[ARM] Fix FIQ issue with ARM926
Jon Eibertzon writes:
> We have noticed that the I-cache is disabled while waiting for
> interrupt in cpu_arm926_do_idle in arch/arm/mm/proc-arm926.S
> and we are curious to know why, because this causes us a great
> performance hit when executing in FIQ-handlers. Is it assumed
> here that every individual FIQ-handler re-enables the I-cache?
The I-cache disable is an errata workaround, so the solution is to
disable FIQs across the section with the I-cache disabled.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
index 5b80b6b..194ef48 100644
--- a/arch/arm/mm/proc-arm926.S
+++ b/arch/arm/mm/proc-arm926.S
@@ -105,9 +105,13 @@
mrc p15, 0, r1, c1, c0, 0 @ Read control register
mcr p15, 0, r0, c7, c10, 4 @ Drain write buffer
bic r2, r1, #1 << 12
+ mrs r3, cpsr @ Disable FIQs while Icache
+ orr ip, r3, #PSR_F_BIT @ is disabled
+ msr cpsr_c, ip
mcr p15, 0, r2, c1, c0, 0 @ Disable I cache
mcr p15, 0, r0, c7, c0, 4 @ Wait for interrupt
mcr p15, 0, r1, c1, c0, 0 @ Restore ICache enable
+ msr cpsr_c, r3 @ Restore FIQ state
mov pc, lr
/*