2.5.53 make modules_install problem

2.5.53 make modules_install problem

[Brad Tilley]

I've finally got a config file that actually builds with the beta kernels, but 
when it gets to 'make modules_install' I get these error after it runs for a 
few minutes:

--------------------------

depmod: *** Unresolved symbols in 
/lib/modules/2.5.53/kernel/net/bridge/bridge.ko
depmod:         nf_unregister_hook
depmod:         _mmx_memcpy
depmod:         eth_type_trans
depmod:         __kfree_skb
depmod:         br_ioctl_hook
depmod:         alloc_skb
depmod:         __pskb_pull_tail
depmod:         ether_setup
depmod:         nf_hooks
depmod:         skb_under_panic
depmod:         dev_set_promiscuity
depmod:         kmalloc
depmod:         unregister_netdevice_notifier
depmod:         __up_wakeup
depmod:         ip_route_output_key
depmod:         netdev_finish_unregister
depmod:         unregister_netdev
depmod:         ip_route_input
depmod:         copy_to_user
depmod:         del_timer
depmod:         register_netdev
depmod:         dev_queue_xmit
depmod:         kfree
depmod:         ___pskb_trim
depmod:         nf_register_hook
depmod:         netif_rx
depmod:         skb_over_panic
depmod:         br_handle_frame_hook
depmod:         skb_clone
depmod:         register_netdevice_notifier
depmod:         copy_from_user
depmod:         jiffies
depmod:         dev_get_by_index
depmod:         printk
depmod:         add_timer
depmod:         irq_stat
depmod:         do_softirq
depmod:         nf_hook_slow
depmod:         __down_failed
depmod:         __dev_get_by_name
depmod: *** Unresolved symbols in 
/lib/modules/2.5.53/kernel/net/decnet/decnet.ko
depmod:         rtnetlink_links
depmod:         neigh_create
depmod:         send_sig
depmod:         sock_wfree
depmod:         __down_failed_trylock
depmod:         _mmx_memcpy
depmod:         dev_load
depmod:         sk_run_filter
depmod:         __get_user_2
depmod:         __get_user_4
depmod:         schedule_timeout
depmod:         __wake_up
depmod:         __kfree_skb
depmod:         alloc_skb
depmod:         sock_no_socketpair
depmod:         schedule
depmod:         nf_hooks
depmod:         skb_under_panic
depmod:         kmalloc
depmod:         skb_realloc_headroom
depmod:         register_gifconf
depmod:         unregister_netdevice_notifier
depmod:         __up_wakeup
depmod:         __rta_fill
depmod:         create_proc_entry
depmod:         add_wait_queue_exclusive
depmod:         rtnl_sem
depmod:         dev_mc_delete
depmod:         sock_unregister
depmod:         __get_free_pages
depmod:         netdev_finish_unregister
depmod:         neigh_table_clear
depmod:         default_wake_function
depmod:         __lock_sock
depmod:         neigh_destroy
depmod:         memcpy_toiovec
depmod:         remove_wait_queue
depmod:         sk_alloc
depmod:         neigh_parms_alloc
depmod:         __dst_free
depmod:         sysctl_intvec
depmod:         skb_copy
depmod:         panic
depmod:         copy_to_user
depmod:         __release_sock
depmod:         sysctl_string
depmod:         proc_dostring
depmod:         __kill_fasync
depmod:         netlink_set_err
depmod:         neigh_parms_release
depmod:         dev_remove_pack
depmod:         kmem_cache_create
depmod:         nf_setsockopt
depmod:         netlink_broadcast
depmod:         proc_dointvec
depmod:         unregister_sysctl_table
depmod:         del_timer
depmod:         net_ratelimit
depmod:         sk_free
depmod:         dev_mc_upload
depmod:         mod_timer
depmod:         dev_queue_xmit
depmod:         dst_alloc
depmod:         nf_getsockopt
depmod:         datagram_poll
depmod:         kfree
depmod:         ___pskb_trim
depmod:         dev_get_by_name
depmod:         remove_proc_entry
depmod:         __dev_get_by_index
depmod:         dev_base
depmod:         sock_no_sendpage
depmod:         skb_over_panic
depmod:         netlink_unicast
depmod:         add_wait_queue
depmod:         proc_net
depmod:         skb_clone
depmod:         memcpy_fromiovec
depmod:         rtnl_unlock
depmod:         sprintf
depmod:         kmem_cache_destroy
depmod:         rtnl
depmod:         dev_add_pack
depmod:         register_netdevice_notifier
depmod:         copy_from_user
depmod:         jiffies
depmod:         num_physpages
depmod:         dev_get_by_index
depmod:         rtnl_lock
depmod:         proc_dointvec_minmax
depmod:         printk
depmod:         add_timer
depmod:         irq_stat
depmod:         sock_register
depmod:         neigh_table_init
depmod:         dev_mc_add
depmod:         dev_ioctl
depmod:         register_sysctl_table
depmod:         do_softirq
depmod:         nf_hook_slow
depmod:         neigh_lookup
depmod:         sock_no_mmap
depmod:         dst_destroy
depmod:         neigh_ifdown
depmod:         sock_init_data
depmod:         sock_rfree
depmod:         loopback_dev
depmod:         __dev_get_by_name
depmod: *** Unresolved symbols in 
/lib/modules/2.5.53/kernel/sound/soundcore.ko
depmod:         kmalloc
depmod:         unregister_chrdev
depmod:         register_chrdev
depmod:         vfree
depmod:         sys_close
depmod:         request_module
depmod:         kfree
depmod:         vmalloc
depmod:         sprintf
depmod:         printk
make: *** [_modinst_post] Error 1

