View Change

5 comments:

• File src/soc/amd/common/block/gpio_banks/gpio.c:

  • Patch Set #4, Line 33: soc_get_gpio_event_table(&table, &items);

    If you put this call in program_gpios() before the for(... […]

    Done

  • Patch Set #4, Line 53: configure_gevent_smi

    Should this be a weak function or something that fails to compile if not provided by the SoC?

    I don't recall my exact line of thinking now, seems reasonable to remove this.

  • Patch Set #4, Line 241: route_sci(event);

    Is this always going to be true for all SoCs using this common code?

    I decided to just call it a hook rather than predict what may come in the future. Also, it occurred to me that route_sci writes one of the registers in SMI's AcpiMmio space and doesn't really belong in this file.

• File src/soc/amd/common/block/include/amdblocks/gpio.h:

  • Patch Set #5, Line 304: void soc_get_gpio_event_table(const struct soc_amd_event **table, size_t *items);

    line over 80 characters

    No longer an error, right? Or at this point is it easier to shorten the line?

• File src/soc/amd/stoneyridge/gpio.c:

  • Patch Set #4, Line 49: int

    size_t?

    Done

To view, visit change 32651. To unsubscribe, or for help writing mail filters, visit settings.