Skip to contentSkip to conversation

arm: Adding support for atomic half word exchange

PATCH Sarbojit Ganguly 1 participant 1 message Aug 20, 2015

lore
Share & tools
lore
#0 lore raw
Quoted 23 lines
Quoted 23 lines

My apologies, the e-mail editor was not configured properly. CC'ed to relevant maintainers and reposting once again with proper formatting.

Since 16 bit half word exchange was not there and MCS based qspinlock by Waiman's xchg_tail() requires an atomic exchange on a half word, here is a small modification to __xchg() code to support the exchange. ARMv6 and lower does not have support for LDREXH, so we need to make sure things do not break when we're compiling on ARMv6.

---
 arch/arm/include/asm/cmpxchg.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
(quoted diff) +0 -0

b/arch/arm/include/asm/cmpxchg.h index 1692a05..547101d 100644

arch/arm/include/asm/cmpxchg.h +1 -0
@@ -50,6 +50,24 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size
                        : "r" (x), "r" (ptr)
                        : "memory", "cc");
                break;
+#if !defined (CONFIG_CPU_V6)
Quoted 8 lines

arch/arm/include/asm/cmpxchg.h static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size · line 59

Quoted 6 lines
arch/arm/include/asm/cmpxchg.h +6 -0
+               /*
+                * Halfword exclusive exchange
+                * This is new implementation as qspinlock
+                * wants 16 bit atomic CAS.
+                * This is not supported on ARMv6.
+                */

I don't think you need this comment. We don't use qspinlock on arch/arm/.

Yes, till date mainline ARM does not support but I've ported Qspinlock on ARM hence I think that comment might be required.

Quoted 15 lines
Quoted 4 lines
arch/arm/include/asm/cmpxchg.h +4 -0
+       case 2:
+               asm volatile("@ __xchg2\n"
+               "1:     ldrexh  %0, [%3]\n"
+               "       strexh  %1, %2, [%3]\n"

arch/arm/include/asm/cmpxchg.h static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size · line 73

Quoted 10 lines
arch/arm/include/asm/cmpxchg.h +7 -0
+               "       teq     %1, #0\n"
+               "       bne     1b"
+               : "=&r" (ret), "=&r" (tmp)
+               : "r" (x), "r" (ptr)
+               : "memory", "cc");
+               break;
+#endif
        case 4:
                asm volatile("@ __xchg4\n"
                "1:     ldrex   %0, [%3]\n"
Quoted 1 line

We have the same issue with the byte exclusives, so I think you need to extend the guard you're adding to cover that case too (which is a bug in current mainline).

Ok, I will work on this and release a v2 soon.

Quoted 1 line

Will

- Sarbojit ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ¥Šwÿº{.nÇ+‰·¥Š{±þG«�éÿŠ{ayºʇڙë,j­¢f£¢·hš�ï�êÿ‘êçz_è®(­éšŽŠÝ¢j"�ú¶m§ÿÿ¾«þG«�éÿ¢¸?™¨è­Ú&£ø§~�á¶iO•æ¬z·švØ^¶m§ÿÿà ÿ¶ìÿ¢¸?–I¥