Messages in this thread Patch in this message | | | Date | Wed, 20 Jan 1999 15:07:03 -0500 (EST) | From | Alexander Viro <> | Subject | Re: VFAT oops (the fun just keeps on coming) |
| |
On Wed, 20 Jan 1999, CaT wrote:
> > Here we go, here we go, here we go... > > I think reality is playing a nasty, evil and bastardly > trick on me to make me delurk. I wish it stopped... [snip] > > >>EIP: c0145015 <vfat_free_ino+d/84> > Trace: c01450f5 <vfat_remove_entry+69/e0> > Trace: c014561b <vfat_rename+20b/4ac> > Trace: c012a109 <vfs_rename+245/254> > Trace: c012a24f <sys_rename+137/1b0> > Trace: c01087f8 <system_call+34/38> > Code: c0145015 <vfat_free_ino+d/84> 00000000 <_EIP>: > Code: c0145015 <vfat_free_ino+d/84> 0: 66 c7 43 24 00 movw $0x0,0x24(%ebx) > Code: c014501a <vfat_free_ino+12/84> 5: 00 > Code: c014501b <vfat_free_ino+13/84> 6: a1 f0 ba 2a c0 movl 0xc02abaf0,%eax > Code: c0145020 <vfat_free_ino+18/84> b: 89 42 34 movl %eax,0x34(%edx) > Code: c0145023 <vfat_free_ino+1b/84> e: 89 43 34 movl %eax,0x34(%ebx) > Code: c0145026 <vfat_free_ino+1e/84> 11: a1 f0 ba 00 00 movl 0xbaf0,%eax
WTF??? Somebody called vfat_remove_entry() with NULL victim? vfat_rename()? <looking at the source> Oh, shit!
Linus, sorry, I had a bad braino in the last part of vfat patch. Darn... call of vfat_remove_entry() should go one level deeper. Shame on me. Fix:
--- linux/fs/vfat/namei.c.old Wed Jan 20 15:03:01 1999 +++ linux/fs/vfat/namei.c Wed Jan 20 15:03:42 1999 @@ -1656,8 +1656,8 @@ } else { drop_aliases(new_dentry); } + res = vfat_remove_entry(new_dir,&sinfo,new_inode); } - res = vfat_remove_entry(new_dir,&sinfo,new_inode); if (res) goto rename_done;
Please, apply it. Embarrased, Al
- 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/
| |