Partition badness

Partition badness

[Sean MacLennan]

I tried to run 2.6.28-rc7 and got the following for every partition,
both nor and nand, after the first partition:

ffc00000.nor: Found 1 x16 devices at 0x0 in 16-bit bank
 Amd/Fujitsu Extended Query Table at 0x0040
ffc00000.nor: CFI does not contain boot bank location. Assuming top.
number of CFI chips: 1
cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
RedBoot partition parsing not available
Creating 4 MTD partitions on "ffc00000.nor":
0x00000000-0x00020000 : "splash"
0x00300000-0x00340000 : "fpga"
------------[ cut here ]------------
Badness at c0058928 [verbose debug info unavailable]
NIP: c0058928 LR: c01320c4 CTR: 00000000
REGS: cf821b50 TRAP: 0700   Not tainted  (2.6.28-rc7-pika)
MSR: 00029000 <EE,ME>  CR: 24000042  XER: 2000005f
TASK = cf81f960[1] 'swapper' THREAD: cf820000
GPR00: cf995a00 cf821c00 cf81f960 cf987630 00000000 c02abd18 cf9959ff c02abd18 
GPR08: cf89b664 cf99d4bc 00000d9f 00000001 84000048 00000000 0ffac000 007fff99 
GPR16: 00400450 00800000 007fff00 0ffa93c4 00000000 ffffffff 00000000 00000001 
GPR24: 00000000 c0310000 00000004 00000001 c030ef2c cf987630 cf98fbe0 cf987630 
NIP [c0058928] bdi_register+0x38/0xe4
LR [c01320c4] add_disk+0x104/0x150
Call Trace:
[cf821c00] [c00ba6f0] __sysfs_add_one+0x24/0xc8 (unreliable)
[cf821c90] [c01320c4] add_disk+0x104/0x150
[cf821cc0] [c0185f8c] add_mtd_blktrans_dev+0x180/0x2a8
[cf821ce0] [c0186c18] mtdblock_add_mtd+0x60/0x80
[cf821cf0] [c0185bbc] blktrans_notify_add+0x40/0x80
[cf821d10] [c01825a0] add_mtd_device+0xa0/0x150
[cf821d30] [c0183d4c] add_mtd_partitions+0x4b0/0x648
[cf821d60] [c02edc14] of_flash_probe+0x1c8/0x4e0
[cf821dc0] [c01bd2a0] of_platform_device_probe+0x58/0x80
[cf821de0] [c0162f50] driver_probe_device+0xb8/0x1cc
[cf821e00] [c0163108] __driver_attach+0xa4/0xa8
[cf821e20] [c0161d98] bus_for_each_dev+0x58/0x94
[cf821e50] [c0162b84] driver_attach+0x24/0x34
[cf821e60] [c01627a8] bus_add_driver+0xb0/0x1f0
[cf821e80] [c01635b8] driver_register+0x5c/0x160
[cf821ea0] [c01bd4d4] of_register_driver+0x54/0x70
[cf821eb0] [c02e649c] of_flash_init+0x20/0x30
[cf821ec0] [c0001420] do_one_initcall+0x34/0x198
[cf821fd0] [c02d0824] kernel_init+0x94/0xf8
[cf821ff0] [c000d244] kernel_thread+0x50/0x6c
Instruction dump:
93e1008c 90010094 80030050 7c7f1b78 90c10014 90e10018 30c0ffff 7d660110 
9101001c 91210020 91410024 7ca72b78 <0f0b0000> 2f8b0000 38a00000 7c661b78 

This is on a PIKA warp with a PPC processor. I haven't really digged
into it yet (I was actually looking for a usb problem) but I am hoping
somebody recognizes the problem.

Yes there is a large gap between the "splash" partition and the fpga
partition. The first partition is at the start of flash and the others
at the end.

Cheers,
   Sean

Re: Partition badness

[Hinko Kocevar]

