Notice

The forum is in read only mode.

Need testers for USBcode

vlad_vy
Offline

29 Nov 2017 05:15 - 29 Nov 2017 05:18 #66249 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

At the moment I havn't idea what is wrong. All looks fine, except firmware flashing result for T8SG. With command line all looks similar for Devo8 and T8SG, but Devo8 firmware flashing result is OK, T8SG firmware flashing result = NOT OK.

Last edit: 29 Nov 2017 05:18 by vlad_vy.

vlad_vy
Offline

29 Nov 2017 06:14 - 30 Nov 2017 11:59 #66252 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

deleted

Last edit: 30 Nov 2017 11:59 by vlad_vy.

vlad_vy
Offline

29 Nov 2017 13:34 #66254 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

From DevoDfuSe.exe v2.31:

Walkera DEVO-12 Radio Controller == DEVO-12 == DV12
Walkera DEVO-08 Radio Controller == DEVO-8 === DV08
Walkera DEVO-06 Radio Controller == DEVO-6 === DV06
Walkera DEVO-10 Radio Controller == DEVO-10 == DV10
Walkera DEVO-07 Radio Controller == DEVO-7 === DV07
Walkera DEVO-7E Radio Controller == DEVO-7E == DV7E
Walkera DEVO-F7 Radio Controller == DEVO-F7 == DVF7
Walkera DEVO-F4 Radio Controller == DEVO-F4 == DVF4
Walkera DFPV-12 Radio Controller == DEVO-F12 == DF12
Walkera DFPV-7S Radio Controller == DEVO-F7S == DF7S
Walkera DEV-12E Radio Controller == DEVO-12E == D12E
Walkera FPV-12E Radio Controller == DEVO-F12E = D12E

vlad_vy
Offline

29 Nov 2017 15:49 #66256 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

I'm not sure that anybody can flash "devo7e-256" firmware with DeviationUpload-0.8.0.jar

Can anybody confirm it?

FDR
Offline

29 Nov 2017 16:17 #66257 by FDR

Replied by FDR on topic Need testers for USBcode

I guess not, that's what I mentioned in the pull request's comment :

BTW in case of the U7E, which has a filename of "deviation-devo7e-256-v5.0.0-be57e84.zip" the results would be "7e" and "256-v5.0.0-be57e84", which is not that good, I think...

vlad_vy
Offline

29 Nov 2017 16:19 #66258 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

It can be flashed anyway, but I think you get firmware update screen.

FDR
Offline

29 Nov 2017 17:03 #66261 by FDR

Replied by FDR on topic Need testers for USBcode

But if the updater thinks about an U7E, that it is a simple 7E, then it will get a wrong flash size, won't it?
Then if the fw size larger then the flash size, it should fail...

Schugy
Offline

07 Dec 2017 12:59 #66376 by Schugy

Replied by Schugy on topic Need testers for USBcode

Got my new F7 but had to upload the last November nightly manually with stock 0.8 uploader. Uploading the whole zip resulted in an aborted progress bar amd a success message. Now I have a single menu line on my F7. Maybe a corrupt devo FS? Is there a way for loopback mounting? No idea about the fs type. Success seems near.

vlad_vy
Offline

07 Dec 2017 14:24 - 07 Dec 2017 14:26 #66381 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

Try copy devo.fs file from archive to Tx USB drive.

Last edit: 07 Dec 2017 14:26 by vlad_vy.

Schugy
Offline

07 Dec 2017 19:32 - 07 Jan 2018 12:52 #66395 by Schugy

Replied by Schugy on topic Need testers for USBcode

Thank you, this was the missing step. Zipped the devo.fs to make it accessible in DeviationUpload but this resulted in a different md5sum. Copying it manually worked. Now I can modify the devo.fs file with buildfs.exe only in wine (wineconsole --backend=user cmd) which is super inconvenient

.
Update: buildfs.py works faster.
But I guess devofs will never make it into the kernel tree or fuse. I added DSMX, J6Pro, PPM and USBHID protocol. Would be great to have language modules.
The good thing is it all worked with a free OS and DeviationUpload 0.8. Great stuff.

Last edit: 07 Jan 2018 12:52 by Schugy. Reason: using buildfs.py now

Moeder
Offline

29 Apr 2018 18:39 #68922 by Moeder

Replied by Moeder on topic Need testers for USBcode

Did you make any progress on this issue? I'm a little annoyed that I have to boot into Windows to flash my jumpers...

vlad_vy wrote: At the moment I havn't idea what is wrong. All looks fine, except firmware flashing result for T8SG. With command line all looks similar for Devo8 and T8SG, but Devo8 firmware flashing result is OK, T8SG firmware flashing result = NOT OK.

