Mares Sirius

[Ivan Kemka]

Is planed to support Mares Sirius?

[Jef Driesen]

On 31/05/2023 16:00, Ivan Kemka wrote:
> Is planed to support Mares Sirius?

Do you have one? There is a good chance the communication protocol is compatible
with the previous models. Most likely candidates are the Icon HD and Genius. So
give it a try with logging enabled and send me the log for analysis.

Jef

[Nicki Groiss]

Hello,

i´ve a Sirius. If you tell me what i´ve to setup for the right log, i will do it on iOs and Windows.

Regards

Nick

[Jef Driesen]

On 17/07/2023 15:49, Nicki Groiss wrote:
> i´ve a Sirius. If you tell me what i´ve to setup for the right log, i will do it
> on iOs and Windows.

Try downloading your dives with the logging enabled as explained here:

http://libdivecomputer.org/subsurface.html#desktop-dives (Windows)
http://libdivecomputer.org/subsurface.html#mobile (iOS)

Select the Mares Icon HD as the dive computer model. If that doesn't work, try
again with the Mares Genius as the model.

On Windows, it's probably also a good idea to include the subsurface logs:

https://subsurface.github.io/faq/#where-do-i-find-the-log-files-on-windows

Jef

[Subsurface Divelog]

Hello,

attached you will find the logfile. 

I also use the Mares Sirius, it would be great, that i can use subsurface for my diving computer.

Selecting Mares Icon HD or Genius is not working.

Best regards

Nick

[Jean-Baptiste Martin]

Hello everyone. 

I am considering purchasing the MARES SIRIUS. But the lack of compatibility with Subsurface is a blocking point. So I'm very interested in this thread. Being completely incompetent on the subject, I rely to the developers.

Do you think that Sirius will soon be supported by Subsurface?

Thanks in advance.

Jean-Baptiste.

[John Bowersmith]

Hello All,

While we wait for the developers to build Sirius compatibility into Subsurface I thought I'd outline my workflow for getting some data from My Sirius into Subsurface. I'd like to point out that this isn't a Subsurface problem only, nearly no logging apps have Sirius compatibility yet. 

Also, this is a totally cumbersome manual process using multiple tools, some of which are only available for PC.

You will need:

Mares Sirius Computer-- duh

Subsurface App -- double duh

Mares OTA Programmer App -- mobile only

An email address

Mares Dive Organizer App (PC Only) -- I use bootcamp

Steps:

1. Connect Sirius to OTA Programmer App
2. Export logs to email using OTA Programmer
   1. You will get a zip file full of otherwise useless .bin files
3. Download zip file from your email
4. DO NOT UNZIP
5. Open Mares Dive Organizer
6. Using the "Computer" selection in Dive Organizer, import your dive log 
   1. You must select the entire zip file and Dive Organizer will allow you to select dives to import
7. Select all dives (or dives you want to export) in Dive Organizer and export them
   1. Dive Organizer will spit out an XML file
8. Open Subsurface
9. Import the exported XML file from Dive Organizer into Subsurface
10. Profit?

This is far from perfect. It doesn't pull in AI data, tank data, EAN O2% data or water-type data

This is kind of a PITA but hey, at least it's something.

-John

[Jean-Baptiste Martin]

Thanks John. That's not perfect, but better than nothing !

[Jef Driesen]

On 4/11/2023 20:25, John Bowersmith wrote:
> While we wait for the developers to build Sirius compatibility into Subsurface I
> thought I'd outline my workflow for getting some data from My Sirius into
> Subsurface. I'd like to point out that this isn't a Subsurface problem only,
> nearly no logging apps have Sirius compatibility yet.

We need help from users owning a Sirius to make that happen. I have already
looked into adding support for the Sirius, but so far without much progress. So
if you have one, you can certainly help!

