Hi,
how to load FPGA emergency without erase head at adress 40000?
I try in IPL command Fpga 100000, but no luck.
thanks
Hi,
how to load FPGA emergency without erase head at adress 40000?
I try in IPL command Fpga 100000, but no luck.
thanks
The "fpga 100000" command loads only the EMG-fpga in memory ...
- For the 9308 (MMI 3G) put the FW (K0257) SD Card in the 1st SD Slot and start from IPL with "boot 1c0000" in the emergency app.
After the automated loading of the SD Card Driver, then type from the QNX prompt as FOLLOWS:
FlashIt -d -a -f 40000 /mnt/sdcard10t11/MU9308/fpga/41/default/systemFPGA.hbbin
- it comes back with "erasing "," progamming "and" Validating "
- if there is an Other unit type, you have to correct the path from the SD Card
- - good Luck!
Последний раз редактировалось Tschako; 20.12.2014 в 17:17.
its another unit... normal ifs its KO...
I need to first emergency FPGA loaded, then emergency began work on USB, or CD.
Because if I run emergency, not begins boot from CD.
Commands to delete the header FPGA can not enter, because unit start quickly and then stop.
I need EM FPGA load, then boot EM IFS.
Thanks
if you start in IPL, then call "fpga 100000" will load the emg-fpga from hex 100000 directly for the sockets.
Then call "boot 1c0000" will load the driver for CD and SD and the emg-app to do your work or start the automated emergency routine directly from the correct SD Card.
I try this, but not work....
its BMW unit and when boot 6dc0000....emergency run, but after a while starts normal ifs and then stop.
In Another unit that works .... can run through diagnosis emergency ...
Then dump begins:
FPGA emergency loaded
start 6dc0000 image ...
The damaged unit but does not communicate diagnostics ....
but it must be a command to load Em FPGA, without deleting the header.
OK, I have no experience with the BWM version... But a failure like this could have something todo with a weak memory...
Try to run the mem check for sure...
Привет, корифеи.
Подскажите по такой проблеме:
голова 3ГП убита, как говорят, кривой инсталляцией и активацией карт. Иногда даже загружается юзеринтерфейс, но сразу отключается.
Было желание запустить автовосстановление с помощью широкоизвестной команды flashit -a 40000 -e 1000
но:
IPL: Audi 3G 9308 (SH7785) / QNX 6.3.0 [PL_qnx_ipl-3g_9308-sh7785_08253A, built 2008/06/18_15-16-25_UTC]
CPU: SH7785 (>= 5th cut, clk mode 20. 4xDDR), 792MHz
RAM: 512 MB (probed, DBCONF=e30002)
FLASH: Spansion S29GL512N (32bit): 512 sectors x 256K
FPGA: PL_MMI3G_PLUS_FPGA_9411_D1_11384A (0ms)
Entering CLI, type '?' or 'help' for help
=> flashit -a 40000 -e 1000
Unknown command 'flashit' - try 'help'
Ну может я ре прав, но 9308 помоему не 3gp. Если пациент грузится и можно войти в зеленое меню, то может emergency repair?
так у тебя же IPL, ты сделай boot ..... и повтори
Может у кого есть дамп eeprom от RNS850 europe.
jalib
Я тоже написал и сразу осознал, что мне тоже не нравится 9308.
Но я понятия не имею, как должно выглядеть на 3Г+
Голова не у меня, далеко. Я просто пытаюсь помочь. Пока все мутно, да еще из за корректности перевода могут быть проблемы.
Так все таки, у 3Г+ такая же строчка, про 9308?
Или в голову запихали что то не то? Собственно машина - А8. ХЗ что там в них ставили.
Логи есть: https://www.sendspace.com/file/b1yvlw
И они как бы от 3ГП.
Последний раз редактировалось ДЮК; 27.12.2014 в 20:21.
flashit -a 40000 -e 1000 not work in IPL....
Commads work only when normal run or emergency.
boot 1c0000 and write command "flashit -a 40000 -e 1000"....
Hi guys and thanks for this very useful thread. I'm new here and I'm very sorry but I have to write in english cause for me the russian is too much difficult.
Reading this thread from the first post, it wasn't so easy using translators that change the sinthax of QNX commands, but finally i was able to find the solutions.
Now i have just recovered the first one of my two 9498 bricked units
On the second one I did an error erasing the partition dev/fs0p3. I tried with the "flashit -a 40000 -e 1000" without success. The emergency update starts but cannot write the lost partition. I think the partition must be created and mounted but how?
here the first part of Boot log any help will be very apreciated.
