[PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module

linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
@ 2013-11-26  4:17 Bo Shen
  2013-11-26  4:17 ` [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata Bo Shen
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: Bo Shen @ 2013-11-26  4:17 UTC (permalink / raw)
  To: linux-arm-kernel

As the priv is not assigned to card->drvdata, it is NULL, so when
unload module, it will cause NULL pointer oops.
Assign priv to card->drvdata to fix this issue.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
---
 sound/soc/atmel/sam9x5_wm8731.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
index 992ae38..6dc33ac 100644
--- a/sound/soc/atmel/sam9x5_wm8731.c
+++ b/sound/soc/atmel/sam9x5_wm8731.c
@@ -99,6 +99,7 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
 
 	card->dev = &pdev->dev;
 	card->owner = THIS_MODULE;
+	card->drvdata = priv;
 	card->dai_link = dai;
 	card->num_links = 1;
 	card->dapm_widgets = sam9x5_dapm_widgets;
-- 
1.7.9.5

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata
  2013-11-26  4:17 [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Bo Shen
@ 2013-11-26  4:17 ` Bo Shen
  2013-11-26 11:42   ` Richard Genoud
  2013-11-26 11:32 ` [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Richard Genoud
  2013-11-26 13:22 ` Mark Brown
  2 siblings, 1 reply; 10+ messages in thread
From: Bo Shen @ 2013-11-26  4:17 UTC (permalink / raw)
  To: linux-arm-kernel

When call snd_soc_register_card, it will set driver data to this
device through dev_set_drvdata, then in driver, no need to call
platform_set_drvdata again, so remove it.

Signed-off-by: Bo Shen <voice.shen@atmel.com>
---
 sound/soc/atmel/sam9x5_wm8731.c |    2 --
 1 file changed, 2 deletions(-)

diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
index 6dc33ac..4af6836 100644
--- a/sound/soc/atmel/sam9x5_wm8731.c
+++ b/sound/soc/atmel/sam9x5_wm8731.c
@@ -154,8 +154,6 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
 	of_node_put(codec_np);
 	of_node_put(cpu_np);
 
-	platform_set_drvdata(pdev, card);
-
 	ret = snd_soc_register_card(card);
 	if (ret) {
 		dev_err(&pdev->dev,
-- 
1.7.9.5

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata
  2013-11-26  4:17 ` [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata Bo Shen
@ 2013-11-26 11:42   ` Richard Genoud
  0 siblings, 0 replies; 10+ messages in thread
From: Richard Genoud @ 2013-11-26 11:42 UTC (permalink / raw)
  To: linux-arm-kernel

2013/11/26 Bo Shen <voice.shen@atmel.com>:
> When call snd_soc_register_card, it will set driver data to this
> device through dev_set_drvdata, then in driver, no need to call
> platform_set_drvdata again, so remove it.
>
> Signed-off-by: Bo Shen <voice.shen@atmel.com>
> ---
>  sound/soc/atmel/sam9x5_wm8731.c |    2 --
>  1 file changed, 2 deletions(-)
>
> diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
> index 6dc33ac..4af6836 100644
> --- a/sound/soc/atmel/sam9x5_wm8731.c
> +++ b/sound/soc/atmel/sam9x5_wm8731.c
> @@ -154,8 +154,6 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
>         of_node_put(codec_np);
>         of_node_put(cpu_np);
>
> -       platform_set_drvdata(pdev, card);
> -
>         ret = snd_soc_register_card(card);
>         if (ret) {
>                 dev_err(&pdev->dev,
> --
> 1.7.9.5
>
Ok for me

Tested-by: Richard Genoud <richard.genoud@gmail.com>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-26  4:17 [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Bo Shen
  2013-11-26  4:17 ` [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata Bo Shen
@ 2013-11-26 11:32 ` Richard Genoud
  2013-11-27  1:13   ` Bo Shen
  2013-11-26 13:22 ` Mark Brown
  2 siblings, 1 reply; 10+ messages in thread
From: Richard Genoud @ 2013-11-26 11:32 UTC (permalink / raw)
  To: linux-arm-kernel

