On Thu, Aug 11, 2011 at 11:05:42AM +0300, Kalle Valo wrote:

Hi Stephen,

On 08/11/2011 08:55 AM, Stephen Rothwell wrote:

quoted

After merging the final tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/staging/built-in.o: In function `aggr_recv_delba_req_evt':
(.opd+0x23a60): multiple definition of `aggr_recv_delba_req_evt'
drivers/net/built-in.o:(.opd+0x79980): first defined here
drivers/staging/built-in.o: In function `.wlan_node_reclaim':
(.text+0x303360): multiple definition of `.wlan_node_reclaim'
drivers/net/built-in.o:(.text+0xdd0618): first defined here
drivers/staging/built-in.o: In function `aggr_init':
(.opd+0x23a00): multiple definition of `aggr_init'
drivers/net/built-in.o:(.opd+0x79968): first defined here
drivers/staging/built-in.o: In function `.aggr_recv_delba_req_evt':
(.text+0x302614): multiple definition of `.aggr_recv_delba_req_evt'
drivers/net/built-in.o:(.text+0xdc8b60): first defined here
drivers/staging/built-in.o: In function `.wlan_node_table_cleanup':
(.text+0x303888): multiple definition of `.wlan_node_table_cleanup'
drivers/net/built-in.o:(.text+0xdd0a6c): first defined here

You get the idea. :-(  So, the ath6kl driver now exists in the wireless
tree and the staging tree.

I applied this patch for today (I have elided the part from "rm -rf
drivers/staging/ath6kl"):

Oh man, sorry about this (again). I did compile both drivers at the same
time, but only as modules and didn't realise the conflicts also with the
staging driver.

Yes, the cleaned up version of ath6kl is now in wireless-next and the
staging version can be removed in 3.2. Greg, how do you want to handle
removal of the staging version? Should I send a patch to you removing
it? Or should I send the patch to John?

Just tell me.  Should I really delete drivers/staging/ath6kl/ for the
3.2 kernel release?  If so, I'll gladly do so in my tree so linux-next
will be fine for issues like this in the future.

greg k-h