[PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[Tiezhu Yang]

This reverts commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
7A1000 controller") to fix the following error on the Loongson LS7A
platform:

rcu: INFO: rcu_preempt self-detected stall on CPU
<SNIP>
NMI backtrace for cpu 0
CPU: 0 PID: 68 Comm: kworker/0:2 Not tainted 5.8.0+ #3
Hardware name:  , BIOS
Workqueue: events azx_probe_work [snd_hda_intel]
<SNIP>
Call Trace:
[<ffffffff80211a64>] show_stack+0x9c/0x130
[<ffffffff8065a740>] dump_stack+0xb0/0xf0
[<ffffffff80665774>] nmi_cpu_backtrace+0x134/0x140
[<ffffffff80665910>] nmi_trigger_cpumask_backtrace+0x190/0x200
[<ffffffff802b1abc>] rcu_dump_cpu_stacks+0x12c/0x190
[<ffffffff802b08cc>] rcu_sched_clock_irq+0xa2c/0xfc8
[<ffffffff802b91d4>] update_process_times+0x2c/0xb8
[<ffffffff802cad80>] tick_sched_timer+0x40/0xb8
[<ffffffff802ba5f0>] __hrtimer_run_queues+0x118/0x1d0
[<ffffffff802bab74>] hrtimer_interrupt+0x12c/0x2d8
[<ffffffff8021547c>] c0_compare_interrupt+0x74/0xa0
[<ffffffff80296bd0>] __handle_irq_event_percpu+0xa8/0x198
[<ffffffff80296cf0>] handle_irq_event_percpu+0x30/0x90
[<ffffffff8029d958>] handle_percpu_irq+0x88/0xb8
[<ffffffff80296124>] generic_handle_irq+0x44/0x60
[<ffffffff80b3cfd0>] do_IRQ+0x18/0x28
[<ffffffff8067ace4>] plat_irq_dispatch+0x64/0x100
[<ffffffff80209a20>] handle_int+0x140/0x14c
[<ffffffff802402e8>] irq_exit+0xf8/0x100

Because AZX_DRIVER_GENERIC can not work well for Loongson LS7A HDA
controller, it needs some workarounds which are not merged into the
upstream kernel at this time, so it should revert this patch now.

Fixes: 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller")
Cc: <stable@vger.kernel.org> # 5.9-rc1+
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
---

v2: update commit message

 sound/pci/hda/hda_intel.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index e34a4d5..0f86e37 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2745,8 +2745,6 @@ static const struct pci_device_id azx_ids[] = {
 	  .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI },
 	/* Zhaoxin */
 	{ PCI_DEVICE(0x1d17, 0x3288), .driver_data = AZX_DRIVER_ZHAOXIN },
-	/* Loongson */
-	{ PCI_DEVICE(0x0014, 0x7a07), .driver_data = AZX_DRIVER_GENERIC },
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, azx_ids);
-- 
2.1.0

Re: [PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[Takashi Iwai]