2013/11/26 Bo Shen <voice.shen@atmel.com>:
> As the priv is not assigned to card->drvdata, it is NULL, so when
> unload module, it will cause NULL pointer oops.
> Assign priv to card->drvdata to fix this issue.
>
> Signed-off-by: Bo Shen <voice.shen@atmel.com>
> ---
>  sound/soc/atmel/sam9x5_wm8731.c |    1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
> index 992ae38..6dc33ac 100644
> --- a/sound/soc/atmel/sam9x5_wm8731.c
> +++ b/sound/soc/atmel/sam9x5_wm8731.c
> @@ -99,6 +99,7 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
>
>         card->dev = &pdev->dev;
>         card->owner = THIS_MODULE;
> +       card->drvdata = priv;
>         card->dai_link = dai;
>         card->num_links = 1;
>         card->dapm_widgets = sam9x5_dapm_widgets;
> --
> 1.7.9.5
>
There's no oops any more !

Tested-by: Richard Genoud <richard.genoud@gmail.com>

This should also go in -stable don't you think ?

Maybe it's a bug on my kernel, but did you try to reload the module
and play something ?
On my board once I unload/reload it, there's no sound comming out
(at least, it's better than a oops ! :) )

Richard.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-26 11:32 ` [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Richard Genoud
@ 2013-11-27  1:13   ` Bo Shen
  2013-11-27 10:09     ` Richard Genoud
  0 siblings, 1 reply; 10+ messages in thread
From: Bo Shen @ 2013-11-27  1:13 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Richard,

On 11/26/2013 07:32 PM, Richard Genoud wrote:
> 2013/11/26 Bo Shen <voice.shen@atmel.com>:
>> As the priv is not assigned to card->drvdata, it is NULL, so when
>> unload module, it will cause NULL pointer oops.
>> Assign priv to card->drvdata to fix this issue.
>>
>> Signed-off-by: Bo Shen <voice.shen@atmel.com>
>> ---
>>   sound/soc/atmel/sam9x5_wm8731.c |    1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/sound/soc/atmel/sam9x5_wm8731.c b/sound/soc/atmel/sam9x5_wm8731.c
>> index 992ae38..6dc33ac 100644
>> --- a/sound/soc/atmel/sam9x5_wm8731.c
>> +++ b/sound/soc/atmel/sam9x5_wm8731.c
>> @@ -99,6 +99,7 @@ static int sam9x5_wm8731_driver_probe(struct platform_device *pdev)
>>
>>          card->dev = &pdev->dev;
>>          card->owner = THIS_MODULE;
>> +       card->drvdata = priv;
>>          card->dai_link = dai;
>>          card->num_links = 1;
>>          card->dapm_widgets = sam9x5_dapm_widgets;
>> --
>> 1.7.9.5
>>
> There's no oops any more !
>
> Tested-by: Richard Genoud <richard.genoud@gmail.com>
>
> This should also go in -stable don't you think ?

After this patch go into mainline tree, I think we can ask it to go in 
-stable tree.

> Maybe it's a bug on my kernel, but did you try to reload the module
> and play something ?

When we reload the module, as the codec is not configured outside, so it 
use default value, the 'Output Mixer HiFi Playback Switch' is in off 
state.If you want to play something, you need make it in on state. 
Please take the following command as a reference:
--->8---
amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback Switch' on
---8<---

> On my board once I unload/reload it, there's no sound comming out
> (at least, it's better than a oops ! :) )
>
> Richard.
>

Best Regards,
Bo Shen

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-27  1:13   ` Bo Shen
@ 2013-11-27 10:09     ` Richard Genoud
  2013-11-28  1:20       ` Bo Shen
  0 siblings, 1 reply; 10+ messages in thread
From: Richard Genoud @ 2013-11-27 10:09 UTC (permalink / raw)
  To: linux-arm-kernel

