >>
> >> Regards,
> >> Akihiko Odaki
> >>
> >>> return;
> >>> }
> >>> - virtio_gpu_remap_dmabuf(res);
> >>> - if (!res->remapped) {
> >>> +
> >>> + res->dmabuf_fd = fd;
> >>> + if (!virtio_gpu_remap_dmabuf(res, &local_err)) {
> >>> + error_report_err(local_err);
> >>> return;
> >>> }
> >>> pdata = res->remapped;
> >
