porting for CH32V203 #33

TianpeiLee · 2024-07-13T14:59:39Z

Hello, I have adapted on the CH32V203, and added the MRS project folder of WCH, and I verified it on the hardware provided by airbot, it works very well, thank you very much for the support of airbot.

I've changed the definition of a couple of variables by adding the keyword "volatile" in front of them to increase their compatibility under gcc, and this will not affect compilation under armcc

tridge · 2024-08-21T02:17:09Z

I've been looking at adding support for this MCU in AM32-bootloader:
https://github.com/tridge/AM32-bootloader
are you on the AM32 discord? If so please ping me and we could create a thread under development
I have a couple of CH32V203 dev boards, but need some help with getting openocd working, setup of toolchain etc

tridge · 2024-09-01T00:18:39Z

here is a rebase of this PR on top of current main with some fixes:
https://github.com/tridge/AM32/tree/pr-ch32v203-wip

tridge · 2024-09-02T01:48:55Z

I've now got the bootloader working. This AM32 tree:
https://github.com/tridge/AM32/tree/pr-ch32v203-wip
now works with this bootloader:
am32-firmware/AM32-bootloader#4

tridge · 2024-09-04T04:01:05Z

Src/main.c

@@ -1680,6 +1683,10 @@ void runBrushedLoop()

 int main(void)
 {
+#ifdef DEBUG_SDI


we should remove this extra debug code before merge

yeah, all debug related code can be deleted

tridge · 2024-09-04T04:01:54Z

Src/main.c

@@ -1832,6 +1839,12 @@ int main(void)
        }
 #endif

+#ifdef MCU_CH32V203


we should add a NEED_INPUT_READY define for F031 and V203, instead of copying the code

Yes, NEED_INPUT_READY would be bettter, it was distinguished by name before.

tridge · 2024-09-04T04:02:05Z

Src/main.c

@@ -1852,6 +1865,7 @@ int main(void)
                for (int i = 0; i < 64; i++) {
                    dma_buffer[i] = 0;
                }
+                PRINT("reset1:%d\r\n",signaltimeout);


debug should be removed

tridge · 2024-09-04T04:02:11Z

Src/main.c

@@ -1865,6 +1879,7 @@ int main(void)
                for (int i = 0; i < 64; i++) {
                    dma_buffer[i] = 0;
                }
+                PRINT("reset2:%d\r\n",signaltimeout);


remove debug

tridge · 2024-09-04T04:02:29Z

Src/main.c

@@ -1952,15 +1967,37 @@ int main(void)
        }

 #ifndef MCU_F031
+#ifdef WCH


commented out code should be removed

tridge · 2024-09-04T04:03:51Z

Src/main.c

-            NVIC_SetPriority(COM_TIMER_IRQ, 0);
-            NVIC_SetPriority(COMPARATOR_IRQ, 0);
-        }
+             NVIC_SetPriority(IC_DMA_IRQ_NAME, 0);


the whitespace changes here should be reverted

tridge · 2024-09-04T04:04:14Z

Src/main.c

@@ -1993,6 +2030,11 @@ int main(void)
            adc_ordinary_software_trigger_enable(ADC1, TRUE);
            //  converted_degrees = (4000 - ADC_raw_temp) / 20;
            converted_degrees = getConvertedDegrees(ADC_raw_temp);
+#endif
+#ifdef WCH
+//            ADC_DMA_Callback( ); //�ж��б�����


comments in english please

tridge · 2024-09-04T04:05:04Z

Src/signal.c

@@ -160,7 +160,13 @@ void transfercomplete()
                    buffersize = 2;
                    computeServoInput();
                }
+#ifdef MCU_CH32V203


why not make these changes part of receiveDShotDma() ?

It can indeed be moved to receiveDShotDma()

tridge · 2024-09-04T04:05:18Z

Src/sounds.c

@@ -97,7 +97,12 @@ void playStartupTune()
    __disable_irq();

    uint8_t value = *(uint8_t*)(EEPROM_START_ADD + 48);
-    if (value != 0xFF) {
+#ifdef WCH


why is this change needed?

