morris
Junior Fellow member
Posts: 37
|
Post by morris on
Jun 16, 2020 eighteen:14:48 GMT -5
Hey Guys,
No news to study on the 656. I have’t spend time on it lately. I received some new nand chips, and so I’m working on a nr818. This 1 had a endless list of bad blocks on the nand. So i replaced it and flashed it with a establish nand dump on the cyberspace. Initially information technology booted and had sound. But I can’t get it to update to a newer version (As available per internet connectedness) And too not via USB. It each time hangs with DSP2 98 percent. So I guess I will exist trivial with this auto for now
If anyone has a good know NR818 nand dump i would gladly take information technology
br Maurice
|
|
|
Post by dobrykamil on
Jul 19, 2020 x:12:xx GMT -5
Hi
I recently fixed TX-NR656 which had similar issues. No sound output and information technology wouldn’t update via usb or network. Fw update process was ok until information technology reached 9% and somewhen stopped with 0-83 or similar mistake. Turned out, firmware for 2nd DSP was corrupted.
|
|
|
Post past tibimakai on
Jul twenty, 2020 10:00:09 GMT -5
I recollect, that I had the aforementioned 0-83 error on the 656 and RZ610 and I had to run the firmware update twice, than it completed. How did y’all fix the corrupted DSP?
|
|
|
Post past chris178 on
Nov 16, 2020 22:01:21 GMT -v
Hi Thanks for your input on above nr656 repair progress. I’ve found this place also after my going on the same model with pretty much very aforementioned symptoms. All modules seems to be working beside DSP (CS49844A). I’ve got this in this condition so, not certain what was previous history…. Did you lot become anywhere with your unit of measurement Maurice? Afterwards some (quite long….) troubleshooting with oscilloscope – it looks as my DSP is struggling to load its FW from SPI nand flash. Also MMPU is resting it every one sec. Data bursts on DSP wink SPI lines suggests that DSP reads flash initially with 30MHz clock, then 6MHz clock, but all stops after near 50kB of information is read from the flash. Looks wrong.
I don’t know what the process is as to how it is supposed to communicate dorsum to MMPU, but it looks as when FW is loaded it shoud pull downwardly ~DSP_IRQ. Just I guess this would only happen if DSP is set to work…
Also my remark/observation as to DSP update procedure. By schematic – at that place isn’t whatsoever concrete connectedness (nor buffered) from Cyberspace CPU (ARM CPU that handles USB stick) to the DSP wink scrap. The just path equally I encounter information technology is: USB/Cyberspace -> ARM -> NET_NOMI/NIMO SPI -> MMPU -> DIG_SDI/SDO SPI -> DSP -> FLASH SPI -> FLASH IC… So I would paradigm when DSP functions normal it is put into “SPI repeater” mode to plan DSP flash to carry out normal user FW upgrade.
I’one thousand a scrap stuck now as I can’t find a way to extract DSP Wink image from official Onkyo firmware bundle… (it’s in .of2)
A copy/dump of working FLASH would help a lot and shed more than light to the case. Surely accelerate faultfinding procedure.
I’ve just spotted something interesting! – in Onkyo FW .of2 file where I think should be firmware for Cirrus Logic DSP – there is RIFF WAVE information….
Could information technology be that something “plays” it to DSP for programming?
This idea came equally I’ve seen some strange Yamaha arrangements with FW updates over SPDIF coax….. very interesting.
|
|
|
Post by chris178 on
Nov 17, 2020 17:54:51 GMT -5
A bit of update – I’ve managed to confirm that ARM CPU (NET) is sending (in fact playing, every bit previously suspected) the update (RIFF waveform) to the DSP CPU over I2S lines. A playout can be observed on DSP CPU pins 21-22, 26-27, during USB->DSP ane update process.
Update process usually goes up to 98%, then there is a quite a filibuster before mistake 1-83 is returned. I would imagine in this time DSP should transform received playout into binary and physically flash the NAND.
Then kick itself and probably feedback current FW version.
Only in my unit there is another contradiction – MMPU constantly resets DSP (pulses on DSP_RST every 1sec) during the update process. This suggests MMPU is not aware of the update….. (if some kind of watchdog kicks in every 1sec. this should exist at least disabled for the duration of the update process). I’m confused, but at least this perhaps explains why Maurice yet could not run normal FW upgrade afterward he revived DSP FLASH (by direct programming).
Maurice or Dobrykamil, please could y’all share your working version of DSP dump?
