mbox series

[SRU,N/O,0/2] alsa: Headphone and Speaker couldn't output sound intermittently

Message ID 20240827031408.130666-1-hui.wang@canonical.com
Headers show
Series alsa: Headphone and Speaker couldn't output sound intermittently | expand

Message

Hui Wang Aug. 27, 2024, 3:14 a.m. UTC
BugLink: https://bugs.launchpad.net/bugs/2077690

The patches are in the linux-next already, it will be in the unstable
kernel automatically, and the patches will be in stable sooner or
later, so it will be in the Jammy kernel automatically in the future.

And our oem project requires the patches to be in the hwe-6.8 ASAP,
Here I submit the SRU to Noble and Oracular.

[Impact]
On a Dell machine, we plug a speaker or headset into the front audio
port, play sound to the speaker or headset, we get a 1/6 chance that
the speaker or headset couldn't output sound.


[Fix]
pick 2 commits from linux-next

[Test]
booting with the patched kernel,
 1. plug a speaker or headset in the front port
 2. play sound
 3. check if we can hear the sound or not
 4. repeat the step 1 ~ 3 20 times
 5. there is no failure

[Where problems could occur]
The patches change the depop and headset_type_detect routine, so it
is possible to introduce regression on pop noise and headset detection,
for realtek codec of alc285 and alc256 family, but the regression
possibility is very low, we tested the patches on a couple of Dell and
Lenovo machines, everything worked as well as before.


Kailang Yang (2):
  ALSA: hda/realtek - Fixed ALC256 headphone no sound
  ALSA: hda/realtek - FIxed ALC285 headphone no sound

 sound/pci/hda/patch_realtek.c | 76 +++++++++++++++++++++++------------
 1 file changed, 51 insertions(+), 25 deletions(-)

Comments

Thibault Ferrante Aug. 27, 2024, 2:07 p.m. UTC | #1
On 27-08-2024 05:14, Hui Wang wrote:
> BugLink: https://bugs.launchpad.net/bugs/2077690
> 
> The patches are in the linux-next already, it will be in the unstable
> kernel automatically, and the patches will be in stable sooner or
> later, so it will be in the Jammy kernel automatically in the future.
> 
> And our oem project requires the patches to be in the hwe-6.8 ASAP,
> Here I submit the SRU to Noble and Oracular.
> 
> [Impact]
> On a Dell machine, we plug a speaker or headset into the front audio
> port, play sound to the speaker or headset, we get a 1/6 chance that
> the speaker or headset couldn't output sound.
> 
> 
> [Fix]
> pick 2 commits from linux-next
> 
> [Test]
> booting with the patched kernel,
>   1. plug a speaker or headset in the front port
>   2. play sound
>   3. check if we can hear the sound or not
>   4. repeat the step 1 ~ 3 20 times
>   5. there is no failure
> 
> [Where problems could occur]
> The patches change the depop and headset_type_detect routine, so it
> is possible to introduce regression on pop noise and headset detection,
> for realtek codec of alc285 and alc256 family, but the regression
> possibility is very low, we tested the patches on a couple of Dell and
> Lenovo machines, everything worked as well as before.
> 
> 
> Kailang Yang (2):
>    ALSA: hda/realtek - Fixed ALC256 headphone no sound
>    ALSA: hda/realtek - FIxed ALC285 headphone no sound
> 
>   sound/pci/hda/patch_realtek.c | 76 +++++++++++++++++++++++------------
>   1 file changed, 51 insertions(+), 25 deletions(-)
> 

Acked-by: Thibault Ferrante <thibault.ferrante@canonical.com>
Agathe Porte Aug. 28, 2024, 9:16 a.m. UTC | #2
2024-08-27 05:15 CEST, Hui Wang:
> BugLink: https://bugs.launchpad.net/bugs/2077690
> 
> The patches are in the linux-next already, it will be in the unstable
> kernel automatically, and the patches will be in stable sooner or
> later, so it will be in the Jammy kernel automatically in the future.
> 
> And our oem project requires the patches to be in the hwe-6.8 ASAP,
> Here I submit the SRU to Noble and Oracular.
> 
> [Impact]
> On a Dell machine, we plug a speaker or headset into the front audio
> port, play sound to the speaker or headset, we get a 1/6 chance that
> the speaker or headset couldn't output sound.
> 
> 
> [Fix]
> pick 2 commits from linux-next
> 
> [Test]
> booting with the patched kernel,
>  1. plug a speaker or headset in the front port
>  2. play sound
>  3. check if we can hear the sound or not
>  4. repeat the step 1 ~ 3 20 times
>  5. there is no failure
> 
> [Where problems could occur]
> The patches change the depop and headset_type_detect routine, so it
> is possible to introduce regression on pop noise and headset detection,
> for realtek codec of alc285 and alc256 family, but the regression
> possibility is very low, we tested the patches on a couple of Dell and
> Lenovo machines, everything worked as well as before.
> 
> 
> Kailang Yang (2):
>   ALSA: hda/realtek - Fixed ALC256 headphone no sound
>   ALSA: hda/realtek - FIxed ALC285 headphone no sound
> 
>  sound/pci/hda/patch_realtek.c | 76 +++++++++++++++++++++++------------
>  1 file changed, 51 insertions(+), 25 deletions(-)

