sb_edac: make minimal use of channel_mask
authorAristeu Rozanski <aris@redhat.com>
Mon, 2 Jun 2014 18:15:28 +0000 (15:15 -0300)
committerMauro Carvalho Chehab <m.chehab@samsung.com>
Thu, 26 Jun 2014 18:46:46 +0000 (15:46 -0300)
channel_mask will be used in the future to determine which group of memory
modules is causing the errors since when mirroring, lockstep and close page
are enabled you can't. While that doesn't happen, use the channel_mask to
determine the channel instead of relying on the MC event/exception.

Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Aristeu Rozanski <aris@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
drivers/edac/sb_edac.c

index 939b54c..4342926 100644 (file)
@@ -1678,6 +1678,9 @@ static void sbridge_mce_output_error(struct mem_ctl_info *mci,
         * EDAC core should be handling the channel mask, in order to point
         * to the group of dimm's where the error may be happening.
         */
+       if (!pvt->is_lockstep && !pvt->is_mirrored && !pvt->is_close_pg)
+               channel = first_channel;
+
        snprintf(msg, sizeof(msg),
                 "%s%s area:%s err_code:%04x:%04x socket:%d channel_mask:%ld rank:%d",
                 overflow ? " OVERFLOW" : "",