2009-07-03 03:00:23     Write data to BF518 SD with exception context

2009-07-03 03:00:23     Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76731   

Hi sir,

    When copy a small file from sdram to SD card,the system output many message and dead after about 20 seconds.From cache write back to SD card happen this problem.The  Sd card file format is fat32.How to find out the uClinux bug and avoid

this problem?

Output message:

root:/>

root:/> dmesg

Linux version 2.6.30-ADI-2010R1-pre (xiechl@localhost.localdomain) (gcc version 4.1.2 (ADI svn)) #18 Wed Jul 1 22:06:08 CST 2009

Board Memory: 64MB

Kernel Managed Memory: 64MB

Memory map:

  fixedcode = 0x00000400-0x00000490

  text      = 0x00001000-0x0011ddf0

  rodata    = 0x0011ddf0-0x001988f8

  bss       = 0x00199000-0x001ab890

  data      = 0x001ab890-0x001c0000

    stack   = 0x001be000-0x001c0000

  init      = 0x001c0000-0x00665000

  available = 0x00665000-0x03eff000

  DMA Zone  = 0x03f00000-0x04000000

Hardware Trace Active and Enabled

Boot Mode: 3

Blackfin support (C) 2004-2009 Analog Devices, Inc.

Compiled for ADSP-BF518 Rev 0.0

Blackfin Linux support by   blackfin.uclinux.org/

Processor Speed: 400 MHz core clock and 80 MHz System Clock

boot memmap: 0000000000665000 - 0000000003eff000 (usable)

On node 0 totalpages: 16127

free_area_init_node: node 0, pgdat 001bc9c4, node_mem_map 00667000

  DMA zone: 126 pages used for memmap

  DMA zone: 0 pages reserved

  DMA zone: 16001 pages, LIFO batch:0

NOMPU: setting up cplb tables

Instruction Cache Enabled for CPU0

  External memory: cacheable in instruction cache

Data Cache Enabled for CPU0

  External memory: cacheable (write-back) in data cache

Built 1 zonelists in Zone order, mobility grouping off.  Total pages: 16001

Kernel command line: console=ttyBF0,57600

NR_IRQS:119

Configuring Blackfin Priority Driven Interrupts

PID hash table entries: 256 (order: 8, 1024 bytes)

Console: colour dummy device 80x25

console [ttyBF0] enabled

Dentry cache hash table entries: 8192 (order: 3, 32768 bytes)

Inode-cache hash table entries: 4096 (order: 2, 16384 bytes)

Kernel managed physical pages: 16127

Memory available: 57388k/65536k RAM, (4756k init code, 1139k kernel code, 649k data, 1024k dma, 576k reserved)

Calibrating delay loop... 796.67 BogoMIPS (lpj=1593344)

Security Framework initialized

Mount-cache hash table entries: 512

Blackfin Scratchpad data SRAM: 4 KB

Blackfin L1 Data A SRAM: 16 KB (16 KB free)

Blackfin L1 Data B SRAM: 16 KB (16 KB free)

Blackfin L1 Instruction SRAM: 48 KB (42 KB free)

net_namespace: 296 bytes

NET: Registered protocol family 16

Blackfin DMA Controller

ezbrd_init(): registering device resources

bio: create slab <bio-0> at 0

NET: Registered protocol family 2

IP route cache hash table entries: 1024 (order: 0, 4096 bytes)

TCP established hash table entries: 2048 (order: 2, 16384 bytes)

TCP bind hash table entries: 2048 (order: 1, 8192 bytes)

TCP: Hash tables configured (established 2048 bind 2048)

TCP reno registered

NET: Registered protocol family 1

Setting up Blackfin MMR debugfs

msgmni has been set to 112

io scheduler noop registered (default)

