Follow-up from here:
https://lists.rtems.org/pipermail/devel/2015-February/009974.html
When talking to Sebastian Huber about the behavior of
_ARMV4_Exception_fiq_default, he mentioned that it shouldn't re-enable FIQs
again. This patch sets the F bit of the SPSR so that when it gets loaded back
to the CPSR in save_more_context it won't enable the FIQs.
So far I've tested it on the TMS570 and it seems to be working fine (we were
getting endless FIQs before, now we end up in bsp_reset). However, I'm not sure
if there are other BSPs out there that rely on the previous behavior.
diff --git a/cpukit/score/cpu/arm/armv4-exception-default.S
b/cpukit/score/cpu/arm/armv4-exception-default.S
index a0ee46c..a10de30 100644
--- a/cpukit/score/cpu/arm/armv4-exception-default.S
+++ b/cpukit/score/cpu/arm/armv4-exception-default.S
@@ -99,6 +99,14 @@ _ARMV4_Exception_fiq_default:
stmdb sp!, {r0-r12}
mov r4, #7
+ /*
+ * Don't enable FIQs yet. Set the FIQ disable bit in the SPSR
+ * (which we'll load into the CPSR in save_more_context).
+ */
+ mrs r2, spsr
+ orr r2, #ARM_PSR_F
+ msr spsr_c, r2
+
save_more_context:
/* Save more context */
_______________________________________________
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel
