> > Please set:
 > > CONFIG_USB_DEVICE_CLASS=y
 > > there should be no patch needed. The option already defaults to yes in
 > > the latest kernel.
 
 > USB device class-devices (DEPRECATED) (USB_DEVICE_CLASS) [N/y/?] (NEW) y
 >  
 > Why is it deprecated if (nut/usb/etc fails) when its not used? Does nut  
 > or udev have to play catch-up to adhere to the new kernel changes?
 
 The usb_device class was created to replace the /proc usbfs device nodes
 which can't handle access-control-lists for logged-in users.
 The device nodes worked fine, but plugging into the device events to
 trigger userspace device handling created a bunch of problems with the
 event timing. We have several open bugs, that can't be fixed properly
 with using the usb_device-devices.
 
 We changed the driver core to allow us to export device nodes directly
 by bus-devices, instead of forcing us to create artificial devices like
 the usb_device-class, just to export device nodes to userspace. If
 systems set CONFIG_USB_DEVICE_CLASS=n (which defaults to yes now), one
 new udev rule is needed, to replace the functionality of the
 usb_device-class devices. Also the next version of HAL (which is where
 nut receives the events from) will depend on the new device nodes, and
 will not longer use usb_device devices. No changes to nut should be
 needed.
 
 Thanks,
 Kay

So what's a pasty white nerd to do? You can't go out on the beach, because  
the goodlooking people will laugh at you, and kick sand in your face.
[...]
But now you _can_ do something: you can download the latest -rc kernel,  
and smile smugly to yourself, knowing that you are running the latest and  
greatest on your machine. And suddenly it doesn't even matter that summer  
is coming, because you can just sit in the basement, and close the blinds,  
and bask in the warm light from your LCD, rather than the harsh glare of  
the daystar..
[...]
So stop worrying about those dangerous ultraviolet rays, and instead get  
your Vitamin D in the form God (and the pharmaceutical industry) intended:  
small easily swallowed pills. Beaches are overrated anyway, the sand gets  
into the laptop fan and soon it won't work.
 
May you have a great summer,
 
                Linus

By popular demand, here is release -v20 of the CFS scheduler. It is a  
full backport of the latest & greatest v2.6.23-rc3 CFS code to  
v2.6.22.5, v2.6.21.7 and v2.6.20.16. The patches can be downloaded from  
the usual place:
 
    http://people.redhat.com/mingo/cfs-scheduler/
 
Changes since -v19: there have been lots of small regression fixes,
speedups, debug enhancements and tidy-ups - many of which can be
user-visible.
 
See the detailed shortlog below - there are nearly 100 changes - they do  
add up to a significant total linecount change. There was no crash bug  
or hang bug found in the CFS code since v19 was released. (in fact the  
last crash/hang bug in CFS was found and fixed in v7, more than 3 months  
ago, and even that crash only happened in an uncommon sw-suspend setup,  
not during normal use. So CFS has turned out to be a pretty robust  
codebase.)
 
Nevertheless, if you had any problems (performance or behavioral) with
v19 it's worth checking v20 out - and if v19 worked great for you it's
worth checking out that v20 still works great =B-)
 
Also, the backported CFS scheduler enables people to test suspected  
scheduler regressions on older codebases too, to filter out the effects  
of other changes.
 
v20 has been build and boot tested on both 32-bit and 64-bit x86, ontop  
of all 3 backport kernel bases. As usual, any sort of feedback,  
bugreport, fix and suggestion is more than welcome!
 
        Ingo
 
------------------->
Adrian Bunk (1):
      sched: make global code static
 
Al Viro (1):
      take sched_debug.c out of nasal demon territory
 
Alexey Dobriyan (2):
      Fix leaks on /proc/{*/sched,sched_debug,timer_list,timer_stats}
      sched: remove binary sysctls from kernel.sched_domain
 
Avi Kivity (1):
      sched: arch preempt notifier mechanism
 
Ingo Molnar (68):
      sched: make cpu_clock() not use the rq clock
      sched: remove cache_hot_time
      sched: calc_delta_mine(): use fixed limit
      sched: uninline calc_delta_mine()
      sched: uninline inc/dec_nr_running()
      sched: ->task_new cleanup
      sched: move load-calculation functions
      sched: add schedstat_set() API
      sched: use schedstat_set() API
      sched: reduce debug code
      sched: batch sleeper bonus
      sched: reorder update_cpu_load(rq) with the ->task_tick() call
      sched: uninline rq_clock()
      sched: schedule() speedup
      sched: clean up delta_mine
      sched: delta_exec accounting fix
      sched: add [__]update_rq_clock(rq)
      sched: eliminate rq_clock() use
      sched: remove rq_clock()
      sched: eliminate __rq_clock() use
      sched: remove __rq_clock()
      sched: remove 'now' use from assignments
      sched: remove the 'u64 now' parameter from print_cfs_rq()
      sched: remove the 'u64 now' parameter from update_curr()
      sched: remove the 'u64 now' parameter from update_stats_wait_start()
      sched: remove the 'u64 now' parameter from update_stats_enqueue()
      sched: remove the 'u64 now' parameter from __update_stats_wait_end()
      sched: remove the 'u64 now' parameter from update_stats_wait_end()
      sched: remove the 'u64 now' parameter from update_stats_curr_start()
      sched: remove the 'u64 now' parameter from update_stats_dequeue()
      sched: remove the 'u64 now' parameter from update_stats_curr_end()
      sched: remove the 'u64 now' parameter from __enqueue_sleeper()
      sched: remove the 'u64 now' parameter from enqueue_sleeper()
      sched: remove the 'u64 now' parameter from enqueue_entity()
      sched: remove the 'u64 now' parameter from dequeue_entity()
      sched: remove the 'u64 now' parameter from set_next_entity()
      sched: remove the 'u64 now' parameter from pick_next_entity()
      sched: remove the 'u64 now' parameter from put_prev_entity()
      sched: remove the 'u64 now' parameter from update_curr_rt()
      sched: remove the 'u64 now' parameter from ->enqueue_task()
      sched: remove the 'u64 now' parameter from ->dequeue_task()
      sched: remove the 'u64 now' parameter from ->pick_next_task()
      sched: remove the 'u64 now' parameter from pick_next_task()
      sched: remove the 'u64 now' parameter from ->put_prev_task()
      sched: remove the 'u64 now' parameter from ->task_new()
      sched: remove the 'u64 now' parameter from update_curr_load()
      sched: remove the 'u64 now' parameter from inc_load()
      sched: remove the 'u64 now' parameter from dec_load()
      sched: remove the 'u64 now' parameter from inc_nr_running()
      sched: remove the 'u64 now' parameter from dec_nr_running()
      sched: remove the 'u64 now' parameter from enqueue_task()
      sched: remove the 'u64 now' parameter from dequeue_task()
      sched: remove the 'u64 now' parameter from deactivate_task()
      sched: remove the 'u64 now' local variables
      sched debug: remove the 'u64 now' parameter from print_task()/_rq()
      sched: move the __update_rq_clock() call to scheduler_tick()
      sched: remove __update_rq_clock() call from entity_tick()
      sched: clean up set_curr_task_fair()
      sched: optimize activate_task()
      sched: optimize update_rq_clock() calls in the load-balancer
      sched: make the multiplication table more accurate
      sched: round a bit better
      sched: fix update_stats_enqueue() reniced codepath
      sched: refine negative nice level granularity
      sched: improve rq-clock overflow logic
      sched: fix typo in the FAIR_GROUP_SCHED branch
      sched debug: dont print kernel address in /proc/sched_debug
      sched: fix sleeper bonus
 
Josh Triplett (2):
      sched: mark sysrq_sched_debug_show() static
      sched: mark print_cfs_stats static
 
Nick Piggin (1):
      sched: debug feature - make the sched-domains tree runtime-tweakable
 
Oleg Nesterov (1):
      sched: run_rebalance_domains: s/SCHED_IDLE/CPU_IDLE/
 
Peter Williams (3):
      sched: tidy up left over smpnice code
      sched: simplify move_tasks()
      sched: fix bug in balance_tasks()
 
Randy Dunlap (1):
      sched: fix kernel-doc warnings
 
Satoru Takeuchi (1):
      sched: remove unused rq->load_balance_class
 
Ulrich Drepper (1):
      sched: clean up sched_getaffinity()