Serial: Blackfin serial driver

bfin-uart.1: ttyBF0 at MMIO 0xffc00400 (irq = 29) is a BFIN-UART

brd: module loaded

bfin_mii_bus: probed

bfin_mac: attached PHY driver [Generic PHY] (mii_bus:phy_addr=0:03, irq=-1, mdc_clk=2500000Hz(mdc_div=15)@sclk=80MHz)

bfin_mac bfin_mac.0: Blackfin on-chip Ethernet MAC driver, Version 1.1

bfin-spi bfin-spi.0: Blackfin on-chip SPI Controller Driver, Version 1.0, regs_base@ffc00500, dma channel@7

bfin-spi bfin-spi.1: Blackfin on-chip SPI Controller Driver, Version 1.0, regs_base@ffc03400, dma channel@5

rtc-bfin rtc-bfin: rtc core: registered rtc-bfin as rtc0

i2c /dev entries driver

i2c-bfin-twi i2c-bfin-twi.0: Blackfin BF5xx on-chip I2C TWI Contoller, regs_base@ffc01400

bfin-wdt: initialized: timeout=20 sec (nowayout=0)

drivers/mmc/host/bfin_sdh.c sdh_init 664

drivers/mmc/host/bfin_sdh.c sdh_probe 504

dma_alloc_init: dma_page @ 0x038d3000 - 256 pages at 0x03f00000

TCP cubic registered

NET: Registered protocol family 17

Distributed Switch Architecture driver version 0.1

eth0[0]: detected a KSZ8893M switch

dsa slave smi: probed

rtc-bfin rtc-bfin: setting system clock to 1996-09-09 13:27:06 UTC (842275626)

Freeing unused kernel memory: 4756k freed

mmc0: new SD card at address d921

mmcblk0: mmc0:d921 SD02G 1.89 GiB

mmcblk0:

eth1: link up, 100 Mb/s, full duplex

Trying 100/FULL

PHY: 0:03 - Link is Up - 100/Full

Exception context:

Undefined instruction

- May be used to emulate instructions that are not defined for

   a particular processor implementation.

Kernel OOPS in progress

Deferred Exception context

CURRENT PROCESS:

COMM=pdflush PID=84

CPU = 0

invalid mm

return address: [0x00093a7e]; contents of:

0x00093a50:  a3b0  9331  4f18  e121  0110  5008  e131  0000

0x00093a60:  e134  003f  3600  e132  001f  e130  0001  e13b

0x00093a70:  0041  3610  345d  e139  001f  e137  0001 [3c9b]

0x00093a80:  e093  006b  2005  3220  9131  67d1  9331  9134

ADSP-BF518-0.0 400(MHz CCLK) 80(MHz SCLK) (mpu off)

Linux version 2.6.30-ADI-2010R1-pre (xiechl@localhost.localdomain) (gcc version 4.1.2 (ADI svn)) #18 Wed Jul 1 22:06:08 CST 2009

SEQUENCER STATUS:        Not tainted

SEQSTAT: 00060021  IPEND: 8008  IMASK: ffff  SYSCFG: 0006

  EXCAUSE   : 0x21

  physical IVG3 asserted : <0xffa00624> { _trap + 0x0 }

  physical IVG15 asserted : <0xffa00e3c> { _evt_system_call + 0x0 }

  logical irq   6 mapped  : <0xffa002f8> { _timer_interrupt + 0x0 }

  logical irq  21 mapped  : <0x000c4b4c> { _bfin_rtc_interrupt + 0x0 }

  logical irq  24 mapped  : <0x000ce9fc> { _sdh_dma_irq + 0x0 }

  logical irq  27 mapped  : <0x000c7f20> { _bfin_twi_interrupt_entry + 0x0 }

  logical irq  29 mapped  : <0x000b2400> { _bfin_serial_dma_rx_int + 0x0 }

  logical irq  30 mapped  : <0x000b278c> { _bfin_serial_dma_tx_int + 0x0 }

  logical irq  35 mapped  : <0x000bac44> { _bfin_mac_interrupt + 0x0 }

  logical irq  58 mapped  : <0x000cea20> { _sdh_stat_irq + 0x0 }

RETE: <0x00000000> /* Maybe null pointer? */

RETN: <0x038dfce8> /* kernel dynamic memory */

RETX: <0x00000480> /* Maybe fixed code section */

RETS: <0x0009418c> { _radix_tree_gang_lookup_tag_slot + 0x74 }

PC  : <0x00093a7e> { ___lookup_tag + 0x4a }

DCPLB_FAULT_ADDR: <0x038dfce8> /* kernel dynamic memory */

ICPLB_FAULT_ADDR: <0x00093a7e> { ___lookup_tag + 0x4a }

PROCESSOR STATE:

R0 : 00000110    R1 : 00000110    R2 : 00000000    R3 : 004e1d08

R4 : 00000fff    R5 : 0000000e    R6 : 00000000    R7 : 00000000

P0 : 004e1dcc    P1 : 00000000    P2 : 0000000c    P3 : 038dfdcc

P4 : 02339210    P5 : 00000002    FP : 038dfe04    SP : 038dfc0c

LB0: 00096d1a    LT0: 00096d1a    LC0: 00000000

LB1: 00093efc    LT1: 00093ece    LC1: 00000000

B0 : 00000110    L0 : 00000000    M0 : 0000003f    I0 : 00000001

B1 : 0000001f    L1 : 00000000    M1 : 0000000e    I1 : 00000000

B2 : 00000110    L2 : 00000000    M2 : 038dfdcc    I2 : 0000001f

B3 : 00000041    L3 : 00000000    M3 : 00000001    I3 : 00000002

A0.w: 0000b383   A0.x: 00000000   A1.w: 00000fad   A1.x: 00000000

USP : 00000000  ASTAT: 02002000

Hardware Trace:

   0 Target : <0x00004ea4> { _trap_c + 0x0 }

     Source : <0xffa005b8> { _exception_to_level5 + 0x94 } CALL pcrel

   1 Target : <0xffa00524> { _exception_to_level5 + 0x0 }

     Source : <0xffa003fc> { _bfin_return_from_exception + 0x18 } RTX

   2 Target : <0xffa003e4> { _bfin_return_from_exception + 0x0 }

     Source : <0xffa00478> { _ex_trap_c + 0x4c } JUMP.S

   3 Target : <0xffa0042c> { _ex_trap_c + 0x0 }

     Source : <0xffa0067e> { _trap + 0x5a } JUMP (P4)

   4 Target : <0xffa00624> { _trap + 0x0 }

     Source : <0x00093a7a> { ___lookup_tag + 0x46 } 0xe137

   5 Target : <0x00093a34> { ___lookup_tag + 0x0 }

     Source : <0x00094188> { _radix_tree_gang_lookup_tag_slot + 0x70 } CALL pcrel

   6 Target : <0x00094174> { _radix_tree_gang_lookup_tag_slot + 0x5c }

     Source : <0x0009416a> { _radix_tree_gang_lookup_tag_slot + 0x52 } JUMP.S

   7 Target : <0x00094118> { _radix_tree_gang_lookup_tag_slot + 0x0 }

     Source : <0x0002e23a> { _find_get_pages_tag + 0x26 } CALL pcrel

   8 Target : <0x0002e228> { _find_get_pages_tag + 0x14 }

     Source : <0x0002e226> { _find_get_pages_tag + 0x12 } JUMP.S

   9 Target : <0x0002e214> { _find_get_pages_tag + 0x0 }

     Source : <0x0003530e> { _pagevec_lookup_tag + 0x1a } CALL pcrel

  10 Target : <0x000352f4> { _pagevec_lookup_tag + 0x0 }

     Source : <0x00033480> { _write_cache_pages + 0xb0 } CALL pcrel

  11 Target : <0x00033428> { _write_cache_pages + 0x58 }

     Source : <0x0003340a> { _write_cache_pages + 0x3a } IF CC JUMP

  12 Target : <0x00033408> { _write_cache_pages + 0x38 }

     Source : <0x00033670> { _write_cache_pages + 0x2a0 } JUMP.S

  13 Target : <0x00033668> { _write_cache_pages + 0x298 }

     Source : <0x000333f4> { _write_cache_pages + 0x24 } IF !CC JUMP

  14 Target : <0x000333d0> { _write_cache_pages + 0x0 }

     Source : <0x0005eb24> { _mpage_writepages + 0x2c } CALL pcrel

  15 Target : <0x0005eaf8> { _mpage_writepages + 0x0 }

     Source : <0x0007d27c> { _fat_writepages + 0x8 } JUMP.L

Kernel Stack

Stack info:

