[PATCH EDAC v26 54/66] edac: Only expose csrows/channels on legacy API if they're populated

From: Mauro Carvalho Chehab
Date: Fri May 18 2012 - 12:48:28 EST


This patch actually fixes a bug with the legacy API, where, at the
same csrow, some channels may have different DIMMs. This can happen
on FB-DIMM/RAMBUS and modern Intel controllers.

This is the case, for example, of Nehalem machines:

$ ./edac-ctl --layout
+-----------------------------------+
| mc0 |
| channel0 | channel1 | channel2 |
-------+-----------------------------------+
slot2: | 0 MB | 0 MB | 0 MB |
slot1: | 1024 MB | 0 MB | 0 MB |
slot0: | 1024 MB | 1024 MB | 1024 MB |
-------+-----------------------------------+

Before this patch, non-filled memories were shown. Now, only what's
filled is there:

grep . /sys/devices/system/edac/mc/mc0/csrow*/ch?*
/sys/devices/system/edac/mc/mc0/csrow0/ch0_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch0_dimm_label:CPU#0Channel#0_DIMM#0
/sys/devices/system/edac/mc/mc0/csrow0/ch1_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch1_dimm_label:CPU#0Channel#0_DIMM#1
/sys/devices/system/edac/mc/mc0/csrow1/ch0_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow1/ch0_dimm_label:CPU#0Channel#1_DIMM#0
/sys/devices/system/edac/mc/mc0/csrow2/ch0_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow2/ch0_dimm_label:CPU#0Channel#2_DIMM#0

Thanks-to: Aristeu Rozanski Filho <arozansk@xxxxxxxxxx>
Reviewed-by: Aristeu Rozanski <arozansk@xxxxxxxxxx>
Cc: Doug Thompson <norsk5@xxxxxxxxx>
Signed-off-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx>
---
drivers/edac/edac_mc_sysfs.c | 24 ++++++++++++++++++++++++
1 files changed, 24 insertions(+), 0 deletions(-)

diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index daa418b..0f67190 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -347,6 +347,16 @@ static struct device_attribute *dynamic_csrow_ce_count_attr[] = {
&dev_attr_legacy_ch5_ce_count.attr
};

+static inline int nr_pages_per_csrow(struct csrow_info *csrow)
+{
+ int chan, nr_pages = 0;
+
+ for (chan = 0; chan < csrow->nr_channels; chan++)
+ nr_pages += csrow->channels[chan].dimm->nr_pages;
+
+ return nr_pages;
+}
+
/* Create a CSROW object under specifed edac_mc_device */
static int edac_create_csrow_object(struct mem_ctl_info *mci,
struct csrow_info *csrow, int index)
@@ -371,6 +381,9 @@ static int edac_create_csrow_object(struct mem_ctl_info *mci,
return err;

for (chan = 0; chan < csrow->nr_channels; chan++) {
+ /* Only expose populated DIMMs */
+ if (!csrow->channels[chan].dimm->nr_pages)
+ continue;
err = device_create_file(&csrow->dev,
dynamic_csrow_dimm_attr[chan]);
if (err < 0)
@@ -405,6 +418,9 @@ static int edac_create_csrow_objects(struct mem_ctl_info *mci)
struct csrow_info *csrow;

for (i = 0; i < mci->nr_csrows; i++) {
+ csrow = &mci->csrows[i];
+ if (!nr_pages_per_csrow(csrow))
+ continue;
err = edac_create_csrow_object(mci, &mci->csrows[i], i);
if (err < 0)
goto error;
@@ -414,7 +430,11 @@ static int edac_create_csrow_objects(struct mem_ctl_info *mci)
error:
for (--i; i >= 0; i--) {
csrow = &mci->csrows[i];
+ if (!nr_pages_per_csrow(csrow))
+ continue;
for (chan = csrow->nr_channels - 1; chan >= 0; chan--) {
+ if (!csrow->channels[chan].dimm->nr_pages)
+ continue;
device_remove_file(&csrow->dev,
dynamic_csrow_dimm_attr[chan]);
device_remove_file(&csrow->dev,
@@ -433,7 +453,11 @@ static void edac_delete_csrow_objects(struct mem_ctl_info *mci)

for (i = mci->nr_csrows - 1; i >= 0; i--) {
csrow = &mci->csrows[i];
+ if (!nr_pages_per_csrow(csrow))
+ continue;
for (chan = csrow->nr_channels - 1; chan >= 0; chan--) {
+ if (!csrow->channels[chan].dimm->nr_pages)
+ continue;
debugf1("Removing csrow %d channel %d sysfs nodes\n",
i, chan);
device_remove_file(&csrow->dev,
--
1.7.8

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/