We found a failure when used iperf tool for wifi performance testing,
there are some MSIs received while clearing the interrupt status,
these MSIs cannot be serviced.

The interrupt status can be cleared even the MSI status still remaining,
as an edge-triggered interrupts, its interrupt status should be cleared
before dispatching to the handler of device.

Signed-off-by: qizhong cheng <qizhong.cheng@mediatek.com>
---
v2:
 - Update the subject line.
 - Improve the commit log and code comments.

 drivers/pci/controller/pcie-mediatek.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c
index 2f3f974977a3..2856d74b2513 100644
--- a/drivers/pci/controller/pcie-mediatek.c
+++ b/drivers/pci/controller/pcie-mediatek.c
@@ -624,12 +624,17 @@ static void mtk_pcie_intr_handler(struct irq_desc *desc)
 		if (status & MSI_STATUS){
 			unsigned long imsi_status;
 
+			/*
+			 * The interrupt status can be cleared even the MSI
+			 * status still remaining, hence as an edge-triggered
+			 * interrupts, its interrupt status should be cleared
+			 * before dispatching handler.
+			 */
+			writel(MSI_STATUS, port->base + PCIE_INT_STATUS);
 			while ((imsi_status = readl(port->base + PCIE_IMSI_STATUS))) {
 				for_each_set_bit(bit, &imsi_status, MTK_MSI_IRQS_NUM)
 					generic_handle_domain_irq(port->inner_domain, bit);
 			}
-			/* Clear MSI interrupt status */
-			writel(MSI_STATUS, port->base + PCIE_INT_STATUS);
 		}
 	}
 
-- 
2.25.1

Il 10/02/22 02:21, qizhong cheng ha scritto:
> We found a failure when used iperf tool for wifi performance testing,
> there are some MSIs received while clearing the interrupt status,
> these MSIs cannot be serviced.
> 
> The interrupt status can be cleared even the MSI status still remaining,
> as an edge-triggered interrupts, its interrupt status should be cleared
> before dispatching to the handler of device.
> 
> Signed-off-by: qizhong cheng <qizhong.cheng@mediatek.com>

Hello Qizhong,

This commit is fixing an issue, which means that you *have to* add a proper
Fixes tag.

I believe that this is fixing commit
43e6409db64d ("PCI: mediatek: Add MSI support for MT2712 and MT7622").

Please add the tag and send a v3, after which:
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>


> ---
> v2:
>   - Update the subject line.
>   - Improve the commit log and code comments.
> 
>   drivers/pci/controller/pcie-mediatek.c | 9 +++++++--
>   1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c
> index 2f3f974977a3..2856d74b2513 100644
> --- a/drivers/pci/controller/pcie-mediatek.c
> +++ b/drivers/pci/controller/pcie-mediatek.c
> @@ -624,12 +624,17 @@ static void mtk_pcie_intr_handler(struct irq_desc *desc)
>   		if (status & MSI_STATUS){
>   			unsigned long imsi_status;
>   
> +			/*
> +			 * The interrupt status can be cleared even the MSI
> +			 * status still remaining, hence as an edge-triggered
> +			 * interrupts, its interrupt status should be cleared
> +			 * before dispatching handler.
> +			 */
> +			writel(MSI_STATUS, port->base + PCIE_INT_STATUS);
>   			while ((imsi_status = readl(port->base + PCIE_IMSI_STATUS))) {
>   				for_each_set_bit(bit, &imsi_status, MTK_MSI_IRQS_NUM)
>   					generic_handle_domain_irq(port->inner_domain, bit);
>   			}
> -			/* Clear MSI interrupt status */
> -			writel(MSI_STATUS, port->base + PCIE_INT_STATUS);
>   		}
>   	}
>   
> 

On Thu, Feb 10, 2022 at 09:21:25AM +0800, qizhong cheng wrote:
> We found a failure when used iperf tool for wifi performance testing,
> there are some MSIs received while clearing the interrupt status,
> these MSIs cannot be serviced.
> 
> The interrupt status can be cleared even the MSI status still remaining,
> as an edge-triggered interrupts, its interrupt status should be cleared
> before dispatching to the handler of device.
> 
> Signed-off-by: qizhong cheng <qizhong.cheng@mediatek.com>
> ---
> v2:
>  - Update the subject line.
>  - Improve the commit log and code comments.
> 
>  drivers/pci/controller/pcie-mediatek.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 

<formletter>

This is not the correct way to submit patches for inclusion in the
stable kernel tree.  Please read:
    https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how to do this properly.

</formletter>