[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [QEMU PATCH v5 09/13] virtio-gpu: Handle resource blob commands



On 2023/09/20 14:50, Huang Rui wrote:
On Sat, Sep 16, 2023 at 12:37:29AM +0800, Akihiko Odaki wrote:
On 2023/09/16 1:04, Akihiko Odaki wrote:
On 2023/09/15 20:11, Huang Rui wrote:
From: Antonio Caggiano <antonio.caggiano@xxxxxxxxxxxxx>

Support BLOB resources creation, mapping and unmapping by calling the
new stable virglrenderer 0.10 interface. Only enabled when available and
via the blob config. E.g. -device virtio-vga-gl,blob=true

Signed-off-by: Antonio Caggiano <antonio.caggiano@xxxxxxxxxxxxx>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@xxxxxxxxxxxxx>
Signed-off-by: Xenia Ragiadakou <xenia.ragiadakou@xxxxxxx>
Signed-off-by: Huang Rui <ray.huang@xxxxxxx>
---

V4 -> V5:
      - Use memory_region_init_ram_ptr() instead of
        memory_region_init_ram_device_ptr() (Akihiko)

   hw/display/virtio-gpu-virgl.c  | 213 +++++++++++++++++++++++++++++++++
   hw/display/virtio-gpu.c        |   4 +-
   include/hw/virtio/virtio-gpu.h |   5 +
   meson.build                    |   4 +
   4 files changed, 225 insertions(+), 1 deletion(-)

diff --git a/hw/display/virtio-gpu-virgl.c
b/hw/display/virtio-gpu-virgl.c
index 312953ec16..563a6f2f58 100644
--- a/hw/display/virtio-gpu-virgl.c
+++ b/hw/display/virtio-gpu-virgl.c
@@ -17,6 +17,7 @@
   #include "trace.h"
   #include "hw/virtio/virtio.h"
   #include "hw/virtio/virtio-gpu.h"
+#include "hw/virtio/virtio-gpu-bswap.h"
   #include "ui/egl-helpers.h"
@@ -78,9 +79,24 @@ static void virgl_cmd_create_resource_3d(VirtIOGPU *g,
       virgl_renderer_resource_create(&args, NULL, 0);
   }
+static void virgl_resource_destroy(VirtIOGPU *g,
+                                   struct virtio_gpu_simple_resource
*res)
+{
+    if (!res)
+        return;
+
+    QTAILQ_REMOVE(&g->reslist, res, next);
+
+    virtio_gpu_cleanup_mapping_iov(g, res->iov, res->iov_cnt);
+    g_free(res->addrs);
+
+    g_free(res);
+}
+
   static void virgl_cmd_resource_unref(VirtIOGPU *g,
                                        struct virtio_gpu_ctrl_command
*cmd)
   {
+    struct virtio_gpu_simple_resource *res;
       struct virtio_gpu_resource_unref unref;
       struct iovec *res_iovs = NULL;
       int num_iovs = 0;
@@ -88,13 +104,22 @@ static void virgl_cmd_resource_unref(VirtIOGPU *g,
       VIRTIO_GPU_FILL_CMD(unref);
       trace_virtio_gpu_cmd_res_unref(unref.resource_id);
+    res = virtio_gpu_find_resource(g, unref.resource_id);
+
       virgl_renderer_resource_detach_iov(unref.resource_id,
                                          &res_iovs,
                                          &num_iovs);
       if (res_iovs != NULL && num_iovs != 0) {
           virtio_gpu_cleanup_mapping_iov(g, res_iovs, num_iovs);
+        if (res) {
+            res->iov = NULL;
+            res->iov_cnt = 0;
+        }
       }
+
       virgl_renderer_resource_unref(unref.resource_id);
+
+    virgl_resource_destroy(g, res);

This may leak memory region.

The memory region should be freed under virgl_cmd_resource_unmap_blob()
which is calling memory_region_del_subregion(&b->hostmem, res->region).
Because this region is created by map_blob(). Do we have the case to call
virgl_cmd_resource_unref() without calling virgl_cmd_resource_unmap_blob()
for blob memory?

Calling virgl_cmd_resource_unmap_blob() and virgl_cmd_resource_unref() in order is a guest's responsibility, and we are required to prepare for broken guests.



 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.