On 27/11/2013 02:13, Bo Shen wrote:
> Hi Richard,
Hi !
> 
> On 11/26/2013 07:32 PM, Richard Genoud wrote:
>> 2013/11/26 Bo Shen <voice.shen@atmel.com>:
>>> As the priv is not assigned to card->drvdata, it is NULL, so when
>>> unload module, it will cause NULL pointer oops.
>>> Assign priv to card->drvdata to fix this issue.
>>>
>>> Signed-off-by: Bo Shen <voice.shen@atmel.com>
>>> ---
>>>   sound/soc/atmel/sam9x5_wm8731.c |    1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/sound/soc/atmel/sam9x5_wm8731.c
>>> b/sound/soc/atmel/sam9x5_wm8731.c
>>> index 992ae38..6dc33ac 100644
>>> --- a/sound/soc/atmel/sam9x5_wm8731.c
>>> +++ b/sound/soc/atmel/sam9x5_wm8731.c
>>> @@ -99,6 +99,7 @@ static int sam9x5_wm8731_driver_probe(struct
>>> platform_device *pdev)
>>>
>>>          card->dev = &pdev->dev;
>>>          card->owner = THIS_MODULE;
>>> +       card->drvdata = priv;
>>>          card->dai_link = dai;
>>>          card->num_links = 1;
>>>          card->dapm_widgets = sam9x5_dapm_widgets;
>>> -- 
>>> 1.7.9.5
>>>
>> There's no oops any more !
>>
>> Tested-by: Richard Genoud <richard.genoud@gmail.com>
>>
>> This should also go in -stable don't you think ?
> 
> After this patch go into mainline tree, I think we can ask it to go in
> -stable tree.
So, you should add "Cc: stable at vger.kernel.org" just after your signed-off (maybe with a note that it applies on 3.12+)
[ actually, that's more a question than an advice :) ]

> 
>> Maybe it's a bug on my kernel, but did you try to reload the module
>> and play something ?
> 
> When we reload the module, as the codec is not configured outside, so it
> use default value, the 'Output Mixer HiFi Playback Switch' is in off
> state.If you want to play something, you need make it in on state.
> Please take the following command as a reference:
> --->8---
> amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback
> Switch' on
> ---8<---
It doesn't seems to be related to the playback switches.
I check the TK/TD/TF lines, and nothing happens on them after the module has been unloaded/reloaded.
I enabled I2C/dma logs (on a 3.13-rc1 + your 2 patches kernel + at91sam9g35ek) and it seems that the DMA doesn't fill the SSC :

# # before unloading the module :
# alsactl -f /etc/asound.state restore
# aplay tone-stereo-100Hz_right-440Hz-left.wav -Dplug:default
Playing WAVE 'tone-stereo-100Hz_right-440Hz-left.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x10, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xe, to go 1
at91_i2c f8010000.i2c: wrote 0x42, to go 0
at91_i2c f8010000.i2c: transfer complete
dma dma0chan4: atc_control (3)
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x5f, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x12, to go 1
at91_i2c f8010000.i2c: wrote 0x1, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x57, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x47, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xa, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
dma dma0chan4: prep_dma_cyclic: TO DEVICE buf@0x27b30000 - 8 (65536/8192)
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: tx_submit: started 2
dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0000
dma dma0chan4:   channel: s0x0 d0x0 ctrl0x0:0x0 cfg0x100022dd l0x0
dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0010
dma dma0chan4:   channel: s0x27b30418 d0xf0010024 ctrl0x1100020c:0x20200010 cfg0x100022dd l0x27981d5c
dma dma0chan4: issue_pending
at_hdmac ffffec00.dma-controller: interrupt: status = 0x00000010, 0x00ff00ff, 0x00000010
dma dma0chan4: new cyclic period llp 0x27981da8
at_hdmac ffffec00.dma-controller: interrupt: status = 0x00000010, 0x00ff00ff, 0x00000010
dma dma0chan4: new cyclic period llp 0x27981df4
[...]
at_hdmac ffffec00.dma-controller: interrupt: status = 0x00000010, 0x00ff00ff, 0x00000010
dma dma0chan4: new cyclic period llp 0x27981ed8
at_hdmac ffffec00.dma-controller: interrupt: status = 0x00000010, 0x00ff00ff, 0x00000010
dma dma0chan4: new cyclic period llp 0x27981f24
at_hdmac ffffec00.dma-controller: interrupt: status = 0x00000010, 0x00ff00ff, 0x00000010
dma dma0chan4: new cyclic period llp 0x27981d10
dma dma0chan4: atc_control (0)
dma dma0chan4: descriptor 2 complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xa, to go 1
at91_i2c f8010000.i2c: wrote 0x8, to go 0
at91_i2c f8010000.i2c: transfer complete
# 
# # unloading the module
# rmmod snd_soc_sam9x5_wm8731i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x57, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x5f, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x7f, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x12, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xc, to go 1
at91_i2c f8010000.i2c: wrote 0x7f, to go 0
at91_i2c f8010000.i2c: transfer complete

