Commit graph

55 commits

Author SHA1 Message Date
Dario Nieuwenhuis
93a3323ed9 stm32-metapac: remove peripehral_pins! macrotable. 2022-02-23 20:22:16 +01:00
Dario Nieuwenhuis
f14cacbf93 stm32-metapac: change af from string to u8 2022-02-23 19:54:46 +01:00
Dario Nieuwenhuis
2c8fee59d6 stm32-metapac: remove peripehral_dma_channels! macrotable. 2022-02-23 19:16:37 +01:00
Dario Nieuwenhuis
8160af6af9 stm32: replace peripheral_rcc! macrotable with build.rs 2022-02-09 00:58:17 +01:00
Dario Nieuwenhuis
4d73d87b40 stm32-metapac: add option to generate chip metadata as a rust const. 2022-02-09 00:28:05 +01:00
Dario Nieuwenhuis
8ef5b404f1 Update stm32-data 2022-02-07 23:13:44 +01:00
Dario Nieuwenhuis
aaf25bbac3 Update stm32-data 2022-02-07 20:40:35 +01:00
Dario Nieuwenhuis
de19fe5c05 Update stm32-data, update build scripts for new schema. 2022-02-07 02:23:02 +01:00
Dario Nieuwenhuis
b0fabfab5d Update stm32-data: rcc regs info comes from yamls now. 2021-11-29 02:28:02 +01:00
Dario Nieuwenhuis
78d109f5d5 stm32-metapac: add pac feature to allow building only the macrotables. 2021-11-24 01:04:23 +01:00
Dario Nieuwenhuis
d06cb0a264 stm32-metapac-gen: use actually common common.rs instead of emitting it at every single chip file. 2021-11-24 00:42:14 +01:00
Dario Nieuwenhuis
dfb6d407a1 stm32: rename core features from _cmX to -cmX, cleanup gen. 2021-11-23 23:49:06 +01:00
Dario Nieuwenhuis
039621c56d stm32-metapac-gen: fix broken build.rs 2021-11-22 02:38:01 +01:00
Dario Nieuwenhuis
0db4da10f8 stm32-metapac-gen: separate data structs 2021-11-22 02:37:46 +01:00
Bob McWhirter
f12b70535b Adjust for STM32U5. 2021-11-02 12:05:24 -04:00
Matous Hybl
4fbac40120 Workaround duplicity of DMA channel declaration when the target chip specifies more than one request, by processing only the first declared request for the channel. 2021-10-22 11:36:47 +02:00
Ulf Lilleengen
c79485c286 Support for STM32L1
* Add RCC
* Fix more issues with dash in chip names
* Update stm32-data version
* Add blinky and spi example
2021-09-21 14:50:23 +02:00
Ulf Lilleengen
14aa4265db Ensure linear stride 2021-09-15 13:58:01 +02:00
Ulf Lilleengen
701dd9fa36 Calculate port number based on address and stride
This allows out of order letter to port mapping
2021-09-15 13:58:01 +02:00
Ulf Lilleengen
840a83e196 Add support for chip definitions with a dash 2021-09-15 13:58:01 +02:00
Dario Nieuwenhuis
f2623e7e9b Update lots of deps 2021-09-11 01:35:23 +02:00
Ben Gamari
e2f71ffbbd Add support for STM32G0 2021-08-20 00:15:11 +02:00
Dario Nieuwenhuis
174c51f097 stm32/metapac: check GPIO RCC regs are always found. 2021-08-19 23:59:50 +02:00
Dario Nieuwenhuis
dd62790f36 stm32-metapac: assume RCC is always present 2021-08-19 22:16:27 +02:00
Ulf Lilleengen
4df63f5379 Add per-core EXTI support
* Generate a core index put into the PAC for the peripherals to use as
  index into registers.
* Add EXTI v2 which uses CORE_INDEX to index exti registers
2021-08-17 16:22:47 +02:00
Timo Kröger
f240edc579 Remove broken panic message 2021-08-04 12:51:24 +02:00
Timo Kröger
e21faaeb47 cargo fmt 2021-08-04 12:43:51 +02:00
Timo Kröger
b36337b3d2 Refactor bit search logic when no clock specified
Always search for a enable bit, even when no clock is specified in the
yaml. Try to derive the clock name from the register name.

