Re: [PATCH/RFC v4 net-next] ravb: Add dma queue interrupt support

[Sergei Shtylyov]

On 02/21/2016 06:42 PM, Sergei Shtylyov wrote:

[...]
From: Kazuya Mizuguchi <kazuya.mizuguchi...@renesas.com>

This patch supports the following interrupts.

- One interrupt for multiple (descriptor, error, management)
- One interrupt for emac
- Four interrupts for dma queue (best effort rx/tx, network control
rx/tx)

This patch improve efficiency of the interrupt handler by adding the
interrupt handler corresponding to each interrupt source described
above. Additionally, it reduces the number of times of the access to
EthernetAVB IF.

Signed-off-by: Kazuya Mizuguchi <kazuya.mizuguchi...@renesas.com>
Signed-off-by: Yoshihiro Kaneko <ykaneko0...@gmail.com>

[...]

diff --git a/drivers/net/ethernet/renesas/ravb_main.c
b/drivers/net/ethernet/renesas/ravb_main.c
index ac43ed9..076f25f 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
[...]
+                          struct net_device *ndev, struct device *dev,
+                          const char *ch)
+{
+       char *name;
+       int error;
+
+       name = devm_kasprintf(dev, GFP_KERNEL, "%s:%s", ndev->name, ch);
+       error = request_irq(irq, handler, IRQF_SHARED, name, ndev);

     Not sure if we need IRQF_SHARED on those IRQs, they're not really
shareable...

I don't know whether this causes something bad.
I think this controller is supporting a shared IRQ.

    Based on the high-level trigger, I'd rather suspect not. Anyway, all the
SoC IRQs are dedicated to a certain (single) source.

I don't want to change that if there is no fatal issue in the use of
IRQF_SHARED.
However, I will remove the flag if it is simple waste...

    It's not a waste but it just shouldn't be needed.

   Actually for RX/TX DMA you're routing 2 interrupts to the same handler, so 
it's needed, sorry.

[...]

Thanks,
kaneko

MBR, Sergei