ALSA: core: Add 384kHz Support
PATCH 1/1 ALSA: core: Add 384kHz Support
From: Brian Austin <[email protected]>
Adding support for 384kHz audio sample rates
--- include/sound/pcm.h | 3 +++ sound/core/pcm_native.c | 4 ++-- 2 files changed, 5 insertions(+), 2 deletions(-)
include/sound/pcm.h +3 -0
@@ -129,6 +129,7 @@ struct snd_pcm_ops {#define SNDRV_PCM_RATE_96000 (1<<10)/* 96000Hz */ #define SNDRV_PCM_RATE_176400 (1<<11)/* 176400Hz */ #define SNDRV_PCM_RATE_192000 (1<<12)/* 192000Hz */ +#define SNDRV_PCM_RATE_384000 (1<<13)/* 384000Hz */ #define SNDRV_PCM_RATE_CONTINUOUS (1<<30)/* continuous range */ #define SNDRV_PCM_RATE_KNOT (1<<31)/* supports more non-continuos rates */
@@ -141,6 +142,8 @@ struct snd_pcm_ops { SNDRV_PCM_RATE_88200|SNDRV_PCM_RATE_96000)#define SNDRV_PCM_RATE_8000_192000 (SNDRV_PCM_RATE_8000_96000|SNDRV_PCM_RATE_176400|\ SNDRV_PCM_RATE_192000) +#define SNDRV_PCM_RATE_8000_384000 (SNDRV_PCM_RATE_8000_192000|SNDRV_PCM_RATE_384000) +#define _SNDRV_PCM_FMTBIT (fmt) (1ULL << (__force int)SNDRV_PCM_FORMAT_##fmt)#define SNDRV_PCM_FMTBIT_S8 _SNDRV_PCM_FMTBIT(S8)#define SNDRV_PCM_FMTBIT_U8 _SNDRV_PCM_FMTBIT(U8)
sound/core/pcm_native.c +2 -2
@@ -1969,12 +1969,12 @@ static int snd_pcm_hw_rule_sample_bits(struct snd_pcm_hw_params *params,return snd_interval_refine (hw_param_interval (params, rule->var), &t); } -#if SNDRV_PCM_RATE_5512 !=1 <<0 ||SNDRV_PCM_RATE_192000 !=1 <<12 +#if SNDRV_PCM_RATE_5512 !=1 <<0 ||SNDRV_PCM_RATE_384000 !=1 <<13 #error "Change this table"#endif static unsigned int rates[] = {5512 ,8000 ,11025 ,16000 ,22050 ,32000 ,44100 , -48000 ,64000 ,88200 ,96000 ,176400 ,192000 }; +48000 ,64000 ,88200 ,96000 ,176400 ,192000 ,384000 };const struct snd_pcm_hw_constraint_list snd_pcm_known_rates = { .count =ARRAY_SIZE (rates),
Signature · 1 line
1.9.1
Quoted 1 line · [email protected] wrote:
Adding support for 384kHz audio sample rates
384 kHz already is supported (see the KNOT bit).
Quoted 4 lines
#define
SNDRV_PCM_RATE_96000(1<<10) /* 96000Hz */
#defineSNDRV_PCM_RATE_176400(1<<11) /* 176400Hz */
#defineSNDRV_PCM_RATE_192000(1<<12) /* 192000Hz */
+#defineSNDRV_PCM_RATE_384000(1<<13) /* 384000Hz */
These bits are shortcuts for often-used rates.
384 kHz is not often used.
Regards,
Clemens
Quoted 4 lines · On Wed, 27 Jan 2016 14:53:45 +0100, <[email protected]> wrote:
From: Brian Austin <[email protected]>
Adding support for 384kHz audio sample rates
The most important information is missing: why this is needed.
