** Description changed:

  [Impact]
  CVE-2026-3886 is an integer overflow in QEMU's virtio-gpu 2D resource
  allocation path, allowing guest-to-host code execution from a
  low-privileged process inside the guest. CVSS 8.8 (AV:L/AC:L/PR:L/UI:N/
  S:C/C:H/I:H/A:H). Reported via ZDI (ZDI-26-332); fixed upstream in
  March 2026 and shipped in QEMU 11.0.
  
  [Affected version]
  qemu 1:10.2.1+ds-1ubuntu3.1 (resolute-updates/main), Ubuntu 26.04.
  
  [Verification]
  hw/display/virtio-gpu.c in the shipped source still contains the
  pre-fix implementation:
  
-     static uint32_t calc_image_hostmem(pixman_format_code_t pformat,
-                                        uint32_t width, uint32_t height)
-     {
-         /* ... skip integer overflow check. ... */
-         int bpp = PIXMAN_FORMAT_BPP(pformat);
-         int stride = ((width * bpp + 0x1f) >> 5) * sizeof(uint32_t);
-         return height * stride;
-     }
+     static uint32_t calc_image_hostmem(pixman_format_code_t pformat,
+                                        uint32_t width, uint32_t height)
+     {
+         /* ... skip integer overflow check. ... */
+         int bpp = PIXMAN_FORMAT_BPP(pformat);
+         int stride = ((width * bpp + 0x1f) >> 5) * sizeof(uint32_t);
+         return height * stride;
+     }
  
  Upstream widens all arithmetic to uint64_t and adds an explicit
  size > UINT32_MAX rejection, returning bool with hostmem and
  rowstride_bytes as out-parameters.
  
  No CVE-named patch is present in debian/patches/; the series file
  contains no security patches.
  
  `pro fix CVE-2026-3886` reports: "Ubuntu security engineers are
  investigating this issue. 1 package is still affected: qemu"
  
  [Status elsewhere]
  - Debian:  fixed in the 13.6 point release
  - SUSE:    SUSE-SU-2026:22550-1 (7 July 2026)
- - Fedora:  fixed via the QEMU 11.0 rebase
- - RHEL/Oracle: ELSA-2026-50337
+ - Oracle: ELSA-2026-50337
  
  [Notes]
  Public technical analysis of this bug class has been available since
  late April 2026. Requesting triage and an SRU for 26.04.

** Description changed:

  [Impact]
  CVE-2026-3886 is an integer overflow in QEMU's virtio-gpu 2D resource
  allocation path, allowing guest-to-host code execution from a
  low-privileged process inside the guest. CVSS 8.8 (AV:L/AC:L/PR:L/UI:N/
  S:C/C:H/I:H/A:H). Reported via ZDI (ZDI-26-332); fixed upstream in
  March 2026 and shipped in QEMU 11.0.
  
  [Affected version]
  qemu 1:10.2.1+ds-1ubuntu3.1 (resolute-updates/main), Ubuntu 26.04.
  
  [Verification]
  hw/display/virtio-gpu.c in the shipped source still contains the
  pre-fix implementation:
  
      static uint32_t calc_image_hostmem(pixman_format_code_t pformat,
                                         uint32_t width, uint32_t height)
      {
          /* ... skip integer overflow check. ... */
          int bpp = PIXMAN_FORMAT_BPP(pformat);
          int stride = ((width * bpp + 0x1f) >> 5) * sizeof(uint32_t);
          return height * stride;
      }
  
  Upstream widens all arithmetic to uint64_t and adds an explicit
  size > UINT32_MAX rejection, returning bool with hostmem and
  rowstride_bytes as out-parameters.
  
  No CVE-named patch is present in debian/patches/; the series file
  contains no security patches.
  
  `pro fix CVE-2026-3886` reports: "Ubuntu security engineers are
  investigating this issue. 1 package is still affected: qemu"
  
  [Status elsewhere]
  - Debian:  fixed in the 13.6 point release
- - SUSE:    SUSE-SU-2026:22550-1 (7 July 2026)
  - Oracle: ELSA-2026-50337
  
  [Notes]
  Public technical analysis of this bug class has been available since
  late April 2026. Requesting triage and an SRU for 26.04.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2162697

Title:
  CVE-2026-3886: virtio-gpu integer overflow (calc_image_hostmem)
  unfixed in qemu 1:10.2.1+ds-1ubuntu3.1

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/2162697/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to