Because after it is erased, its default value is 0xe339e339

ok, better to have a define ERASED_FLASH_BYTE, we could default it to 0xff at the top of sounds.c, but use ifndef to allow a Mcu port to override it

yeah would be better

tridge

the PR is looking pretty good now, but needs a rebase on current main branch

tridge · 2024-09-12T21:01:13Z

Src/main.c

@@ -224,7 +224,7 @@ an settings option)
 #endif

 #define VERSION_MAJOR 2
-#define VERSION_MINOR 8
+#define VERSION_MINOR 12


the PR needs to be rebased on current main branch, these defines no longer exist

tridge · 2024-09-12T21:18:47Z

Src/functions.c

+#ifdef WCH
+
+void delayMicros(uint32_t micros)
+{


i'm not a fan of the busy loop method, depends too much on the clock selected. Works for current v203 but will fail on other variants. Don't we have a timer we can use?

All timers have been used

Originally, this delay should use the core timer, so that it can be used as long as it is a chip with the same core. However, the V203 has a total of 4 peripheral timers.

1 for scheduled task, 2 for commutation tasks, 1 for dshot signal detection task, and 1 for PWM output task，so.....

tridge · 2024-09-12T21:37:39Z

@TianpeiLee I have rebased your code and fixed some issues with it in this branch:
https://github.com/tridge/AM32/tree/pr-ch32v203-wip

fix compilation errors

tridge · 2024-09-13T21:16:40Z

the commit "v203_pr" removed thousands of files, removing support for other MCUs, and removing the CI tests
please do a hard reset against my branch
https://github.com/tridge/AM32/tree/pr-ch32v203-wip

tridge · 2024-09-20T03:17:19Z

@TianpeiLee I also just noticed this PR is missing the handling of the driving_brake_strength in Src/main.c
you need to search for '#ifdef STMICRO' and make sure ch32 is handled in all the same places

TianpeiLee · 2024-09-24T06:37:41Z

@tridge Hi,How about resubmitting the pr directly from your branch? Not only is my modification and update very slow every time, but it also fails due to network reasons.

TianpeiLee added 2 commits July 2, 2024 16:04

porting for CH32V203

6a534d8

fix bugs

4628ce4

tridge suggested changes Sep 4, 2024

View reviewed changes

Format and compatibility modifications

09acbad

TianpeiLee requested a review from tridge September 11, 2024 06:48

tridge suggested changes Sep 12, 2024

View reviewed changes

tridge reviewed Sep 12, 2024

View reviewed changes

TianpeiLee and others added 2 commits September 13, 2024 15:28

v203_pr

df27d59

Update main.c

57061c4

fix compilation errors

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

porting for CH32V203 #33

porting for CH32V203 #33

TianpeiLee commented Jul 13, 2024

tridge commented Aug 21, 2024

tridge commented Sep 1, 2024

tridge commented Sep 2, 2024

tridge Sep 4, 2024

TianpeiLee Sep 4, 2024

tridge Sep 4, 2024

TianpeiLee Sep 4, 2024

tridge Sep 4, 2024

tridge Sep 4, 2024

tridge Sep 4, 2024

tridge Sep 4, 2024

tridge Sep 4, 2024

tridge Sep 4, 2024

TianpeiLee Sep 4, 2024

tridge Sep 4, 2024

TianpeiLee Sep 4, 2024

tridge Sep 4, 2024

TianpeiLee Sep 4, 2024

tridge left a comment

tridge Sep 12, 2024

tridge Sep 12, 2024

TianpeiLee Sep 13, 2024

TianpeiLee Sep 13, 2024

tridge commented Sep 12, 2024

tridge commented Sep 13, 2024

tridge commented Sep 20, 2024

TianpeiLee commented Sep 24, 2024

porting for CH32V203 #33

Are you sure you want to change the base?

porting for CH32V203 #33

Conversation

TianpeiLee commented Jul 13, 2024

tridge commented Aug 21, 2024

tridge commented Sep 1, 2024

tridge commented Sep 2, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tridge left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tridge commented Sep 12, 2024

tridge commented Sep 13, 2024

tridge commented Sep 20, 2024

TianpeiLee commented Sep 24, 2024