cpu/stm32f7/4/2: add flashpage and flashpage_raw#11681
cpu/stm32f7/4/2: add flashpage and flashpage_raw#11681fjmolinas wants to merge 12 commits intoRIOT-OS:masterfrom
Conversation
fjmolinas
added
Platform: ARM
fjmolinas
changed the title
aabadie
added
Area: cpu
aabadie
left a comment
aabadie
left a comment
There was a problem hiding this comment.
Looks good at first sight. This is a nice work, very useful.
I have a comment regarding flashpage_init, because I think it could be called from the cpu_init of the stm32.
The new code in flashpage.c should be uncrustified.
fjmolinas
force-pushed
the
pr_flashpage_stm32f4
branch
2 times, most recently
from
fd71050 to
04e3143
Compare
cpu/stm32_common/periph/flashpage.c
Outdated
| #if defined(CPU_FAM_STM32F2) || defined(CPU_FAM_STM32F4) | ||
| static inline int flashbank_sector(void *addr) { | ||
| uint8_t sn = (uint8_t)(((uint32_t)addr - CPU_FLASH_BASE) / FLASHSECTOR_SIZE_MIN); | ||
| if(sn > 3 && sn < 8) { |
There was a problem hiding this comment.
This looks like magic number. Can you add a comment and a reference ? Maybe there's a macro available from the CMSIS ?
There was a problem hiding this comment.
There is not Macro available in CMSIS, tell me if the comment is enough or if I should define Macros for this :)
cpu/stm32_common/periph/flashpage.c
Outdated
| } | ||
| #endif | ||
|
|
||
| #if !(defined(CPU_FAM_STM32F2) || defined(CPU_FAM_STM32F4)) |
There was a problem hiding this comment.
| #if !(defined(CPU_FAM_STM32F2) || defined(CPU_FAM_STM32F4)) | |
| #if defined(CPU_FAM_STM32F0) || defined(CPU_FAM_STM32F1) \ | |
| defined(CPU_FAM_STM32L0) || defined(CPU_FAM_STM32L1) \ | |
| defined(CPU_FAM_STM32L4) |
Better be explicit
There was a problem hiding this comment.
Should this be done for the other cases not related to this PR?
There was a problem hiding this comment.
I would say yes but not in this PR.
fjmolinas
force-pushed
the
pr_flashpage_stm32f4
branch
from
04e3143 to
8195b9c
Compare
fjmolinas
force-pushed
the
pr_flashpage_stm32f4
branch
from
5904d2f to
cc20674
Compare
fjmolinas
force-pushed
the
pr_flashpage_stm32f4
branch
from
8ea9b4c to
32ce3fe
Compare
|
@aabadie I rebased and added support for STM32F7 |
fjmolinas
changed the title
-stm32f4 doesn't have pages but sectors, flashpage is used as a wrapper around sectors
-stm32f2 doesn't have pages but sectors, flashpage is used as a wrapper around sectors
-stm32f7 doesn't have pages but sectors, flashpage is used as as wrapper around sectors
fjmolinas
force-pushed
the
pr_flashpage_stm32f4
branch
from
32ce3fe to
caba796
Compare
fjmolinas
added
the
State: WIP
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you want me to ignore this issue, please mark it with the "State: don't stale" label. Thank you for your contributions. |
stale
bot
added
the
State: stale
2 participants
Contribution description
This PR adds
flashpageandflashpage_rawfor stm32f2/4/7stm32f2 and stm32f4. Theybothuse sectors as the way of dividing flash, instead of the usual page. These sector do not have the same size, and they can vary according to the specific CPU.This PR implements
flashpageandflashpage_rawas a wrapper over sectors. To still keep the page notion, an arbitrarypageof length 1kB has been defined. When an erase is requested the sector where thispageis located is only erased if thepagewas in not in a erase state anymore, i.e. '0xFF'.NOTE1: this "wrapper" concept is only for erasing, when it comes to writing it's pretty much the same than for other STM32.
NOTE2: there is an api change since aflashpage_init()is introduced. Not sure if that is the best option.NOTE3: I tried to make it work forstm32f7but I have problems with unexpected latency when performing flash operations, so delaying it out; but the general approach works as well forstm32f7.Testing procedure
Tested on
nucleo-f207zg,nucleo-f446re,nucleo-f767zi&stm32f429i-disc1. On these or otherstm32f2andstm32f4run:make clean -C tests/periph_flashpage/ BOARD=nucleo-f207zg flash testeverything should pass.
Issues/PRs references