Messages in this thread | | | Date | Wed, 5 May 1999 16:51:03 +0100 | From | Steve Dodd <> | Subject | Invalid inode->i_sb in clear_inode()? (was Re: Kernel Oops, 2.2.7) |
| |
On Wed, May 05, 1999 at 06:53:56AM -0400, Michael B. Trausch wrote:
> Code: c0131556 <clear_inode+36/50> 00000000 <_EIP>: <=== > Code: c0131556 <clear_inode+36/50> 0: 8b 40 24 movl 0x24(%eax),%eax <=== > Code: c0131559 <clear_inode+39/50> 3: 85 c0 testl %eax,%eax > Code: c013155b <clear_inode+3b/50> 5: 74 06 je c0131563 <clear_inode+43/50> > Code: c013155d <clear_inode+3d/50> 7: 53 pushl %ebx > Code: c013155e <clear_inode+3e/50> 8: ff d0 call *%eax > Code: c0131560 <clear_inode+40/50> a: 83 c4 04 addl $0x4,%esp > Code: c0131563 <clear_inode+43/50> d: c7 83 84 00 00 movl $0x0,0x84(%ebx)
Okay, I reckon that looks like this bit of code:
;0239 if (inode->i_sb && inode->i_sb->s_op && inode->i_sb->s_op->clear_inode) .LM200: movl 104(%ebx),%eax testl %eax,%eax je .L869 movl 24(%eax),%eax <==== testl %eax,%eax je .L869 movl 36(%eax),%eax testl %eax,%eax je .L869 ;0240 inode->i_sb->s_op->clear_inode(inode); .LM201:
Making a wild stab in the dark, I'd say that inode->i_sb points somewhere silly for this inode. I don't know why that should happen. This is probably a job for Mr. Viro, but I'll keep looking up the stack trace in case anything obvious leaps out at me.
Nothing in your system is marginal is it, as far as you know? Overheating CPU, bad memory or whatever /might/ conceivably be causing this (via random memory hosing), but I've seen this exact report before..
You did say in your mail that you haven't been smoking any weird patches, right? :)
-- If smoking is so bad for you, how come it cures kippers?
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |