2014-11-12 12:50 AM
Hi!
I
necessity to use
the standard libraries
for
microcontroller
STM32L052K8
.If there is
, where
can I download
?thanks
#stm32l02014-11-12 01:09 AM
There is none and probably never will be.
https://my.st.com/public/STe2ecommunities/mcu/Lists/STM32Java/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2FSTM32Java%2FSTM32L053%20Standard%20Peripheral%20Library&FolderCTID=0x01200200770978C69A1141439FE559EB459D758000F9A0E3A95BA69146A17C2E80209ADC21¤tviews=768 My personal recommendation is to avoid any ''library'' for the low-level work. The ''Snippets'' show you how. JW2014-11-12 02:21 AM
I have a project being developed with a microcontroller STM32F051K8 CoIDE to be translated for STM32L052K8 . The libraries STM32SnippetsL0 can replace the standard libraries ?
2014-11-12 02:42 AM
> I have a project being developed with a microcontroller STM32F051K8 CoIDE to be translated for STM32L052K8 .
I understand your problem, but I have no easy solution for it. You'll have to either rewrite your project for ''Cube'' libraries (this is what ST would recommend at this moment), develop a (possibly stripped-down) version of the standard peripheral library for the 'L0 (this is what I'd expect from Clive to recommend ;) ), or to rewrite your project for direct register access, without the ''library'' (this is what I would recommend). There is also a community-driven initiative out there to write a replacement for the Standard Peripheral Library for STM32, but I don't know its status (and I don't even know its name :| ) > The libraries STM32SnippetsL0 can replace the standard libraries ? No. Snippets are no libraries, they are examples of how to program STM32 low level without any libraries, just using the definitions provided in the stm32xxxx.h header. JW2014-11-12 10:11 AM
Yeah, perhaps...
I don't have much time or effort invested in the L0, though I did go to the seminar the other week. I guess you have to decide how much effort it is to port it to the HAL libraries, code a subset of the SPL, or go commando and code a register level solution. Given the trade-offs, I might tend toward the latter. The current trend seems to be to make the abstraction thicker whilst make the the speed/resources smaller. I see the Cortex-M0 as a bit of a design regression. You need to determine what your design needs, and commit resources at that.