Messages in this thread | | | Date | Mon, 5 Jan 2009 14:18:36 -0800 | From | "Paul E. McKenney" <> | Subject | Re: [BUG] NULL pointer deref with rcutorture |
| |
On Mon, Jan 05, 2009 at 09:31:53PM +0100, Eric Sesterhenn wrote: > * Paul E. McKenney (paulmck@linux.vnet.ibm.com) wrote: > > On Mon, Jan 05, 2009 at 09:01:45PM +0100, Eric Sesterhenn wrote: > > > hi, > > > > > > * Paul E. McKenney (paulmck@linux.vnet.ibm.com) wrote: > > > > On Mon, Jan 05, 2009 at 07:56:55PM +0100, Eric Sesterhenn wrote: > > > > > > > > Wow!!! Am I reading this correctly? Does the above "call" instruction > > > > -really- call one byte into itself? That is what the hex for the x86 > > > > instruction -looks- like it is doing, but I cannot see what would have > > > > possessed the compiler to generate this code. > > > > > > Compiler is gcc version 4.2.4 (Ubuntu 4.2.4-1ubuntu3) > > > > I am using 4.1.3, for whatever it is worth. (Ancient, I know!) > > > > > > When I compile on a 32-bit x86 machine, I don't see the above "call" > > > > instruction. Other than that, the code I see looks consistent. > > > > > > > > > 9f0: eb 1d jmp a0f <rcu_stutter_wait+0x27> > > > > > 9f2: 83 3d 00 00 00 00 00 cmpl $0x0,0x0 > > > > > 9f9: b8 01 00 00 00 mov $0x1,%eax > > > > > 9fe: 75 0a jne a0a <rcu_stutter_wait+0x22> > > > > > a00: b8 e8 03 00 00 mov $0x3e8,%eax > > > > > a05: e8 fc ff ff ff call a06 <rcu_stutter_wait+0x1e> > > > > > a0a: e8 fc ff ff ff call a0b <rcu_stutter_wait+0x23> > > > > > a0f: 83 3d 6c 00 00 00 00 cmpl $0x0,0x6c > > > > > ^---------- this line > > > > > > > > This looks like the first test in the "while" loop. > > > > > > > > > a16: 75 09 jne a21 <rcu_stutter_wait+0x39> > > > > > a18: 83 3d 00 00 00 00 00 cmpl $0x0,0x0 > > > > > a1f: 75 09 jne a2a <rcu_stutter_wait+0x42> > > > > > a21: 83 3d 50 1a 00 00 00 cmpl $0x0,0x1a50 > > > > > a28: 74 c8 je 9f2 <rcu_stutter_wait+0xa> > > > > > a2a: 5d pop %ebp > > > > > a2b: c3 ret > > > > > > > > The corresponding C code is as follows: > > > > > > > > static void > > > > rcu_stutter_wait(void) > > > > { > > > > while ((stutter_pause_test || !rcutorture_runnable) && !fullstop) { > > > > if (rcutorture_runnable) > > > > schedule_timeout_interruptible(1); > > > > else > > > > schedule_timeout_interruptible(round_jiffies_relative(HZ)); > > > > } > > > > } > > > > > > > > I don't see much opportunity for a page fault here... This is the > > > > binary I get when I compile it, though not as a module: > > > > > > > > 0000085a <rcu_stutter_wait>: > > > > 85a: 55 push %ebp > > > > 85b: 89 e5 mov %esp,%ebp > > > > 85d: eb 1d jmp 87c <rcu_stutter_wait+0x22> > > > > 85f: 83 3d 00 00 00 00 00 cmpl $0x0,0x0 > > > > 866: b8 01 00 00 00 mov $0x1,%eax > > > > 86b: 75 0a jne 877 <rcu_stutter_wait+0x1d> > > > > 86d: b8 e8 03 00 00 mov $0x3e8,%eax > > > > 872: e8 fc ff ff ff call 873 <rcu_stutter_wait+0x19> > > > > 877: e8 fc ff ff ff call 878 <rcu_stutter_wait+0x1e> > > > > 87c: 83 3d 14 00 00 00 00 cmpl $0x0,0x14 > > > > 883: 75 09 jne 88e <rcu_stutter_wait+0x34> > > > > 885: 83 3d 00 00 00 00 00 cmpl $0x0,0x0 > > > > 88c: 75 09 jne 897 <rcu_stutter_wait+0x3d> > > > > 88e: 83 3d 08 1a 00 00 00 cmpl $0x0,0x1a08 > > > > 895: 74 c8 je 85f <rcu_stutter_wait+0x5> > > > > 897: 5d pop %ebp > > > > 898: c3 ret > > > > > > > > I confess, I am confused!!! > > > > > > on the other box with a different gcc version > > > > > > gcc version 4.3.2 (Ubuntu 4.3.2-1ubuntu11) > > > > > > d1902e90 is the start of rcu_stutter_wait > > > > > > [ 533.391719] d087e000 d1902e90 > > > [ 533.392294] rcu-torture:--- Start of test: nreaders=2 nfakewriters=4 stat_interval=0 verbose=0 test_no_idle_hz=0 shuffle_interval=3 stutter=5 irqreader=1 > > > [ 541.000139] BUG: unable to handle kernel paging request at d1902efd > > > [ 541.000423] IP: [<d1902efd>] 0xd1902efd > > > [ 541.000660] *pde = 0f08f067 *pte = 00000000 > > > [ 541.000867] Oops: 0000 [#1] DEBUG_PAGEALLOC > > > [ 541.001126] last sysfs file: /sys/block/sda/size > > > [ 541.001246] Modules linked in: nfsd exportfs nfs lockd nfs_acl auth_rpcgss sunrpc ipv6 fuse unix [last unloaded: rcutorture] > > > [ 541.002235] > > > [ 541.002334] Pid: 5292, comm: rcu_torture_wri Not tainted (2.6.28 #84) > > > [ 541.002470] EIP: 0060:[<d1902efd>] EFLAGS: 00010296 CPU: 0 > > > [ 541.002598] EIP is at 0xd1902efd > > > [ 541.002767] EAX: 00000000 EBX: d19073c0 ECX: 00000000 EDX: 00000000 > > > [ 541.002900] ESI: 0000000a EDI: 00000000 EBP: c7b63fb8 ESP: c7b63fb8 > > > [ 541.003033] DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068 > > > [ 541.003160] Process rcu_torture_wri (pid: 5292, ti=c7b63000 task=c7b09710 task.ti=c7b63000) > > > [ 541.003400] Stack: > > > [ 541.003497] c7b63fd0 d19032c1 00000000 00000000 00000000 d1903200 c7b63fe0 c013d80a > > > [ 541.004022] c013d7d0 00000000 00000000 c0103cf3 cef6ee70 00000000 00000000 00000000 > > > [ 541.004022] 00000201 000004b4 > > > [ 541.004022] Call Trace: > > > [ 541.004022] [<c013d80a>] ? kthread+0x3a/0x70 > > > [ 541.004022] [<c013d7d0>] ? kthread+0x0/0x70 > > > [ 541.004022] [<c0103cf3>] ? kernel_thread_helper+0x7/0x14 > > > [ 541.004022] Code: Bad EIP value. > > > [ 541.004022] EIP: [<d1902efd>] 0xd1902efd SS:ESP 0068:c7b63fb8 > > > [ 541.004022] ---[ end trace cb3b10c2bb94b4e3 ]--- > > > > > > > > > 00000e90 <rcu_stutter_wait>: > > > e90: 55 push %ebp > > > e91: 89 e5 mov %esp,%ebp > > > e93: 90 nop > > > e94: 8d 74 26 00 lea 0x0(%esi,%eiz,1),%esi > > > e98: a1 98 00 00 00 mov 0x98,%eax > > > e9d: 85 c0 test %eax,%eax > > > e9f: 75 09 jne eaa <rcu_stutter_wait+0x1a> > > > ea1: a1 00 00 00 00 mov 0x0,%eax > > > ea6: 85 c0 test %eax,%eax > > > ea8: 75 36 jne ee0 <rcu_stutter_wait+0x50> > > > eaa: a1 88 1a 00 00 mov 0x1a88,%eax > > > eaf: 85 c0 test %eax,%eax > > > eb1: 75 2d jne ee0 <rcu_stutter_wait+0x50> > > > eb3: 8b 15 00 00 00 00 mov 0x0,%edx > > > eb9: 85 d2 test %edx,%edx > > > ebb: 74 2b je ee8 <rcu_stutter_wait+0x58> > > > ebd: b8 01 00 00 00 mov $0x1,%eax > > > ec2: e8 fc ff ff ff call ec3 <rcu_stutter_wait+0x33> > > > ec7: a1 98 00 00 00 mov 0x98,%eax > > > ecc: 85 c0 test %eax,%eax > > > ece: 74 d1 je ea1 <rcu_stutter_wait+0x11> > > > ed0: a1 88 1a 00 00 mov 0x1a88,%eax > > > ed5: 85 c0 test %eax,%eax > > > ed7: 