i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0xff, to go 0
at91_i2c f8010000.i2c: transfer complete
dma dma0chan4: free_chan_resources: (descs allocated=64)
dma dma0chan4:   freeing descriptor c885dd10
dma dma0chan4:   freeing descriptor c885dd5c
dma dma0chan4:   freeing descriptor c885dda8
dma dma0chan4:   freeing descriptor c885ddf4
[...]
dma dma0chan4:   freeing descriptor c888bed8
dma dma0chan4:   freeing descriptor c888bf24
dma dma0chan4:   freeing descriptor c888bf70
dma dma0chan4:   freeing descriptor c888d000
dma dma0chan4:   freeing descriptor c888d04c
dma dma0chan4: free_chan_resources: done
dma dma0chan5: free_chan_resources: (descs allocated=64)
dma dma0chan5:   freeing descriptor c888d098
dma dma0chan5:   freeing descriptor c888d0e4
dma dma0chan5:   freeing descriptor c888d130
[...]
dma dma0chan5:   freeing descriptor c888f2ac
dma dma0chan5:   freeing descriptor c888f2f8
dma dma0chan5:   freeing descriptor c888f344
dma dma0chan5:   freeing descriptor c888f390
dma dma0chan5: free_chan_resources: done
# # reloading the module
# modprobe snd_soc_sam9x5_wm8731
dmaengine: private_candidate: dma0chan0 busy
dmaengine: private_candidate: dma0chan1 busy
dmaengine: private_candidate: dma0chan2 busy
dmaengine: private_candidate: dma0chan3 busy
dma dma0chan4: alloc_chan_resources
dma dma0chan4: alloc_chan_resources: allocated 64 descriptors
dmaengine: __dma_request_channel: success (dma0chan4)
dmaengine: private_candidate: dma0chan0 busy
dmaengine: private_candidate: dma0chan1 busy
dmaengine: private_candidate: dma0chan2 busy
dmaengine: private_candidate: dma0chan3 busy
dmaengine: private_candidate: dma0chan4 busy
dma dma0chan5: alloc_chan_resources
dma dma0chan5: alloc_chan_resources: allocated 64 descriptors
dmaengine: __dma_request_channel: success (dma0chan5)
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x1e, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x0, to go 1
at91_i2c f8010000.i2c: wrote 0x9a, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x2, to go 1
at91_i2c f8010000.i2c: wrote 0x9a, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x4, to go 1
at91_i2c f8010000.i2c: wrote 0x74, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x6, to go 1
at91_i2c f8010000.i2c: wrote 0x74, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x8, to go 1
at91_i2c f8010000.i2c: wrote 0x12, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0xff, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xe, to go 1
at91_i2c f8010000.i2c: wrote 0x42, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x1e, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0x7f, to go 0
at91_i2c f8010000.i2c: transfer complete
sam9x5-snd-wm8731 sound.5: wm8731-hifi <-> f0010000.ssc mapping ok
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xe, to go 1
at91_i2c f8010000.i2c: wrote 0x42, to go 0
at91_i2c f8010000.i2c: transfer complete
# # restoring codec state
# alsactl -f /etc/asound.state restore
# aplay tone-stereo-100Hz_right-440Hz-left.wav -Dplug:default
Playing WAVE 'tone-stereo-100Hz_right-440Hz-left.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x10, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xe, to go 1
at91_i2c f8010000.i2c: wrote 0x42, to go 0
at91_i2c f8010000.i2c: transfer complete
dma dma0chan4: atc_control (3)
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0x5f, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0x12, to go 1
at91_i2c f8010000.i2c: wrote 0x1, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0x57, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xff, to go 1
at91_i2c f8010000.i2c: wrote 0x47, to go 0
at91_i2c f8010000.i2c: transfer complete
i2c i2c-0: master_xfer[0] W, addr=0x1a, len=2
i2c i2c-0: at91_xfer: processing 1 messages:
at91_i2c f8010000.i2c: transfer: write 2 bytes.
at91_i2c f8010000.i2c: wrote 0xa, to go 1
at91_i2c f8010000.i2c: wrote 0x0, to go 0
at91_i2c f8010000.i2c: transfer complete
dma dma0chan4: prep_dma_cyclic: TO DEVICE buf at 0x27950000 - 8 (65536/8192)
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: scanned 1 descriptors on freelist
dma dma0chan4: tx_submit: started 2
dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0000
dma dma0chan4:   channel: s0x27b3eb20 d0xf0010024 ctrl0x91000590:0x20200010 cfg0x100022dd l0x27981d10
dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0010
dma dma0chan4:   channel: s0x27950010 d0xf0010024 ctrl0x11000008:0x20200010 cfg0x100022dd l0x27a58344
dma dma0chan4: issue_pending
[ blocks here forever (or at least a looooong time) ]


Anyway, this should not prevent this patch from being mainlined :)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-27 10:09     ` Richard Genoud
@ 2013-11-28  1:20       ` Bo Shen
  2013-11-29 14:59         ` Richard Genoud
  0 siblings, 1 reply; 10+ messages in thread
From: Bo Shen @ 2013-11-28  1:20 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Richard,

On 11/27/2013 06:09 PM, Richard Genoud wrote:

[Snip]

> dma dma0chan4: prep_dma_cyclic: TO DEVICE buf at 0x27950000 - 8 (65536/8192)
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: scanned 1 descriptors on freelist
> dma dma0chan4: tx_submit: started 2
> dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0000
> dma dma0chan4:   channel: s0x27b3eb20 d0xf0010024 ctrl0x91000590:0x20200010 cfg0x100022dd l0x27981d10
> dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0010
> dma dma0chan4:   channel: s0x27950010 d0xf0010024 ctrl0x11000008:0x20200010 cfg0x100022dd l0x27a58344
> dma dma0chan4: issue_pending
> [ blocks here forever (or at least a looooong time) ]

How can I reproduce this issue? Please help provide detail information 
about your operation.

On my side, I just do:
--->8---
while true;
do
	insmod snd-soc-atmel-pcm-dma.ko && insmod snd-soc-atmel_ssc_dai.ko && 
insmod snd-soc-wm8731.ko && insmod snd-soc-sam9x5-wm8731.ko;
	amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback 
Switch' on;
	aplay sample.wav;
	rmmod snd-soc-sam9x5-wm8731.ko && rmmod snd-soc-wm8731.ko && rmmod 
snd-soc-atmel_ssc_dai.ko && rmmod snd-soc-atmel-pcm-dma.ko;
done
---8<---

It works well on my side on at91sam9g35ek board, based on v3.13-rc1

Best Regards,
Bo Shen

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-28  1:20       ` Bo Shen
@ 2013-11-29 14:59         ` Richard Genoud
  0 siblings, 0 replies; 10+ messages in thread
From: Richard Genoud @ 2013-11-29 14:59 UTC (permalink / raw)
  To: linux-arm-kernel