IPL: Audi 3G 9308 (SH7785) / QNX 6.3.0 [PL_qnx_ipl-3g_9308-sh7785_08253A, built 2008/06/18_15-16-25_UTC]
CPU: SH7785 (>= 5th cut, clk mode 20. 4xDDR), 792MHz
RAM: 512 MB (probed, DBCONF=e30002)
FLASH: Spansion S29GL512N (32bit): 512 sectors x 256K
FPGA: PL_MMI3G_PLUS_FPGA_9411_D1_11384A (0ms)
*** Old-fashioned mark found on IFS @0x1c0000: .zero=0x30000013
*** Old-fashioned mark found on IFS @0x680000: .zero=0x10000000
starting image @680000 ...
Startup: PL_qnx_startup-sh7785_11292A, built 2011/07/19_08-03-38_UTC
RAM : 128 MB @0x08000000
RAM : 128 MB @0x40000000
NVRAM: 36 MB @0x5dc00000
RAM : 220 MB @0x50000000
FLASH: 128 MB @0x00000000
FPGA IRQ: Interrupt Controller ID:33 Ver:2.1.0 ILVL:7 Lines:255
System page at phys:0800a000 user:0800a000 kern:8800a000
Starting next program at v8804aae0
Welcome QNX6.3.2 (QNXLabel RL_qnx_os_632_PSP3_08041A)- Harman/Becker MMI3GP Build 9498 E1-12321A
USB-stuff (QNXLabel RL_qnx_os_632_PSP3_08041A)
(devf t1::f3s_table_find:247) fs0p3 corrupt U: 74
checkPersistence: /mnt/efs-persist has not been mounted by devf-generic after 30 seconds: formatting /dev/fs0p3
checkPersistence: trying to backup the corrupted flash: ('.' == 1M)
.........
Unable to access "/mnt/efs-persist" (2)
# NV: Waiting for the minirm init done indication file /dev/shmem/nvminirm.finished'
NVLOADER: Booting DSP (DRAM Setup time = 1000 us)...
NVLOADER: Going to call nvTsrmRawLoad for file /etc/NvDspCodecs.elf
NVLOADER: Done
Enabling DMA using channel 11
Overriding default Interrupt Priority
Current debug level = 0x7fffffff
dmaChannel = 11, interrupt thread priority = 21
devg-NVMEM: Mapped vaddr 0x38100000 to paddr 0x5d397000
[pcidma] PL_qnx_pcidma_FMod08363A_09274A, built 2009/07/02_11-10-37_UTC
multicored [000000154.647]:version 4.6.d, verbosity 1 (supported 4), scope 0x00000001, console: /dev/ser1 (fd=3), FILE_LOGGING, TESTCO
NTROL, DYNAMIC_LOAD
multicored [000000154.650]:using 750 buffers of size 1460 (total 1095000 bytes)
multicored [000000154.694]:version 4.6.d, verbosity 2 (supported 4), scope 0x00000001, console: /dev/ser1 (fd=3), FILE_LOGGING, TESTCO
NTROL, DYNAMIC_LOAD
multicored [000000154.716]:udp: broadcaster created
multicored [000000154.722]:attached to filesystem at /hbsystem/multicore, port 851
multicored [000000154.740]:MACHINE:SH(1), total MEM 499122176, 1: 271582976 SH7785 (SH4A) 792MHz, FPU
multicored [000000154.745]:sloop active
multicored [000000154.747]:lo0 127.0.0.1 255.0.0.0
multicored [000000154.747]flog0 0.0.0.0 0.0.0.0
multicored [000000154.774]:listener_activate()
multicored [000000154.776]:listener on port 851 active, fd=7
Mask = c3
GPIO 7 set to Low
Overriding default Interrupt Priority
dev-nvram: PL_qnx_dev-nvram_07225A, built 2007/06/01_15-44-08_UTC
dev-nvram: page_size:4096
dev-nvram: signalling register 10ee:9411:0:40:40 [vid:did:barffs:mask]
dev-nvram: CPU: SH7785 (5th cut or higher, clk_mode 20) [10300700:250, 133735fa]
dev-nvram: map '/dev/nvram/navi' to 0x5dc00000-0x5fffefff
dev-nvram: map '/dev/nvram/.guard' to 0x5ffff000-0x5fffffff
dev-nvram: SelfRefresh was inactive
UART: Mod:0E Date:09503A Ver:2.2.2
devc-ser8250hb: UART supports FiFo counter
devc-ser8250hb: UART supports 32bit access
devc-ser8250hb: RX_FIFO_SIZE:2048 TX_FIFO_SIZE:2048
devc-ser8250hb: set IBUF size to 16384