First, try downloading your dives with the logging enabled as explained here:

http://libdivecomputer.org/subsurface.html#desktop-dives (Windows)
http://libdivecomputer.org/subsurface.html#mobile (iOS)

Select the Mares Genius as the dive computer model.

On Windows, it's probably also a good idea to include the subsurface logs:

https://subsurface.github.io/faq/#where-do-i-find-the-log-files-on-windows

> 1. Connect Sirius to OTA Programmer App
> 2. Export logs to email using OTA Programmer
> 1. You will get a zip file full of otherwise useless .bin files
> 3. Download zip file from your email

Can you send me an example of those OTA bin files? I suspect these contain the
raw dive data, and that may help adding support for the Sirius.

Jef

[John Bowersmith]

Hi Jef,

I tried to download using the Genius setting but ran into problems. I will do it again and take notes of the experience to help you. I'll post the results of my test here.

I'm positive the .bin files contain the raw dive data. I asked one of my devs from work if he could take a look at the files as a "side quest" but you know how "side quests" go. I've attached the entire zip export from OTA Programmer.

What was interesting to me is that even the Mares software can't process the individual bin files, it can only work with the zip. I suspect, not being a dev, that one of these bin files is also telling the Mares software how to parse the data. That's why I just stuck the whole zip here.

Would it be helpful to you for me to provide some better documentation about what the Mares software is doing too?

Best,

John

[Nicki Groiss]

Hey Guy´s,

I've also owner of a Sirius and i've sent Jef already some Files. 

The Sirius is the evolution of the Genius, what is working well for me.

@Jef, if you need more, i can start the logging on the android device again.

Also, the new Computers what Mares will show on the Boat, will be on the same Software as the Sirius (Puk 4 and Quad CI)

[Jean-Baptiste Martin]

Hello to all the community.
Any news on Sirius support? Were you able to find a solution?
Thanks in advance.

[John Bowersmith]

Sadly, the computer I had running bootcamp died and I no longer have a way to run the Mares software, so I am stuck again :(

[Cosma Tassinari]

Hi, I also have a Sirius and I'm avid user of Subsurface.

I'm attaching the exports both directly from OTA programmer (zip) and Dive organizer (xml).

If I can help in any other way let me know.

[A K]

Still a great workaround John! I went from Dive organizer to divelogs.de as recommended by Subsurface as well. Then at least I got correct nitrox levels and the dive chart. Then from divelogs to Subsurface.

Btw, all this files created by the OTA Programmer? Is that something the Subsurface developers can use to get Sirius support for subsurface?
I really like Sirius but I won't buy a tank module until this is sorted out. Then I just continue to use my old Suunto D4 to monitor tank level.

[Jef Driesen]

On 2024-02-05 11:50, Cosma Tassinari wrote:
> Hi, I also have a Sirius and I'm avid user of Subsurface.
> I'm attaching the exports both directly from OTA programmer (zip) and
> Dive organizer (xml).
>
> If I can help in any other way let me know.

I have now implemented the changes needed to support the Sirius, and
have a first test build available. You can download it here:

https://libdivecomputer.org/builds/experimental/windows/sirius.exe

It's a command-line application which you need to execute with these
options:

sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o dive.%n.bin
-f raw -t ble

Send me the sirius.log file.

Jef

[Cosma Tassinari]

Maybe I did something wrong, or there were other steps to do before that I didn't do, because the sirius.log is empty.

This is what comes out of the cmd:

C:\Users\cosma>sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o dive.%n.bin -f raw -t ble
DATETIME 2024-04-11T09:45:43Z (1712828743)
VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
Opening the I/O stream (ble, null).
ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]
INFO: Discover: address=47E413191606, name=
INFO: Open: address=47E413191606
ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]

[A K]

Ok, now something happened!
First my Siruis booted itself an re-initialized, then when I tried again it stayed connected to my computer and I've gotten a big logfile.
it is attached. It seems like all the dives on the computer has been downloaded as bins.
Hope this helps Jef!

What now?

@cosmo - i have Windows 11. I just set the Sirius and BT-mode and then adda at new bt-device in the windows settings.

[Jef Driesen]

On 2024-04-11 11:48, Cosma Tassinari wrote:
> Maybe I did something wrong, or there were other steps to do before
> that I didn't do, because the sirius.log is empty.

I forgot to mention you need to activate the bluetooth mode of your
Sirius dive computer, and then pair it using the Windows bluetooth tools
before starting the command-line tool.

> This is what comes out of the cmd:
>
> C:\Users\cosma>sirius.exe -vv -l sirius.log -f iconhd -m 0xff download
> -o dive.%n.bin -f raw -t ble
> DATETIME 2024-04-11T09:45:43Z (1712828743)
> VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
> Opening the I/O stream (ble, null).
> ERROR: Dati non validi (13) [in src/ble-win32.c:46
> (win32_ble_get_regproperty)]
> INFO: Discover: address=47E413191606, name=
> INFO: Open: address=47E413191606
> ERROR: Dati non validi (13) [in src/ble-win32.c:46
> (win32_ble_get_regproperty)]

The applications fails to get the bluetooth device name of a device.
That shouldn't be a fatal error. Is there no further output shown?
Unless this device is your Sirius, there are also no other bluetooth
devices discovered.

Jef

[Cosma Tassinari]

Ok, now I got my pc to see the Sirius, but I get the same output in cmd

C:\Users\cosma>sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o dive.%n.bin -f raw -t ble

DATETIME 2024-04-11T11:30:58Z (1712835058)

VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
Opening the I/O stream (ble, null).
ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]
INFO: Discover: address=47E413191606, name=
INFO: Open: address=47E413191606
ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]

[Jef Driesen]

On 11/04/2024 13:32, Cosma Tassinari wrote:
> Ok, now I got my pc to see the Sirius, but I get the same output in cmd
>
> C:\Users\cosma>sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o
> dive.%n.bin -f raw -t ble
> DATETIME 2024-04-11T11:30:58Z (1712835058)
> VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
> Opening the I/O stream (ble, null).
> ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]
> INFO: Discover: address=47E413191606, name=
> INFO: Open: address=47E413191606
> ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]

I'm not sure what is going wrong here. There should be more log output. Can you
send me the sirius.log file?

What output do you get when you only try to scan for bluetooth devices:

sirius.exe -vv -l scan.log scan -t ble

Jef

[Cosma Tassinari]

Here is the scan result, but the sirius.log file is empty...

C:\Users\cosma\Desktop>sirius.exe -vv -l scan.log scan -t ble
DATETIME 2024-04-12T12:52:04Z (1712926324)
VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)

ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]
INFO: Discover: address=47E413191606, name=

47:E4:13:19:16:06       (null)
INFO: Discover: address=D0802B184AE0, name=Designer Compact Keyboard
D0:80:2B:18:4A:E0       Designer Compact Keyboard
INFO: Discover: address=943469D5CDFF, name=Sirius
94:34:69:D5:CD:FF       Sirius

[Jef Driesen]

On 12/04/2024 14:54, Cosma Tassinari wrote:
> Here is the scan result, but the sirius.log file is empty...

Very strange the file is empty. I suspect it's because you are running the
application from a directory where it does not have write permissions.

Can you try to run it from a different directory than "Desktop"? For example the
Documents or Downloads directory.

> C:\Users\cosma\Desktop>sirius.exe -vv -l scan.log scan -t ble
> DATETIME 2024-04-12T12:52:04Z (1712926324)
> VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
> ERROR: Dati non validi (13) [in src/ble-win32.c:46 (win32_ble_get_regproperty)]
> INFO: Discover: address=47E413191606, name=
> 47:E4:13:19:16:06       (null)
> INFO: Discover: address=D0802B184AE0, name=Designer Compact Keyboard
> D0:80:2B:18:4A:E0       Designer Compact Keyboard
> INFO: Discover: address=943469D5CDFF, name=Sirius
> 94:34:69:D5:CD:FF       Sirius

The problem is that you have a bluetooth device, where retrieving the name
fails, and because of that empty name, the application tries to connect to that
device instead of your Sirus. You can force the correct device by appending the
MAC address of your Sirius at the end of the command, like this:

sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o dive.%n.bin -f raw -t

ble 943469D5CDFF

(All on one line.)

Jef

[Cosma Tassinari]

C:\Users\cosma\Downloads>sirius.exe -vv -l sirius.log -f iconhd -m 0xff download -o dive.%n.bin -f raw -t ble 943469D5CDFF
DATETIME 2024-04-12T13:38:48Z (1712929128)
VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
Opening the I/O stream (ble, 943469D5CDFF).
INFO: Open: address=943469D5CDFF
INFO: Service: handle=0001, uuid=00001801-0000-1000-8000-00805f9b34fb
INFO:   Characteristic: handle=0002, uuid=00002a05-0000-1000-8000-00805f9b34fb, flags=I
INFO:           Descriptor: handle=0004, uuid=00002902-0000-1000-8000-00805f9b34fb, type=2
INFO:   Characteristic: handle=0005, uuid=00002b2a-0000-1000-8000-00805f9b34fb, flags=R
INFO:   Characteristic: handle=0007, uuid=00002b29-0000-1000-8000-00805f9b34fb, flags=RW
INFO: Service: handle=0009, uuid=00001800-0000-1000-8000-00805f9b34fb
INFO:   Characteristic: handle=000a, uuid=00002a00-0000-1000-8000-00805f9b34fb, flags=RW
INFO:   Characteristic: handle=000c, uuid=00002a01-0000-1000-8000-00805f9b34fb, flags=R
INFO: Service: handle=000e, uuid=0000180a-0000-1000-8000-00805f9b34fb
INFO:   Characteristic: handle=000f, uuid=00002a29-0000-1000-8000-00805f9b34fb, flags=R
INFO:   Characteristic: handle=0011, uuid=00002a23-0000-1000-8000-00805f9b34fb, flags=R
INFO: Service: handle=0013, uuid=544e326b-5b72-c6b0-1c46-41c1bc448118
INFO:   Characteristic: handle=0014, uuid=99a91ebd-b21f-1689-bb43-681f1f55e966, flags=R(WWR)
DEBUG: RX: 99a91ebd-b21f-1689-bb43-681f1f55e966
INFO:   Characteristic: handle=0016, uuid=1d1aae28-d2a8-91a1-1242-9d2973fbe571, flags=RN
INFO:           Descriptor: handle=0018, uuid=00002902-0000-1000-8000-00805f9b34fb, type=2
DEBUG: TX: 1d1aae28-d2a8-91a1-1242-9d2973fbe571
INFO:   Characteristic: handle=0019, uuid=d8b3ab7c-4101-ec80-c441-9b0914f6ebc3, flags=R
INFO: Service: handle=001b, uuid=1d14d6ee-fd63-4fa1-bfa4-8f47b42119f0
INFO:   Characteristic: handle=001c, uuid=f7bf3564-fb6d-4e53-88a4-5e37e0326063, flags=W
INFO:   Characteristic: handle=001e, uuid=984227f3-34fc-4045-a5d0-2c581f81a153, flags=W(WWR)
ERROR: Periodo di timeout del semaforo scaduto (-2147024775) [in src/ble-win32.c:504 (win32_ble_notify)]
ERROR: Failed to enable notifications for the Tx characteristic. [in src/ble.c:756 (dc_ble_open)]
ERROR: Failed to open the ble device. [in examples/common.c:608 (dctool_ble_open)]
ERROR: Error opening the I/O stream. [in examples/dctool_download.c:162 (download)]
ERROR: Input/output error

[Cosma Tassinari]

Don't exactly know why, but removing and reconnecting the Sirius (after having done it multiple times), this time has worked.

