• Haim Dreyfuss's avatar
    iwlwifi: pcie: Set affinity mask for rx interrupt vectors per cpu · 7c8d91eb
    Haim Dreyfuss authored
    In order to utilize the host's CPUs in the most efficient way
    we bind each rx interrupt vector to each CPU on the host.
    Each rx interrupt is prioritized to execute only on the designated CPU
    rather than any CPU.
    Processor affinity takes advantage of the fact that some remnants of
    a process that was run on a given processor may remain in that
    processor's memory state for example, data in the CPU cache after
    another process is run on that CPU. Scheduling that process to execute
    on the same processor could result in an efficient use of process by
    reducing performance-degrading situations such as cache misses
    and parallel processing.
    Signed-off-by: default avatarHaim Dreyfuss <haim.dreyfuss@intel.com>
    Signed-off-by: default avatarLuca Coelho <luciano.coelho@intel.com>
    7c8d91eb
trans.c 86.5 KB