UART: Mod:0E Date:09503A Ver:2.2.2
devc-ser8250hb: UART supports FiFo counter
devc-ser8250hb: UART supports 32bit access
devc-ser8250hb: RX_FIFO_SIZE:512 TX_FIFO_SIZE:512
UART: Mod:0E Date:09503A Ver:2.2.2
devc-ser8250hb: UART supports FiFo counter
devc-ser8250hb: RX_FIFO_SIZE:32 TX_FIFO_SIZE:32
Debug filehandle = 5
For further debug messages have a look at multicore output consoles '/hbsystem/multicore/navi/g'
[pcidma] PL_qnx_pcidma_FMod08363A_09441A, built 2009/10/26_07-46-16_UTC
mam0 available
NAVI: sys_alloc_shm_area(/dev/nvram/navi) open() ok: 9
NAVI: sys_alloc_shm_area(/dev/nvram/navi) filesize() = 37744640
NAVI: sys_alloc_shm_area(/dev/nvram/navi) mmap to 38100000
NAVI: sys_alloc_shm_area(/dev/nvram/navi): physical of 38100000 at 5dc00000
NV_ENABLE_DYNAMIC_PRIORITIES is enabled
max-consecutive-errors set to 2
timeout retries config string == |2|
timeout-retries set to 2
io-media-nvidia built: Jul 20 2012 14:33:47 against Official Tree
nv_global.verbose = 0
Virtual Address for CDDA DirectIO = 842F000
0,"nVAudio_init: IN
0,"nVAudio_class_init: IN
nVAudio_class_init: OUT success
0,"nvidia_render_manager_init();
0,"nVAudio_init: IN
0,"nVAudio_class_init: IN
nVAudio_class_init: OUT success
0,"nVAudio_class_init: IN
nVAudio_class_init: OUT success
io-media-nvidia/a2dp_player: The 'metadata-timeout' setting is missing or has an invalid value, defaulting to 4000ms
0,"nVAudio_init: IN
0,"nVAudio_class_init: IN
nVAudio_class_init: OUT success
0,"nvidia_render_manager_init();
0,"nVAudio_init: IN
[t(p) usr/sbin/io-display 98332 7 10 7033aaf6]
[t(c) usr/sbin/io-display 98332 5 10 70334f72]
WARNING: NON PERFORMANT MODULAR STARTUP!!!
Path=0 - XILINX ATAPI
target=0 lun=0 CD-ROM(5) - FJ-TEN DVD-ROM DV-05FT2 Rev: 1B20
Path=1 - XILINX ATAPI
target=1 lun=0 Direct-Access(0) - TOSHIBA MK6050GA Rev: TJ10
cd0 mounted after 1 try
[t(p) proc/boot/procnto-instr 1 8 10 8803cc12]
[t(c) ./MMI3GNavigation 155689 23 10 703351de]
***************************************
*** ***
*** Online Speech ***
*** ***
*** EncodeStream ***
*** v1.0.02 ***
*** ***
***************************************
DSIVersion: 0.0.0
DSISource: DSI 2012_26_26
DSI-Build-Version: Mon Jun 25 11:24:44 CEST 2012 / DSIGEN v1.138.4
ReleaseLabel: Int_AU3G_EU_SOP9_12272P
start JDSI bundle
HBTRACE: jdsi.verboselevel = null => verbose-level = 0 because property not set or invalid
jdsi.loglevel : 2
No External Component found
.......................
You have to rewrite the efs-persist image in emg-app mode like following this steps:
1. Prepare a SD card with an binary image file from an ifs-persist partition (name: efs-persist.bin)
2. Put SD card in slot #1 (sdcard10t11)
3. start unit in IPL mode
4. enter command "boot 1c0000" to boot emg-app
5. enter command "echo dummy > /HBpersistence/DeveloperMode" to extend the timeout for watchdog reboot
6. enter "flashit -a 6200000 -d -f /mnt/sdcard10t11/efs-persist.bin" to flash the partition image
7. reboot the unit after the verification message
8. the unit now will recover the details for the persistence partition
Thats it!
Последний раз редактировалось Tschako; 06.01.2015 в 02:06.
Many thanks Tschako, sorry but you say this :1
what do you mean? I did "boot 1c0000" but w/o success
Sorry I missed the second part of the post,
thanks a lot I TRY IMMEDIATLY