On 11/24/2011 10:50 AM, Randy Dunlap wrote: > On 11/23/2011 10:43 PM, Stephen Rothwell wrote: >> Hi all, >> >> Changes since 20111123: > > > When CONFIG_EXOFS is not enabled: > > fs/built-in.o: In function `objio_alloc_io_state': > objio_osd.c:(.text+0xcb525): undefined reference to `ore_get_rw_state' > fs/built-in.o: In function `_write_done': > objio_osd.c:(.text+0xcb58d): undefined reference to `ore_check_io' > fs/built-in.o: In function `_read_done': > objio_osd.c:(.text+0xcb5bf): undefined reference to `ore_check_io' > fs/built-in.o: In function `objio_alloc_lseg': > (.text+0xcb9a4): undefined reference to `ore_verify_layout' > fs/built-in.o: In function `objio_free_result': > (.text+0xcbb0e): undefined reference to `ore_put_io_state' > fs/built-in.o: In function `objio_read_pagelist': > (.text+0xcbc9b): undefined reference to `ore_read' > fs/built-in.o: In function `objio_write_pagelist': > (.text+0xcbd4e): undefined reference to `ore_write' > > > Full randconfig file is attached. > Thanks Randy. I completely missed that one. The missing enablement is not CONFIG_EXOFS (that, I tested). It is the CONFIG_MISC_FILESYSTEMS. Because CONFIG_MISC_FILESYSTEMS is not set then exofs/Kconfig is not examined and it cannot do it's magic in selecting everything. I'm posting a patch as a separate mail reply. What I did was revert to the old way of forcing all users of the ORE to select everything that it requires. Instead of the elegant way it was. The other choice is, of course, to move ore to the lib/ directory but that one I fear even more. So here ... Thanks Boaz -- To unsubscribe from this list: send the line "unsubscribe linux-next" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
