Re: 2.4.18-pre9-xfs-shawn4 - hpfs bug

[Shawn Starr <spstarr@sh0n.net>]

Well, size is being passed in to grow_buffers, something outside is not
less the 512 and greater then PAGE_SIZE (a constant).

Shawn.

On Wed, 2002-02-13 at 05:09, Dag Bakke wrote:
> On Tue, Feb 12, 2002 at 09:16:10AM -0500, Shawn Starr wrote:
> > Some others have reported problems with superblock reading on boot.
> > There may be some code that has changed superblock reading but this
> > might be different.
> 
> Booting with BUG() enabled, I get:
> 
> kernel BUG at buffer.c:2630! 
> 
> The relevant section is:
> 
>    2618 static int grow_buffers(kdev_t dev, unsigned long block, int size)
>    2619 {
>    2620         struct page * page;
>    2621         struct block_device *bdev;
>    2622         unsigned long index;
>    2623         int sizebits;
>    2624 
>    2625         /* Size must be multiple of hard sectorsize */
>    2626         if (size & (get_hardsect_size(dev)-1))
>    2627                 BUG();
>    2628         /* Size must be within 512 bytes and PAGE_SIZE */
>    2629         if (size < 512 || size > PAGE_SIZE)
>    2630                 BUG();
> 
> 
> My disks are partitioned/formatted as follows:
> 
>     hda1        Boot        Primary   Linux ext2                          49.36 
>     hda2                    Primary   Linux XFS                        10001.95
>     hda3                    Primary   Linux raid autodetect            10001.95
>                             Pri/Log   Free Space                       20925.12
> 
>     hdc1                    Primary   Linux ext2                          49.36 
>     hdc2                    Primary   Linux ext2                       10001.95
>     hdc3                    Primary   Linux raid autodetect            10001.95
>                             Pri/Log   Free Space                       20925.12
> 
> I.e. my hpfs disk is not connected at this time.
> 
> I have compiled-in support for:
> CONFIG_VIDEO_PROC_FS=y
> CONFIG_EXT3_FS=y
> CONFIG_FAT_FS=y
> CONFIG_MSDOS_FS=y
> CONFIG_VFAT_FS=y
> CONFIG_TMPFS=y
> CONFIG_ISO9660_FS=y
> CONFIG_ZISOFS=y
> CONFIG_HPFS_FS=y
> CONFIG_PROC_FS=y
> CONFIG_DEVFS_FS=y
> CONFIG_DEVPTS_FS=y
> CONFIG_EXT2_FS=y
> CONFIG_UDF_FS=y
> CONFIG_XFS_FS=y
> CONFIG_XFS_DMAPI=y
> CONFIG_HAVE_XFS_DMAPI=y
> CONFIG_NFS_FS=y
> CONFIG_NFS_V3=y
> CONFIG_NFSD=y
> CONFIG_NFSD_V3=y
> CONFIG_ZISOFS_FS=y
> CONFIG_ZLIB_FS_INFLATE=y
> CONFIG_USB_DEVICEFS=y
> 
> 
> Disregarding hpfs for a moment:
> I see this during boot with working and non-working kernels:
> FAT: bogus logical sector size 0
> FAT: bogus logical sector size 0
> 
> Not sure what the FAT code complains about, but it could possibly be the
> two raid partitions (raid0/xfs). 
> 
> Anything else I can provide? kdb appears to require a little bit of
> rewiring/replugging. Got no response to keyboardwhacking when I hit the
> kdb prompt. Don't know if my USB keyboard makse a difference, but it
> probably does.... :-)
> If you want kdb stuff, please provide a little roadmap. 
> 
> Regards,
> 
> Dag B
> 
>  
> > On Tue, 2002-02-12 at 08:00, Dag Bakke wrote:
> > > Hi.
> > > 
> > > Compiling in support for hpfs in 2.4.18-pre9-xfs-shawn4 causes panic on
> > > boot.
> > > (I do have a some devicedriver patches added to your patch, but nothing related to
> > > core, vfs or hpfs.)
> > > 
> > > Thanks,
> > > 
> > > Dag B
> > > 
> > > 
> > > root@dagb:~# ksymoops -L -m /boot/System.map -K <~dagb/kernelcrash.txt 
> > > ksymoops 2.4.3 on i686 2.4.18-pre9-xfs-shawn4.  Options used
> > >      -V (default)
> > >      -K (specified)
> > >      -L (specified)
> > >      -o /lib/modules/2.4.18-pre9-xfs-shawn4/ (default)
> > >      -m /boot/System.map (specified)
> > > 
> > > No modules in ksyms, skipping objects
> > > invalid operand:        0000
> > > CPU:    0
> > > EIP:  0010:[<c0136d10>] Not tainted
> > > Using defaults from ksymoops -t elf32-i386 -a i386
> > > EFLAGS: 00010286
> > > Warning (Oops_read): Code line not seen, dumping what data is available
> > > 
> > > >>EIP; c0136d10 <grow_buffers+50/100>   <=====
> > > 
> > > Stack:  00000302  00000000  00000000  c13a91c0  00002240  c0134eb7  00000302  00000000
> > >         00000000  cffe7ed8  cfe6a000  cfe6a044  c01350aa  00000302  00000000  00000000
> > >         cffe7ed8  cfe6a000  c13a91c0  c019e96c  00000302  00000000  00000000  00000000
> > > Call Trace: [<c0134eb7>] [<c01350aa>] [<c019e96e>] [<c01a7a8d>] [<c01378a9>] [<c0137c27>] [<c010503a>] [<c010504c>] [<c01054e8>]
> > > Code: 0f 0b b9 ff ff ff ff 89 fa b6 00 90 8d 74 26 06 8b 44 24 20
> > > 
> > > Trace; c0134eb6 <getblk+26/40>
> > > Trace; c01350aa <bread+1a/c0>
> > > Trace; c019e96e <hpfs_map_sector+1e/40>
> > > Trace; c01a7a8c <hpfs_read_super+15c/730>
> > > Trace; c01378a8 <insert_super+38/40>
> > > Trace; c0137c26 <read_super+56/b0>
> > > Trace; c010503a <prepare_namespace+a/10>
> > > Trace; c010504c <init+c/110>
> > > Trace; c01054e8 <kernel_thread+28/40>
> > > Code;  c0136d10 <grow_buffers+50/100>
> > > 00000000 <_EIP>:
> > > Code;  c0136d10 <grow_buffers+50/100>
> > >    0:   0f 0b                     ud2a   
> > > Code;  c0136d12 <grow_buffers+52/100>
> > >    2:   b9 ff ff ff ff            mov    $0xffffffff,%ecx
> > > Code;  c0136d16 <grow_buffers+56/100>
> > >    7:   89 fa                     mov    %edi,%edx
> > > Code;  c0136d18 <grow_buffers+58/100>
> > >    9:   b6 00                     mov    $0x0,%dh
> > > Code;  c0136d1a <grow_buffers+5a/100>
> > >    b:   90                        nop    
> > > Code;  c0136d1c <grow_buffers+5c/100>
> > >    c:   8d 74 26 06               lea    0x6(%esi,1),%esi
> > > Code;  c0136d20 <grow_buffers+60/100>
> > >   10:   8b 44 24 20               mov    0x20(%esp,1),%eax
> > > 
> > > 
> > > 1 warning issued.  Results may not be reliable.
> > > 
> > 
> > 
> > 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>