It turns out I was having the same issue as A K, Sirius kept connecting and disconnecting. It just didn't reboot...

Anyway, here's the results.

[Jef Driesen]

On 2024-04-12 18:37, Cosma Tassinari wrote:
> Don't exactly know why, but removing and reconnecting the Sirius (after
> having done it multiple times), this time has worked.
> It turns out I was having the same issue as A K, Sirius kept connecting
> and disconnecting. It just didn't reboot...
>
> Anyway, here's the results.

The download worked perfectly. I still have to double check a few
things, but it means support for the Sirius will become available soon.

Right now I'm looking for someone with a Genius or Horizon to confirm
whether the Sirius changes remain compatible with these previous models
or not.

Jef

[Nicki Groiss]

Hi Jef,

i'll test it until tomorrow with the Sirius and the Genius, also with the Puk 4.

Regards

Nick

--
You received this message because you are subscribed to a topic in the Google Groups "Subsurface Divelog" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/subsurface-divelog/MWRdzl6jqRk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to subsurface-dive...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/830acc8adc24d467aa7cdd089e83f0bb%40libdivecomputer.org.

[Nicki Groiss]

Hi Jef,

faster than i tought.

here the Log from the Sirius running FW 1.03.26

An the Puk 4 what have failed

sirius.exe -vv -l puk4.log -f iconhd -m 0xff download -o dive.%n.bin -f raw -t ble
DATETIME 2024-04-16T08:09:04Z (1713254944)
VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
Opening the I/O stream (ble, null).
INFO: Discover: address=D7C079A708EF, name=M585/M590
INFO: Discover: address=C07E9247BD35, name=MX Vertical
INFO: Discover: address=D959441B9D05, name=M585/M590
INFO: Discover: address=FD2831E23453, name=ERGO K860
INFO: Discover: address=705464A2DFFA, name=Puck4
ERROR: No dive computer found. [in examples/common.c:599 (dctool_ble_open)]

ERROR: Error opening the I/O stream. [in examples/dctool_download.c:162 (download)]

ERROR: No device found

The Genius connection i can try tomorrow, what should i do?

Regards

Nick

Am Di., 16. Apr. 2024 um 08:50 Uhr schrieb Jef Driesen <j...@libdivecomputer.org>:

[Jef Driesen]

On 2024-04-16 10:11, Nicki Groiss wrote:
> here the Log from the Sirius running FW 1.03.26

The download failed establishing the bluetooth connection. Others have
reported this specific error too. I'm not sure why it fails, but in
previous post people reported success after removing the pairing and
setting it up again.

Note that for the genius, you should also omit the "-m 0xff" option
(because that enables the sirius variant of the protocol):

sirius.exe -vv -l genius.log -f iconhd download -o dive.%n.bin -f raw -t
ble

If that also fails, try this build to confirm it's indeed the sirius
changes that cause the download to fail:

https://libdivecomputer.org/builds/experimental/windows/dctool-ble.exe

But of course, first that bluetooth connection needs to work.

> An the Puk 4 what have failed
> sirius.exe -vv -l puk4.log -f iconhd -m 0xff download -o dive.%n.bin -f
> raw -t ble
> DATETIME 2024-04-16T08:09:04Z (1713254944)
> VERSION 0.9.0-devel (8a3799e9dd0aa0d219236108038aa62ac8eab5a6)
> Opening the I/O stream (ble, null).
> INFO: Discover: address=D7C079A708EF, name=M585/M590
> INFO: Discover: address=C07E9247BD35, name=MX Vertical
> INFO: Discover: address=D959441B9D05, name=M585/M590
> INFO: Discover: address=FD2831E23453, name=ERGO K860
> INFO: Discover: address=705464A2DFFA, name=Puck4
> ERROR: No dive computer found. [in examples/common.c:599
> (dctool_ble_open)]
> ERROR: Error opening the I/O stream. [in examples/dctool_download.c:162
> (download)]
> ERROR: No device found

The Puck 4 isn't recognized as a dive computer. You can manually select
the correct device by adding the MAC address (705464A2DFFA) as the last
option:

sirius.exe -vv -l puk4.log -f iconhd -m 0xff download -o dive.%n.bin -f

raw -t ble 705464A2DFFA

If the download fails, also try without the "-m 0xff" option. At this
point I simply don't know yet which variant of the protocol the Puck 4
uses, and we need to try to find out.