SP: [0x038dff18] <0x038dff18> /* kernel dynamic memory */

FP: (0x038dff88)

Memory from 0x038dff10 to 038e0000

038dff10: 038dff60 <00034290>[001a4680] 038dff64  001a4694  001aef4c  039ddae0  001a4640

038dff30: 00000000  00000000  00000000  00000000  00000000  00000400  00000000  00000000

038dff50: 00000000  00000000  00000000  00000031  00000b38  0000059c <000348c4> 038dffcc

038dff70: 038dffc0  001b2508  001aef4c  001a4644  001a4640  00000000 (038dffa0)<0000ae00>

038dff90: 00000000  00000000 <00034a2c> 00034a10 (00000000) 00000000  00000000  00000000

038dffb0: 00000000  00034a10  00000000 <038de000> 03e993e0  000341ec  00000000  038dffcc

038dffd0: 038dffcc  0023ea7a <0001ece8> 00000000  00000000  00000000 <0000146e> 00000000

038dfff0: 00000000  00000000  ffffffff  00000006  464c457f

Return addresses in stack:

    address : <0x00034290> { _background_writeout + 0xa4 }

    address : <0x000348c4> { ___pdflush + 0xc4 }

   frame  1 : <0x0000ae00> { _set_user_nice + 0xa0 }

    address : <0x00034a2c> { _pdflush + 0x1c }

    address : <0x038de000> /* kernel dynamic memory */

    address : <0x0001ece8> { _kthread + 0x54 }

    address : <0x0000146e> { _kernel_thread_helper + 0x6 }

Modules linked in:

Kernel panic - not syncing: Kernel exception

Hardware Trace:

Stack info:

SP: [0x038dfb24] <0x038dfb24> /* kernel dynamic memory */

FP: (0x038dfd98)

Memory from 0x038dfb20 to 038e0000

038dfb20: 038dfdcc [0016d544]<0000f574> 0019fc0c  0016d544  0019e49e  0019e49e  0019e49e

038dfb40: 038dfb60  038dfb60 <0000521a> 038dfc0c  ffe02014  038dfdcc  0016cd44  038dfdd4

038dfb60: 001adfdc  0000003f  001adfdc  0000003f  00030001  0016cd44  00030001  001adfdc

038dfb80: 0000003f  603b6800 <0008b72c> 00030001  03e8b0d8  00000001  01000001  00000000

038dfba0: 6c620910  000f4940  0000c810  00000010  00000001 <0008b844> 002e67c0  002e67c0

038dfbc0: 030e581c  03e8b0c8  00000000  00000005  00000001  00000070  03ea3a20  03ea3a20

038dfbe0: 00000010 <0005b49e><ffa005bc> 0019b000  00008008  00000006  0000000e  00000fff

038dfc00: 0000000e  00000fff  00000480  00000480  00008008  00060021  00000000  038dfce8

038dfc20: 00000480  00093a7e <0009418c> 00000110  02002000  00093efc  00096d1a  00093ece

038dfc40: 00096d1a  00000000  00000000  00000fad  00000000  0000b383  00000000  00000041

038dfc60: 00000110  0000001f  00000110  00000000  00000000  00000000  00000000  00000001

038dfc80: 038dfdcc  0000000e  0000003f  00000002  0000001f  00000000  00000001  00000000

038dfca0: 038dfe04  00000002  02339210  038dfdcc  0000000c  00000000  004e1dcc  00000000

038dfcc0: 00000000  0000000e  00000fff  004e1d08  00000000  00000110  00000110  00000110

038dfce0: 004e1dcc  00000006  00000006  038dfe04  02339210  038dfd24  038dfdcc  00000000

038dfd00: 00000000  0000000e  00000fff  0000000e  00000fff  038dfde8  0000000e  038dfd24

038dfd20: 00000000  038dfd2c <0002e23e> 038dfdcc  038dff38  038dfdcc  00000000  00000000

038dfd40: 004e1d0c  00000000  038dfde8  038dff3c  038dfde8  0000000e  00000000  004e1d08

038dfd60:<00035312> 038dfdcc  038dff38  ffffffff  00000000  00000000  00000000  00000400

