ASoC: fsl_ssi: free irq before irq_dispose_mapping()

Commit Message

Fabio Estevam Dec. 1, 2014, 8:40 p.m. UTC

On Mon, Dec 1, 2014 at 6:30 PM, Lars-Peter Clausen <[e-mail]> wrote:
> On 12/01/2014 09:11 PM, Timur Tabi wrote:
>>
>> On 12/01/2014 02:01 PM, Arnd Bergmann wrote:
>>>>
>>>> >Does this mean that fsl_ssi.c should not be calling
>>>> >irq_of_parse_and_map?  How else should it get the IRQ?
>>>
>>> platform_get_irq()
>>
>>
>> Ok, but that function also calls irq_create_of_mapping().  So it still
>> appears that the only way to get the IRQ is to map it, but then we can't
>> use
>> devm_request_irq().
>>
>
> Hm... that's new. But it's not really a driver issue anymore if it is done
> in the core. So I guess for now just use platform_get_irq() and ignore the
> other issue.

With the suggested changes below, the removal of the driver works fine on a mx6:

root@freescale /$ modprobe   snd-soc-fsl-ssi
root@freescale /$ modprobe snd-soc-imx-wm8962
[  319.517679] input: WM8962 Beep Generator as
/devices/soc0/soc/2100000.aips-bus/21a0000.i2c/i2c-0/0-001a/input/input7
[  319.543225] imx-wm8962 sound: wm8962 <-> 202c000.ssi mapping ok
root@freescale /$ rmmod  snd-soc-imx-wm8962
root@freescale /$ rmmod   snd-soc-fsl-ssi

 sound/soc/fsl/fsl_ssi.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

Comments

Timur Tabi Dec. 1, 2014, 8:42 p.m. UTC | #1

On 12/01/2014 02:40 PM, Fabio Estevam wrote:
>> >Hm... that's new. But it's not really a driver issue anymore if it is done
>> >in the core. So I guess for now just use platform_get_irq() and ignore the
>> >other issue.

> With the suggested changes below, the removal of the driver works fine on a mx6:

Would the mapping continue to exist after the driver is unloaded?  Can 
you try multiple loads/unloads and see if interrupts still work?

Fabio Estevam Dec. 1, 2014, 9:03 p.m. UTC | #2

On Mon, Dec 1, 2014 at 6:42 PM, Timur Tabi <[e-mail]> wrote:

> Would the mapping continue to exist after the driver is unloaded?  Can you
> try multiple loads/unloads and see if interrupts still work?

I tried multiple loads/unloads and audio works fine with those changes.

About the ssi irq we have:

- With the ssi driver loaded:
root@freescale /home$ cat /proc/interrupts | grep ssi
 79:          0          0          0          0       GIC  79  202c000.ssi

- After removing the ssi driver:
root@freescale /home$ rmmod   snd-soc-fsl-ssi
root@freescale /home$ cat /proc/interrupts | grep ssi
root@freescale /home$

,so it seems to behave properly.

Patch

diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c
index 32a31d9..c528f16 100644
--- a/sound/soc/fsl/fsl_ssi.c
+++ b/sound/soc/fsl/fsl_ssi.c
@@ -1361,7 +1361,7 @@  static int fsl_ssi_probe(struct platform_device *pdev)
         return PTR_ERR(ssi_private->regs);
     }

-    ssi_private->irq = irq_of_parse_and_map(np, 0);
+    ssi_private->irq = platform_get_irq(pdev, 0);
     if (!ssi_private->irq) {
         dev_err(&pdev->dev, "no irq for node %s\n", np->full_name);
         return -ENXIO;
@@ -1387,7 +1387,7 @@  static int fsl_ssi_probe(struct platform_device *pdev)
     if (ssi_private->soc->imx) {
         ret = fsl_ssi_imx_probe(pdev, ssi_private, iomem);
         if (ret)
-            goto error_irqmap;
+            return ret;
     }

     ret = snd_soc_register_component(&pdev->dev, &fsl_ssi_component,
@@ -1458,10 +1458,6 @@  error_asoc_register:
     if (ssi_private->soc->imx)
         fsl_ssi_imx_clean(pdev, ssi_private);

-error_irqmap:
-    if (ssi_private->use_dma)
-        irq_dispose_mapping(ssi_private->irq);
-
     return ret;
 }

@@ -1478,9 +1474,6 @@  static int fsl_ssi_remove(struct platform_device *pdev)
     if (ssi_private->soc->imx)
         fsl_ssi_imx_clean(pdev, ssi_private);

-    if (ssi_private->use_dma)
-        irq_dispose_mapping(ssi_private->irq);
-
     return 0;
 }