Improve MKS Robin support #19333

Robin-DUBREUIL · 2020-09-10T10:00:25Z

Updates on the MKS Robin pins configuration file

Hi everyone and thank you for working on the wonderful piece of software that Marlin is.
I managed to update pins_MKS_ROBIN.h to take advantage of most of the board capabilities.

Changes

Support for the 16Kb (2KB) onboard I2C EEPROM:tested working.
Support for the 64Kb (8KB) onboard SPI FLASH: not tested.
Support for the ColorUI interface on most MKS ROBIN TFT controllers (FSMC): tested working only with ILI9328 on MKS Robin TFT v1.1R.
SD now uses 4.5MHz SDIO (no card detect pin, needs edits to take adavantage of it on the MKS Robin V2.4): tested working on MKS Robin v2.2 and v2.3, latest bugfix required.
Added some hints on the required hardware mod to use SPI2 for additional devices.
Cleanup and refactoring.
Miscellaneous minor tweaks.

This code can still be improved as i do not use the LVGL interface because it lacks UBL and advanced pause feature so things may still be missing for the LVGL interface users. More work can be done to add or fix those features on MKS Robin derivatives, as i don't have those boards i prefer to let others improve their pin configuration file on the basis on my work.
Also, i've seen that a lot of work is being done to refactor the TFT interface (see #19192) and it sounds awesome, but it would need to update this file again.

### Updates on the MKS Robin pins configuration file Hi everyone and thank you for working on the wonderful piece of software that Marlin is. I managed to update pins_MKS_ROBIN.h to take advantage of most of the board capabilities. ### Changes - Support for the 16Kb (2KB) onboard I2C EEPROM: **tested working.** - Support for the 64Kb (8KB) onboard SPI FLASH: **not tested.** - Support for the emulated DOGM interface on most MKS ROBIN TFT controllers (FSMC): **tested working only with ILI9328 on MKS Robin TFT v1.1R.** - SD now uses 4.5MHz SDIO (no card detect pin, needs edits to take adavantage of it on the MKS Robin V2.4): **tested working on MKS Robin v2.2, latest bugfix required.** - Added some hints on the required hardware mod to use SPI2 for additional devices. - Cleanup and refactoring.

thisiskeithb · 2020-09-10T10:24:05Z

tested working on MKS Robin v2.2, latest bugfix required

~~Should some of these changes have been made to the Robin 2 pins instead? https://github.com/MarlinFirmware/Marlin/blob/bugfix-2.0.x/Marlin/src/pins/stm32f4/pins_MKS_ROBIN2.h~~

Ignore me. These boards are difficult to keep track of when they all have the same model & numbers in the name.

Robin-DUBREUIL · 2020-09-10T13:04:09Z

Hi,
MKS Robin and Robin 2 boards are completely different. The kind of board my PR deals with are the first Robin based on the STM32F103ZET6 MCU and it appears that mine is a v2.2. Robin 2 boards are newer and more powerful boards based on the STM32F407ZET6 MCU, so apparently none of the changes i made could possibly affect those boards (different HAL, pin mappings, features...).

sjasonsmith · 2020-09-10T14:05:43Z

@thisiskeithb that is the naming confusion in mentioned the other day. Robin v2 != Robin2

sjasonsmith · 2020-09-10T14:14:02Z

@mattdog01 does this seem related to the SDIO issues you’ve been having with a Robin? @TheCodeExorcist, which PlatformIO environment are you using this with? mks_robin or mks_robin_stm23?

rhapsodyv · 2020-09-10T15:29:43Z

Also, i've seen that a lot of work is being done to refactor the TFT interface (see #19192) and it sounds awesome, but it would need to update this file again.

Sure, one of the idea from the TFT refactoring, is to remove the need of board know about the UI. The board will just setup FSMC / SPI interface, and it's done.
There's another place where the display types are defined, with their specs: driver, resolution, orientation, etc. The user just select the UI and the Display type.