Jef

[Cosma Tassinari]

Hey there, sorry to bother.
Any update on Sirius compatibility?

[Jef Driesen]

On 22/05/2024 11:39, Cosma Tassinari wrote:
> Hey there, sorry to bother.
> Any update on Sirius compatibility?

Yes, the Sirius support is available in libdivecomputer master branch already.
You'll be able to use it in subsurface once those changes are also merged there.

Jef

[Michael Keller]

Hi all.

This is fixed in Subsurface > 6.0.5187, available from https://subsurface-divelog.org/latest-release

Ngā mihi

  Michael Keller

[Cosma Tassinari]

Hi,

I just downloaded the latest version (CICD): 6.0.5194, both for Windows and Android, but I cannot find Sirius computer under the brand Mares...

[Michael Keller]

Hi Cosma.

On 31/05/24 19:24, Cosma Tassinari wrote:
> I just downloaded the latest version (CICD): 6.0.5194, both for
> Windows and Android, but I cannot find Sirius computer under the brand
> Mares...

Apologies, I just realised that this thread is about a new Mares model.
The libdivecomputer bug that was fixed in 6.0.5187 was affecting the
currently supported Mares models, but the change did not add support for
new models.

The next time that support for new models will be added to Subsurface
will likely be around August.


Cheers

  Michael Keller

[Cosma Tassinari]

Hi there, sorry to bother but back from vacations, lots of dives to download :-)

Any news on the release date for the Sirius support?

[Michael Keller]

Hi Cosma.

On 1/09/24 23:19, Cosma Tassinari wrote:

Hi there, sorry to bother but back from vacations, lots of dives to download :-)

Any news on the release date for the Sirius support?

Try the latest version from https://subsurface-divelog.org/latest-release/:

Ngā mihi

  Michael Keller

[A K]

Works like a charm! Thanks!

Had to remove Sirius bluetooth connection and reconnect in windows before.

--

You received this message because you are subscribed to a topic in the Google Groups "Subsurface Divelog" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/subsurface-divelog/MWRdzl6jqRk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to subsurface-dive...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/3d3c4d45-ddbc-4d66-99a7-961422899f65%40042.ch.

[Erik Larsson]

I tried to connect my Sirius to my PC today, with no luck. It won't appear when searching using Subsurface. When searching with windows, it will sometimes appear. When I try connecting with windows, Sirius shows connected for a few seconds but nothing happens on my PC. Any advice?

[Rafał Multan]

Hi,

The import was working some time ago with Sirius. Device was also auto discovered. It doesn't work anymore (some data parsing error). I would be happy to provide a fix for it, but I'm struggling to debug libdivecomputer  in Subsurface. I can debug Subsurface, but debugger doesn't step into libdc code. I presume that there are no symbols for libdivecomputer generated during a build from Subsurface. Michael Keller could you provide some help on how to build libdc with debug symbols?  

