With CONFIG_FAIR_CGROUP_SCHED=y, following commands on 2.6.24-rc1 crash
the system.

	$ mount -t cgroup none /cgroups
	$ ./ns_exec -cm /bin/ls

"ns_exec -cm" calls clone() to clone the mount namespace and then
executes the '/bin/ls' program in the cloned child.

Some observations that Serge and I made (we have been able to reproduce
reliably, but crash logs have not been very useful)

	a. If we skip the 'mount' command, there is no crash.

	b. If CONFIG_FAIR_CGROUP_SCHED=n again, there is no crash (even with 
	   'mount' command).

       	c. mounting just the cpu or just the ns subsystem does not
           lead to a crash.  You can even

           mount -t cgroup -o cpu none /mnt1
           mount -t cgroup -o ns none /mnt2

           and you still don't get a crash

        d. mount -t cgroup -o cpu,ns none /cgroup

           will always cause a crash with subsequent ns_exec

ns_exec.c and config file are attached. Let us know if you need more info.

Suka

---
Crash log:

Red Hat Enterprise Linux release 4.90 (Tikanga)
Kernel 2.6.24-rc1 on an x86_64

elm3a241 login: Unable to handle kernel NULL pointer dereferenceUnable to handle kernel NULL pointer dereference at 0000000000000000 RIP:
 at 0000000000000000 RIP:
 [<0000000000000000>]
 [<0000000000000000>]
PGD 102d4d067 PGD 102d4d067 PUD 102c88067 PUD 102c88067 PMD 0 PMD 0

Oops: 0000 [1] Oops: 0000 [1] SMP SMP

CPU 2 CPU 2

Modules linked in:Modules linked in:

Pid: 3273, comm: ns_exec Not tainted 2.6.24-rc1 #9
Pid: 3273, comm: ns_exec Not tainted 2.6.24-rc1 #9
RIP: 0010:[<0000000000000000>] RIP: 0010:[<0000000000000000>]  [<0000000000000000>]
 [<0000000000000000>]
RSP: 0018:ffff8101006a6af0  EFLAGS: 00055292
RSP: 0018:ffff8101006a6af0  EFLAGS: 00055292
RAX: 0000000000000000 RBX: ffff810100d11140 RCX: ffff810101de4000
RAX: 0000000000000000 RBX: ffff810100d11140 RCX: ffff810101de4000
RDX: 0000000000000000 RSI: ffff810100d1a880 RDI: ffff810001037c00
RDX: 0000000000000000 RSI: ffff810100d1a880 RDI: ffff810001037c00
RBP: ffff810102c136c0 R08: ffff810101de4000 R09: ffff810101d31bb8
RBP: ffff810102c136c0 R08: ffff810101de4000 R09: ffff810101d31bb8
R10: 0000000000000000 R11: 00000000ffffffff R12: ffff8101034075b8
R10: 0000000000000000 R11: 00000000ffffffff R12: ffff8101034075b8
R13: ffff8101029d6028 R14: ffff810103407500 R15: ffffffff80869d00
R13: ffff8101029d6028 R14: ffff810103407500 R15: ffffffff80869d00
FS:  00002b80c2a396f0(0000) GS:ffff81010068f3c0(0000) knlGS:0000000000000000
FS:  00002b80c2a396f0(0000) GS:ffff81010068f3c0(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000000 CR3: 00000001028be000 CR4: 00000000000006e0
CR2: 0000000000000000 CR3: 00000001028be000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process ns_exec (pid: 3273, threadinfo ffff810101de4000, task ffff810100d11140)
Process ns_exec (pid: 3273, threadinfo ffff810101de4000, task ffff810100d11140)
Stack: Stack:  0000000000000000 0000000000000000 0000000000000001 0000000000000001 6c2f343662696c2f 6c2f343662696c2f 2d78756e696c2d64 2d78756e696c2d64

 732e34362d363878 732e34362d363878 ffffffff00322e6f ffffffff00322e6f 762f003561736376 762f003561736376 0000357363762f63 0000357363762f63

 0000000000000000 0000000000000000 0000000000000000 0000000000000000 762f73007665642f 762f73007665642f 0000317363762f63 0000317363762f63

Call Trace:
Call Trace:




Code: Code:  Bad RIP value. Bad RIP value.

RIP RIP  [<0000000000000000>]
 [<0000000000000000>]
 RSP <ffff8101006a6af0>
 RSP <ffff8101006a6af0>
CR2: 0000000000000000
CR2: 0000000000000000