Robin-DUBREUIL · 2020-09-10T15:39:19Z

@sjasonsmith, I built this on PlatformmIO for Code OSS withe the standard mks_robin environment. I suspect that the mks_robin_stm32 environment is kind of a relic from the past as i don't see any practical use case for it nowadays...
However, i forgot to mention that i edited W25Qxx.cpp to be able to use the ENABLED() macro here in pins_MKS_ROBIN.h:
#define HAS_SPI_FLASH #if ENABLED(HAS_SPI_FLASH)
So, you have to change those two lines this way in attempt to be able to compile marlin with a stock W25Qxx.cpp:
#define HAS_SPI_FLASH 1 #if HAS_SPI_FLASH
Nevermind, i just edited the file accordingly.

Reverting back changes to make file compilable with a stock W25Qxx.cpp file.

sjasonsmith · 2020-09-10T15:44:16Z

mks_robin_stm32 is very new. Different overlapping efforts for TFT support resulted in it being first implemented for that environment, then @rhapsodyv later expanding it to support more environments.

Robin-DUBREUIL · 2020-09-10T15:49:41Z

@sjasonsmith I better understand now where this environment comes from, i had some severe compilation issues with it this is why I pushed some efforts to get a functionnal ColorUI on the standard mks_robin environment, on the basis of what has been done on other boards and by reading some of the Marlin code.

mattdog01 · 2020-09-10T16:07:06Z

@mattdog01 does this seem related to the SDIO issues you’ve been having with a Robin? @TheCodeExorcist, which PlatformIO environment are you using this with? mks_robin or mks_robin_stm23?

@sjasonsmith I suspect it does. I have the MKS-Robin v2.3 :)

FYI I ran a couple hour print last night with the latest bugfix available (last night) and it still had the random print head movements to the ends of the axis and back but with a new one, the extruder tried to drive a what seemed like 1000+mm of filament through the hot end in a couple seconds and of course just shredded the filament and then continued to the end of the print. It didn't reset but it did have issues reading the SD card. Sorry if that is too much info.

Robin-DUBREUIL · 2020-09-10T16:20:13Z

@mattdog01 I compiled my firmware last night too with the latest bugfix available and my custom config file. My printer worked like a charm all night long so you should definitely give it a try. I precised that the latest bugfix is needed only because of this fix recently made for boards without SD detect pin: #19236

mattdog01 · 2020-09-10T16:57:54Z

@TheCodeExorcist I'm not a coder and know little about how the files should be / are arranged, so I will need to wait for it to get pulled into bugfix for me to get your fixes. How long does that normally take?
I looked at the piece of code you have on the other issue. Does that just need to be pasted into the pins_mks_robin file? I'm assuming at that location?
Any config setting changes? Build in mks_robin_stm32?

mattdog01 · 2020-09-10T17:10:41Z

@TheCodeExorcist I could't find any of this text in my pins_MKS_ROBIN.h in order the know where to change the code there. Best if I just look and not touch.

W25Q64 64Mb (8MB) SPI flash
#define HAS_SPI_FLASH

Robin-DUBREUIL · 2020-09-10T17:23:13Z

@mattdog01, many things have been added over the original pins_MKS_ROBIN.h so the easiest way to go is to simply replace the original file by my modified one. You have to build with the mks_robin environment as i did not achieve to build it with mks_robin_stm32 and as i don't see any advantage of using mks_robin_stm32 over mks_robin i don't think i will try to make it work. Your config should remain the same as long as you use FSMC_GRAPHICAL_TFT (for upscaled Marlin UI) or TFT_320x240 (for Color UI) and if you don't, switch to one. If you choose TFT_320x240 to take advantage of Color UI, you may get a non working screen and if that happens, try commenting this line #define TFT_ILI9328 and decommenting the next driver //#define TFT_R61505 and so on until you find the driver your TFT uses...

sjasonsmith · 2020-09-10T17:30:49Z

i don't see any advantage of using mks_robin_stm32 over mks_robin

The hope is to eventually move everything from HAL/STM32F1 to HAL/STM32, making mks_robin_stm32 the future. That said, there is no immediate timeline for that migration. PlatformIO has deprecated the Maple framework used by HAL/STM32F1, and from a maintenance perspective there are advantages to only maintaining a single STM32 HAL.