--------------------------------

What must I do to fix this? Please CC me as I'm not on the list.

Thanks,
Brad

Re: 2.5.53 make modules_install problem

[Alex Goddard]

On Fri, 27 Dec 2002, Brad Tilley wrote:

[Snip]

> What must I do to fix this?

Obtain the newest module-init-tools from:
http://www.kernel.org/pub/linux/kernel/people/rusty/modules

You may also find it helpful in the future to find a lkml archive 
( http://groups.google.com keeps one under linux.kernel, IIRC.  I use 
http://marc.theaimsgroup.com ), and search it for any problem you're 
having.  I mention this only because someone reports this "problem" about 
once a week, so it's been answered a bunch already.

I don't mean that as a flame or anything.  But if whatever problem you're
having has already been discussed, and addressed before, a little research
on your part will get you your answer more quickly than waiting for
someone on the lkml to answer it (which may never happen).

-- 
Alex Goddard
agoddard@purdue.edu

Re: 2.5.53 make modules_install problem

[carbonated beverage]

On Fri, Dec 27, 2002 at 05:56:17PM +0000, Alex Goddard wrote:
> I don't mean that as a flame or anything.  But if whatever problem you're
> having has already been discussed, and addressed before, a little research
> on your part will get you your answer more quickly than waiting for
> someone on the lkml to answer it (which may never happen).

Of course, updating Documentation/Changes would help, too. :)

See patches I've posted. :)

<speaking too soon>
(I've gotten *one* bit of feedback on it, so I'm asusming no protest means
no problems.)
</speaking too soon>

-- DN
Daniel

Re: 2.5.53 make modules_install problem

[Brad Tilley]

> Obtain the newest module-init-tools from:
> http://www.kernel.org/pub/linux/kernel/people/rusty/modules
>

That package fixed the problem. Thank you for telling me about it.

> You may also find it helpful in the future to find a lkml archive
> ( http://groups.google.com keeps one under linux.kernel, IIRC.  I use
> http://marc.theaimsgroup.com ), and search it for any problem you're
> having.  I mention this only because someone reports this "problem" about
> once a week, so it's been answered a bunch already.
>
> I don't mean that as a flame or anything.  But if whatever problem you're
> having has already been discussed, and addressed before, a little research
> on your part will get you your answer more quickly

I'm use to the 64 bit, compiled the compiler, tough guy, kernel hacker 
attitude. So, now offense was taken ;-)

Re: 2.5.53 make modules_install problem

[khromy]

On Fri, Dec 27, 2002 at 03:43:42PM -0800, carbonated beverage wrote:
> On Fri, Dec 27, 2002 at 05:56:17PM +0000, Alex Goddard wrote:
> > I don't mean that as a flame or anything.  But if whatever problem you're
> > having has already been discussed, and addressed before, a little research
> > on your part will get you your answer more quickly than waiting for
> > someone on the lkml to answer it (which may never happen).
> 
> Of course, updating Documentation/Changes would help, too. :)

Yep, that'd be nice..

-- 
L1:	khromy		;khromy(at)lnuxlab.ath.cx