hw/intc/arm_gicv3: Check correct HCR_EL2 bit when routing IRQ

[Qemu-devel] [PATCH for-3.0?] hw/intc/arm_gicv3: Check correct HCR_EL2 bit when routing IRQ

Peter Maydell posted 1 patch 7 years, 3 months ago

Download mbox

Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20180723180337.17378-1-peter.maydell@linaro.org

Test docker-mingw@fedora passed

Test checkpatch passed

Test docker-quick@centos7 passed

hw/intc/arm_gicv3_cpuif.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

Expand all Fold all

[Qemu-devel] [PATCH for-3.0?] hw/intc/arm_gicv3: Check correct HCR_EL2 bit when routing IRQ

Posted by Peter Maydell 7 years, 3 months ago

In icc_dir_write() we were incorrectly checking HCR_EL2.FMO
when determining whether IRQ should be routed to EL2; this should
be HCR_EL2.IMO (compare the GICv3 pseudocode ICC_DIR_EL1[]).
Use the correct mask.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
Cut-n-paste bug we've had since forever. Linux always sets
FMO and IMO to the same thing, so we haven't noticed this.
Probably should go into 3.0, though...

 hw/intc/arm_gicv3_cpuif.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/intc/arm_gicv3_cpuif.c b/hw/intc/arm_gicv3_cpuif.c
index 5c89be1af07..2a60568d82c 100644
--- a/hw/intc/arm_gicv3_cpuif.c
+++ b/hw/intc/arm_gicv3_cpuif.c
@@ -1550,7 +1550,7 @@ static void icc_dir_write(CPUARMState *env, const ARMCPRegInfo *ri,
      * tested in cases where we know !IsSecure is true.
      */
     route_fiq_to_el2 = env->cp15.hcr_el2 & HCR_FMO;
-    route_irq_to_el2 = env->cp15.hcr_el2 & HCR_FMO;
+    route_irq_to_el2 = env->cp15.hcr_el2 & HCR_IMO;
 
     switch (arm_current_el(env)) {
     case 3:
-- 
2.17.1

Re: [Qemu-devel] [PATCH for-3.0?] hw/intc/arm_gicv3: Check correct HCR_EL2 bit when routing IRQ

Posted by Richard Henderson 7 years, 3 months ago

On 07/23/2018 11:03 AM, Peter Maydell wrote:
> In icc_dir_write() we were incorrectly checking HCR_EL2.FMO
> when determining whether IRQ should be routed to EL2; this should
> be HCR_EL2.IMO (compare the GICv3 pseudocode ICC_DIR_EL1[]).
> Use the correct mask.
> 
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> ---
> Cut-n-paste bug we've had since forever. Linux always sets
> FMO and IMO to the same thing, so we haven't noticed this.
> Probably should go into 3.0, though...
> 
>  hw/intc/arm_gicv3_cpuif.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~