Re: linux-next: Tree for May 27 (infiniband: qib)

Re: linux-next: Tree for May 27 (infiniband: qib)

[Randy Dunlap]

On Thu, 27 May 2010 15:26:28 +1000 Stephen Rothwell wrote:

> Hi all,
> 
> We are in the merge window again.  I remind you all not to add stuff for
> 2.6.36 to your linux-next trees until after 2.6.35-rc1.
> 
> Changes since 20100526:


when CONFIG_INFINIBAND_QIB=y and CONFIG_DCA=m:

drivers/built-in.o: In function `qib_update_sdma_dca':
qib_iba7322.c:(.text+0x409219): undefined reference to `dca3_get_tag'
drivers/built-in.o: In function `qib_update_rhdrq_dca':
qib_iba7322.c:(.text+0x40a1f9): undefined reference to `dca3_get_tag'
drivers/built-in.o: In function `qib_setup_7322_cleanup':
qib_iba7322.c:(.text+0x40f91f): undefined reference to `dca_remove_requester'
drivers/built-in.o: In function `qib_init_iba7322_funcs':
(.text+0x419b6f): undefined reference to `dca_add_requester'


However, it looks like qib needs to handle DCA config in a way that
is similar to how it is handled in drivers/net/{myri10ge,igb,ixgbe}/
instead of assuming that DCA is enabled.


And please fix the linux-next 2010-may-25 reported qib problem:
  http://lkml.org/lkml/2010/5/25/321

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: linux-next: Tree for May 27 (infiniband: qib)

[Roland Dreier]

 > However, it looks like qib needs to handle DCA config in a way that
 > is similar to how it is handled in drivers/net/{myri10ge,igb,ixgbe}/
 > instead of assuming that DCA is enabled.

Looks like we're just going to rip out DCA support for now.

 > And please fix the linux-next 2010-may-25 reported qib problem:
 >   http://lkml.org/lkml/2010/5/25/321

I think that should be fixed in my for-next branch already (at least I
have a patch from Ralph called "IB/qib: Fix undefined symbol error when
CONFIG_PCI_MSI=n" in there).

 - R.
-- 
Roland Dreier <rolandd@cisco.com> || For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html

Re: linux-next: Tree for May 27 (infiniband: qib)

[Randy Dunlap]

On 05/27/10 14:45, Roland Dreier wrote:
>  > However, it looks like qib needs to handle DCA config in a way that
>  > is similar to how it is handled in drivers/net/{myri10ge,igb,ixgbe}/
>  > instead of assuming that DCA is enabled.
> 
> Looks like we're just going to rip out DCA support for now.
> 
>  > And please fix the linux-next 2010-may-25 reported qib problem:
>  >   http://lkml.org/lkml/2010/5/25/321
> 
> I think that should be fixed in my for-next branch already (at least I
> have a patch from Ralph called "IB/qib: Fix undefined symbol error when
> CONFIG_PCI_MSI=n" in there).

Sounds good.  Was is posted to linux-rdma?

thanks,
-- 
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: linux-next: Tree for May 27 (infiniband: qib)

[Ralph Campbell]

On Thu, 2010-05-27 at 15:06 -0700, Randy Dunlap wrote:
> On 05/27/10 14:45, Roland Dreier wrote:
> >  > However, it looks like qib needs to handle DCA config in a way that
> >  > is similar to how it is handled in drivers/net/{myri10ge,igb,ixgbe}/
> >  > instead of assuming that DCA is enabled.
> > 
> > Looks like we're just going to rip out DCA support for now.
> > 
> >  > And please fix the linux-next 2010-may-25 reported qib problem:
> >  >   http://lkml.org/lkml/2010/5/25/321
> > 
> > I think that should be fixed in my for-next branch already (at least I
> > have a patch from Ralph called "IB/qib: Fix undefined symbol error when
> > CONFIG_PCI_MSI=n" in there).
> 
> Sounds good.  Was is posted to linux-rdma?
> 
> thanks,

Yes, I posted it to linux-rdma.

http://www.spinics.net/lists/linux-rdma/msg04140.html

http://www.spinics.net/lists/linux-rdma/msg04187.html