@mattdog01 is having some reliability issues with his SD reads using mks_robin_stm32, so this will be a good data point to determine whether the same problems exist when he is using your version with the mks_robin environment.

Robin-DUBREUIL · 2020-09-10T17:38:44Z

@sjasonsmith Ok, so with that said every STM32 based board should move to the unified STM32 HAL with no real deadline for now. Is the MKS Robin use of the unified STM32 HAL a standard way of doing it (all boards should do the same way) or something still hacky ?
It is in fact interesting to see if @mattdog01 have the same issues on mks_robin because this new HAL seems to be in use for a few boards only so we have less feedback on running marlin with it. I think this should still be considered unsafe for production, right ?

Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

sjasonsmith · 2020-09-10T17:44:11Z

Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

-#ifdef ARDUINO_ARCH_STM32F1
-  #define BEEPER_PIN                        PC13
-#else
-  #define BEEPER_PIN                        -1
-#endif
+//
+// Piezzoelectric speaker
+//
+#define BEEPER_PIN                          PC13


I wonder why this was -1 when using the STM32 HAL?

This might introduce a timer conflict for mks_robin_stm32, but I'm not too worried about that. If that happens a little extra work will be needed to possibly move the TONE_TIMER, but that would be the right thing to do.

I verified that PC13 isn't attached to any timer pins. The Maple framework will take over any timer attached to BEEPER_PIN if SPEAKER is enabled, which can make enabling it risky for some boards. It seems ok for this board.

The beeper works fine when compiled with mks_robin on my board, i should try compiling with mks_robin_stm32 to see if everything is still fine but for now i have other issues to solve to make marlin compile on this framework with my config...

I wonder why this was -1 when using the STM32 HAL?

There was a bug that caused Marlin with HAL STM32 hangs when speaker enabled.
I have not tested this for a while, so I can't tell if this bug still present.

If you tested before I upgraded to ststm32 8.0 that might be fixed. I think I remember TONE not working on one timers, but it worked with the newer framework.

The only problem in BEEPER_PIN that I know is when SPEAKER is enabled. I'm using the BEEPER in a STM32F4, on HAL STM32 with no problem, in the current version.

Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

sjasonsmith · 2020-09-10T17:52:10Z

@TheCodeExorcist I only just discovered that mks_robin_stm32 environment. It slipped in with a TFT PR, so I didn't notice it happened (I tend to ignore display-related work).

I haven't really looked at it close enough to have an opinion on it. I would not assume it is perfect, especially with one of the few early users reporting SD reliability issues.

Nobody has really formulated a plan for moving things over. I would prefer an approach where we get one or two boards really stable and well-defined, which then serves as a template for the rest. The real test will be moving the SKR E3 Mini over, to see if a week can go by without a ton of new issues reported!

commenting TMC SOFTWARE SERIAL by default

Robin-DUBREUIL · 2020-09-10T18:03:18Z

Nobody has really formulated a plan for moving things over. I would prefer an approach where we get one or two boards really stable and well-defined, which then serves as a template for the rest. The real test will be moving the SKR E3 Mini over, to see if a week can go by without a ton of new issues reported!

This would be indeed a quite fireproof test, i'll try to see what i can do on the Robin side.

sjasonsmith · 2020-09-10T18:16:49Z

In the meantime, it is fine to continue fixing things in the STM32F1 versions, I just wouldn't recommend large development investments in it.

mattdog01 · 2020-09-10T20:24:08Z

@TheCodeExorcist So far so good. I went to your repo copied and pasted the contents of pins_MKS_ROBIN.h to my pins_MKS_ROBIN.h . It compiled just fine, The new UI looks good. SD card was ready to be used, no need to use the custom commands to get it working.
@sjasonsmith I think I will need to specify a different set of servo pins and move my servo wires to that location because my servo for bed leveling doesn't work anymore.

mattdog01 · 2020-09-10T20:34:46Z

@sjasonsmith No, way easier than that. Just move the connector to the next servo connection on the board. Works great.

Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

…_patch-1 # Conflicts: # Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

sjasonsmith · 2020-12-12T20:43:27Z

I have rebased this on the current state of bugfix-2.0.x. I guarantee I have made mistakes. @rhapsodyv is going to review the code, the TFT and SD definitions in particular.

I did not force push this. I rebased, then merged the remote branch into my rebase. This means my rebase is reversible if I made a horrible mess of things.

sjasonsmith · 2020-12-12T20:53:01Z

@rhapsodyv I don't know how to configure TFT displays. I left the various display sections in the file, for you to review. I have no idea whether these are still necessary or are correct.

…atch-1 # Conflicts: # Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

sjasonsmith · 2021-01-01T23:01:01Z

I resolved merge conflicts on this again. I think it can only move forward if @rhapsodyv or @TheCodeExorcist respond and provide feedback. I had to do a lot of "educating guessing" when I updated this 20 days ago.

rhapsodyv · 2021-01-04T01:46:52Z

@mattdog01 are you using the pins from here? or from bugfix?

sjasonsmith · 2021-01-04T05:13:05Z

Marlin/src/libs/W25Qxx.cpp

-  /**
-   * STM32F1 APB2 = 72MHz, APB1 = 36MHz, max SPI speed of this MCU if 18Mhz
-   * STM32F1 has 3 SPI ports, SPI1 in APB2, SPI2/SPI3 in APB1
-   * so the minimum prescale of SPI1 is DIV4, SPI2/SPI3 is DIV2
-   */
-  #if SPI_DEVICE == 1
-    #define SPI_CLOCK_MAX SPI_CLOCK_DIV4
-  #else
-    #define SPI_CLOCK_MAX SPI_CLOCK_DIV2
-  #endif
  uint8_t clock;
  switch (spiRate) {
-    case SPI_FULL_SPEED:    clock = SPI_CLOCK_MAX;  break;


This change seems like it does not belong with this PR. @rhapsodyv do you know anything about this?

Sure. In fact we can just keep as it is today. I'm changing those spi calls anyway.

Keep it as it is here today? Or keep it as it is in bugfix today, before these changes?

mattdog01 · 2021-01-04T18:28:50Z

@mattdog01 are you using the pins from here? or from bugfix?

@rhapsodyv I was using the bugfix files without any changes except in the configuration files.

ETE-Design · 2021-01-07T18:59:54Z

@TheCodeExorcist You write "SD now uses 4.5MHz SDIO (no card detect pin, needs edits to take adavantage of it on the MKS Robin V2.4): tested working on MKS Robin v2.2 and v2.3, latest bugfix required." I have an MKS Robin Lite who dosen't seem to have card pin, what needed to be changed to use it with that board if possible?

Robin-DUBREUIL · 2021-01-07T21:51:57Z

Hi, The MKS Robin Lite uses a cheaper variant of the STM32F1 microcontroller which has less IO so the pinout probably differs from the original MKS Robin board in many ways. You should try to understand how this pin file is written and explore the open source MKS schematics that concerns your board to write a relevant pin file for the Robin lite, if nobody did it already. Have fun. 7 janv. 2021 8:00:15 PM ETE-Design <[email protected]>:

…

@TheCodeExorcist[https://github.com/TheCodeExorcist] You write "SD now uses 4.5MHz SDIO (no card detect pin, needs edits to take adavantage of it on the MKS Robin V2.4): tested working on MKS Robin v2.2 and v2.3, latest bugfix required." I have an MKS Robin Lite who dosen't seem to have card pin, what needed to be changed to use it with that board if possible? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub[#19333 (comment)], or unsubscribe[https://github.com/notifications/unsubscribe-auth/APAXBSEANSQISOLYYFU5RDTSYYAD3ANCNFSM4REWC2IQ]. [data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAEgAAABICAYAAABV7bNHAAAABHNCSVQICAgIfAhkiAAAACtJREFUeJztwQENAAAAwqD3T20ON6AAAAAAAAAAAAAAAAAAAAAAAAAAAD4MUUgAARy2AfAAAAAASUVORK5CYII=###24x24:true###][Image de pistage][https://github.com/notifications/beacon/APAXBSCIZXBLVQRCHWD3DILSYYAD3A5CNFSM4REWC2I2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOFUKHBAI.gif]

ETE-Design · 2021-01-07T22:58:36Z

@TheCodeExorcist Thanks for taking your time to answer me thougt the PR is closed... Mabye you can help me a little?
How to find/know #define SDIO_CLOCK and #define SDIO_READ_RETRIES
Also what is WIFI_IO0_PIN for?