038dfd80: 038dfde8  038dff3c  ffffffff  0000000e  038dfdcc <00033484>(038dff38)<00033484>

038dfda0: 038dff3c  006c8c60  00000000  0000000e  ffffffff  03e8b164  00000000  00000001

038dfdc0: 00000000  00000000  00000000  ffffffff  03e8b164  00000000  00000001  00000000

038dfde0: 00000000  00000000  00000000 <ffa002c6><00057c18><0007e352><00033dc6><00671ce0>

038dfe00: 006f8cb0  00000000  038dff38 <0005eb28> 038dff38  038dff38  03b60c00  00000007

038dfe20: 00000000  00000000  004e1d08  038dff3c  038dff38  0005e60c  038dfe3c  00000000

038dfe40: 00000000  0007d048  00000001 <000336d4> 00000007  03b60c00  00000007  004e1ecc

038dfe60:<00053ac8> 004e1c70  004e1dc4  00000000  004e1dc4 <00053bec> 004e1c70  038dff38

038dfe80:<038de000> 00000000  0023ec1b  00000000  ffff4b7a  00000000  00564000  001b2a38

038dfea0: 001b2508 <00057b1c> 00564000  0000003f  001a4644  00615f30 <00057c18><00053ee6>

038dfec0: 004e1c70  03e8b164 <038de000> 00000000  0023ec1b  00000000 <0000ac14> 001adaec

038dfee0:<00033906> 001acfa0 <00054136> 03b60c00  001b2a38  038dff38  03b60c3c  00000000

038dff00: 00000000  038dff60  001acfa0 <0003425e> 038dff60 <00034290> 001a4680  038dff64

038dff20: 001a4694  001aef4c  039ddae0  001a4640  00000000  00000000 (00000000) 00000000

038dff40: 00000000  00000400  00000000  00000000  00000000  00000000  00000000  00000031

038dff60: 00000b38  0000059c <000348c4> 038dffcc  038dffc0  001b2508  001aef4c  001a4644

038dff80: 001a4640  00000000  038dffa0 <0000ae00> 00000000  00000000 <00034a2c> 00034a10

038dffa0: 00000000  00000000  00000000  00000000  00000000  00034a10  00000000 <038de000>

038dffc0: 03e993e0  000341ec  00000000  038dffcc  038dffcc  0023ea7a <0001ece8> 00000000

038dffe0: 00000000  00000000 <0000146e> 00000000  00000000  00000000  ffffffff  00000006

038e0000: 464c457f

Return addresses in stack:

