Re: [v2 1/3] virtio-gpu: Add support for VIRTIO_GPU_BLOB_FLAG_USE_USERPTR flag

[Honglei Huang]

On 2025/11/21 10:56, Akihiko Odaki wrote:

On 2025/11/21 11:47, Honglei Huang wrote:

Add support for the USE_USERPTR blob flag in virtio-gpu to enable
user pointer mapping for blob resources. This allows guest applications
to use user-allocated memory for GPU resources more efficiently.

Changes include:
- Add VIRTIO_GPU_BLOB_FLAG_USE_USERPTR flag definition
- Enhance blob resource creation to handle userptr flag properly
- Remove arbitrary nr_entries limit (16384) in mapping creation
- Add conditional handling for userptr vs regular blob mapping

I don't see the added conditional handling.

Sorry, the additional handing is replaced by the fixing of value check.
I will correct this commit message in the next version.

- Support guest_blob_mapped parameter for virgl renderer
- Fix value check issue in virtio-gpu

This enables more flexible memory management between guest and host
for GPU virtualization scenarios.

Signed-off-by: Honglei Huang <hongh...@amd.com>
---
  hw/display/virtio-gpu-virgl.c               | 2 +-
  hw/display/virtio-gpu.c                     | 7 -------
  include/standard-headers/linux/virtio_gpu.h | 1 +
  3 files changed, 2 insertions(+), 8 deletions(-)

diff --git a/hw/display/virtio-gpu-virgl.c b/hw/display/virtio-gpu- virgl.c

index 07f6355ad6..c927275c79 100644
--- a/hw/display/virtio-gpu-virgl.c
+++ b/hw/display/virtio-gpu-virgl.c

@@ -705,7 +705,7 @@ static void virgl_cmd_resource_create_blob(VirtIOGPU *g,           ret = virtio_gpu_create_mapping_iov(g, cblob.nr_entries, sizeof(cblob),

                                              cmd, &res->base.addrs,

                                              &res->base.iov, &res- >base.iov_cnt);

-        if (!ret) {
+        if (ret != 0) {
              cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
              return;
          }
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
index 43e88a4daf..956dc811fa 100644
--- a/hw/display/virtio-gpu.c
+++ b/hw/display/virtio-gpu.c
@@ -808,13 +808,6 @@ int virtio_gpu_create_mapping_iov(VirtIOGPU *g,
      size_t esize, s;
      int e, v;
-    if (nr_entries > 16384) {
-        qemu_log_mask(LOG_GUEST_ERROR,
-                      "%s: nr_entries is too big (%d > 16384)\n",
-                      __func__, nr_entries);
-        return -1;
-    }
-
      esize = sizeof(*ents) * nr_entries;
      ents = g_malloc(esize);
      s = iov_to_buf(cmd->elem.out_sg, cmd->elem.out_num,

diff --git a/include/standard-headers/linux/virtio_gpu.h b/include/ standard-headers/linux/virtio_gpu.h

index 00cd3f04af..b85e781a2d 100644
--- a/include/standard-headers/linux/virtio_gpu.h
+++ b/include/standard-headers/linux/virtio_gpu.h
@@ -405,6 +405,7 @@ struct virtio_gpu_resource_create_blob {
  #define VIRTIO_GPU_BLOB_FLAG_USE_MAPPABLE     0x0001
  #define VIRTIO_GPU_BLOB_FLAG_USE_SHAREABLE    0x0002
  #define VIRTIO_GPU_BLOB_FLAG_USE_CROSS_DEVICE 0x0004
+#define VIRTIO_GPU_BLOB_FLAG_USE_USERPTR      0x0008
      /* zero is invalid blob mem */
      uint32_t blob_mem;
      uint32_t blob_flags;