[RFC,0/3] target: remove some unused stats

Message

David Disseldorp Oct. 17, 2018, 3:48 p.m. UTC

This patchset removes a couple of unused error stat counters and a
redundant cumulative counter.
I've tagged this patchset RFC, as it may be considered a kernel<->user
(configfs) API change.

Cheers, David
---
 drivers/target/iscsi/iscsi_target_stat.c | 40 ----------------------------------------
 include/target/iscsi/iscsi_target_stat.h |  4 ----
 2 files changed, 44 deletions(-)

Comments

David Disseldorp Oct. 29, 2018, 11 p.m. UTC | #1

On Wed, 17 Oct 2018 17:48:17 +0200, David Disseldorp wrote:

> This patchset removes a couple of unused error stat counters and a
> redundant cumulative counter.
> I've tagged this patchset RFC, as it may be considered a kernel<->user
> (configfs) API change.

Ping, any thoughts on this patchset?

Cheers, David

Mike Christie Oct. 30, 2018, 3:54 p.m. UTC | #2

On 10/29/2018 06:00 PM, David Disseldorp wrote:
> On Wed, 17 Oct 2018 17:48:17 +0200, David Disseldorp wrote:
> 
>> This patchset removes a couple of unused error stat counters and a
>> redundant cumulative counter.
>> I've tagged this patchset RFC, as it may be considered a kernel<->user
>> (configfs) API change.
> 
> Ping, any thoughts on this patchset?
> 

I think these stats were supposed to match the iSCSI MIB definitions. It
was not clear why we didn't just fix them so they report the correct values?

David Disseldorp Nov. 16, 2018, 2:59 p.m. UTC | #3

Hi Mike

On Tue, 30 Oct 2018 10:54:07 -0500, Mike Christie wrote:

> >> This patchset removes a couple of unused error stat counters and a
> >> redundant cumulative counter.
> >> I've tagged this patchset RFC, as it may be considered a kernel<->user
> >> (configfs) API change.  
> > 
> > Ping, any thoughts on this patchset?
> >   
> 
> I think these stats were supposed to match the iSCSI MIB definitions. It
> was not clear why we didn't just fix them so they report the correct values?

I finally got a chance to look through rfc#4544. It does indeed look
like these counters correspond to:
IscsiInstanceSsnErrorStatsEntry ::= SEQUENCE {
    iscsiInstSsnDigestErrors       Counter32,
    iscsiInstSsnCxnTimeoutErrors   Counter32,
    iscsiInstSsnFormatErrors       Counter32
}

There doesn't appear to be anything stopping us from plumbing them into
the corresponding error paths, so I'll withdraw this patchset.

Cheers, David