be2iscsi driver

be2iscsi driver

[Jayamohan Kallickal]

Hi

   Thanks to Eike and Mike Christie for the reviews.
   This patch takes care of the many issues raised.

   I request you to pl consider  this for inclusion
   in the Linux Kernel.

  This patch supports blk-iopoll and is pulled from 

  git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-2.6-block.git 
 								 blk-iopoll

  The ServerEngines be2iscsi driver works in Full Hardware Offload
  mode utilizing the hooks provided by  open-iscsi layer.
 
  The series of  patches are described below:

  [PATCH 1/8]: The completion of IO's is done using a different path
 
  [PATCH 2/8]: Adding checks for iscsi_queuecommand/xmit_task  to handle
	       Tx Suspended state.

  [PATCH 3/8]: Modified scsi Layer and be2iscsi Makefile/Kconfig
	   	    				 to add be2iscsi

  [PATCH 4/8]: This file contains all routines for Initiallization/Teardown,
  	       registering with upper layer, IO Handling, allocation/free etc.

  [PATCH 5/8]: These are the low level routines to talk to the chip for
  	   	    				Management commands.

  [PATCH 6/8]: The iscsi hooks to interact with open-iscsi layer.

  [PATCH 7/8]: The Firmware Commands that are used for Management action
	   	    		  like setup /tear down a connection etc.

  [PATCH 8/8]: The modified MAINTAINERS file with be2iscsi added.


Thank You

Jayamohan Kallickal

Signed-off-by:  Jayamohan Kallickal <jayamohank@serverengines.com>

---

be2iscsi driver

[Jayamohan Kallickal]

Hi

   Thanks to John Stoffel for the review. 
   
   This patch takes care of the concern raised by Mike Christie that if an
   allocation fails then a lower size is not tried.	

   I request you to pl consider  this for inclusion
   in the Linux Kernel.

  This patch supports blk-iopoll and is pulled from Jens Axboe's tree at 

  git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-2.6-block.git 
 								 blk-iopoll

  The ServerEngines be2iscsi driver works in Full Hardware Offload
  mode utilizing the hooks provided by  open-iscsi layer.
 
  The series of  patches are described below:

  [PATCH 1/8]: The completion of IO's is done using a different path
 
  [PATCH 2/8]: Adding checks for iscsi_queuecommand/xmit_task  to handle
	       Tx Suspended state.

  [PATCH 3/8]: Modified scsi Layer and be2iscsi Makefile/Kconfig
	   	    				 to add be2iscsi

  [PATCH 4/8]: This file contains all routines for Initiallization/Teardown,
  	       registering with upper layer, IO Handling, allocation/free etc.

  [PATCH 5/8]: These are the low level routines to talk to the chip for
  	   	    				Management commands.

  [PATCH 6/8]: The iscsi hooks to interact with open-iscsi layer.

  [PATCH 7/8]: The Firmware Commands that are used for Management action
	   	    		  like setup /tear down a connection etc.

  [PATCH 8/8]: The modified MAINTAINERS file with be2iscsi added.


Thank You

Jayamohan Kallickal

Signed-off-by:  Jayamohan Kallickal <jayamohank@serverengines.com>

---

be2iscsi driver

[Jayamohan Kallickal]

Hi
  
   This patch contains  a fix for powerpc compile issue  and has
    been tested for compile on powerpc with allyesconfig option.	
 
   I request you to pl consider  this for inclusion
   in the Linux Kernel.

  This patch requires  Mike's Patch 
  (commit f54b4b17046a762b80787edda3bb25e84ff0809c and
   240e4e54b2b324681cda93b10e750ff9d0251b41 on 
  git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6.git) 
  Since Mike's patches have already been applied, I am not including them here.

  This patch supports blk-iopoll and is pulled from Jens Axboe's tree at 

  git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-2.6-block.git 
 								 blk-iopoll
 
  The series of  patches are described below:

  [PATCH 1/6]: Modified scsi Layer and be2iscsi Makefile/Kconfig
	   	    				 to add be2iscsi

  [PATCH 2/6]: This file contains all routines for Initiallization/Teardown,
  	       registering with upper layer, IO Handling, allocation/free etc.

  [PATCH 3/6]: These are the low level routines to talk to the chip for
  	   	    				Management commands.

  [PATCH 4/6]: The iscsi hooks to interact with open-iscsi layer.

  [PATCH 5/6]: The Firmware Commands that are used for Management action
	   	    		  like setup /tear down a connection etc.

  [PATCH 6/6]: The modified MAINTAINERS file with be2iscsi added.


Thank You

Jayamohan Kallickal

Signed-off-by:  Jayamohan Kallickal <jayamohank@serverengines.com>

---

Re: be2iscsi driver

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 623 bytes --]

Hi Jayamohan,

On Tue, 8 Sep 2009 14:05:58 +0530 Jayamohan Kallickal <jayamohank@serverengines.com> wrote:
>
>    This patch contains  a fix for powerpc compile issue  and has
>     been tested for compile on powerpc with allyesconfig option.	

Thanks for that.

>   The series of  patches are described below:

This series should be applied in an order that allows the kernel as a
whole to keep building at each step.  The easiest way to guarantee that
would be to put your patch 1 as the last one.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: be2iscsi driver

[James Bottomley]

On Tue, 2009-09-08 at 20:06 +1000, Stephen Rothwell wrote:
> Hi Jayamohan,
> 
> On Tue, 8 Sep 2009 14:05:58 +0530 Jayamohan Kallickal <jayamohank@serverengines.com> wrote:
> >
> >    This patch contains  a fix for powerpc compile issue  and has
> >     been tested for compile on powerpc with allyesconfig option.	
> 
> Thanks for that.
> 
> >   The series of  patches are described below:
> 
> This series should be applied in an order that allows the kernel as a
> whole to keep building at each step.  The easiest way to guarantee that
> would be to put your patch 1 as the last one.

Actually, I just folded these patches in to one last time.  For new
drivers, it really makes no sense to apply anything other than all in
one go.

However, for code inspection purposes, this is pretty useless.  The fix
is buried somewhere in several thousand lines of code.

Jayamohan, could you please just send a patch for what's currently in
the tree that fixes the issue, so we can better verify by inspection
that it's correct?  I'll fold it into the driver submit, so there won't
be any point in the tree where there's a ppc build breakage, but I would
like to see what the fix is without having to rebuild this patch and run
interdiff.

Thanks,

James