(Note that the support of 384kHz itself can be done without this
addition :)
Takashi
Quoted 50 lines
Signed-off-by Brian Austin <[email protected]>--- include/sound/pcm.h | 3 +++ sound/core/pcm_native.c | 4 ++-- 2 files changed, 5 insertions(+), 2 deletions(-)include/sound/pcm.h +3
-0@@ -129,6 +129,7 @@ struct snd_pcm_ops {#define SNDRV_PCM_RATE_96000 (1<<10)/* 96000Hz */ #define SNDRV_PCM_RATE_176400 (1<<11)/* 176400Hz */ #define SNDRV_PCM_RATE_192000 (1<<12)/* 192000Hz */ +#define SNDRV_PCM_RATE_384000 (1<<13)/* 384000Hz */ #define SNDRV_PCM_RATE_CONTINUOUS (1<<30)/* continuous range */ #define SNDRV_PCM_RATE_KNOT (1<<31)/* supports more non-continuos rates */ @@ -141,6 +142,8 @@ struct snd_pcm_ops { SNDRV_PCM_RATE_88200|SNDRV_PCM_RATE_96000)#define SNDRV_PCM_RATE_8000_192000 (SNDRV_PCM_RATE_8000_96000|SNDRV_PCM_RATE_176400|\ SNDRV_PCM_RATE_192000) +#define SNDRV_PCM_RATE_8000_384000 (SNDRV_PCM_RATE_8000_192000|SNDRV_PCM_RATE_384000) +#define _SNDRV_PCM_FMTBIT (fmt) (1ULL << (__force int)SNDRV_PCM_FORMAT_##fmt)#define SNDRV_PCM_FMTBIT_S8 _SNDRV_PCM_FMTBIT(S8)#define SNDRV_PCM_FMTBIT_U8 _SNDRV_PCM_FMTBIT(U8)sound/core/pcm_native.c +2
-2@@ -1969,12 +1969,12 @@ static int snd_pcm_hw_rule_sample_bits(struct snd_pcm_hw_params *params,return snd_interval_refine (hw_param_interval (params, rule->var), &t); } -#if SNDRV_PCM_RATE_5512 !=1 <<0 ||SNDRV_PCM_RATE_192000 !=1 <<12 +#if SNDRV_PCM_RATE_5512 !=1 <<0 ||SNDRV_PCM_RATE_384000 !=1 <<13 #error "Change this table"#endif static unsigned int rates[] = {5512 ,8000 ,11025 ,16000 ,22050 ,32000 ,44100 , -48000 ,64000 ,88200 ,96000 ,176400 ,192000 }; +48000 ,64000 ,88200 ,96000 ,176400 ,192000 ,384000 };const struct snd_pcm_hw_constraint_list snd_pcm_known_rates = { .count =ARRAY_SIZE (rates),Signature · 1 line
1.9.1
Quoted 8 lines · Takashi Iwai
Quoted 4 lines · On Wed, 27 Jan 2016 14:53:45 +0100, <[email protected]> wrote:
From: Brian Austin <[email protected]>
Adding support for 384kHz audio sample rates
The most important information is missing: why this is needed.
So I can specify that I want 384kHz.
Quoted 4 lines
(Note that the support of 384kHz itself can be done without this
addition :)
So we stay at SNDRV_PCM_RATE_192000 always for sample rate defines? And everything else is just covered with SNDRV_PCM_RATE_KNOT?
Quoted 9 lines · Brian Austin
Quoted 7 lines · Takashi Iwai
Quoted 4 lines · On Wed, 27 Jan 2016 14:53:45 +0100, <[email protected]> wrote:
From: Brian Austin <[email protected]>
Adding support for 384kHz audio sample rates
Quoted 9 lines
Quoted 1 line
The most important information is missing: why this is needed.
So I can specify that I want 384kHz.
Quoted 4 lines
(Note that the support of 384kHz itself can be done without this
addition :)So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?
If this is the case then we have to have a contraint in the driver to support higher rates and leave others out? Is that understanding correct?
Quoted 10 lines · Brian Austin
Quoted 7 lines · Takashi Iwai
Quoted 4 lines · On Wed, 27 Jan 2016 14:53:45 +0100, <[email protected]> wrote:
From: Brian Austin <[email protected]>
Adding support for 384kHz audio sample rates
Quoted 8 lines
Quoted 1 line
The most important information is missing: why this is needed.
So I can specify that I want 384kHz.
Quoted 4 lines
(Note that the support of 384kHz itself can be done without this
addition :)So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?
Depends. If a new sample rate is (or will be) demanded by many drivers, it's worth to add it, of course, as it would simplify the code.
That's why I mentioned "why" is most important information; you need to convince others about the necessity of this change, after all. Then you see that "because I-wanna-it" doesn't sound convincing enough, right?