I’m really curious to come across how SPI comms should await similar between DSP and MMPU.. (assuming this will restore comms).
|
|
morris
Junior Fellow member
Posts: 37
|
Post by morris on
November 17, 2020 18:19:12 GMT -five
Hi Chris,
I have the unit stored in a corner of my workshop. At this moment I’1000 working on my house. Keep the mrs happy
brick laying, woodwork. And so no “free” time at this moment. This volition keep going for some months since i’m doing everything myself. I wish I had a working DSP dump. Perhaps if I observe some fourth dimension I volition wait at it. Reading your postal service, you are far more advanced then me on this unit of measurement.
kind regards, Maurice
|
|
|
Post by chris178 on
Nov 17, 2020 19:06:51 GMT -five
Hi Maurice, thanks for reply. At least your work might be a bit more rewarding and straightforward than contrary-engineering and debuging some odd SPI lines
I will proceed posting as works progress. BR Chris
|
|
|
|
Post by chris178 on
Nov 18, 2020 16:00:06 GMT -v
Hi tibimakai, Thank you for this besides. I’ve seen it and this is very informative and valuable source. Ive read through most of it.
I can written report that I’ve managed to revive my unit of measurement. Big credits go to Kamil as he helped with DSP wink working content. This was the key. At present CS49844A is working again and appears to take updates too.
Onkyo’southward (or Cirrus…) build-in update method via RIFF audio file seems to me a bit cloudy… why this style, I don’t understand and not simply over SPI comms… Only it looks that perhaps this DSP IC simply way to boot is via its SPI Flash, and at that place isn’t whatsoever recovery mechanism built into it. Cursory datasheet only mentions booting via secondary SPI interface.
Anyway this is a win and it’s made my twenty-four hour period today!
Give thanks you for your help.
|
|
|
Post by geduxaz on
Dec 5, 2020 eighteen:07:31 GMT -5
Hello. Please help with NR656. Serial stuck at starting kernel… This is afterwards net update. Tried all possible updates, get 0-81 error!
U-Boot SPL 2014.07 (Mar 18 2016 – 11:33:00) detected max. frequency: 800(MHz) –
U-Boot 2014.07 (Mar 18 2016 – 11:33:00), Build: jenkins-2016EntryAVR-9
I2C: prepare DRAM: 256 MiB rstst: 0x1, rstctrl: 0x0 NAND: 256 MiB MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1 Setting partitions for 256MB flash. Setting mtdparts: mtdparts=omap2-nand.0:128k(mlo),128k(mlo2),128k(mlo3),128k(mlo4),1024k(u-kick),512k(u-boot-env),512k(constants),8192k(settings),5120k(kernel),5120k(kernel_2),512k(dts),512k(dts_2),120064k(rootfs),120064k(rootfs_2) Cyberspace: cpsw BOOTCOUNT is 18 Warning: Bootlimit (half-dozen) exceeded. Using altbootcmd. Hit any key to stop autoboot: 0 Error: Maximum kick count reached! INFO: Checking boot partitioning. Booting from nand … INFO: loading kernel image into RAM… Unknown command ‘bstate’ – try ‘help’
NAND read: device 0 offset 0x1480000, size 0x80000
524288 bytes read: OK ### CRAMFS load complete: 38029 bytes loaded to 0x80800000 INFO: dts file load successful INFO: fdt addr set successful INFO: fdt resize successful
Loading from nand0, offset 0xa80000
Image Name: Linux-3.19.0
Image Type: ARM Linux Kernel Paradigm (uncompressed)
Data Size: 4780376 Bytes = 4.6 MiB
Load Address: 80008000
Entry Point: 80008000 INFO: kernel sectionalisation load successful INFO: booting kernel image… ## Booting kernel from Legacy Image at 80200000 …
Image Proper noun: Linux-iii.19.0
Prototype Blazon: ARM Linux Kernel Image (uncompressed)
Data Size: 4780376 Bytes = 4.half dozen MiB
Load Address: 80008000
Entry Bespeak: 80008000
Verifying Checksum … OK ## Flattened Device Tree hulk at 80800000
Booting using the fdt blob at 0x80800000
Loading Kernel Image … OK
reserving fdt retentiveness region: addr=80800000 size=a000
Loading Device Tree to 8fc19000, terminate 8fc25fff … OK
Starting kernel …
|
|
|
Mail by poppabradj on
Jan 8, 2021 eleven:57:12 GMT -5
Hello. Please aid with NR656. Serial stuck at starting kernel… This is after cyberspace update. Tried all possible updates, get 0-81 mistake!