Sean MacLennan wrote:
> I tried to run 2.6.28-rc7 and got the following for every partition,
> both nor and nand, after the first partition:
> 
> ffc00000.nor: Found 1 x16 devices at 0x0 in 16-bit bank
>  Amd/Fujitsu Extended Query Table at 0x0040
> ffc00000.nor: CFI does not contain boot bank location. Assuming top.
> number of CFI chips: 1
> cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
> RedBoot partition parsing not available
> Creating 4 MTD partitions on "ffc00000.nor":
> 0x00000000-0x00020000 : "splash"
> 0x00300000-0x00340000 : "fpga"
> ------------[ cut here ]------------
> Badness at c0058928 [verbose debug info unavailable]
> NIP: c0058928 LR: c01320c4 CTR: 00000000
> REGS: cf821b50 TRAP: 0700   Not tainted  (2.6.28-rc7-pika)
> MSR: 00029000 <EE,ME>  CR: 24000042  XER: 2000005f
> TASK = cf81f960[1] 'swapper' THREAD: cf820000
> GPR00: cf995a00 cf821c00 cf81f960 cf987630 00000000 c02abd18 cf9959ff c02abd18 
> GPR08: cf89b664 cf99d4bc 00000d9f 00000001 84000048 00000000 0ffac000 007fff99 
> GPR16: 00400450 00800000 007fff00 0ffa93c4 00000000 ffffffff 00000000 00000001 
> GPR24: 00000000 c0310000 00000004 00000001 c030ef2c cf987630 cf98fbe0 cf987630 
> NIP [c0058928] bdi_register+0x38/0xe4
> LR [c01320c4] add_disk+0x104/0x150
> Call Trace:
> [cf821c00] [c00ba6f0] __sysfs_add_one+0x24/0xc8 (unreliable)
> [cf821c90] [c01320c4] add_disk+0x104/0x150
> [cf821cc0] [c0185f8c] add_mtd_blktrans_dev+0x180/0x2a8
> [cf821ce0] [c0186c18] mtdblock_add_mtd+0x60/0x80
> [cf821cf0] [c0185bbc] blktrans_notify_add+0x40/0x80
> [cf821d10] [c01825a0] add_mtd_device+0xa0/0x150
> [cf821d30] [c0183d4c] add_mtd_partitions+0x4b0/0x648
> [cf821d60] [c02edc14] of_flash_probe+0x1c8/0x4e0
> [cf821dc0] [c01bd2a0] of_platform_device_probe+0x58/0x80
> [cf821de0] [c0162f50] driver_probe_device+0xb8/0x1cc
> [cf821e00] [c0163108] __driver_attach+0xa4/0xa8
> [cf821e20] [c0161d98] bus_for_each_dev+0x58/0x94
> [cf821e50] [c0162b84] driver_attach+0x24/0x34
> [cf821e60] [c01627a8] bus_add_driver+0xb0/0x1f0
> [cf821e80] [c01635b8] driver_register+0x5c/0x160
> [cf821ea0] [c01bd4d4] of_register_driver+0x54/0x70
> [cf821eb0] [c02e649c] of_flash_init+0x20/0x30
> [cf821ec0] [c0001420] do_one_initcall+0x34/0x198
> [cf821fd0] [c02d0824] kernel_init+0x94/0xf8
> [cf821ff0] [c000d244] kernel_thread+0x50/0x6c
> Instruction dump:
> 93e1008c 90010094 80030050 7c7f1b78 90c10014 90e10018 30c0ffff 7d660110 
> 9101001c 91210020 91410024 7ca72b78 <0f0b0000> 2f8b0000 38a00000 7c661b78 
> 
> This is on a PIKA warp with a PPC processor. I haven't really digged
> into it yet (I was actually looking for a usb problem) but I am hoping
> somebody recognizes the problem.
> 
> Yes there is a large gap between the "splash" partition and the fpga
> partition. The first partition is at the start of flash and the others
> at the end.
> 

I had similar problems when inserting the mtd driver for the second time.
This helped me:
http://forum.soft32.com/linux/bdi-register-sysfs-bdi-device-queue-ftopict469102.html.

HTH,
Hinko

-- 
Hinko Kočevar, OSS developer
ČETRTA POT, d.o.o.
Planina 3, 4000 Kranj, SI EU
tel     ++386 (0) 4 280 66 03
e-mail  hinko.kocevar@cetrtapot.si
http    www.cetrtapot.si

Re: Partition badness

[Sean MacLennan]

On Thu, 04 Dec 2008 09:25:10 +0100
"Hinko Kocevar" <hinko.kocevar@cetrtapot.si> wrote:

> I had similar problems when inserting the mtd driver for the second
> time. This helped me:
> http://forum.soft32.com/linux/bdi-register-sysfs-bdi-device-queue-ftopict469102.html.

That patch was added, and they made it a WARN_ON()!!!!! Which means I
get 8 WARN_ON messages on every boot :(

Thanks! I'll just remove the WARN_ON.

Cheers,
   Sean

Re: Partition badness

[Paulius Zaleckas]

Sean MacLennan wrote:
> On Thu, 04 Dec 2008 09:25:10 +0100
> "Hinko Kocevar" <hinko.kocevar@cetrtapot.si> wrote:
> 
>> I had similar problems when inserting the mtd driver for the second
>> time. This helped me:
>> http://forum.soft32.com/linux/bdi-register-sysfs-bdi-device-queue-ftopict469102.html.
> 
> That patch was added, and they made it a WARN_ON()!!!!! Which means I
> get 8 WARN_ON messages on every boot :(
> 
> Thanks! I'll just remove the WARN_ON.

This WARN_ON is to detect bugs in driver... Actually MTD has to be fixed,
but not this WARN_ON...

> Cheers,
>    Sean
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>