    address : <0x0000f574> { _panic + 0x4c }

    address : <0x0000521a> { _trap_c + 0x376 }

    address : <0x0008b72c> { _get_request + 0xb0 }

    address : <0x0008b844> { _get_request_wait + 0x18 }

    address : <0x0005b49e> { _bio_phys_segments + 0x22 }

    address : <0xffa005bc> { _exception_to_level5 + 0x98 }

    address : <0x0009418c> { _radix_tree_gang_lookup_tag_slot + 0x74 }

    address : <0x0002e23e> { _find_get_pages_tag + 0x2a }

    address : <0x00035312> { _pagevec_lookup_tag + 0x1e }

    address : <0x00033484> { _write_cache_pages + 0xb4 }

   frame  1 : <0x00033484> { _write_cache_pages + 0xb4 }

    address : <0xffa002c6> { _asm_do_IRQ + 0x36 }

    address : <0x00057c18> { ___getblk + 0x10 }

    address : <0x0007e352> { _fat_time_unix2fat + 0x1e6 }

    address : <0x00033dc6> { _task_dirty_inc + 0x1e }

    address : <0x00671ce0> /* kernel dynamic memory */

    address : <0x0005eb28> { _mpage_writepages + 0x30 }

    address : <0x000336d4> { _do_writepages + 0x2c }

    address : <0x00053ac8> { ___sync_single_inode + 0x34 }

    address : <0x00053bec> { ___writeback_single_inode + 0x34 }

    address : <0x038de000> /* kernel dynamic memory */

    address : <0x00057b1c> { ___find_get_block + 0x20 }

    address : <0x00057c18> { ___getblk + 0x10 }

    address : <0x00053ee6> { _generic_sync_sb_inodes + 0x122 }

    address : <0x038de000> /* kernel dynamic memory */

    address : <0x0000ac14> { _dequeue_entity + 0x2c }

    address : <0x00033906> { _get_dirty_limits + 0x16 }

    address : <0x00054136> { _writeback_inodes + 0x6a }

    address : <0x0003425e> { _background_writeout + 0x72 }

    address : <0x00034290> { _background_writeout + 0xa4 }

    address : <0x000348c4> { ___pdflush + 0xc4 }

    address : <0x0000ae00> { _set_user_nice + 0xa0 }

    address : <0x00034a2c> { _pdflush + 0x1c }

    address : <0x038de000> /* kernel dynamic memory */

    address : <0x0001ece8> { _kthread + 0x54 }

    address : <0x0000146e> { _kernel_thread_helper + 0x6 }

--Bill

QuoteReplyEditDelete

2009-07-03 09:55:23     Re: Write data to BF518 SD with exception context

Robin Getz (UNITED STATES)

Message: 76776   

Bill:

Can you do a "make bugreport", and post the bugreport.tgz that it makes?

The next thing to do - is do not use trunk - use the 2009R1 branch.

QuoteReplyEditDelete

2009-07-05 21:36:11     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76813   

Robin,

      Thank you.I post bugreport.tar.gz.When I add  debug message in lib/radix-tree.c  __lookup_tag function,It is ok  when  copy file to sd card.It is problem to link 2009R1.

--Bill

bugreport.tar.gz

QuoteReplyEditDelete

2009-07-05 22:24:13     Re: Write data to BF518 SD with exception context

Yi Li (CHINA)

Message: 76815   

Bill,

If using 2009R1, can you reproduce your problem?

-Yi

QuoteReplyEditDelete

2009-07-05 22:47:00     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76816   

Yi,

   I will try 2009R1 to test.But It is trouble to download   blackfin.uclinux.org/gf/download/frsrelease/380/6509/uClinux-dist-2009R1-RC1.tar.bz2 with "Invalid/empty data or corrupted file" message.

--Bill

QuoteReplyEditDelete

2009-07-07 03:32:02     Re: Write data to BF518 SD with exception context

Yi Li (CHINA)

Message: 76844   

Bill,

I tested this link and it works fine:

  blackfin.uclinux.org/gf/download/frsrelease/380/6527/uClinux-dist-2009R1-RC1.tar.bz2

-Yi

QuoteReplyEditDelete

2009-07-07 04:40:58     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76880   

Yi,

   Thank you.I have been working in 2009R1-RC1,But the 518 mmc host also have this promblem.It is no problem using   blackfin.uclinux.org/gf/download/frsrelease/380/6515/uImage_bf518_ezbrd to write sd card.Compareed both image dmesg information,it is almost  same  ecept for

Memory map:

  fixedcode = 0x00000400-0x00000490

  text      = 0x00001000-0x00116a50

  rodata    = 0x00116a50-0x00169918

  bss       = 0x0016a000-0x0017c358

  data      = 0x0017c358-0x00190000

    stack   = 0x0018e000-0x00190000

  init      = 0x00190000-0x00722000

  available = 0x00722000-0x03eff000

  DMA Zone  = 0x03f00000-0x04000000

It is puzzled to both image file to induce two result to write sd card.

--Bill

QuoteReplyEditDelete

2009-07-07 05:17:49     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76881   

Hi Sir,

  Because of write sd card with exception in 2009R1+518,It is happened exception in lib/radix-tree.c

static unsigned int

__lookup_tag(struct radix_tree_node *slot, void ***results, unsigned long index,

    unsigned int max_items, unsigned long *next_index, unsigned int tag)

{

    unsigned int nr_found = 0;

    unsigned int shift, height;

    height = slot->height;

    if (height == 0)

        goto out;

    shift = (height-1) * RADIX_TREE_MAP_SHIFT;

    while (height > 0) {

        unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK ;

#ifdef  SD_DEBU

   printk(KERN_DEBUG "%s %s %d i=%d RADIX_TREE_MAP_MASK=%d index=%d\n",__FILE__,__FUNCTION__,__LINE__,i,RADIX_TREE_MAP_MASK,index);

