[PATCH] aoe: document sysfs interface

[PATCH] aoe: document sysfs interface

[Aishwarya Pant]

Documentation has been compiled from git commit logs and descriptions in
Documentation/aoe/aoe.txt. This should be useful for scripting and
tracking changes in the ABI.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
---
 Documentation/ABI/testing/sysfs-block-aoe | 44 +++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-block-aoe

diff --git a/Documentation/ABI/testing/sysfs-block-aoe b/Documentation/ABI/testing/sysfs-block-aoe
new file mode 100644
index 000000000000..d197c7770dc0
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-block-aoe
@@ -0,0 +1,44 @@
+What:		/sys/block/etherd*/mac
+Date:		Apr, 2005
+KernelVersion:	v2.6.12
+Contact:	Ed L. Cashin <ed.cashin@acm.org>
+Description:
+		(RO) The ethernet address of the remote Ata over Ethernet (AoE)
+		device.
+
+What:		/sys/block/etherd*/netif
+Date:		Apr, 2005
+KernelVersion:	v2.6.12
+Contact:	Ed L. Cashin <ed.cashin@acm.org>
+Description:
+		(RO) The name of the network interface on the localhost through
+		which we are communicating with the remote AoE device.
+
+What:		/sys/block/etherd*/state
+Date:		Apr, 2005
+KernelVersion:	v2.6.12
+Contact:	Ed L. Cashin <ed.cashin@acm.org>
+Description:
+		(RO) Device status. The state attribute is "up" when the device
+		is ready for I/O and "down" if detected but unusable. The
+		"down,closewait" state shows that the device is still open and
+		cannot come up again until it has been closed.  The "up,kickme"
+		state means that the driver wants to send more commands to the
+		target but found out there were already the max number of
+		commands waiting for a response. It will retry again after being
+		kicked by the periodic timer handler routine.
+
+What:		/sys/block/etherd*/firmware-version
+Date:		Apr, 2005
+KernelVersion:	v2.6.12
+Contact:	Ed L. Cashin <ed.cashin@acm.org>
+Description:
+		(RO) Version of the firmware in the target.
+
+What:		/sys/block/etherd*/payload
+Date:		Dec, 2012
+KernelVersion:	v3.10
+Contact:	Ed L. Cashin <ed.cashin@acm.org>
+Description:
+		(RO) The amount of user data transferred (in bytes) inside each AoE
+		command on the network, network headers excluded.
-- 
2.16.1

Re: [PATCH] aoe: document sysfs interface

[Ed Cashin]

On 02/16/2018 10:39 AM, Aishwarya Pant wrote:
> Documentation has been compiled from git commit logs and descriptions in
> Documentation/aoe/aoe.txt. This should be useful for scripting and
> tracking changes in the ABI.
...
> +What:		/sys/block/etherd*/netif
...
> +Description:
> +		(RO) The name of the network interface on the localhost through
> +		which we are communicating with the remote AoE device.

I'd recommend changing that to, "network interfaces".

Thanks!

-- 
   Ed

Re: [PATCH] aoe: document sysfs interface

[Julia Lawall]

On Fri, 16 Feb 2018, Ed Cashin wrote:

> On 02/16/2018 10:39 AM, Aishwarya Pant wrote:
> > Documentation has been compiled from git commit logs and descriptions in
> > Documentation/aoe/aoe.txt. This should be useful for scripting and
> > tracking changes in the ABI.
> ...
> > +What:		/sys/block/etherd*/netif
> ...
> > +Description:
> > +		(RO) The name of the network interface on the localhost
> > through
> > +		which we are communicating with the remote AoE device.
>
> I'd recommend changing that to, "network interfaces".

Should it then be "names"?

julia

Re: [PATCH] aoe: document sysfs interface

[Ed Cashin]

> On Feb 17, 2018, at 9:37 AM, Julia Lawall <julia.lawall@lip6.fr> wrote:
> 
> 
> 
>> On Fri, 16 Feb 2018, Ed Cashin wrote:
>> 
>>> On 02/16/2018 10:39 AM, Aishwarya Pant wrote:
>>> Documentation has been compiled from git commit logs and descriptions in
>>> Documentation/aoe/aoe.txt. This should be useful for scripting and
>>> tracking changes in the ABI.
>> ...
>>> +What:        /sys/block/etherd*/netif
>> ...
>>> +Description:
>>> +        (RO) The name of the network interface on the localhost
>>> through
>>> +        which we are communicating with the remote AoE device.
>> 
>> I'd recommend changing that to, "network interfaces".
> 
> Should it then be "names"?

Yes, s’il vous plait. Merci.

— 
  Ed