Moeder
Offline

29 Apr 2018 18:53 #68923 by Moeder

Replied by Moeder on topic Need testers for USBcode

Vlad, can you check your devices serial numbers? Maybe we can use this to differentiate between different versions in the uploader:
This is what I get for my Jumper T8SG:

STM32 DFU:
Product ID: 0xdf11
Vendor ID: 0x0483  (STMicroelectronics)
Version: 2.00
Serial Number: DEVO-7E 㐀げ仿⌵㜅Q

And this is the result for the Jumper T8SG V2 plus:

STM32 DFU:
Product ID: 0xdf11
Vendor ID: 0x0483  (STMicroelectronics)
Version: 2.00
Serial Number: DEVO-7E „ÄÄ·âç‰áøÌππ·Ä∏„åÖC

Moeder
Offline

01 May 2018 19:38 - 01 May 2018 21:27 #68959 by Moeder

Replied by Moeder on topic Need testers for USBcode

I do not think it is related to the id read from the device, as the only difference in the 256 boot loader is two bytes almost at the end - not where the device is identified from. I rather suspect it has to do with the flashing and page addressing, as 256k and higher chips are high density devices with 256 pages of 2 Kbyte vs 128 pages of 1 kbyte. ~~I don't know how this relates to the sector overriding stored in Transmitter.yml or the flashing process itself, but I think~~ this is where you could start looking.

vlad_vy wrote: At the moment I havn't idea what is wrong. All looks fine, except firmware flashing result for T8SG. With command line all looks similar for Devo8 and T8SG, but Devo8 firmware flashing result is OK, T8SG firmware flashing result = NOT OK.

Last edit: 01 May 2018 21:27 by Moeder.

Moeder
Offline

01 May 2018 21:39 - 01 May 2018 22:37 #68963 by Moeder

Replied by Moeder on topic Need testers for USBcode

Taking a closer look, this could actually be the reason:
The patched firmware for 256k wrongly reports Interface 0 as 1k, while it should be 2k (see devo8 output)

Device  Interface    Start        End    Size   Count
Devo 7e
                0 08003000   0803ffff    1024   244
                1 00000000   0001ffff    4096   32
                1 00020000   001fffff   65536   30

edit:
Indeed this was the problem! Overriding interface 0 in Transmitter.yml let me successfully flash the t8sg dfu file!

matchRules:
    - "DEVO-7E.*"
    - ".*devo7e.*"
    - "7e"
    - "t8sg"
    - ".*t8sg.*"
overrideSectors:
    - iface: 0
      remove: true
      add:
       - start:     0x8003000
         end:       0x803ffff
         size:      0x800
         count:     128
    - iface: 2
      remove: true
    - iface: 1
      remove: true
      add:
        - start:    0
          end:      0x1FFFF
          size:     0x1000
          count:    32
        - start:    0x20000
          end:      0x1FFFFF
          size:     0x10000
          count:    30

Now we need to overwrite sectors for iface 0 when a firmware >128k in size is selected (devo256 or t8sg*) and maybe issue a warning to ask the user to verify (s)he is using a transmitter other than a regular devo7e.

Last edit: 01 May 2018 22:37 by Moeder. Reason: Success!

vlad_vy
Offline

02 May 2018 09:16 #68972 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

I think "count" has to be 122 (122*2K = 244K), size without bootloader (16K). Devo8 "count" = 120.

Moeder
Offline

02 May 2018 09:54 #68973 by Moeder

Replied by Moeder on topic Need testers for USBcode

Yes you're right, count has to be 122 with 12k bootloader in devo7e and its variants.
I'll try to modify the uploader to read out the changed bootloader bytes to automatically switch to the 256 settings if applicable, as that would be the most foolproof. Haven't done any Java in 15 years, so please be patient

vlad_vy
Offline

02 May 2018 11:10 #68974 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

Probably it can be solved without overriding interface 0, by modifying class DfuMemory:

...
                if (m.matches()) {
                    int count = Integer.parseInt(m.group(1));
                    int size  = Integer.parseInt(m.group(2));
                    if (m.group(3).equals("K")) {
                        size = size * 1024;
                    } else if (m.group(3).equals("M")) {
                        size = size * 1024 * 1024;
                    }
                    if (count == 244) {
                        count = 122;
                        size = size * 2;
                    }
...

vlad_vy
Offline

02 May 2018 13:48 - 11 May 2018 08:21 #68975 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

Tested, it works. Now I can flash T8SG and T8SG V2 Plus with DeviationUpload-0.9.0.jar

Last edit: 11 May 2018 08:21 by vlad_vy.

Moeder
Offline

02 May 2018 17:21 #68976 by Moeder

Replied by Moeder on topic Need testers for USBcode

I still prefer adding an entry in transmitter.yml for „Devo7e-256/T8SG/T8SG plus/IR8M“ so escpecially the users of the latter transceivers do not get confused by the display of devo7e as the identified transmitter. I guess doing the type detection by the flash count is just fine for now. Would you mind pushing your changes to github together with a pr to the upstream repo for further refinements?

vlad_vy
Offline

02 May 2018 17:28 - 02 May 2018 17:49 #68977 by vlad_vy

Replied by vlad_vy on topic Need testers for USBcode

At any case bootloader has "DEVO-7E" as identifier and DfuSe also display "DEVO-7E" for "Devo7e-256/T8SG/T8SG V2 Plus/IR8M".

Pull request placed.

Last edit: 02 May 2018 17:49 by vlad_vy.

Time to create page: 0.516 seconds

Re-enable mobile display