Error:

[19.200223] ERROR: Unsupported object type (0) or version (2.0). [in src/mares_iconhd_parser.c:1077 (mares_genius_foreach)]

ERROR: Dive 1: Error parsing the samples: Data format error

[22.314585] ERROR: Failed to transfer the segment packet. [in src/mares_iconhd.c:505 (mares_iconhd_read_object)]

[22.314641] ERROR: Failed to read the dive data. [in src/mares_iconhd.c:1106 (mares_iconhd_device_foreach_object)]

INFO: Deleting BLE object

[Jef Driesen]

On 23/03/2025 11:15, Rafał Multan wrote:
> The import was working some time ago with Sirius. Device was also auto
> discovered. It doesn't work anymore (some data parsing error). I would be happy
> to provide a fix for it, but I'm struggling to debug libdivecomputer  in
> Subsurface. I can debug Subsurface, but debugger doesn't step into libdc code. I
> presume that there are no symbols for libdivecomputer generated during a build
> from Subsurface. Michael Keller could you provide some help on how to build
> libdc with debug symbols?
>
>
> Error:
>
> [19.200223] ERROR: Unsupported object type (0) or version (2.0). [in
> src/mares_iconhd_parser.c:1077 (mares_genius_foreach)]
>
> ERROR: Dive 1: Error parsing the samples: Data format error
>
> [22.314585] ERROR: Failed to transfer the segment packet. [in
> src/mares_iconhd.c:505 (mares_iconhd_read_object)]
>
> [22.314641] ERROR: Failed to read the dive data. [in src/mares_iconhd.c:1106
> (mares_iconhd_device_foreach_object)]
>
> INFO: Deleting BLE object

This error indicates a change in the mares data format. Have you recently
upgrade the Sirius firmware?

Can you enable the "Save libdivecomputer logfile" and "Force download of all
dives" in the subsurface download dialog, and download your dives again. After
the download is finished you can cancel without importing any dives. Send me the
generated log file, so I can have a look a the changes.

Jef

[Rafał Multan]

I have updated the firmware. I'm happy to help with the changes (the plan was to submit a PR with a fix). I'm struggling a bit with debugging libdc code (I'm pretty sure that subsurface build doesn't include debug symbols for libdc). I have no experience with C++/C/Qt, but I should manage once I get the debugger stepping into libdc code. Any tips on that (I'm using Mac - Apple Silicon) ?  Please find log file attached.

[Jef Driesen]

On 23/03/2025 13:11, Rafał Multan wrote:
> I have updated the firmware. I'm happy to help with the changes (the plan was to
> submit a PR with a fix). I'm struggling a bit with debugging libdc code (I'm
> pretty sure that subsurface build doesn't include debug symbols for libdc). I
> have no experience with C++/C/Qt, but I should manage once I get the debugger
> stepping into libdc code. Any tips on that (I'm using Mac - Apple Silicon) ?

You can probably modify the build options in the build scripts to enable debug
symbols. I never tried that, because I usually debug this kind of problems
outside of subsurface, using the libdivecomputer command-line tool.

>  Please find log file attached.

It looks like the profile version changed from 1.0 to 2.0. The code contains a
safety check to reject unknown versions. If I simply adjust that check to accept
the new version 2.0, the parsing appears to work. But there is probably a reason
why the major version has changed.

What firmware version did you have before the upgrade and after? Are there any
new features in the release notes?

All your dives in the log (with the oldest one from 2024-05-25) already have
format 2.0. Does that mean you already upgraded in May 2025 and now is the first
time you tried to download your dives? Or is the firmware upgrade more recent
and did that update also adjusted your existing dives?

Jef

[Rafał Multan]

I don't know the version before the upgrade, but the last time I upgraded was most likely before 24.11.2024. Current firmware version is 01.06.16, I did the upgrade day before/after last dive. It must have updated previous dives. Mares changelog only says "Logbook enhancement, Code optimization" . 

[Rafał Multan]

Can I build dctool from sources ? Version from the website doesn't work on Apple Silicon.

[Rafał Multan]

I managed to debug libdc from Subsurface. As you mentioned changing the version check allows the entry to be parsed. I downloaded it to my log and it looks good. I think that could be a solution. No idea why Mares changed the version, their website lists some old version as the latest firmware - so the change log is invalid.