Robin-DUBREUIL · 2021-01-07T23:04:47Z

The SDIO values should work flawlessly on any STM32F1 based board, just use the same for yours. The WIFI_IO0_PIN is supposedly useful in an effort to make the ESP12 module (with MKS proprietary formware) work with marlin. I never used nor tested any WiFi functionnality on my board so many things may be needed to make things work properly when it comes to WiFi. 7 janv. 2021 11:58:53 PM ETE-Design <[email protected]>:

…

@TheCodeExorcist[https://github.com/TheCodeExorcist] Thanks for taking your time to answer me thougt the PR is closed... Mabye you can help me a little? How to find/know #define SDIO_CLOCK and #define SDIO_READ_RETRIES Also what is WIFI_IO0_PIN for? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub[#19333 (comment)], or unsubscribe[https://github.com/notifications/unsubscribe-auth/APAXBSFY5VNQIAV7ZAUWVNDSYY4CXANCNFSM4REWC2IQ]. [data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAEgAAABICAYAAABV7bNHAAAABHNCSVQICAgIfAhkiAAAACtJREFUeJztwQENAAAAwqD3T20ON6AAAAAAAAAAAAAAAAAAAAAAAAAAAD4MUUgAARy2AfAAAAAASUVORK5CYII=###24x24:true###][Image de pistage][https://github.com/notifications/beacon/APAXBSFVPJ2AGKLYQHYKSZ3SYY4CXA5CNFSM4REWC2I2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOFULFGCA.gif]

… into bugfix-2.0.x * 'bugfix-2.0.x' of https://github.com/MarlinFirmware/Marlin: (105 commits) [cron] Bump distribution date (2021-01-08) Fix M48 output (MarlinFirmware#20713) Improved MKS Robin support (MarlinFirmware#19333) Preheat before Power Loss Recovery homing (MarlinFirmware#20697) [cron] Bump distribution date (2021-01-07) Custom build_flags by feature (MarlinFirmware#20692) [cron] Bump distribution date (2021-01-06) Multi-Z stepper inverting (MarlinFirmware#20678) Fix Azteeg X3 macro typo (MarlinFirmware#20681) Define SANGUINOLOLU 1.1 enable pins (MarlinFirmware#20682) No BTN_ENC_EN on Anet 10 (MarlinFirmware#20684) Temperature report followup (MarlinFirmware#20687) Adjustable precision in M105 temperature report (MarlinFirmware#20602) Don't apply hotend_offset.z to Z soft endstops (MarlinFirmware#20675) Indent tool_change_prime Clarify solenoid active / magnet-on state Defer "quiet probing" till the last Z bump (MarlinFirmware#20610) Solenoid cleanups [cron] Bump distribution date (2021-01-05) Remove untranslated strings ...

Robin-DUBREUIL marked this pull request as draft September 10, 2020 10:14

Update pins_MKS_ROBIN.h

a79cbb9

Reverting back changes to make file compilable with a stock W25Qxx.cpp file.

sjasonsmith reviewed Sep 10, 2020

View reviewed changes

Update pins_MKS_ROBIN.h

4603812

commenting TMC SOFTWARE SERIAL by default

thinkyhead added 2 commits September 10, 2020 17:11

Apply auto-format

f0edf90

Merge remote-tracking branch 'upstream/bugfix-2.0.x' into pr/19333

ec4da8e

thinkyhead reviewed Sep 10, 2020

View reviewed changes

Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h Outdated Show resolved Hide resolved

thisiskeithb added C: Boards/Pins PR: Improvement labels Sep 10, 2020

Merge remote-tracking branch 'TheCodeExorcist/patch-1' into PRR/19333…

569332e

…_patch-1 # Conflicts: # Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

sjasonsmith added the Needs: Work More work is needed label Dec 12, 2020

Fix failed tests

e842cd4

sjasonsmith assigned rhapsodyv Dec 12, 2020

Merge remote-tracking branch 'upstream/bugfix-2.0.x' into pr/19333

683a6dc

thinkyhead requested a review from rhapsodyv December 22, 2020 13:43

sjasonsmith added 2 commits January 1, 2021 14:50

Merge remote-tracking branch 'upstream/bugfix-2.0.x' into PRR/19333_p…

883ea5e

…atch-1 # Conflicts: # Marlin/src/pins/stm32f1/pins_MKS_ROBIN.h

Remove ONBOARD SD from Robin

0d227ac

rhapsodyv added 2 commits January 3, 2021 22:39

outdated spi configs

42c1496

outdated TFT changes

8628b41

sjasonsmith reviewed Jan 4, 2021

View reviewed changes

thinkyhead added 2 commits January 6, 2021 21:08

Merge remote-tracking branch 'upstream/bugfix-2.0.x' into pr/19333

dfdaa88

Revert W25Qxx.cpp changes

067d5be

thinkyhead merged commit 3009707 into MarlinFirmware:bugfix-2.0.x Jan 7, 2021

thinkyhead changed the title ~~Update pins_MKS_ROBIN.h~~ Improve MKS Robin support Jan 7, 2021

dpreed pushed a commit to dpreed/Marlin_2.0.x that referenced this pull request Feb 5, 2021

Improved MKS Robin support (MarlinFirmware#19333)

8c95cab

kpishere pushed a commit to kpishere/Marlin that referenced this pull request Feb 19, 2021

Improved MKS Robin support (MarlinFirmware#19333)

0ebb9bd

W4tel-BiDi pushed a commit to W4tel-BiDi/Marlin that referenced this pull request Apr 5, 2021

Improved MKS Robin support (MarlinFirmware#19333)

78f2851

thinkyhead pushed a commit to thinkyhead/Marlin that referenced this pull request Apr 29, 2021

Improved MKS Robin support (MarlinFirmware#19333)

9a7db36

thinkyhead pushed a commit that referenced this pull request Apr 30, 2021

Improved MKS Robin support (#19333)

074d3f6

Improve MKS Robin support #19333

Improve MKS Robin support #19333

Conversation

Robin-DUBREUIL commented Sep 10, 2020 • edited Loading

Updates on the MKS Robin pins configuration file

Changes

thisiskeithb commented Sep 10, 2020 • edited Loading

Robin-DUBREUIL commented Sep 10, 2020

sjasonsmith commented Sep 10, 2020

sjasonsmith commented Sep 10, 2020

rhapsodyv commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020 • edited Loading

sjasonsmith commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020

mattdog01 commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020

mattdog01 commented Sep 10, 2020

mattdog01 commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020 • edited Loading

sjasonsmith commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sjasonsmith commented Sep 10, 2020

Robin-DUBREUIL commented Sep 10, 2020 • edited Loading

sjasonsmith commented Sep 10, 2020

mattdog01 commented Sep 10, 2020

mattdog01 commented Sep 10, 2020

sjasonsmith commented Dec 12, 2020

sjasonsmith commented Dec 12, 2020

sjasonsmith commented Jan 1, 2021

rhapsodyv commented Jan 4, 2021

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mattdog01 commented Jan 4, 2021

ETE-Design commented Jan 7, 2021

Robin-DUBREUIL commented Jan 7, 2021 via email

ETE-Design commented Jan 7, 2021

Robin-DUBREUIL commented Jan 7, 2021 via email

Robin-DUBREUIL commented Sep 10, 2020 •

edited

Loading

thisiskeithb commented Sep 10, 2020 •

edited

Loading

Robin-DUBREUIL commented Sep 10, 2020 •

edited

Loading

Robin-DUBREUIL commented Sep 10, 2020 •

edited

Loading

Robin-DUBREUIL commented Sep 10, 2020 •

edited

Loading

Robin-DUBREUIL commented Sep 10, 2020 •

edited

Loading