The attached (radically shortened) program to measure pipe performance oopses
the Linux-2.6.12-rc3 kernel when traced using "strace -f". The problem no longer
occurs in -rc4 or -rc5. It would be good to know whether the bug causing the
problem is known and has been fixed or simply went away due to other changes.
I have been unable to find a Changelog detailing the patches that went
into -rc4.
The bug is reproducible on all -rc3 kernels I've tried.
Unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
00000000
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in:
CPU: 0
EIP: 0060:[<00000000>] Not tainted VLI
EFLAGS: 00010286 (2.6.12-rc3)
EIP is at 0x0
eax: db052000 ebx: 01200011 ecx: 00000000 edx: 00000000
esi: df605060 edi: 00000000 ebp: db052000 esp: db052fc4
ds: 007b es: 007b ss: 0068
Process pipe (pid: 778, threadinfo=db052000 task=df605060)
Stack: 01202011 00000000 00000000 00000000 b7ea0708 bf9da748 00000000 0000007b
c010007b 00000078 b7f290a5 00000073 00000282 bf9da6e0 0000007b
Call Trace:
Code: Bad EIP value.
Best regards,
-Udo.
fre 2005-05-27 klockan 17:02 +0200 skrev Udo A. Steinberg:
> The attached (radically shortened) program to measure pipe performance oopses
> the Linux-2.6.12-rc3 kernel when traced using "strace -f". The problem no longer
> occurs in -rc4 or -rc5. It would be good to know whether the bug causing the
> problem is known and has been fixed or simply went away due to other changes.
> I have been unable to find a Changelog detailing the patches that went
> into -rc4.
>
> The bug is reproducible on all -rc3 kernels I've tried.
Yeah it is known and fixed:
tree 8eece0b3fd959622afdef405dc42dc4a6b63efe7
parent 47c297529bd23d93d2a088d9620bb220763e9cb1
author Alexander Nyberg <[email protected]> Fri, 06 May 2005 06:15:03 -0700
committer Linus Torvalds <[email protected]> Fri, 06 May 2005 06:36:30 -0700
[PATCH] x86 stack initialisation fix
The recent change fix-crash-in-entrys-restore_all.patch
childregs->esp = esp;
p->thread.esp = (unsigned long) childregs;
- p->thread.esp0 = (unsigned long) (childregs+1);
+ p->thread.esp0 = (unsigned long) (childregs+1) - 8;
p->thread.eip = (unsigned long) ret_from_fork;
introduces an inconsistency between esp and esp0 before the task is run the
first time. esp0 is no longer the actual start of the stack, but 8 bytes
off.
This shows itself clearly in a scenario when a ptracer that is set to also
ptrace eventual children traces program1 which then clones thread1. Now
the ptracer wants to modify the registers of thread1. The x86 ptrace
implementation bases it's knowledge about saved user-space registers upon
p->thread.esp0. But this will be a few bytes off causing certain writes to
the kernel stack to overwrite a saved kernel function address making the
kernel when actually running thread1 jump out into user-space. Very
spectacular.
The testcase I've used is:
/* start with strace -f ./a.out */
#include <pthread.h>
#include <stdio.h>
void *do_thread(void *p)
{
for (;;);
}
int main()
{
pthread_t one;
pthread_create(&one, NULL, &do_thread, NULL);
for (;;);
return 0;
}
So, my solution is to instead of just adjusting esp0 that creates an
inconsitent state I adjust where the user-space registers are saved with -8
bytes. This gives us the wanted extra bytes on the start of the stack and
esp0 is now correct. This solves the issues I saw from the original
testcase from Mateusz Berezecki and has survived testing here. I think
this should go into -mm a round or two first however as there might be some
cruft around depending on pt_regs lying on the start of the stack. That
however would have broken with the first change too!
It's actually a 2-line diff but I had to move the comment of why the -8 bytes
are there a few lines up. Thanks to Zwane for helping me with this.
Signed-off-by: Alexander Nyberg <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
i386/kernel/process.c | 13 +++++++------
1 files changed, 7 insertions(+), 6 deletions(-)
Index: arch/i386/kernel/process.c
===================================================================
--- 52672d42102a4d0c8b4fc13b79812290efbd5d3a/arch/i386/kernel/process.c (mode:100644 sha1:85bd56d4431417836a1b6c3992d57bf36e9186ca)
+++ 8eece0b3fd959622afdef405dc42dc4a6b63efe7/arch/i386/kernel/process.c (mode:100644 sha1:96e3ea6b17c7b989c1bafb3c1ce87f7768348693)
@@ -400,11 +400,6 @@ int copy_thread(int nr, unsigned long cl
int err;
childregs = ((struct pt_regs *) (THREAD_SIZE + (unsigned long) p->thread_info)) - 1;
- *childregs = *regs;
- childregs->eax = 0;
- childregs->esp = esp;
-
- p->thread.esp = (unsigned long) childregs;
/*
* The below -8 is to reserve 8 bytes on top of the ring0 stack.
* This is necessary to guarantee that the entire "struct pt_regs"
@@ -415,7 +410,13 @@ int copy_thread(int nr, unsigned long cl
* "struct pt_regs" is possible, but they may contain the
* completely wrong values.
*/
- p->thread.esp0 = (unsigned long) (childregs+1) - 8;
+ childregs = (struct pt_regs *) ((unsigned long) childregs - 8);
+ *childregs = *regs;
+ childregs->eax = 0;
+ childregs->esp = esp;
+
+ p->thread.esp = (unsigned long) childregs;
+ p->thread.esp0 = (unsigned long) (childregs+1);
p->thread.eip = (unsigned long) ret_from_fork;