U-Kicking SPL 2014.07 (Mar xviii 2016 – 11:33:00) detected max. frequency: 800(MHz) –
U-Boot 2014.07 (Mar 18 2016 – 11:33:00), Build: jenkins-2016EntryAVR-9
I2C: ready DRAM: 256 MiB rstst: 0x1, rstctrl: 0x0 NAND: 256 MiB MMC: OMAP SD/MMC: 0, OMAP SD/MMC: ane Setting partitions for 256MB wink. Setting mtdparts: mtdparts=omap2-nand.0:128k(mlo),128k(mlo2),128k(mlo3),128k(mlo4),1024k(u-kicking),512k(u-kick-env),512k(constants),8192k(settings),5120k(kernel),5120k(kernel_2),512k(dts),512k(dts_2),120064k(rootfs),120064k(rootfs_2) Cyberspace: cpsw BOOTCOUNT is eighteen Alarm: Bootlimit (6) exceeded. Using altbootcmd. Hit whatever key to stop autoboot: 0 Error: Maximum boot count reached! INFO: Checking boot partition. Booting from nand … INFO: loading kernel epitome into RAM… Unknown control ‘bstate’ – try ‘help’
NAND read: device 0 offset 0x1480000, size 0x80000
524288 bytes read: OK ### CRAMFS load consummate: 38029 bytes loaded to 0x80800000 INFO: dts file load successful INFO: fdt addr prepare successful INFO: fdt resize successful
Loading from nand0, outset 0xa80000
Image Name: Linux-three.19.0
Image Type: ARM Linux Kernel Image (uncompressed)
Data Size: 4780376 Bytes = 4.6 MiB
Load Address: 80008000
Entry Point: 80008000 INFO: kernel partition load successful INFO: booting kernel epitome… ## Booting kernel from Legacy Image at 80200000 …
Image Name: Linux-three.19.0
Epitome Blazon: ARM Linux Kernel Image (uncompressed)
Data Size: 4780376 Bytes = 4.half-dozen MiB
Load Address: 80008000
Entry Point: 80008000
Verifying Checksum … OK ## Flattened Device Tree blob at 80800000
Booting using the fdt blob at 0x80800000
Loading Kernel Image … OK
reserving fdt retention region: addr=80800000 size=a000
Loading Device Tree to 8fc19000, terminate 8fc25fff … OK
Starting kernel …
===
|
|
|
Mail by poppabradj on
Jan viii, 2021 xi:59:32 GMT -5
I am Having the same Issue .. error 0-81 (I believe I have a corrupted firmware on my TX-NR656 … anyone has a solution or proffer, before I bring information technology to service heart, I am a noob when it comes to this merely I tin can follow Instruction.
The receive works, simply I don’t have access to Network or Bluetooth, OSD doesn’t piece of work on my TV and IN Settings, I don’t take access to many choice: for instance in seven.Miscellenous all I take access to is 1.Tuner, iii.Remote and 5. Lock (and the option is unlocked).
Any assistance would be appreciated.
|
|
|
Mail past geduxaz on
Jan 8, 2021 17:58:18 GMT -v
Just fixed mine!!!
Totaly same 0-81. It means usb non started and cant practise usb/cyberspace update (linux not booting upwards). I managed to kicking into UBOOT and reflash nand with images from decoded oficial firmware. Information technology took 2 weeks in evenings to lear nand programming/ uboot/ uboot env.
|
|
|
Post by tjmotter on
Jan nine, 2021 11:27:59 GMT -v
Just fixed mine!!!
Totaly aforementioned 0-81. It means usb not started and cant do usb/net update (linux not booting up). I managed to kick into UBOOT and reflash nand with images from decoded oficial firmware. It took 2 weeks in evenings to lear nand programming/ uboot/ uboot env.
Would love to hear the process yous developed to practise this.
|
|
|
Mail service by geduxaz on
Jan 12, 2021 17:14:44 GMT -v
Just fixed mine!!!
Totaly same 0-81. It means usb not started and cant do usb/internet update (linux not booting up). I managed to kick into UBOOT and reflash nand with images from decoded oficial firmware. It took ii weeks in evenings to lear nand programming/ uboot/ uboot env.
Would love to hear the process you adult to do this.
Well this was my first repair over all so it took me many tries, flashes, soldering…
I recollect information technology was never booting upward to usb to do firmware update. At outset i was trying to fix native nand like copying parts and irresolute places rootfs-rootfs2 etc.
Afterward i found that uboot enviroment had many scripts. Reading through them it was clear that factory/service can run updates from uboot. Scripts can look into usb or network for update files, and later flash them to needed mtds.
Then i did simple things.
I cheched where my boot stops, tried deleting dts, kernel, rootfs and plant that it looked only rootfs was non booting.
I took rootfs file from decoded firmware to usb. Booted uboot, started usb. Copied rootfs from usb to RAM. Erased nand partirions with rootfs and rootfs-2 Wrote nand rootfs and rootfs2 from RAM.
Exact comands can paste later. Need to accept terminal dump. After that it booted to linux and i got usb. Did firmware update from official files by usb.
What is interesting. This device has hidden output when system is running ok. It shows just sratring kernel….
|
|