#endif

        for (;;) {

            if (tag_get(slot, tag, i))

                break;

            index &= ~((1UL << shift) - 1);

            index += 1UL << shift;

            if (index == 0)

                goto out;    /* 32-bit wraparound */

            i++;

            if (i == RADIX_TREE_MAP_SIZE)

                goto out;

        }

        height--;

        if (height == 0) {    /* Bottom level: grab some items */

            unsigned long j = index & RADIX_TREE_MAP_MASK;

            for ( ; j < RADIX_TREE_MAP_SIZE; j++) {

                index++;

                if (!tag_get(slot, tag, j))

                    continue;

                /*

                 * Even though the tag was found set, we need to

                 * recheck that we have a non-NULL node, because

                 * if this lookup is lockless, it may have been

                 * subsequently deleted.

                 *

                 * Similar care must be taken in any place that

                 * lookup ->slots[x] without a lock (ie. can't

                 * rely on its value remaining the same).

                 */

                if (slot->slots[j]) {

                    results[nr_found++] = &(slot->slots[j]);

                    if (nr_found == max_items)

                        goto out;

                }

            }

        }

        shift -= RADIX_TREE_MAP_SHIFT;

        slot = rcu_dereference(slot->slots[i]);

        if (slot == NULL)

            break;

    }

out:

    *next_index = index;

    return nr_found;

}

It is not  reappear exception write data to sd card when  added  debug code with red. I not found bug in __lookup_tag function.Who can help me fix thi bug? Thank you.

--Bill

QuoteReplyEditDelete

2009-07-07 06:08:55     Re: Write data to BF518 SD with exception context

Yi Li (CHINA)

Message: 76882   

Please make sure you are using 2009R1 toolchain (2009R1_RC9). You can download from:    blackfin.uclinux.org/gf/project/toolchain/frs/?action=FrsReleaseBrowse&frs_package_id=74.

If you are using 2009R1 branch kernel. also 2009R1_RC9 toolchain, and you still see the problem, you can do "make bugreport" under uclinux-dist/ and post bugreport.tgz here.

We tested the SD host on BF518 everyday and it works well, so we need you configuration to reproduce.

-Yi

QuoteReplyEditDelete

2009-07-07 06:48:07     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76883   

Yi,

    Thank you. I will install 2009R1 toolchain.Please help me analyze this assembler code.

Undefined instruction

- May be used to emulate instructions that are not defined for

   a particular processor implementation.

Kernel OOPS in progress

Deferred Exception context

CURRENT PROCESS:

COMM=pdflush PID=77

CPU = 0

invalid mm

return address: [0x0008893c]; contents of:

0x00088910:  05e3  3218  014f  a2f0  3609  3600  915d  0c45

0x00088920:  1843  5cad  a370  5b12  4f18  6fd4  e130  0000

0x00088930:  3410  344d  e133  001f  e134  0001 [3c99] e093

0x00088940:  007d  2003  3218  6fd4  304c  3002  4048  61f9

ADSP-BF518-0.0 400(MHz CCLK) 80(MHz SCLK) (mpu off)

Linux version 2.6.28.10-ADI-2009R1

Built with gcc version 4.1.2 (ADI svn)