Hi,
On 28/11/2013 02:20, Bo Shen wrote:
> Hi Richard,
> 
> On 11/27/2013 06:09 PM, Richard Genoud wrote:
> 
> [Snip]
> 
>> dma dma0chan4: prep_dma_cyclic: TO DEVICE buf at 0x27950000 - 8 (65536/8192)
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: scanned 1 descriptors on freelist
>> dma dma0chan4: tx_submit: started 2
>> dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0000
>> dma dma0chan4:   channel: s0x27b3eb20 d0xf0010024
>> ctrl0x91000590:0x20200010 cfg0x100022dd l0x27981d10
>> dma dma0chan4:   channel 4 : imr = 0xff00ff, chsr = 0xff0010
>> dma dma0chan4:   channel: s0x27950010 d0xf0010024
>> ctrl0x11000008:0x20200010 cfg0x100022dd l0x27a58344
>> dma dma0chan4: issue_pending
>> [ blocks here forever (or at least a looooong time) ]
> 
> How can I reproduce this issue? Please help provide detail information
> about your operation.
> 
> On my side, I just do:
> --->8---
> while true;
> do
>     insmod snd-soc-atmel-pcm-dma.ko && insmod snd-soc-atmel_ssc_dai.ko
> && insmod snd-soc-wm8731.ko && insmod snd-soc-sam9x5-wm8731.ko;
>     amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback
> Switch' on;
>     aplay sample.wav;
>     rmmod snd-soc-sam9x5-wm8731.ko && rmmod snd-soc-wm8731.ko && rmmod
> snd-soc-atmel_ssc_dai.ko && rmmod snd-soc-atmel-pcm-dma.ko;
> done
> ---8<---
> 
> It works well on my side on at91sam9g35ek board, based on v3.13-rc1
Ok, it's working with your script.

I was doing:
modprobe snd-soc-wm8731
modprobe snd-soc-sam9x5-wm8731
amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback Switch' on
aplay sample.wav # works ok
rmmod snd-soc-sam9x5-wm8731
insmod /lib/modules/3.13.0-rc1/kernel/sound/soc/atmel/snd-soc-sam9x5-wm8731.ko
amixer cset numid=14,iface=MIXER,name='Output Mixer HiFi Playback Switch' on
aplay sample.wav # error
aplay: pcm_write:1737: write error: Input/output error

=> I wasn't unloading snd-soc-wm8731


Thanks for your time !

Richard.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-26  4:17 [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Bo Shen
  2013-11-26  4:17 ` [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata Bo Shen
  2013-11-26 11:32 ` [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Richard Genoud
@ 2013-11-26 13:22 ` Mark Brown
  2013-11-27  1:16   ` Bo Shen
  2 siblings, 1 reply; 10+ messages in thread
From: Mark Brown @ 2013-11-26 13:22 UTC (permalink / raw)
  To: linux-arm-kernel

On Tue, Nov 26, 2013 at 12:17:51PM +0800, Bo Shen wrote:

> +	card->drvdata = priv;

You should use snd_soc_card_set_drvdata() for this.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131126/60359891/attachment.sig>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module
  2013-11-26 13:22 ` Mark Brown
@ 2013-11-27  1:16   ` Bo Shen
  0 siblings, 0 replies; 10+ messages in thread
From: Bo Shen @ 2013-11-27  1:16 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Mark,

On 11/26/2013 09:22 PM, Mark Brown wrote:
> On Tue, Nov 26, 2013 at 12:17:51PM +0800, Bo Shen wrote:
>
>> +	card->drvdata = priv;
>
> You should use snd_soc_card_set_drvdata() for this.

OK, I will change to use snd_soc_card_set_drvdata() for this.
Thanks.

Best Regards,
Bo Shen

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2013-11-29 14:59 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-11-26  4:17 [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Bo Shen
2013-11-26  4:17 ` [PATCH 2/2] ASoC: atmel: sam9x5_wm8731: remove platform_set_drvdata Bo Shen
2013-11-26 11:42   ` Richard Genoud
2013-11-26 11:32 ` [PATCH 1/2] ASoC: atmel: sam9x5_wm8731: fix oops when unload module Richard Genoud
2013-11-27  1:13   ` Bo Shen
2013-11-27 10:09     ` Richard Genoud
2013-11-28  1:20       ` Bo Shen
2013-11-29 14:59         ` Richard Genoud
2013-11-26 13:22 ` Mark Brown
2013-11-27  1:16   ` Bo Shen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).