This may be related to one of the stack dumps I reported earlier as occurring 
when booting the 4/30 linux-next tree...

During two attempted boots of the 5/5 tree using the same config as the 4/30 
tree the system locked solid at "testing tracer ftrace".

The 4/30 tree produced the following dump (the 5/5 tree locks solid as soon as 
the test is started):
[   10.238230] Testing tracer ftrace: <4>------------[ cut here ]------------
[   10.356558] WARNING: at kernel/lockdep.c:2712 check_flags+0xfb/0x180()
[   10.356726] Modules linked in:
[   10.356915] Pid: 1, comm: swapper Not tainted 2.6.25-next-20080430 #2
[   10.357718]
[   10.357722] Call Trace:
[   10.357939]  <IRQ>  [<ffffffff8023d5f4>] warn_on_slowpath+0x64/0x90
[   10.358174]  [<ffffffff8025f789>] ? tick_get_tick_sched+0x9/0x30
[   10.358337]  [<ffffffff8022f048>] ? __update_rq_clock+0x68/0x140
[   10.358501]  [<ffffffff8026173d>] ? trace_hardirqs_off+0xd/0x10
[   10.358665]  [<ffffffff8026173d>] ? trace_hardirqs_off+0xd/0x10
[   10.358827]  [<ffffffff802953ce>] ? function_trace_call+0x9e/0xa0
[   10.358992]  [<ffffffff8020c4b4>] ? ftrace_call+0x5/0x2b
[   10.359148]  [<ffffffff8023177a>] ? cpu_clock+0xda/0x190
[   10.359302]  [<ffffffff80262d7b>] check_flags+0xfb/0x180
[   10.359457]  [<ffffffff80265635>] lock_acquire+0x65/0xd0
[   10.359614]  [<ffffffff80231799>] cpu_clock+0xf9/0x190
[   10.359769]  [<ffffffff80293de9>] ftrace_now+0x9/0x10
[   10.359920]  [<ffffffff80295194>] trace_function+0xc4/0x260
[   10.360079]  [<ffffffff80243723>] ? _local_bh_enable+0x63/0xc0
[   10.360240]  [<ffffffff804aeec9>] ? sub_preempt_count+0x9/0x90
[   10.360400]  [<ffffffff80243723>] ? _local_bh_enable+0x63/0xc0
[   10.360561]  [<ffffffff804aeec9>] ? sub_preempt_count+0x9/0x90
[   10.360721]  [<ffffffff8029539c>] function_trace_call+0x6c/0xa0
[   10.360883]  [<ffffffff8020c4b4>] ftrace_call+0x5/0x2b
[   10.36103...
[ message continues ]

cc'ing Ingo ...

On Mon, 5 May 2008 20:04:33 -0400 Daniel Hazelton <dhazelton@enter.net> wro=

--=20
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

This is the same error I was seeing with the 20080505 tree. 

It may not be an error in ftrace, but I will be testing a build of the kernel 
with the ftrace parts turned off.

.config for the kernel is attached - dmesg from failed boots is not available 
(neither qemu or bochs supports x86-64, apparently and netconsole is beyond 
what I have available hardware-wise ATM.)

DRH

-- 
Dialup is like pissing through a pipette. Slow and excruciatingly painful.

About this error - should I bother trying the 5/6 tree? If I'm going to be 
stuck with a lock-up during boot I'm wondering if I should try a test before 
seeing any action on this bug. (I realize I could just turn off the ftrace 
code, but I am attempting to do some "regular use" testing of the -next tree 
with new options (well, those that I think might become defaulted to "on") 
enabled)




-- 
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
--