SEQUENCER STATUS:        Not tainted

SEQSTAT: 00060021  IPEND: 8030  SYSCFG: 0006

  EXCAUSE   : 0x21

  interrupts disabled

  physical IVG5 asserted : <0xffa00b90> { _evt_ivhw + 0x0 }

  physical IVG15 asserted : <0xffa00de0> { _evt_system_call + 0x0 }

  logical irq   6 mapped  : <0xffa00328> { _timer_interrupt + 0x0 }

  logical irq  21 mapped  : <0x000b68cc> { _bfin_rtc_interrupt + 0x0 }

  logical irq  24 mapped  : <0x000c0e70> { _sdh_dma_irq + 0x0 }

  logical irq  27 mapped  : <0x000b93a0> { _bfin_twi_interrupt_entry + 0x0 }

  logical irq  29 mapped  : <0x000a4ed4> { _bfin_serial_dma_rx_int + 0x0 }

  logical irq  30 mapped  : <0x000a5210> { _bfin_serial_dma_tx_int + 0x0 }

  logical irq  35 mapped  : <0x000ace54> { _bfin_mac_interrupt + 0x0 }

  logical irq  58 mapped  : <0x000c0e94> { _sdh_stat_irq + 0x0 }

RETE: <0x00000000> /* Maybe null pointer? */

RETN: <0x03bd7d24> /* kernel dynamic memory */

RETX: <0x00000480> /* Maybe fixed code section */

RETS: <0x000891a0> { _radix_tree_gang_lookup_tag_slot + 0x74 }

PC  : <0x0008893c> { ___lookup_tag + 0x2c }

DCPLB_FAULT_ADDR: <0x03bd7d58> /* kernel dynamic memory */

ICPLB_FAULT_ADDR: <0x0008893c> { ___lookup_tag + 0x2c }

00088910 <___lookup_tag>:

   88910:    e3 05           [--SP] = (R7:4, P5:3);

   88912:    18 32           P3 = R0;

   88914:    4f 01           [--SP] = FP;

   88916:    f0 a2           R0 = [SP + 0x2c];

   88918:    09 36           B1 = R1;

   8891a:    00 36           B0 = R0;

   8891c:    5d 91           P5 = [P3];

   8891e:    45 0c           CC = P5 == 0x0;

   88920:    43 18           IF CC JUMP 0x889a6 <___lookup_tag+0x96>;

   88922:    ad 5c           P2 = P5 + (P5 << 0x1);

   88924:    70 a3           R0 = [SP + 0x34];

   88926:    12 5b           P4 = P2 << 0x1;

   88928:    18 4f           R0 <<= 0x3;

   8892a:    d4 6f           P4 += -0x6;        /* ( -6) */

   8892c:    30 e1 00 00     I0 = 0x0 (X);        /*        I0=0x0 <__ebss_l2>(  0) */

   88930:    10 34           I2 = R0;

   88932:    4d 34           I1 = P5;

   88934:    33 e1 1f 00     I3 = 0x1f (X);        /*        I3=0x1f( 31) */

   88938:    34 e1 01 00     M0 = 0x1 (X);        /*        M0=0x1(  1) */

   8893c:    99 3c           LC1 = I1;

   8893e:    93 e0 7d 00     LSETUP(0x88944 <___lookup_tag+0x34>, 0x88a38 <___lookup_tag+0x128>) LC1;

   88942:    03 20           JUMP.S 0x88948 <___lookup_tag+0x38>;

   88944:    18 32           P3 = R0;

   88946:    d4 6f           P4 += -0x6;        /* ( -6) */

   88948:    4c 30           R1 = P4;

   8894a:    02 30           R0 = R2;

   8894c:    48 40           R0 >>= R1;

   8894e:    f9 61           R1 = 0x3f (X);        /*        R1=0x3f( 63) */

   88950:    08 54           R0 = R0 & R1;

What wrong  is in    8893c:    99 3c           LC1 = I1;

--Bill

QuoteReplyEditDelete

2009-07-07 08:13:00     Re: Write data to BF518 SD with exception context

Robin Getz (UNITED STATES)

Message: 76885   

Bill:

Exactly which toolchain did you use to build that code? (it is wrong, and needs a bug fixed).

-Robin

QuoteReplyEditDelete

2009-07-07 08:15:14     Re: Write data to BF518 SD with exception context

Robin Getz (UNITED STATES)

Message: 76886   

Bill:

When I use the latest toolchain:

> cat foo.S

LC1 = I1;

> bfin-elf-as foo.S -o foo

foo.S: Assembler messages:

foo.S:1: Error: Register mismatch.

Nothing should be generating that assembly - since _it_is_ an illegal instruction.

-Robin

QuoteReplyEditDelete

2009-07-09 06:32:44     Re: Write data to BF518 SD with exception context

Bill Xie (CHINA)

Message: 76964   

Robin,

     Thank you.It is ok with toolchain 2009R1-RC9 to compile uClinux  when write data to sd card.

--Bill

Attachments: