EDAC, sb_edac: Classify memory mirroring modes

Basically, there are full memory mirroring and address range partial
memory mirroring (supported by Haswell EX and Broadwell EX) modes.

a) In full memory mirroring, the memory behind each memory controller
   is mirrored, i.e. the memory is split into two identical mirrors
   (primary and secondary), half of the memory is reserved for redundancy.

b) In address range partial memory mirroring, the memory size (range)
   of primary and secondary behind each memory controller can be user
   defined by the TAD0 register. The rest of memory ranges defined by
   TAD1/TAD2/... in that memory controller are non-mirrored.

For more detail on memory mirroring, see the following link written by Tony Luck:

  https://01.org/lkp/blogs/tonyluck/2016/address-range-partial-memory-mirroring-linux

Currently the sb_edac driver only supports address decoding in full
memory mirroring and non-mirroring modes. In address range partial
memory mirroring mode, it may fail to decode an address that falls in a
non-mirroring area (the following was one of this kind of failed logs).

  mce: Uncorrected hardware memory error in user-access at 566d53a400
  Memory failure: 0x566d53a: Killing einj_mem_uc:4647 due to hardware memory corruption
  Memory failure: 0x566d53a: recovery action for dirty LRU page: Recovered
  mce: [Hardware Error]: Machine check events logged
  EDAC sbridge MC1: HANDLING MCE MEMORY ERROR
  EDAC sbridge MC1: CPU 48: Machine Check Event: 0 Bank 7: ec00000000010090
  EDAC sbridge MC1: TSC 4b914aa5a99dab
  EDAC sbridge MC1: ADDR 566d53a400
  EDAC sbridge MC1: MISC 1443a0c86
  EDAC sbridge MC1: PROCESSOR 0:406f1 TIME 1499712764 SOCKET 2 APIC 80
  EDAC MC1: 0 UE Can't discover the memory rank for ch addr 0x7fb54e900 on any memory ( page:0x0 offset:0x0 grain:32)
  mce: [Hardware Error]: Machine check events logged

Therefore, classify memory mirroring modes and make the address decoding
in address range partial memory mode correct.

Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: linux-edac <linux-edac@vger.kernel.org>
Link: http://lkml.kernel.org/r/20170730180651.30060-1-qiuxu.zhuo@intel.com
Signed-off-by: Borislav Petkov <bp@suse.de>

1 files changed, 47 insertions, 16 deletions

diff --git a/drivers/edac/sb_edac.c b/drivers/edac/sb_edac.c
index 687d0f23b9cc..dc0591654011 100644
--- a/drivers/edac/sb_edac.c
+++ b/drivers/edac/sb_edac.c
@@ -300,6 +300,12 @@ enum domain {
 	SOCK,
 };
 
+enum mirroring_mode {
+	NON_MIRRORING,
+	ADDR_RANGE_MIRRORING,
+	FULL_MIRRORING,
+};
+
 struct sbridge_pvt;
 struct sbridge_info {
 	enum type	type;
@@ -377,8 +383,9 @@ struct sbridge_pvt {
 	struct sbridge_channel	channel[NUM_CHANNELS];
 
 	/* Memory type detection */
-	bool			is_mirrored, is_lockstep, is_close_pg;
+	bool			is_cur_addr_mirrored, is_lockstep, is_close_pg;
 	bool			is_chan_hash;
+	enum mirroring_mode	mirror_mode;
 
 	/* Memory description */
 	u64			tolm, tohm;
@@ -1648,10 +1655,6 @@ static int get_dimm_config(struct mem_ctl_info *mci)
 	enum edac_type mode;
 	u32 reg;
 
