A flaw was found during cache eviction on some Intel processors which may allow a local attacker to infer cache contents and disclose information through this side-channel.

Reference: ---------- -> https://access.redhat.com/solutions/l1d-cache-eviction-and-vector-register-sampling

Additional information: ----------------------- -> https://en.wikipedia.org/wiki/Vectorprocessor -> https://software.intel.com/en-us/articles/introduction-to-intel-advanced-vector-extensions

A flaw was found in Intel processors where a local attacker is able to gain information about registers used for vector calculations by observing register states from other processes running on the system. This results in a race condition where store buffers, which were not cleared, could be read by another process or a CPU sibling. The highest threat from this vulnerability is data confidentiality where an attacker could read arbitrary data as it passes through the processor.

A race condition in specific microprocessors using Intel (R) DDIO cache allocation and RDMA may allow an authenticated user to potentially enable partial information disclosure via adjacent access.