Re: [Announce]: Target_Core_Mod/ConfigFS and LIO-Target v3.0 work

[Nicholas A. Bellinger]

On Sat, 2008-12-13 at 11:23 +0100, Bart Van Assche wrote:
> On Sat, Dec 13, 2008 at 11:08 AM, Nicholas A. Bellinger <nab@xxxxxxxxxx> wrote:
> >> See also:
> >> * February 1, 2008, LIO kernel panic during configuration,
> >> http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/74c8b37f24b84e59/d94c07626bd20521?lnk=gst&q=kernel+panic#d94c07626bd20521.
> >> * February 8, 2008, kernel crash triggered by LIO,
> >> http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/30835aede1028188/5708e16a23367fb4?lnk=gst&q=kernel+crash#5708e16a23367fb4.
> >> * February 13, 2008, LIO target kernel code triggers memory
> >> corruption, http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/ddc1bf7666372972/2150a09f9ed3d1cd?lnk=gst&q=ipoib#2150a09f9ed3d1cd.
> >> * February 18, 2008, LIO target makes entire system hang,
> >> http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/6a76f9efd9409fc5/55bd8840b6a5f757?lnk=gst&q=lio+target+hangs#55bd8840b6a5f757.
> >>
> >
> > I have no idea why why you keep bringing up a minor BUG (completely
> > unrelated to Target_Core_Mod/ConfigFS and LIO-Target v3.0 btw) that was
> > fixed 10 months ago..?  Perhaps if you spent half the time looking at
> > actual lio-core-2.6.git code that you do bringing up minor closed bugs
> > from months ago [ ... ]
> 
> I won't comment on the fact that you consider a kernel crash or a
> system hang as a minor bug.
> 

The problem is that you like to handwave on the technical issues, just
like you are doing here.  :-)  Of course I fix bugs when people report
them, but when people like yourself yell and scream and handwave, it
makes me not want to fix it as quickly if someone wrote a nice and
thoughful email and said 'thank you'.

Anyways, I am not going to debate the development process with you, and
as folks on the LIO-devel list can tell you, I am very quick to produce
patches when a issue is located.  

> I reported four bugs instead of one. Only two of these have been
> reported to be fixed.
> 

Considering you said earlier that you have not actually looked at any
recently LIO code, how could you know these bugs are fixed..?   Back
here in the land of reality, these *TWO* bugs where fixed in back Feb.
One was related to iSCSI discovery, and one related to v2.6.24 kernel
breakage and struct scatterlist->page_link, so what..?  Do you honestly
think handwaving about bugs from 10 months ago will get you anywhere
here..?  If you are so certain these bugs still exist or have any effect
on my upstream work, then please, go ahead and prove it.  No..? I did
not think so.

What does any of this have to do with lio-core-2.6.git,
Target_Core_Mod/ConfigFS and LIO-Target v3.0 btw..?  Are you actually
going to write a thoughtful or relivent comment on the v3.0 design
and/or code, because that would be nice out of your for once.  Otherwise
I am going to ignore you again, just the like conversation where you
said:

"Zero-copy means that data is copied as few times as possible".

when I was attempting to explain the finer pointers of
Target_Core_Mod/ConfigFS design to you and Vlad.  Remember that one..?

http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/8cff61671cd2de6b/37ade00e607dd8c8

Perhaps you should clarify the techincal points I raised in that thread
(that where never answered) when I actually explained *WHY* I ended up
not using SCST as a base for Target_Core_Mod/ConfigFS.  Perhaps folks
would like know why I found SCST Core's memory mapping and allocating
algortihms to be incomplete for my purposes, and that I had to give up
the discussion because neither you or Vlad would acknowledge the issues,
and would rather handwave and steamroll.

Perhaps you would like to address these points here and now, considering
you seem so jazzed on bringing up threads previously threads.  How does
it make sense for you to bring up bugs from 10 months ago that where
fixed then months ago, when you guys don't even want to acknowledge any
techinical issues I bring up to you..? 

Are you going to continue to refuse to acknowledge techincal concerns
about design fundementals about your memory pre-allocation and mapping
logic now that you have posted your patches to LKML..?  Care to address
the issues in public where everyone can see, or would you rather keep
that on the SCST and LIO lists..?

Surely this type of behaviour is frowned upon by a subsystem maintainer.

Best regards,

--nab

> Bart.
> 
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups "Linux-iSCSI.org Target Development" group.
> To post to this group, send email to linux-iscsi-target-dev@xxxxxxxxxxxxxxxx
> To unsubscribe from this group, send email to linux-iscsi-target-dev+unsubscribe@xxxxxxxxxxxxxxxx
> For more options, visit this group at http://groups.google.com/group/linux-iscsi-target-dev?hl=en
> -~----------~----~----~----~------~----~------~--~---
> 

--
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/