Re: linux-next: Tree for January 19 (vhost_net)

Re: linux-next: Tree for January 19 (vhost_net)

[Randy Dunlap]

On Tue, 19 Jan 2010 18:10:41 +1100 Stephen Rothwell wrote:

> Hi all,
> 
> Changes since 20100118:


When CONFIG_TUN is not enabled:

drivers/built-in.o: In function `get_socket':
net.c:(.text+0x2899a3): undefined reference to `tun_get_socket'


so VHOST_NET can either depend on TUN or select TUN, but if it
selects TUN, it will also need to select CRC32.

---
~Randy

Re: linux-next: Tree for January 19 (vhost_net)

[Michael S. Tsirkin]

On Tue, Jan 19, 2010 at 10:41:39AM -0800, Randy Dunlap wrote:
> On Tue, 19 Jan 2010 18:10:41 +1100 Stephen Rothwell wrote:
> 
> > Hi all,
> > 
> > Changes since 20100118:
> 
> 
> When CONFIG_TUN is not enabled:
> 
> drivers/built-in.o: In function `get_socket':
> net.c:(.text+0x2899a3): undefined reference to `tun_get_socket'
> 
> 
> so VHOST_NET can either depend on TUN or select TUN, but if it
> selects TUN, it will also need to select CRC32.
> 
> ---
> ~Randy

Oh, I fixed this already but then I lost the patch:
http://lkml.org/lkml/2009/11/10/137

Sorry about that, I'll repost the fix immediately.

-- 
MST

Re: linux-next: Tree for January 19 (vhost_net)

[Michael S. Tsirkin]

On Tue, Jan 19, 2010 at 10:41:39AM -0800, Randy Dunlap wrote:
> On Tue, 19 Jan 2010 18:10:41 +1100 Stephen Rothwell wrote:
> 
> > Hi all,
> > 
> > Changes since 20100118:
> 
> 
> When CONFIG_TUN is not enabled:
> 
> drivers/built-in.o: In function `get_socket':
> net.c:(.text+0x2899a3): undefined reference to `tun_get_socket'
> 
> 
> so VHOST_NET can either depend on TUN or select TUN, but if it
> selects TUN, it will also need to select CRC32.
> 
> ---
> ~Randy

Just to clarify sure, the broken configuration is VHOST=y TUN=m, I sent
a fixup patch to forbid that configuration.  But both VHOST=y TUN=n and
VHOST=m TUN=n build fine for me on linux-next (commit
af2b07e7e0a9c6bf82ce4d6a77f0412ae54f575b).  Randy, could you please
confirm that VHOST=y TUN=m is what you tested, or alternatively make
your .config available to me so I can reproduce the problem?

Thanks!

-- 
MST

Re: linux-next: Tree for January 19 (vhost_net)

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 371 bytes --]

Hi Michael,

On Wed, 20 Jan 2010 12:16:44 +0200 "Michael S. Tsirkin" <mst@redhat.com> wrote:
>
> Oh, I fixed this already but then I lost the patch:
> http://lkml.org/lkml/2009/11/10/137
> 
> Sorry about that, I'll repost the fix immediately.

Thanks.

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

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: linux-next: Tree for January 19 (vhost_net)

[Randy Dunlap]

On Wed, 20 Jan 2010 13:33:41 +0200 Michael S. Tsirkin wrote:

> On Tue, Jan 19, 2010 at 10:41:39AM -0800, Randy Dunlap wrote:
> > On Tue, 19 Jan 2010 18:10:41 +1100 Stephen Rothwell wrote:
> > 
> > > Hi all,
> > > 
> > > Changes since 20100118:
> > 
> > 
> > When CONFIG_TUN is not enabled:
> > 
> > drivers/built-in.o: In function `get_socket':
> > net.c:(.text+0x2899a3): undefined reference to `tun_get_socket'
> > 
> > 
> > so VHOST_NET can either depend on TUN or select TUN, but if it
> > selects TUN, it will also need to select CRC32.
> > 
> > ---
> > ~Randy
> 
> Just to clarify sure, the broken configuration is VHOST=y TUN=m, I sent
> a fixup patch to forbid that configuration.  But both VHOST=y TUN=n and
> VHOST=m TUN=n build fine for me on linux-next (commit
> af2b07e7e0a9c6bf82ce4d6a77f0412ae54f575b).  Randy, could you please
> confirm that VHOST=y TUN=m is what you tested, or alternatively make
> your .config available to me so I can reproduce the problem?

Hi Michael,

You are correct, VHOST_NET=y, TUN=m.

Thanks.
---
~Randy