-	if (pvt->info.type == HASWELL || pvt->info.type == BROADWELL) {
-		pci_read_config_dword(pvt->pci_ha, HASWELL_HASYSDEFEATURE2, &reg);
-		pvt->is_chan_hash = GET_BITFIELD(reg, 21, 21);
-	}
 	pvt->sbridge_dev->node_id = pvt->info.get_node_id(pvt);
 	edac_dbg(0, "mc#%d: Node ID: %d, source ID: %d\n",
 		 pvt->sbridge_dev->mc,
@@ -1663,22 +1666,45 @@ static int get_dimm_config(struct mem_ctl_info *mci)
 	 */
 	if (pvt->info.type == KNIGHTS_LANDING) {
 		mode = EDAC_S4ECD4ED;
-		pvt->is_mirrored = false;
+		pvt->mirror_mode = NON_MIRRORING;
+		pvt->is_cur_addr_mirrored = false;
 
 		if (knl_get_dimm_capacity(pvt, knl_mc_sizes) != 0)
 			return -1;
-		pci_read_config_dword(pvt->pci_ta, KNL_MCMTR, &pvt->info.mcmtr);
+		if (pci_read_config_dword(pvt->pci_ta, KNL_MCMTR, &pvt->info.mcmtr)) {
+			edac_dbg(0, "Failed to read KNL_MCMTR register\n");
+			return -ENODEV;
+		}
 	} else {
-		pci_read_config_dword(pvt->pci_ras, RASENABLES, &reg);
+		if (pvt->info.type == HASWELL || pvt->info.type == BROADWELL) {
+			if (pci_read_config_dword(pvt->pci_ha, HASWELL_HASYSDEFEATURE2, &reg)) {
+				edac_dbg(0, "Failed to read HASWELL_HASYSDEFEATURE2 register\n");
+				return -ENODEV;
+			}
+			pvt->is_chan_hash = GET_BITFIELD(reg, 21, 21);
+			if (GET_BITFIELD(reg, 28, 28)) {
+				pvt->mirror_mode = ADDR_RANGE_MIRRORING;
+				edac_dbg(0, "Address range partial memory mirroring is enabled\n");
+				goto next;
+			}
+		}
+		if (pci_read_config_dword(pvt->pci_ras, RASENABLES, &reg)) {
+			edac_dbg(0, "Failed to read RASENABLES register\n");
+			return -ENODEV;
+		}
 		if (IS_MIRROR_ENABLED(reg)) {
-			edac_dbg(0, "Memory mirror is enabled\n");
-			pvt->is_mirrored = true;
+			pvt->mirror_mode = FULL_MIRRORING;
+			edac_dbg(0, "Full memory mirroring is enabled\n");
 		} else {
-			edac_dbg(0, "Memory mirror is disabled\n");
-			pvt->is_mirrored = false;
+			pvt->mirror_mode = NON_MIRRORING;
+			edac_dbg(0, "Memory mirroring is disabled\n");
 		}
 
-		pci_read_config_dword(pvt->pci_ta, MCMTR, &pvt->info.mcmtr);
+next:
+		if (pci_read_config_dword(pvt->pci_ta, MCMTR, &pvt->info.mcmtr)) {
+			edac_dbg(0, "Failed to read MCMTR register\n");
+			return -ENODEV;
+		}
 		if (IS_LOCKSTEP_ENABLED(pvt->info.mcmtr)) {
 			edac_dbg(0, "Lockstep is enabled\n");
 			mode = EDAC_S8ECD8ED;
@@ -2092,7 +2118,8 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 
 	pci_read_config_dword(pvt->pci_tad[base_ch], tad_ch_nilv_offset[n_tads], &tad_offset);
 
-	if (pvt->is_mirrored) {
+	if (pvt->mirror_mode == FULL_MIRRORING ||
+	    (pvt->mirror_mode == ADDR_RANGE_MIRRORING && n_tads == 0)) {
 		*channel_mask |= 1 << ((base_ch + 2) % 4);
 		switch(ch_way) {
 		case 2:
@@ -2103,8 +2130,12 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 			sprintf(msg, "Invalid mirror set. Can't decode addr");
 			return -EINVAL;
 		}
-	} else
+
+		pvt->is_cur_addr_mirrored = true;
+	} else {
 		sck_xch = (1 << sck_way) * ch_way;
+		pvt->is_cur_addr_mirrored = false;
+	}
 
 	if (pvt->is_lockstep)
 		*channel_mask |= 1 << ((base_ch + 1) % 4);
@@ -2967,7 +2998,7 @@ 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)
+	if (!pvt->is_lockstep && !pvt->is_cur_addr_mirrored && !pvt->is_close_pg)
 		channel = first_channel;
 
 	snprintf(msg, sizeof(msg),