Filtered by vendor Amd Subscriptions
Filtered by product Radeon Pro V620 Subscriptions
Total 26 CVE
CVE Vendors Products Updated CVSS v3.1
CVE-2024-36324 1 Amd 25 Amd Ryzen™ Ai 300 Series Processors, Radeon Pro V520, Radeon Pro V620 and 22 more 2026-04-15 8.8 High
Improper input validation in AMD Graphics Driver could allow an attacker to supply a specially crafted pointer, potentially leading to arbitrary code execution.
CVE-2021-26377 1 Amd 11 Athlon, Athlon 3000, Radeon Instinct Mi25 and 8 more 2026-04-15 4.1 Medium
Insufficient parameter validation while allocating process space in the Trusted OS (TOS) may allow for a malicious userspace process to trigger an integer overflow, leading to a potential denial of service.
CVE-2021-26381 1 Amd 17 Radeon Pro V520, Radeon Pro V620, Radeon Pro W5000 Series and 14 more 2026-04-15 N/A
Improper system call parameter validation in the Trusted OS may allow a malicious driver to perform mapping or unmapping operations on a large number of pages, potentially resulting in kernel memory corruption.
CVE-2021-46750 1 Amd 11 Athlon, Athlon 3000, Radeon Pro V620 and 8 more 2026-04-15 3 Low
Failure to validate the address and size in TEE (Trusted Execution Environment) may allow a malicious x86 attacker to send malformed messages to the graphics mailbox resulting in an overlap of a TMR (Trusted Memory Region) that was previously allocated by the ASP bootloader leading to a potential loss of integrity.
CVE-2021-26410 1 Amd 15 Radeon Pro V520, Radeon Pro V620, Radeon Pro W5000 Series and 12 more 2026-04-15 N/A
Improper syscall input validation in ASP (AMD Secure Processor) may force the kernel into reading syscall parameter values from its own memory space allowing an attacker to infer the contents of the kernel memory leading to potential information disclosure.
CVE-2023-4969 3 Amd, Imaginationtech, Khronos 261 Athlon 3000g, Athlon 3000g Firmware, Instinct Mi100 and 258 more 2025-06-20 6.5 Medium
A GPU kernel can read sensitive data from another GPU kernel (even from another user or app) through an optimized GPU memory region called _local memory_ on various architectures.