change the c/c++ preprocessor symbols to "__heap_base__=Image$$$$RW_IRAM1$$$$ZI$$$$Limit __heap_end__=Image$$$$RW_IRAM2$$$$Base" and magically it compiles without issue and works (on the one board I tried). This is for MDK5.14.0.0 - so I don't make any promises for earlier versions.
I confirm that ...
Search found 35 matches
- Mon Oct 19, 2015 9:04 am
- Forum: STM32 Support
- Topic: STM32F4-Discovery + Keil v5.11.1.00 + ChibiOS 2.6.4
- Replies: 39
- Views: 23748
- Fri Jun 05, 2015 8:42 pm
- Forum: STM32 Support
- Topic: Misprint in platform_f105_f107.mk
- Replies: 1
- Views: 2037
Misprint in platform_f105_f107.mk
Hello!
When compiling with trunk (rev.8011) I got this error from make:
This file lacks a trailing back slash in line #78. Patch attached.
When compiling with trunk (rev.8011) I got this error from make:
os/hal/ports/STM32/STM32F1xx/platform_f105_f107.mk:78: *** missing separator (did you mean TAB instead of 8 spaces?). Stop.
This file lacks a trailing back slash in line #78. Patch attached.
- Thu Apr 16, 2015 9:43 am
- Forum: STM32 Support
- Topic: CAN2 RCC is absent for STM32F10X_CL devices
- Replies: 1
- Views: 2239
CAN2 RCC is absent for STM32F10X_CL devices
Hello!
In recent ChibiOS 3 I found that the rccEnableCAN2() (and others for CAN2) is absent at all in os/hal/ports/STM32/STM32F1XX/stm32_rcc.h
STM32F10X_CL devices have 2 CANs. In ChibiOS 2.6 it present.
Patch attached.
In recent ChibiOS 3 I found that the rccEnableCAN2() (and others for CAN2) is absent at all in os/hal/ports/STM32/STM32F1XX/stm32_rcc.h
STM32F10X_CL devices have 2 CANs. In ChibiOS 2.6 it present.
Patch attached.
- Wed Apr 01, 2015 2:31 pm
- Forum: Development and Feedback
- Topic: [FEATURE REQUEST] STM32: nvicClearPending()
- Replies: 2
- Views: 2837
- Wed Apr 01, 2015 2:22 pm
- Forum: STM32 Support
- Topic: IRQ disable/enable in usb_lld.c
- Replies: 1
- Views: 2085
IRQ disable/enable in usb_lld.c
Hi!
I found an issue in OTGv1 usb_lld.c: in case when CORTEX_SIMPLIFIED_PRIORITY = TRUE a special assumption that OTG FIFO operations must not be interrupted is not satisfied.
New code:
#if (CORTEX_SIMPLIFIED_PRIORITY == FALSE)
#if (STM32_USB_OTGFIFO_FILL_BASEPRI > 0)
__set_BASEPRI(CORTEX_PRIO ...
I found an issue in OTGv1 usb_lld.c: in case when CORTEX_SIMPLIFIED_PRIORITY = TRUE a special assumption that OTG FIFO operations must not be interrupted is not satisfied.
New code:
#if (CORTEX_SIMPLIFIED_PRIORITY == FALSE)
#if (STM32_USB_OTGFIFO_FILL_BASEPRI > 0)
__set_BASEPRI(CORTEX_PRIO ...
- Wed Apr 01, 2015 1:56 pm
- Forum: Development and Feedback
- Topic: [FEATURE REQUEST] STM32: nvicClearPending()
- Replies: 2
- Views: 2837
[FEATURE REQUEST] STM32: nvicClearPending()
Hello!
Some tasks with extensive IRQ usage needs a way to clear pending interrupts.
I request a simple function in NVIC like this:
Patch attached. Thanks.
Some tasks with extensive IRQ usage needs a way to clear pending interrupts.
I request a simple function in NVIC like this:
Code: Select all
void nvicClearPending(uint32_t n) {
NVIC->ICPR[n >> 5] = 1 << (n & 0x1F);
}
Patch attached. Thanks.
- Mon Mar 23, 2015 2:47 pm
- Forum: ChibiOS/RT
- Topic: System hang with enabled debug
- Replies: 4
- Views: 4727
Re: System hang with enabled debug
with classic periodic time
system works, no hangs. So I assume this is somewhat related to freerunning timer on TIM2.
Code: Select all
#define CH_CFG_ST_TIMEDELTA 0system works, no hangs. So I assume this is somewhat related to freerunning timer on TIM2.
- Mon Mar 23, 2015 2:29 pm
- Forum: ChibiOS/RT
- Topic: System hang with enabled debug
- Replies: 4
- Views: 4727
Re: System hang with enabled debug
No, IRQ priorities at level 0/1 are not used.
Without debug in chconf.h system works, but I encountered some sort of hangs due to stack overflow or something during USB transfers, so decided to enable debug in ChibiOS and right after that I have found this issue (even without my code).
config files ...
Without debug in chconf.h system works, but I encountered some sort of hangs due to stack overflow or something during USB transfers, so decided to enable debug in ChibiOS and right after that I have found this issue (even without my code).
config files ...
- Mon Mar 23, 2015 1:54 pm
- Forum: ChibiOS/RT
- Topic: System hang with enabled debug
- Replies: 4
- Views: 4727
System hang with enabled debug
Hi!
Problem was found in ChibiOS/RT 3 (rev.7788): with enabled debug in chconf.h
#define CH_DBG_SYSTEM_STATE_CHECK TRUE
#define CH_DBG_ENABLE_CHECKS TRUE
#define CH_DBG_ENABLE_ASSERTS TRUE
Interrupt from timer TIM2 (STM32F105) generated after chThdSleepS() calling and before chSysUnlock ...
Problem was found in ChibiOS/RT 3 (rev.7788): with enabled debug in chconf.h
#define CH_DBG_SYSTEM_STATE_CHECK TRUE
#define CH_DBG_ENABLE_CHECKS TRUE
#define CH_DBG_ENABLE_ASSERTS TRUE
Interrupt from timer TIM2 (STM32F105) generated after chThdSleepS() calling and before chSysUnlock ...
- Sun Mar 22, 2015 11:57 am
- Forum: ChibiOS/RT
- Topic: Compilation errors in Keil
- Replies: 4
- Views: 5508
Re: Compilation errors in Keil
Hello Giovanni!
Why the change in chcore_v7m.h? That "U" is required for MISRA compliance.
That's why:
os\rt\ports\ARMCMx\compilers\RVCT\chcoreasm_v7m.s(90): error: A1137E: Unexpected characters at end of line
assembler control string:
--cpu Cortex-M3 -g --apcs=interwork --pd "__MICROLIB ...
Why the change in chcore_v7m.h? That "U" is required for MISRA compliance.
That's why:
os\rt\ports\ARMCMx\compilers\RVCT\chcoreasm_v7m.s(90): error: A1137E: Unexpected characters at end of line
assembler control string:
--cpu Cortex-M3 -g --apcs=interwork --pd "__MICROLIB ...