mkfs.ext3 - segmentation fault

mkfs.ext3 - segmentation fault

[Dmitry Skorinko]

2.6.8-1 + mtd snapshot (about 2 week ago)

#modprobe inftl
#modprobe diskonchip
#cat /proc/mtd
dev:    size   erasesize  name
mtd0: 10000000 00004000 "DiskOnChip 2000 (INFTL Model)"
mtd1: 0ffe8000 00004000 " DiskOnChip BDTL partition"
#fdisk -l /dev/inftla

Disk /dev/inftla: 262 MB, 262946816 bytes
16 heads, 32 sectors/track, 1003 cylinders
Units = cylinders of 512 * 512 = 262144 bytes

      Device Boot      Start         End      Blocks   Id  System
/dev/inftla1   *           1        1003      256752   83  Linux

#mkfs.ext3 /dev/inftla1
Segmentation fault
#dmesg
--skip--
nand_read_oob: from = 0x00027a00, len = 8
nand_read_oob: from = 0x00027c00, len = 8
nand_read_oob: from = 0x00027e00, len = 8
INFTL: deleting empty VUC 2
Deleting EUN 9 from VUC 2
INFTL: INFTL_formatblock(inftl=c7554200,block=9)
nand_erase: start = 0x00024000, len = 16384
nand_isbad_bbt(): bbt info for offs 0x00024000: (block 0) 0x09
Unable to handle kernel NULL pointer dereference at virtual address 00000100
 printing eip:
c8065711
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in: diskonchip reed_solomon nand nand_ids nand_ecc mtdpart 
inftl mtd_blkdevs mtdcore ipv6 af_packet 8139too mii crc32 usbkbd usbcore rtc 
ext3 jbd ide_generic sis5513 ide_disk ide_core sd_mod ata_piix libata 
scsi_mod unix
CPU:    0
EIP:    0060:[<c8065711>]    Not tainted
EFLAGS: 00010246   (2.6.8mymtd26)
EIP is at mtd_erase_callback+0x6/0x3b [mtdpart]
eax: 00004000   ebx: c60f1968   ecx: c7554300   edx: 00000000
esi: 00000140   edi: c7554300   ebp: 00000000   esp: c60c7db4
ds: 007b   es: 007b   ss: 0068
Process inftld (pid: 741, threadinfo=c60c6000 task=c6543180)
Stack: c810a458 c7554300 00000000 00000020 c7554300 c7554200 00000000 00000009
       c810a273 c60f1800 c7554300 00000000 c80b378b c60f1800 c7554300 0000001d
       00000002 c0117cc7 c6118004 c7554200 00000009 00000002 c80b290c c7554200
Call Trace:
 [<c810a458>] nand_erase_nand+0x1e1/0x1f7 [nand]
 [<c810a273>] nand_erase+0xf/0x13 [nand]
 [<c80b378b>] INFTL_formatblock+0x5c/0x116 [inftl]
 [<c0117cc7>] printk+0xfb/0x11b
 [<c80b290c>] INFTL_trydeletechain+0x1dd/0x22a [inftl]
 [<c80b2af1>] INFTL_deleteblock+0x198/0x1a6 [inftl]
 [<c0117cc7>] printk+0xfb/0x11b
 [<c80b2f01>] inftl_writeblock+0x402/0x40e [inftl]
 [<c01d75d9>] __end_that_request_first+0x193/0x1a3
 [<c8020096>] do_blktrans_request+0x96/0xcb [mtd_blkdevs]
 [<c8020208>] mtd_blktrans_thread+0x13d/0x18e [mtd_blkdevs]
 [<c0114c48>] default_wake_function+0x0/0x12
 [<c0105996>] ret_from_fork+0x6/0x20
 [<c0114c48>] default_wake_function+0x0/0x12
 [<c80200cb>] mtd_blktrans_thread+0x0/0x18e [mtd_blkdevs]
 [<c01041e1>] kernel_thread_helper+0x5/0xb
Code: 81 ba 00 01 00 00 cf 56 06 c8 75 1d 8b 41 0c 83 f8 ff 74 09

Help, plz.  Thanks.
Dmitry

Re: mkfs.ext3 - segmentation fault

[Sean Young]

On Mon, Nov 22, 2004 at 12:11:12PM +0300, Dmitry Skorinko wrote:
> 2.6.8-1 + mtd snapshot (about 2 week ago)
> 
> #modprobe inftl
> #modprobe diskonchip

-snip-

> EIP is at mtd_erase_callback+0x6/0x3b [mtdpart]

inftl is not setting the mtd member of struct erase_info. None of the
ftl's seem to be doing this but I do not have a current cvs handy so it
might be fixed.


Sean

Re: mkfs.ext3 - segmentation fault

[Kalev Lember]

Dmitry Skorinko wrote:
<snip>

>EIP is at mtd_erase_callback+0x6/0x3b [mtdpart]
>  
>
<snip>

>Help, plz.  Thanks.
>  
>
I fixed this in the CVS about two weeks ago and sent a note about the 
fix to the list. Try the current CVS.

-- 
Best regards,
Kalev Lember

Re: mkfs.ext3 - segmentation fault

[Dmitry Skorinko]

On Monday 22 November 2004 13:24, you wrote:
> Dmitry Skorinko wrote:
> <snip>
>
> >EIP is at mtd_erase_callback+0x6/0x3b [mtdpart]
>
> <snip>
>
> >Help, plz.  Thanks.
>
> I fixed this in the CVS about two weeks ago and sent a note about the
> fix to the list. Try the current CVS.

try current cvs and have same :((   :

-skip-
nand_read_oob: from = 0x00027c00, len = 8
nand_read_oob: from = 0x00027e00, len = 8
INFTL: deleting empty VUC 2
Deleting EUN 9 from VUC 2
INFTL: INFTL_formatblock(inftl=c126e200,block=9)
nand_erase: start = 0x00024000, len = 16384
nand_isbad_bbt(): bbt info for offs 0x00024000: (block 0) 0x09
Unable to handle kernel NULL pointer dereference at virtual address 00000100
 printing eip:
c803f846
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in: diskonchip reed_solomon nand nand_ids nand_ecc mtdpart 
inftl mtd_blkdevs mtdcore ipv6 af_packet 8139too mii crc32 usbkbd usbcore rtc 
ext3 jbd ide_generic sis5513 ide_disk ide_core unix
CPU:    0
EIP:    0060:[<c803f846>]    Not tainted
EFLAGS: 00010246   (2.6.8mymtd26)
EIP is at mtd_erase_callback+0x6/0x40 [mtdpart]
eax: 00000000   ebx: 00000000   ecx: 0000000e   edx: c126e304
esi: 00000140   edi: c126e304   ebp: 00000000   esp: c620fdd0
ds: 007b   es: 007b   ss: 0068
Process inftld (pid: 749, threadinfo=c620e000 task=c620d790)
Stack: c80dfa4d c126e304 00000009 c126e200 c126e334 00000000 00000000 00000020
       c80df831 c6211400 c126e304 00000000 c8032971 c6211400 c126e304 c6260004
       c126e200 00000009 c620fe5c c126e304 c126e200 00000009 c8031a64 c8031a6e
Call Trace:
 [<c80dfa4d>] nand_erase_nand+0x20d/0x230 [nand]
 [<c80df831>] nand_erase+0x11/0x20 [nand]
 [<c8032971>] INFTL_formatblock+0x71/0x140 [inftl]
 [<c8031a64>] INFTL_trydeletechain+0x214/0x280 [inftl]
 [<c8031a6e>] INFTL_trydeletechain+0x21e/0x280 [inftl]
 [<c8031c7c>] INFTL_deleteblock+0x1ac/0x1c0 [inftl]
 [<c80320b9>] inftl_writeblock+0x429/0x440 [inftl]
 [<c01f42f7>] __end_that_request_first+0x1c7/0x1e0
 [<c800d0b4>] gcc2_compiled.+0xb4/0x100 [mtd_blkdevs]
 [<c800d25d>] mtd_blktrans_thread+0x15d/0x1c0 [mtd_blkdevs]
 [<c800d100>] mtd_blktrans_thread+0x0/0x1c0 [mtd_blkdevs]
 [<c01174d0>] default_wake_function+0x0/0x20
 [<c01174d0>] default_wake_function+0x0/0x20
 [<c0104279>] kernel_thread_helper+0x5/0xc
Code: 81 b8 00 01 00 00 f0 f7 03 c8 75 1a 8b 4a 0c 83 f9 ff 74 09

Dmitry

Re: mkfs.ext3 - segmentation fault

[Kalev Lember]

[-- Attachment #1: Type: text/plain, Size: 125 bytes --]

Dmitry Skorinko wrote:

>try current cvs and have same :((   :
>  
>
Try the attached patch.

-- 
Best regards,
Kalev Lember

[-- Attachment #2: inftlmount.c.diff --]
[-- Type: text/plain, Size: 775 bytes --]

Index: inftlmount.c
===================================================================
RCS file: /home/cvs/mtd/drivers/mtd/inftlmount.c,v
retrieving revision 1.15
diff -u -r1.15 inftlmount.c
--- inftlmount.c	5 Nov 2004 21:55:55 -0000	1.15
+++ inftlmount.c	22 Nov 2004 13:17:53 -0000
@@ -389,8 +389,6 @@
 	struct erase_info *instr = &inftl->instr;
 	int physblock;
 
-	instr->mtd = inftl->mbd.mtd;
-
 	DEBUG(MTD_DEBUG_LEVEL3, "INFTL: INFTL_formatblock(inftl=%p,"
 		"block=%d)\n", inftl, block);
 
@@ -400,6 +398,7 @@
 	   _first_? */
 
 	/* Use async erase interface, test return code */
+	instr->mtd = inftl->mbd.mtd;
 	instr->addr = block * inftl->EraseSize;
 	instr->len = inftl->mbd.mtd->erasesize;
 	/* Erase one physical eraseblock at a time, even though the NAND api

Re: mkfs.ext3 - segmentation fault

[Dmitry Skorinko]

On Monday 22 November 2004 16:21, you wrote:
> Dmitry Skorinko wrote:
> >try current cvs and have same :((   :
>
> Try the attached patch.

ok
thank you very much

Dmitry

Re: mkfs.ext3 - segmentation fault

[Dmitry Skorinko]

22 Ноябрь 2004 18:44, Dmitry Skorinko написал:
> On Monday 22 November 2004 16:21, you wrote:
> > Dmitry Skorinko wrote:
> > >try current cvs and have same :((   :
> >
> > Try the attached patch.
> 

I wanted to say that it's working. thanx
Dmitry