On Tue, 25 Aug 2020 11:39:48 +0200,
Tiezhu Yang wrote:
> 
> This reverts commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
> 7A1000 controller") to fix the following error on the Loongson LS7A
> platform:
> 
> rcu: INFO: rcu_preempt self-detected stall on CPU
> <SNIP>
> NMI backtrace for cpu 0
> CPU: 0 PID: 68 Comm: kworker/0:2 Not tainted 5.8.0+ #3
> Hardware name:  , BIOS
> Workqueue: events azx_probe_work [snd_hda_intel]
> <SNIP>
> Call Trace:
> [<ffffffff80211a64>] show_stack+0x9c/0x130
> [<ffffffff8065a740>] dump_stack+0xb0/0xf0
> [<ffffffff80665774>] nmi_cpu_backtrace+0x134/0x140
> [<ffffffff80665910>] nmi_trigger_cpumask_backtrace+0x190/0x200
> [<ffffffff802b1abc>] rcu_dump_cpu_stacks+0x12c/0x190
> [<ffffffff802b08cc>] rcu_sched_clock_irq+0xa2c/0xfc8
> [<ffffffff802b91d4>] update_process_times+0x2c/0xb8
> [<ffffffff802cad80>] tick_sched_timer+0x40/0xb8
> [<ffffffff802ba5f0>] __hrtimer_run_queues+0x118/0x1d0
> [<ffffffff802bab74>] hrtimer_interrupt+0x12c/0x2d8
> [<ffffffff8021547c>] c0_compare_interrupt+0x74/0xa0
> [<ffffffff80296bd0>] __handle_irq_event_percpu+0xa8/0x198
> [<ffffffff80296cf0>] handle_irq_event_percpu+0x30/0x90
> [<ffffffff8029d958>] handle_percpu_irq+0x88/0xb8
> [<ffffffff80296124>] generic_handle_irq+0x44/0x60
> [<ffffffff80b3cfd0>] do_IRQ+0x18/0x28
> [<ffffffff8067ace4>] plat_irq_dispatch+0x64/0x100
> [<ffffffff80209a20>] handle_int+0x140/0x14c
> [<ffffffff802402e8>] irq_exit+0xf8/0x100
> 
> Because AZX_DRIVER_GENERIC can not work well for Loongson LS7A HDA
> controller, it needs some workarounds which are not merged into the
> upstream kernel at this time, so it should revert this patch now.
> 
> Fixes: 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller")
> Cc: <stable@vger.kernel.org> # 5.9-rc1+
> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
> ---
> 
> v2: update commit message

Applied now.  Thanks.


Takashi

Re: [PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[Huacai Chen]

Hi, all

This patch should be backported to 5.4.

Huacai

On Tue, Aug 25, 2020 at 6:03 PM Takashi Iwai <tiwai@suse.de> wrote:
>
> On Tue, 25 Aug 2020 11:39:48 +0200,
> Tiezhu Yang wrote:
> >
> > This reverts commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
> > 7A1000 controller") to fix the following error on the Loongson LS7A
> > platform:
> >
> > rcu: INFO: rcu_preempt self-detected stall on CPU
> > <SNIP>
> > NMI backtrace for cpu 0
> > CPU: 0 PID: 68 Comm: kworker/0:2 Not tainted 5.8.0+ #3
> > Hardware name:  , BIOS
> > Workqueue: events azx_probe_work [snd_hda_intel]
> > <SNIP>
> > Call Trace:
> > [<ffffffff80211a64>] show_stack+0x9c/0x130
> > [<ffffffff8065a740>] dump_stack+0xb0/0xf0
> > [<ffffffff80665774>] nmi_cpu_backtrace+0x134/0x140
> > [<ffffffff80665910>] nmi_trigger_cpumask_backtrace+0x190/0x200
> > [<ffffffff802b1abc>] rcu_dump_cpu_stacks+0x12c/0x190
> > [<ffffffff802b08cc>] rcu_sched_clock_irq+0xa2c/0xfc8
> > [<ffffffff802b91d4>] update_process_times+0x2c/0xb8
> > [<ffffffff802cad80>] tick_sched_timer+0x40/0xb8
> > [<ffffffff802ba5f0>] __hrtimer_run_queues+0x118/0x1d0
> > [<ffffffff802bab74>] hrtimer_interrupt+0x12c/0x2d8
> > [<ffffffff8021547c>] c0_compare_interrupt+0x74/0xa0
> > [<ffffffff80296bd0>] __handle_irq_event_percpu+0xa8/0x198
> > [<ffffffff80296cf0>] handle_irq_event_percpu+0x30/0x90
> > [<ffffffff8029d958>] handle_percpu_irq+0x88/0xb8
> > [<ffffffff80296124>] generic_handle_irq+0x44/0x60
> > [<ffffffff80b3cfd0>] do_IRQ+0x18/0x28
> > [<ffffffff8067ace4>] plat_irq_dispatch+0x64/0x100
> > [<ffffffff80209a20>] handle_int+0x140/0x14c
> > [<ffffffff802402e8>] irq_exit+0xf8/0x100
> >
> > Because AZX_DRIVER_GENERIC can not work well for Loongson LS7A HDA
> > controller, it needs some workarounds which are not merged into the
> > upstream kernel at this time, so it should revert this patch now.
> >
> > Fixes: 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller")
> > Cc: <stable@vger.kernel.org> # 5.9-rc1+
> > Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
> > ---
> >
> > v2: update commit message
>
> Applied now.  Thanks.
>
>
> Takashi

Re: [PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[Tiezhu Yang]

On 09/08/2020 08:37 AM, Huacai Chen wrote:
> Hi, all
>
> This patch should be backported to 5.4.