Acked-by: Agathe Porte <agathe.porte@canonical.com>
Stefan Bader Aug. 30, 2024, 3:09 p.m. UTC | #3
On 27.08.24 05:14, Hui Wang wrote:
> BugLink: https://bugs.launchpad.net/bugs/2077690
> 
> The patches are in the linux-next already, it will be in the unstable
> kernel automatically, and the patches will be in stable sooner or
> later, so it will be in the Jammy kernel automatically in the future.
> 
> And our oem project requires the patches to be in the hwe-6.8 ASAP,
> Here I submit the SRU to Noble and Oracular.
> 
> [Impact]
> On a Dell machine, we plug a speaker or headset into the front audio
> port, play sound to the speaker or headset, we get a 1/6 chance that
> the speaker or headset couldn't output sound.
> 
> 
> [Fix]
> pick 2 commits from linux-next
> 
> [Test]
> booting with the patched kernel,
>   1. plug a speaker or headset in the front port
>   2. play sound
>   3. check if we can hear the sound or not
>   4. repeat the step 1 ~ 3 20 times
>   5. there is no failure
> 
> [Where problems could occur]
> The patches change the depop and headset_type_detect routine, so it
> is possible to introduce regression on pop noise and headset detection,
> for realtek codec of alc285 and alc256 family, but the regression
> possibility is very low, we tested the patches on a couple of Dell and
> Lenovo machines, everything worked as well as before.
> 
> 
> Kailang Yang (2):
>    ALSA: hda/realtek - Fixed ALC256 headphone no sound
>    ALSA: hda/realtek - FIxed ALC285 headphone no sound
> 
>   sound/pci/hda/patch_realtek.c | 76 +++++++++++++++++++++++------------
>   1 file changed, 51 insertions(+), 25 deletions(-)
> 

Applied to noble:linux/master-next. Thanks.

-Stefan
Timo Aaltonen Sept. 2, 2024, 12:29 p.m. UTC | #4
Hui Wang kirjoitti 27.8.2024 klo 6.14:
> BugLink: https://bugs.launchpad.net/bugs/2077690
> 
> The patches are in the linux-next already, it will be in the unstable
> kernel automatically, and the patches will be in stable sooner or
> later, so it will be in the Jammy kernel automatically in the future.
> 
> And our oem project requires the patches to be in the hwe-6.8 ASAP,
> Here I submit the SRU to Noble and Oracular.
> 
> [Impact]
> On a Dell machine, we plug a speaker or headset into the front audio
> port, play sound to the speaker or headset, we get a 1/6 chance that
> the speaker or headset couldn't output sound.
> 
> 
> [Fix]
> pick 2 commits from linux-next
> 
> [Test]
> booting with the patched kernel,
>   1. plug a speaker or headset in the front port
>   2. play sound
>   3. check if we can hear the sound or not
>   4. repeat the step 1 ~ 3 20 times
>   5. there is no failure
> 
> [Where problems could occur]
> The patches change the depop and headset_type_detect routine, so it
> is possible to introduce regression on pop noise and headset detection,
> for realtek codec of alc285 and alc256 family, but the regression
> possibility is very low, we tested the patches on a couple of Dell and
> Lenovo machines, everything worked as well as before.
> 
> 
> Kailang Yang (2):
>    ALSA: hda/realtek - Fixed ALC256 headphone no sound
>    ALSA: hda/realtek - FIxed ALC285 headphone no sound
> 
>   sound/pci/hda/patch_realtek.c | 76 +++++++++++++++++++++++------------
>   1 file changed, 51 insertions(+), 25 deletions(-)
> 

This is in v6.11-rc5 already