74 da je eb3 <rcu_stutter_wait+0x23> > > > ed9: 8d b4 26 00 00 00 00 lea 0x0(%esi,%eiz,1),%esi > > > ee0: 5d pop %ebp > > > ee1: c3 ret > > > ee2: 8d b6 00 00 00 00 lea 0x0(%esi),%esi > > > ee8: b8 fa 00 00 00 mov $0xfa,%eax > > > eed: e8 fc ff ff ff call eee <rcu_stutter_wait+0x5e> > > > > Here we are again calling one byte into the current instruction!!! > > > > Or am I misinterpreting this code? > > > > > ef2: 8d b6 00 00 00 00 lea 0x0(%esi),%esi > > > ef8: e8 fc ff ff ff call ef9 <rcu_stutter_wait+0x69> > > > efd: 8d 76 00 lea 0x0(%esi),%esi > > > ^------------- here > > > > > > This one looks more like it can explain a page fault > > > > I don't understand why there are indirections in the assembly given the > > C code for rcu_stutter_wait(). > > > > > f00: eb 96 jmp e98 <rcu_stutter_wait+0x8> > > > f02: 8d b4 26 00 00 00 00 lea 0x0(%esi,%eiz,1),%esi > > > f09: 8d bc 27 00 00 00 00 lea 0x0(%edi,%eiz,1),%edi > > ok, after trying to find out if the ubuntu gccs are broken, i stumbled > upon this: > http://forum.soft32.com/linux/Strange-problem-disassembling-shared-lib-ftopict439936.html > > Seems the difference is that you dont compile it as a module and the > jump is perfectly normal, it gets overwritten when the stuff is loaded > objdump -dr gives me > > 00000e90 <rcu_stutter_wait>: > e90: 55 push %ebp > e91: 89 e5 mov %esp,%ebp > e93: 90 nop > e94: 8d 74 26 00 lea 0x0(%esi,%eiz,1),%esi > e98: a1 98 00 00 00 mov 0x98,%eax > e99: R_386_32 .bss > e9d: 85 c0 test %eax,%eax > e9f: 75 09 jne eaa <rcu_stutter_wait+0x1a> > ea1: a1 00 00 00 00 mov 0x0,%eax > ea2: R_386_32 rcutorture_runnable > ea6: 85 c0 test %eax,%eax > ea8: 75 36 jne ee0 <rcu_stutter_wait+0x50> > eaa: a1 88 1a 00 00 mov 0x1a88,%eax > eab: R_386_32 .bss > eaf: 85 c0 test %eax,%eax > eb1: 75 2d jne ee0 <rcu_stutter_wait+0x50> > eb3: 8b 15 00 00 00 00 mov 0x0,%edx > eb5: R_386_32 rcutorture_runnable > eb9: 85 d2 test %edx,%edx > ebb: 74 2b je ee8 <rcu_stutter_wait+0x58> > ebd: b8 01 00 00 00 mov $0x1,%eax > ec2: e8 fc ff ff ff call ec3 <rcu_stutter_wait+0x33> > ec3: R_386_PC32 schedule_timeout_interruptible > ec7: a1 98 00 00 00 mov 0x98,%eax > ec8: R_386_32 .bss > ecc: 85 c0 test %eax,%eax > ece: 74 d1 je ea1 <rcu_stutter_wait+0x11> > ed0: a1 88 1a 00 00 mov 0x1a88,%eax > ed1: R_386_32 .bss > ed5: 85 c0 test %eax,%eax > ed7: 74 da je eb3 <rcu_stutter_wait+0x23> > ed9: 8d b4 26 00 00 00 00 lea 0x0(%esi,%eiz,1),%esi > ee0: 5d pop %ebp > ee1: c3 ret > ee2: 8d b6 00 00 00 00 lea 0x0(%esi),%esi > ee8: b8 fa 00 00 00 mov $0xfa,%eax > eed: e8 fc ff ff ff call eee <rcu_stutter_wait+0x5e> > eee: R_386_PC32 round_jiffies_relative > ef2: 8d b6 00 00 00 00 lea 0x0(%esi),%esi > ef8: e8 fc ff ff ff call ef9 <rcu_stutter_wait+0x69> > ef9: R_386_PC32 schedule_timeout_interruptible > efd: 8d 76 00 lea 0x0(%esi),%esi > > here is the deref ------------------------^
Ah!!! We are getting a page fault while cleaning up the stack frame?
Ouch!
> f00: eb 96 jmp e98 <rcu_stutter_wait+0x8> > f02: 8d b4 26 00 00 00 00 lea 0x0(%esi,%eiz,1),%esi > f09: 8d bc 27 00 00 00 00 lea 0x0(%edi,%eiz,1),%edi
Thanx, Paul
| |