Hi,

Commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
7A1000 controller") has been not yet merged into 5.4, so no
need to backport.

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/pci/hda/hda_intel.c?h=v5.4

Thanks,
Tiezhu

>
> Huacai
>
> On Tue, Aug 25, 2020 at 6:03 PM Takashi Iwai <tiwai@suse.de> wrote:
>> On Tue, 25 Aug 2020 11:39:48 +0200,
>> Tiezhu Yang wrote:
>>> This reverts commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
>>> 7A1000 controller") to fix the following error on the Loongson LS7A
>>> platform:
>>>
>>> rcu: INFO: rcu_preempt self-detected stall on CPU
>>> <SNIP>
>>> NMI backtrace for cpu 0
>>> CPU: 0 PID: 68 Comm: kworker/0:2 Not tainted 5.8.0+ #3
>>> Hardware name:  , BIOS
>>> Workqueue: events azx_probe_work [snd_hda_intel]
>>> <SNIP>
>>> Call Trace:
>>> [<ffffffff80211a64>] show_stack+0x9c/0x130
>>> [<ffffffff8065a740>] dump_stack+0xb0/0xf0
>>> [<ffffffff80665774>] nmi_cpu_backtrace+0x134/0x140
>>> [<ffffffff80665910>] nmi_trigger_cpumask_backtrace+0x190/0x200
>>> [<ffffffff802b1abc>] rcu_dump_cpu_stacks+0x12c/0x190
>>> [<ffffffff802b08cc>] rcu_sched_clock_irq+0xa2c/0xfc8
>>> [<ffffffff802b91d4>] update_process_times+0x2c/0xb8
>>> [<ffffffff802cad80>] tick_sched_timer+0x40/0xb8
>>> [<ffffffff802ba5f0>] __hrtimer_run_queues+0x118/0x1d0
>>> [<ffffffff802bab74>] hrtimer_interrupt+0x12c/0x2d8
>>> [<ffffffff8021547c>] c0_compare_interrupt+0x74/0xa0
>>> [<ffffffff80296bd0>] __handle_irq_event_percpu+0xa8/0x198
>>> [<ffffffff80296cf0>] handle_irq_event_percpu+0x30/0x90
>>> [<ffffffff8029d958>] handle_percpu_irq+0x88/0xb8
>>> [<ffffffff80296124>] generic_handle_irq+0x44/0x60
>>> [<ffffffff80b3cfd0>] do_IRQ+0x18/0x28
>>> [<ffffffff8067ace4>] plat_irq_dispatch+0x64/0x100
>>> [<ffffffff80209a20>] handle_int+0x140/0x14c
>>> [<ffffffff802402e8>] irq_exit+0xf8/0x100
>>>
>>> Because AZX_DRIVER_GENERIC can not work well for Loongson LS7A HDA
>>> controller, it needs some workarounds which are not merged into the
>>> upstream kernel at this time, so it should revert this patch now.
>>>
>>> Fixes: 61eee4a7fc40 ("ALSA: hda: Add support for Loongson 7A1000 controller")
>>> Cc: <stable@vger.kernel.org> # 5.9-rc1+
>>> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
>>> ---
>>>
>>> v2: update commit message
>> Applied now.  Thanks.
>>
>>
>> Takashi

Re: [PATCH v2] Revert "ALSA: hda: Add support for Loongson 7A1000 controller"

[Greg KH]

On Tue, Sep 08, 2020 at 11:09:31AM +0800, Tiezhu Yang wrote:
> On 09/08/2020 08:37 AM, Huacai Chen wrote:
> > Hi, all
> > 
> > This patch should be backported to 5.4.
> 
> Hi,
> 
> Commit 61eee4a7fc40 ("ALSA: hda: Add support for Loongson
> 7A1000 controller") has been not yet merged into 5.4, so no
> need to backport.
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/pci/hda/hda_intel.c?h=v5.4

It showed up in 5.4.62, so yes, it does need to go there.

thanks,

greg k-h