This change picked up USART2 for stm32wle which was missing the clock.
2021-08-04 12:42:35 +02:00
Timo Kröger
37536695e1 Refactor find_reg_for_field() 2021-08-04 11:22:54 +02:00
Timo Kröger
c985c031d4 Use a regex to match find enable and rst regs 2021-08-03 17:58:28 +02:00
Timo Kröger
4e47b15daf Ignore the clock number for enable bit search
The number has different meanings depending on family:
stm32f0: RCC_APB2ENR - APB peripheral clock enable register 2    CLOCK: APB1
stm32f4: RCC_APB2ENR - RCC APB2 peripheral clock enable register CLOCK: APB2

Ignore the clock number and search all registers for a matching enable bit.
2021-08-03 17:58:27 +02:00
Timo Kröger
fba8b86005 BTreeMaps to preserve ordering of generated code
Makes diffing of changes easier
2021-08-03 17:58:26 +02:00
Bob McWhirter
63b32b39e1 Use an em bikeshed instead of an underscore bikeshed. 2021-08-02 13:29:06 -04:00
Bob McWhirter
5f9447abb4 Put the implicit memory.x behind a memory_x feature on embassy-stm32. 2021-08-02 13:21:30 -04:00
Bob McWhirter
4da8ed471b Fallback to total RAM or FLASH size if otherwise not present. 2021-08-02 11:23:55 -04:00
Bob McWhirter
f6c5f039c8 Emit a default memory.x alongside device.x from metapac. 2021-08-02 11:23:55 -04:00
Thales Fragoso
16d5294817 Get rid of some warnings 2021-07-29 18:48:43 -03:00
Dario Nieuwenhuis
8d76c582f5 Update chiptool 2021-07-29 13:39:19 +02:00
Bob McWhirter
b910551c9a Generate more rows in the interrupts! table.
Adjust DMA/BDMA to use the new style.
2021-07-27 12:52:01 -04:00
Bob McWhirter
650f867b1c Add a single-column variant to gpio_rcc! macro table
which includes just the set of registers that need to be
considered.

Then match against those registers with a single `modify(...)`
2021-07-23 11:32:20 -04:00
Bob McWhirter
13873df30b Auto-enable all GPIOs during init(). 2021-07-23 11:32:19 -04:00
Bob McWhirter
d68f2617e6 Add a Dbgmcu struct capable of enabling all relevant DBGMCU.cr fields.
Includes the addition of a `dbgmcu!(...)` macro table which currently takes
the form of

	(cr, $fn_name:ident)

where `$fn_name` is something like `set_dbgsleep_d1` etc.

The method is unsafe, since it's performing unsafe PAC operations.

Two examples modified to demonstrate its usage.
2021-07-22 14:18:48 -04:00
Dario Nieuwenhuis
54b5012c56 stm32/dma: update codegen+macrotables for new stm32-data 2021-07-17 07:35:59 +02:00
Dario Nieuwenhuis
dff4b42a33 stm32/dma: Fix macrotable column count not matching in DMAMUX-less BDMAv1. 2021-07-17 04:54:34 +02:00
Bob McWhirter
01cc733c61 Prep for new stm32-data with dmamux differentiation, but not yet using. 2021-07-16 15:44:22 -04:00
Bob McWhirter
1254e99be1 Be able to scrub out BDMA-centric IRQs.
Further refine distinction between DMA and BDMA interrupts in the table.
2021-07-16 14:41:20 -04:00
Thales Fragoso
2f08c7ced5 stm32: Allow for RccPeripheral without reset field
This fix build on F0, since it doesn't have DMARST. This change makes
RccPeripheral::reset a no-op on peripherals where a reset field couldn't
be found
2021-07-15 13:25:51 -03:00
Bob McWhirter
92247369e7 Remove some unused traits.
Move some fns to associated consts.
2021-07-13 10:09:35 -04:00
Bob McWhirter
2e10ab2e5c Let's count channels per DMA peripheral, shall we now? 2021-07-13 10:09:35 -04:00
Bob McWhirter
06e899b14c Adjust to DMA1EN in the rcc for l0. 2021-07-13 10:09:35 -04:00