[alsa-devel] Remove depop delay for suspend and resume
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Thanks. Kailang
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks,
Takashi
Hi Takashi,
I know your concerm. Thanks.
Hi Mengdong,
Could you only remove it on you test image?
Thanks. Kailang.
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks,
Takashi
------Please consider the environment before printing this e-mail.
-----Original Message----- From: Kailang [mailto:kailang@realtek.com] Sent: Friday, November 29, 2013 3:08 PM To: Takashi Iwai; Lin, Mengdong
Hi Takashi,
I know your concerm. Thanks.
Hi Mengdong,
Could you only remove it on you test image?
Thank you for the info! I only tested Baytrail-M platform.
Is it possible to add a flag and set the flag a in board-specific fixup?
Regards Mengdong
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks,
Takashi
------Please consider the environment before printing this e-mail.
At Fri, 29 Nov 2013 08:13:05 +0000, Lin, Mengdong wrote:
-----Original Message----- From: Kailang [mailto:kailang@realtek.com] Sent: Friday, November 29, 2013 3:08 PM To: Takashi Iwai; Lin, Mengdong
Hi Takashi,
I know your concerm. Thanks.
Hi Mengdong,
Could you only remove it on you test image?
Thank you for the info! I only tested Baytrail-M platform.
Is it possible to add a flag and set the flag a in board-specific fixup?
Yes, feel free to hack struct alc_spec.
Takashi
Regards Mengdong
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks,
Takashi
------Please consider the environment before printing this e-mail.
-----Original Message----- From: Takashi Iwai [mailto:tiwai@suse.de] Sent: Friday, November 29, 2013 4:23 PM
Thank you for the info! I only tested Baytrail-M platform.
Is it possible to add a flag and set the flag a in board-specific fixup?
Yes, feel free to hack struct alc_spec.
Okay. I'll modify the patch as you suggested, and add a board fixup for Baytrail.
Thanks Mengdong
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks,
Takashi
------Please consider the environment before printing this e-mail.
Hi Kailang/Takashi,
I revised and submitted the patches. Please have a review.
[PATCH 1/2] ALSA: hda/realtek - Remove depop delay for suspend and resume if applicable [PATCH 2/2] ALSA: hda - fixup ALC262 to remove depop delay on Intel BayleyBay board
Thanks Mengdong
-----Original Message----- From: alsa-devel-bounces@alsa-project.org [mailto:alsa-devel-bounces@alsa-project.org] On Behalf Of Lin, Mengdong Sent: Friday, November 29, 2013 5:10 PM To: Takashi Iwai Cc: alsa-devel@alsa-project.org; Kailang Subject: Re: [alsa-devel] Remove depop delay for suspend and resume
-----Original Message----- From: Takashi Iwai [mailto:tiwai@suse.de] Sent: Friday, November 29, 2013 4:23 PM
Thank you for the info! I only tested Baytrail-M platform.
Is it possible to add a flag and set the flag a in board-specific fixup?
Yes, feel free to hack struct alc_spec.
Okay. I'll modify the patch as you suggested, and add a board fixup for Baytrail.
Thanks Mengdong
At Fri, 29 Nov 2013 10:48:20 +0800, Kailang wrote:
Hi Takashi,
Intel test alc262. They find the suspend and resume time were > 50ms. So, I delete the delay for alc_eapd_shutup and alc_resume. If we want to have depop functions, we will follow like alc283.
Which delay value to take is pretty specific to machine. Removing delay will likely cause problem on other machines.
So, if we'd remove delays, do it only for tested machines, not do blindly to all machines using the same codec chips. The pop noise isn't only determined by a codec chip but rather influenced by the whole hardware analog design.
thanks, > > > > Takashi
participants (3)
-
Kailang
-
Lin, Mengdong
-
Takashi Iwai