[Erik Larsson]

6.0.5378: Works on android

[Mary]

I am having some problems now, both on desktop (linux mint) as well as android (graphene os on a pixel 8). I am on the latest firmware (01.06.16), and have updated to enable freedive mode. These problems existed before I changed that though. It seems like both devices see the computer fine, and there is no issue with bluetooth or the connection.

On desktop version 6.0.5387-CICD (and the two previous) I get the message "Error parsing the header: Data format error" for all 3 of the dives on the dive computer, and nothing is imported.

On mobile it just says "no new dives downloaded from dive computer".

Attached is the log from desktop as "subsurface.log". I tried to get the dump as well, but it gave me an error. The log from this attempt to also attached as "subsurface1.log".

[Jef Driesen]

On 2/05/2025 00:10, Mary wrote:
> I am having some problems now, both on desktop (linux mint) as well as android
> (graphene os on a pixel 8). I am on the latest firmware (01.06.16), and have
> updated to enable freedive mode. These problems existed before I changed that
> though. It seems like both devices see the computer fine, and there is no issue
> with bluetooth or the connection.
>
> On desktop version 6.0.5387-CICD (and the two previous) I get the message "Error
> parsing the header: Data format error" for all 3 of the dives on the dive
> computer, and nothing is imported.
>
> On mobile it just says "no new dives downloaded from dive computer".

It looks like there are some more changes necessary. The dive header version in
the data also changed to 2.0. I'm looking into it.

> Attached is the log from desktop as "subsurface.log". I tried to get the dump as
> well, but it gave me an error. The log from this attempt to also attached as
> "subsurface1.log".

The Sirius doesn't support downloading a memory dump, so that's expected.

Jef

[Maci]

Okay thanks. Im available if you need anything more logs from the computer.

Dont know if its worth noting, but when connecting it says "Model=Sirius firmware=0 serial=3090". "3090" doesnt appear anywhere in the much longer SN of my computer, if this should be a reference to that.

[Maci]

Hi, this still isn't fixed in the latest dev release. Does someone have an update on this?

[Maci]

It is working now on the latest dev release, but with errors. 

It only gets through 29 out of 41 dives before disconnecting with an error.

Attached is the log.

[Jef Driesen]

On 11/06/2025 21:30, Maci wrote:
> It is working now on the latest dev release, but with errors.
>
> It only gets through 29 out of 41 dives before disconnecting with an error.
>
> Attached is the log.

[92,828392] INFO: Write: size=16, data=401D3003000000000000000000000000
[92,923634] INFO: Read: size=17, data=411D3003B40200000000000000000000EA
[92,924365] INFO: Write: size=2, data=AC09
[93,020355] INFO: Read: size=18, data=AA801D3003200000080000000000000000EA
[93,020362] ERROR: Unexpected packet header (80). [in mares_iconhd.c:521
(mares_iconhd_read_object)]
[93,020365] ERROR: Failed to read the dive data. [in mares_iconhd.c:1139
(mares_iconhd_device_foreach_object)]

This is an error I haven't seen before. Does it happen every time you try to
download, or just one time?

Jef

[Maci]

Hi, yes, every time. Both on desktop and on mobile. Always at dive 29.

[Maci]

sorry, forgot to attach the log

[Maci]

I just checked now after coming home from diving today. It is still failing on what is now dive 32. Dive 32 was from a day that I was doing snorkeling, with the watch set in "freedive mode". This freedive mode feature was not available at launch, but became available in a relatively recent firmware update. I am wondering if subsurface hasn't been updated to include this yet. 

When in freedive mode, the computer groups all dives into a single entry. For example, "dive 32" as it is shown in the log on the Sirius itself is actually a collection of 15 "free dives" over the course of an hour or so.

Attached again is the log, though I doubt that it will show any new info. Let me know if there is anything else I can provide. Thanks

On Thursday, June 12, 2025 at 1:08:50 PM UTC Maci wrote: