linux/poky.git/meta/recipes-graphics, branch scarthgap-5.0.11

xwayland: fix CVE-2025-49180

2025-07-09T15:43:32+00:00

A flaw was found in the RandR extension, where the RRChangeProviderProperty function does not properly validate input. This issue leads to an integer overflow when computing the total size to allocate. (From OE-Core rev: 15881f41f8c00c5f0a68628c2d49ca1aa1999c2e) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

xwayland: fix CVE-2025-49179

2025-07-09T15:43:32+00:00

A flaw was found in the X Record extension. The RecordSanityCheckRegisterClients function does not check for an integer overflow when computing request length, which allows a client to bypass length checks. (From OE-Core rev: de28bff9b54b2725d8c06c4760e0ed2b59d3fa61) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

xwayland: fix CVE-2025-49178

2025-07-09T15:43:32+00:00

A flaw was found in the X server's request handling. Non-zero 'bytes to ignore' in a client's request can cause the server to skip processing another client's request, potentially leading to a denial of service. (From OE-Core rev: 4c6df8320497c2ebf09902a62b6a3f3b061be917) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

xwayland: fix CVE-2025-49177

2025-07-09T15:43:32+00:00

A flaw was found in the XFIXES extension. The XFixesSetClientDisconnectMode handler does not validate the request length, allowing a client to read unintended memory from previous requests (From OE-Core rev: 0b2afd59ce8c35083c1cb3596a2f7d4eaa7bd1c8) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

xwayland: fix CVE-2025-49176

2025-07-09T15:43:32+00:00

A flaw was found in the Big Requests extension. The request length is multiplied by 4 before checking against the maximum allowed size, potentially causing an integer overflow and bypassing the size check. (From OE-Core rev: 0a2c5179e1f08ccd0fcaccb6f95c892ebafac8a8) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

xwayland: fix CVE-2025-49175

2025-07-09T15:43:32+00:00

A flaw was found in the X Rendering extension's handling of animated cursors. If a client provides no cursors, the server assumes at least one is present, leading to an out-of-bounds read and potential crash. (From OE-Core rev: fec7644b70452794fabfb7d967e2124918215440) Signed-off-by: Archana Polampalli Signed-off-by: Steve Sakoman

freetype: follow-up patch for CVE-2025-27363

2025-04-07T13:34:44+00:00

Per [1] there are two patches needed - [2] which is already included in 3.13.1 and [3] which is only in 2.13.3. Backport the second patch. [1] https://gitlab.freedesktop.org/freetype/freetype/-/issues/1322 [2] https://gitlab.freedesktop.org/freetype/freetype/-/commit/ef636696524b081f1b8819eb0c6a0b932d35757d [3] https://gitlab.freedesktop.org/freetype/freetype/-/commit/73720c7c9958e87b3d134a7574d1720ad2d24442 (From OE-Core rev: 41f855ea5a2018d08e0e9457d710032e96fe669b) Signed-off-by: Peter Marko Signed-off-by: Steve Sakoman

xserver-xorg: mark CVEs fixed in 21.1.16 as fixed

2025-03-26T15:48:51+00:00

These are tracked as versionless redhat CVEs in NVD DB. (From OE-Core rev: 84b1631bcbead1409ff44a1ed430244784c382be) Signed-off-by: Peter Marko Signed-off-by: Steve Sakoman

xserver-xorg: upgrade 21.1.15 -> 21.1.16

2025-03-15T13:40:08+00:00

Includes security fix * CVE-2025-26594 * CVE-2025-26595 * CVE-2025-26596 * CVE-2025-26597 * CVE-2025-26598 * CVE-2025-26599 * CVE-2025-26600 * CVE-2025-26601 Ref: https://lists.x.org/archives/xorg-announce/2025-February/003584.html https://lists.x.org/archives/xorg-announce/2025-February/003585.html (From OE-Core rev: 60d1d54a1c23f007434a1f2c1940fd8ccabe656f) Signed-off-by: Vijay Anusuri Signed-off-by: Richard Purdie (cherry picked from commit b5491688a045e52b2a1a00d04b746ed6af456784) Signed-off-by: Steve Sakoman

xserver-xf86-config: remove obsolete configuration files

2025-03-15T13:40:08+00:00

For reasons we have explicit xorg.conf files for a number of the qemu machines, but not all of them. These mainly disabled screen blanking (which is now down with a separate fragment) but also explictly set the device driver to fbdev which meant they didn't use the modesettings driver as they should (with the virtio framebuffer from qemu). This is the root cause of why the xserver 21.1.16 upgrade doesn't work on a number of machines: the /sys probing changed and the fbdev driver now refuses to use the PCI framebuffer device as there are better drivers, but we've explictly told xorg to use the wrong driver. For more details, see https://gitlab.freedesktop.org/xorg/xserver/-/issues/1798. (From OE-Core rev: ccbb0f5491e13d61015872fba93417b91c3213a2) Signed-off-by: Ross Burton Signed-off-by: Richard Purdie (cherry picked from commit 8c8039bf4c2d011e3d12c970ce45036b184902a9) Signed-off-by: Steve Sakoman