Why don't you use GNU assembly? It should look something like this
.equ locked,1.equ unlocked,0@ lock_mutex@ Declare for use from C as extern void lock_mutex(void * mutex); .global lock_mutexlock_mutex: LDR r1, =locked1: LDREX r2, [r0] CMP r2, r1 @ Test if mutex is locked or unlocked BEQ 2f @ If locked - wait for it to be released, from 2 STREXNE r2, r1, [r0] @ Not locked, attempt to lock it CMPNE r2, #1 @ Check if Store-Exclusive failed BEQ 1b @ Failed - retry from 1 # Lock acquired DMB @ Required before accessing protected resource BX lr2: @ Take appropriate action while waiting for mutex to become unlocked @ WAIT_FOR_UPDATE B 1b @ Retry from 1@ unlock_mutex@ Declare for use from C as extern void unlock_mutex(void * mutex); .global unlock_mutexunlock_mutex: LDR r1, =unlocked DMB @ Required before releasing protected resource STR r1, [r0] @ Unlock mutex @ SIGNAL_UPDATE BX lrThen its dump looks like this
$ arm-linux-gnueabihf-objdump -d mutex.omutex.o: file format elf32-littlearmDisassembly of section .text:00000000 <lock_mutex>: 0: e3a01001 mov r1, #1 4: e1902f9f ldrex r2, [r0] 8: e1520001 cmp r2, r1 c: 0a000004 beq 24 <lock_mutex+0x24> 10: 11802f91 strexne r2, r1, [r0] 14: 13520001 cmpne r2, #1 18: 0afffff9 beq 4 <lock_mutex+0x4> 1c: f57ff05f dmb sy 20: e12fff1e bx lr 24: eafffff6 b 4 <lock_mutex+0x4>00000028 <unlock_mutex>: 28: e3a01000 mov r1, #0 2c: f57ff05f dmb sy 30: e5801000 str r1, [r0] 34: e12fff1e bx lrHowever what I'm wondering if you did managed to configure both cores to be included in core coherency. To my knowledge you can specify which cores participate in ldrex/strex operations.
