Yeah you fumbled and things looked a bit odd on Github until I  understood 
what you had attempted, no biggie :-) (wrote you some notes there)

---
The Bitfiles are generated and distributed via a online build system in apt 
packages:

Whats not so apparent is that when a PR to the mksocfpga repo is merged 
(always by someone different than the submitter as a rule),
this merge starts first an online bitfile builder that builds the bitfiles 
for both the Quartus and the Vivado projects, 
afterwards these bitfiles are piped into a package build system 
that first puts the new bitfiles into debian packages(named socfpga-bit and 
socfpga-rbf), and lastly exports them to the machinekit repo.
(if all goes well)

'This whole build process takes somewhere around 2 hours.

---

Next time a machinekit(socfpga) user then does:
sudo apt update && sudo apt upgrade
The new bitfiles are then pulled into his/hers system and automatically 
placed in the correct folders.




On Monday, 30 September 2019 23:21:07 UTC+2, justin White wrote:
>
> Forked and created a new PR, git is not my specialty. "dc1f" wasn't 
> "released", dc1G is the correct config. I don't see any bitfiles in 
> mksocfpga, only pin files. If you want me to PR the bitfile let me know 
> where it goes.
>
> On Mon, Sep 30, 2019 at 3:52 AM Michael Brown <[email protected] 
> <javascript:>> wrote:
>
>> @Justin
>> I just created a PR <https://github.com/machinekit/mksocfpga/pull/109> 
>> including 
>> the boards bitfile, I recon the lase edit of the  PIN config file 
>> <https://github.com/machinekit/mksocfpga/pull/109/files#diff-e3c15966cdcbd9f076f42fe47e6aa88f>is
>>  
>> in order else place your corrections as a commit message.
>> Best wishes
>> Michael B.
>>
>> On Monday, 23 September 2019 05:34:48 UTC+2, justin White wrote:
>>>
>>> I've decided to release the board I've been working on pretty much as 
>>> is, just with some open source considerations. It was intended for a 
>>> specific machine, but I rung out all of the I/O possibillities I could, no 
>>> DE10 GPIO pin went unused. There is an onboard 5v regulator that will power 
>>> the nano from GPIO and has a PTC fused connector to power about 3A worth of 
>>> external whatever. Specs:
>>>
>>> 9-25v VIN, 
>>> 5v regulator powers Nano from GPIO
>>> 6 differential stepgen interfaces with 5v enable (for external drivers)
>>> 6 differential encoder inputs (single ended encoders pull down encoders 
>>> work fine as well with no extra wiring)
>>> 16 sourcing outputs at supplied field voltage Outputs are done at 
>>> whatever field voltage supplies the board (recommend 24v)
>>> 2 high current opto-mosfet outputs
>>> 16 inputs arranged with single 3-pin connectors each to simplify NPN or 
>>> PNP type switch wiring. Inputs upto 30v
>>> 1 RS422 connector interface for SmartSerial. (not well tested, may be 
>>> issues with MK SS)
>>> On PCB terminal blocks for ground and field V+ that simplify wiring in 
>>> smaller machines
>>> a 3A PTC fused connector for powering external devices from the overkill 
>>> 5V/5A regulator (Nano+onboard components probably don't use more than 2.5A 
>>> @ 5v)
>>> 2 scaled analog input interfaces (4 channels each). 5v interface for 
>>> using potentiometers and such at 5v_ref, and one 4 channel interface that 
>>> is hardware scaled to accept 0-10v external input. (ADC hal component in 
>>> repo) 
>>>
>>> The stepgens or outputs could probably be configured in hm2 firmware to 
>>> support PWM. Stepgens would provide differential PWM @ 5v, outputs would be 
>>> single ended PWM @ supplied field voltage haven't tested PWM yet but 
>>> there's not much to it.
>>>
>>> There are hal files, a gladevcp GUI, and display python file that will 
>>> set the DE10-FB image up as a test platform for the board. The hal files 
>>> are examples of pin masking and pin inversion that is done in hal to make 
>>> the i/o intuitive. It could use some sort of hm2 overlay type thing but 
>>> that is beyond me. There is also 2 versions of an ADC hal component that 
>>> will convert the 12bit data from the onboard ADC into a usable scaled 
>>> voltage input in hal.
>>>
>>> The board isn't super cheap, that wasn't the intention but compared  to 
>>> the BBB hardware it's probably not too bad. It's a fairly large board 
>>> (200x155), but that's because I prefer Phoenix connectors and overall 
>>> wiring cleanliness over small form factor stuff. Still working on the git, 
>>> but it's up.
>>>
>>> https://github.com/ShadeTechnik/socfpga-developement-OSHW
>>>
>>> Testing a stepgen and encoder:
>>>
>>> https://www.dropbox.com/s/qa4ro9r0io0dlvf/Video%20Sep%2022%2C%209%2054%2040%20PM.mov?dl=0
>>>
>>>
>>>
>>>
>>>
>>>  
>>>
>>> -- 
>> website: http://www.machinekit.io blog: http://blog.machinekit.io 
>> github: https://github.com/machinekit
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "Machinekit" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/machinekit/a966b8b2-ac8e-4c8f-831e-c481ef8ddc44%40googlegroups.com
>>  
>> <https://groups.google.com/d/msgid/machinekit/a966b8b2-ac8e-4c8f-831e-c481ef8ddc44%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
website: http://www.machinekit.io blog: http://blog.machinekit.io github: 
https://github.com/machinekit
--- 
You received this message because you are subscribed to the Google Groups 
"Machinekit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/machinekit/4c4cdbef-854f-488a-a73c-20a660f0a97b%40googlegroups.com.

Reply via email to