Page 2 of 2

Re: HALMX roadmap.

Posted: Wed Sep 09, 2015 5:35 am
by sheepdoll
stevech wrote:
A lot of professionals jump-start learning it by attending training classes and workshops. I did it without such.
That is what I did. Why I have so many demo boards. Attend the class or webinar, get a free board. Works great in class, where one has a full IAR or Keil tool. Not so good when one gets home days months or weeks later. I had completly forgotten about the 2012 Arduino class I attended till I found the F0 "Arduino." code on my computer.

Re: HALMX roadmap.

Posted: Sat Sep 19, 2015 3:40 pm
by monsonite
Sheepdoll & All,

I am attending a STM32F7xx HAL course in Cambridge (UK) this Thursday. Free to attend, and they give you a STM32F7 Discovery to take home :)

I really need to switch over to the HAL way of thinking, because although so far using the Standard Peripheral Libraries has worked for M3 and M4 they are not supported on the M7.

Atmel have a blistering 300MHz M7 available in a 144 pin LQFP. http://www.atmel.com/products/microcont ... sam-e.aspx

I guess that is a good candidate for a future descendant of Arduino.


Ken

Re: HALMX roadmap.

Posted: Sun May 08, 2016 2:31 pm
by jonr
A couple more sites to watch:

Lots of useful routines using HAL:

https://github.com/MaJerle/stm32fxxx_hal_libraries

Should soon release support for the STM32L4 (don't know if they will use HAL):

https://www.tindie.com/products/onehors ... out-board/

Supposedly the mbed code uses the HAL.

The STM32L476G-DISCO is only $19.50.

Re: HALMX roadmap.

Posted: Sun May 08, 2016 11:08 pm
by stevech
monsonite wrote:Sheepdoll & All,

I am attending a STM32F7xx HAL course in Cambridge (UK) this Thursday. Free to attend, and they give you a STM32F7 Discovery to take home :)

I really need to switch over to the HAL way of thinking, because although so far using the Standard Peripheral Libraries has worked for M3 and M4 they are not supported on the M7.

Atmel have a blistering 300MHz M7 available in a 144 pin LQFP. http://www.atmel.com/products/microcont ... sam-e.aspx

I guess that is a good candidate for a future descendant of Arduino.


Ken
I elected the HAL + CubeMX route despite pressure to use ye ole Standard Peripheral Libraries. Documentation of HAL libraries is good. After 1+ years with HAL/CubeMX, I'm glad I took this route, for a professional project.
I self-taught CubeMX. That took about a week as it was my first GUI pin-mapper and full up I/O system configuration tool. And it generates all the I/O initialization code and that's very time consuming if you go the bare metal route. That makes a HUGE difference. No more struggling with complex peripherals like DMA on UARTs, GPIO, capture timers. No more DIY driver and ISR code. Getting ST's version of ChanFS (FATFS) working with SDIO (not SPI) and MLC NAND flash was very simple. Works well with 4GB.

I do wish I could have found a decent CubeMX tutorial video. That would have helped a LOT. HAL libraries' documentation is good enough.
Prerequisite: read, read, read the STM23Fxxx reference manual so you get the big picture.

CubeMX is more complex and difficult to learn than is the HAL docs. I rarely need to look at the HAL code per se.

Gotta use a supported targeted compiler/IDE with CubeMX. Last I looked the list was
ST' distributed Eclipse + GCC for Windows/LInux, not sure about Mac. Free.
IAR
Keil
Atollic- Eclipse + GCC. Beware stability of this; see user forum.

Visual GDB claims to be able to ingest a project file generated by CubeMX.

Re: HALMX roadmap.

Posted: Sun May 08, 2016 11:15 pm
by stevech
monsonite wrote: Atmel have a blistering 300MHz M7 available in a 144 pin LQFP. http://www.atmel.com/products/microcont ... sam-e.aspx

I guess that is a good candidate for a future descendant of Arduino.


Ken
Microchip's acquisition of Atmel does not bode well for SAMs, me thinks. Microchip will cherry-pick Atmel's product line and leave much of it behind. Unless Microchip has changed their philosophy of avoiding industry standards like ARM, we have: Goodbye SAMs. Microchip is probably focused on Atmel's presence in automotive.

Re: HALMX roadmap.

Posted: Mon May 09, 2016 12:59 am
by Slammer
Microchip policy against Atmel's product line is a mystery yet. I think, the main cleanup will be done in 8 bit world, as there are too many chips with similar specifications. In 32bit portofolio, the Microchip has no ARM products, there is no reason to kill ARM business. On the other side, I believe that Microchip will improve the supply chain of Atmel chips... (at least for the cherry picked...) :roll:

Re: HALMX roadmap.

Posted: Mon May 09, 2016 1:54 am
by GrumpyOldPizza
jonr wrote:A
Should soon release support for the STM32L4 (don't know if they will use HAL):

https://www.tindie.com/products/onehors ... out-board/
The alpha should be on github late this week or early next week. But then again I am always optimisitc ;-)

Re: HALMX roadmap.

Posted: Wed May 17, 2017 2:38 pm
by grafalex
Hi All,

Is the port alive?

I see 3 separate branches:
- rogerclarkmelbourne's one seems too be original. Has USB Serial , I2C and SPI modules. Work is mostly happen around F103CB (which is one I am interested in)
- sheeldoll's fork has more recent CMCIS for NucleoF103
- ekawahyu's branch is focused on NodinoF072. Has number of examples and Tone core module. CMCIS is located on a HALMX/drivers lever, not on each variant directory. Same thing for USB device libraries.

Each version has something interesting for me. But none of them is looking like a master branch, which is frustrating.

I found also https://github.com/danieleff/STM32GENERIC which seems to be quite good structured.

I'd like to try porting my project to a HAL based Arduino. But having so different variants I really can't decide where to go.

Any insights on future plans for HALMX project?

Re: HALMX roadmap.

Posted: Wed May 17, 2017 4:45 pm
by sheepdoll
This port has been superseded by the "Official" STM port and the STM32Generic port.

This was more of a proof of concept that it was possible to combine HAL and Arduino libraries.

There is another part on my github https://github.com/sheepdoll/XSTMScrips.git which is one of the scripts I was creating to build new variants. I use the printer language postscript for general purpose scripting as postscript has some powerful parsing options.
The STM port took this idea and changed the scripts to use python to create the needed mapping files directly from the CubeMX xml and setup files.

As for the branches. This was an attempt as a way to have multiple development boards in the Arduino tools menu without clutter.

I have sort of moved more of my effort to coding directly in Eclipse without using the Arduino IDE.

Re: HALMX roadmap.

Posted: Thu May 18, 2017 5:59 pm
by grafalex
Thank you for the answer.
What do you mean under 'official' STM port?