Takashi
Quoted 7 lines · Takashi Iwai
Quoted 2 lines
So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?Depends. If a new sample rate is (or will be) demanded by many drivers, it's worth to add it, of course, as it would simplify the code.
Quoted 8 lines
That's why I mentioned "why" is most important information; you need to convince others about the necessity of this change, after all. Then you see that "because I-wanna-it" doesn't sound convincing enough, right?
Takashi
I really, really, really want it :)
I understand. Going forward, from our perspective, 384 and other high sample rates are going to be defaults for devices as the market is moving that way. I just wanted to make it easier to use those instead of doing all the contraint coding. Now my understanding on the KNOT define may be wrong. I can add the other rates to this, but for now just wanted to add one we use currently.
Does that make sense?
Quoted 16 lines · Brian Austin
Quoted 13 lines · Takashi Iwai
Quoted 2 lines
So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?Depends. If a new sample rate is (or will be) demanded by many drivers, it's worth to add it, of course, as it would simplify the code.
That's why I mentioned "why" is most important information; you need to convince others about the necessity of this change, after all. Then you see that "because I-wanna-it" doesn't sound convincing enough, right?
Quoted 8 lines
Quoted 2 lines
Takashi
I really, really, really want it :)
I understand. Going forward, from our perspective, 384 and other high sample rates are going to be defaults for devices as the market is moving that way. I just wanted to make it easier to use those instead of doing all the contraint coding.
Yeah that's the reason I could *guess*, but it wasn't mentioned.
Quoted 5 lines
Now my understanding on the KNOT define
may be wrong. I can add the other rates to this, but for now just wanted to add one we use currently.Does that make sense?
Just resubmit the patch with mo' better advertisement :)
Takashi
Quoted 29 lines · Takashi Iwai
Quoted 16 lines · Brian Austin
Quoted 13 lines · Takashi Iwai
Quoted 2 lines
So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?Depends. If a new sample rate is (or will be) demanded by many drivers, it's worth to add it, of course, as it would simplify the code.
That's why I mentioned "why" is most important information; you need to convince others about the necessity of this change, after all. Then you see that "because I-wanna-it" doesn't sound convincing enough, right?
Quoted 8 lines
Quoted 2 lines
Takashi
I really, really, really want it :)
I understand. Going forward, from our perspective, 384 and other high sample rates are going to be defaults for devices as the market is moving that way. I just wanted to make it easier to use those instead of doing all the contraint coding.
Yeah that's the reason I could *guess*, but it wasn't mentioned.
Quoted 7 lines
Quoted 5 lines
Now my understanding on the KNOT define
may be wrong. I can add the other rates to this, but for now just wanted to add one we use currently.Does that make sense?
Just resubmit the patch with mo' better advertisement :)
For symmetry reasons maybe also include the matching 44.1kHz based rate. Unless you think that is not going to be a thing.
Quoted 34 lines · Lars-Peter Clausen
Quoted 33 lines · Takashi Iwai
Quoted 16 lines · Brian Austin
Quoted 13 lines · Takashi Iwai
Quoted 2 lines
So we stay at
SNDRV_PCM_RATE_192000always for sample rate defines? And everything else is just covered withSNDRV_PCM_RATE_KNOT?Depends. If a new sample rate is (or will be) demanded by many drivers, it's worth to add it, of course, as it would simplify the code.
That's why I mentioned "why" is most important information; you need to convince others about the necessity of this change, after all. Then you see that "because I-wanna-it" doesn't sound convincing enough, right?
Quoted 8 lines
Quoted 2 lines
Takashi
I really, really, really want it :)
I understand. Going forward, from our perspective, 384 and other high sample rates are going to be defaults for devices as the market is moving that way. I just wanted to make it easier to use those instead of doing all the contraint coding.
Yeah that's the reason I could *guess*, but it wasn't mentioned.
Quoted 4 lines
Now my understanding on the KNOT define
may be wrong. I can add the other rates to this, but for now just wanted to add one we use currently.
Quoted 6 lines
Quoted 3 lines
Quoted 1 line
Does that make sense?
Just resubmit the patch with mo' better advertisement :)
For symmetry reasons maybe also include the matching 44.1kHz based rate. Unless you think that is not going to be a thing.
Everything I have seen using >192k sample rates has been using the 48k based rate families. My guess would be that the matching 44.1k rate is unlikely to be a thing.
Thanks,
Charles