Re: SYSFS "errors"

Em Tue, 19 Feb 2013 12:43:46 +0100
Borislav Petkov [off-list ref] escreveu:

On Tue, Feb 19, 2013 at 08:11:49AM -0300, Mauro Carvalho Chehab wrote:

quoted

quoted

quoted

I remember I saw some discussions about it in the past at bluesmoke ML,
saying that -ENODEV is the expected behavior when this is not supported.

Changing from -ENODEV to "N/A" will break anything that would be relying
on the previous behavior. So, I think that such change will for sure break
userspace.

If we're willing to change it, not creating the "sdram_scrub_rate" sysfs 
node is less likely to affect userspace.

This will break scripts which assume this file's presence implicitly.

If are there any script like that, then yes.

[ … ]

quoted

@@ -1017,6 +1010,14 @@ int edac_create_sysfs_mci_device(struct mem_ctl_info *mci)
 		return err;
 	}
 
+	if (mci->set_sdram_scrub_rate && mci->get_sdram_scrub_rate) {

This will break cpc925_edac.c because it defines a
->get_sdram_scrub_rate but not a ->set_sdram_scrub_rate.

True. Patch below fixes it.

I think a maybe better fix would be to figure out the sysfs file
permissions based on the presence of the two functions and *then* add
the attribute.

This way, the only visible change to userspace is the corrected sysfs
file permissions.

I'm not sure if is there a way to pass fs permissions to something similar
to device_create_file().

On both cases, an error will happen at open:
	- if file doesn't exist (this approach), it will return -ENOENT;
	- if file is opened with wrong permissions, open will return -EPERM.

However, if the file is not created, readdir() won't show the file.
So, if userspace first seeks for the sdram_scrub_rate, it won't fail.
That makes the logic below a little safer, IMO.


[PATCH EDAC] edac: only create sdram_scrub_rate where supported

Currently, sdram_scrub_rate sysfs node is created even if the device
doesn't support get/set the scub rate. Change the logic to only
create this device node when the operation is supported.

If only read or only write is supported, it will keep returning -ENODEV
just like before.

Reported-by: Felipe Balbi <redacted>
Signed-off-by: Mauro Carvalho Chehab <redacted>

diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 9c58da6..29b66a2 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c

@@ -7,7 +7,7 @@
  *
  * Written Doug Thompson <norsk5@xmission.com> www.softwarebitmaker.com
  *
- * (c) 2012 - Mauro Carvalho Chehab <mchehab@redhat.com>
+ * (c) 2012-2013 - Mauro Carvalho Chehab <mchehab@redhat.com>
  *	The entire API were re-written, and ported to use struct device
  *
  */

@@ -882,7 +882,6 @@ static struct attribute *mci_attrs[] = {
 	&dev_attr_ce_noinfo_count.attr,
 	&dev_attr_ue_count.attr,
 	&dev_attr_ce_count.attr,
-	&dev_attr_sdram_scrub_rate.attr,
 	&dev_attr_max_location.attr,
 	NULL
 };

@@ -1017,6 +1016,14 @@ int edac_create_sysfs_mci_device(struct mem_ctl_info *mci)
 		return err;
 	}
 
+	if (mci->set_sdram_scrub_rate || mci->get_sdram_scrub_rate) {
+		err = device_create_file(&mci->dev,
+					 &dev_attr_sdram_scrub_rate);
+		if (err) {
+			edac_dbg(1, "failure: create sdram_scrub_rate\n");
+			goto fail2;
+		}
+	}
 	/*
 	 * Create the dimm/rank devices
 	 */

@@ -1061,6 +1068,7 @@ fail:
 			continue;
 		device_unregister(&dimm->dev);
 	}
+fail2:
 	device_unregister(&mci->dev);
 	bus_unregister(&mci->bus);
 	kfree(mci->bus.name);