* [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc
@ 2014-06-17 4:41 Axel Lin
2014-06-23 5:32 ` Oder Chiou
2014-06-30 18:42 ` Mark Brown
0 siblings, 2 replies; 3+ messages in thread
From: Axel Lin @ 2014-06-17 4:41 UTC (permalink / raw)
To: Mark Brown; +Cc: Oder Chiou, alsa-devel, Liam Girdwood
The implementation of rt5677_pll_calc() has the same logic of rl6231_pll_calc().
The only difference is the lower boundary checking for freq_in.
This patch calls rl6231_pll_calc() instead of open-coded.
The k_bp of struct rt5677_pll_code is always false, thus also remove the
code to check pll_code.k_bp.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
---
Hi Oder,
This patch removes the usage of k_bp from struct rt5677_pll_code because it is
always false. I think if k_bp is necessay in the further development, maybe
it's better to add it to rl6231_pll_code. How do you think?
Regards,
Axel
sound/soc/codecs/rt5677.c | 69 ++++++-----------------------------------------
sound/soc/codecs/rt5677.h | 8 ------
2 files changed, 8 insertions(+), 69 deletions(-)
diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c
index 45f99a8..e7c401f 100644
--- a/sound/soc/codecs/rt5677.c
+++ b/sound/soc/codecs/rt5677.c
@@ -2988,62 +2988,12 @@ static int rt5677_set_dai_sysclk(struct snd_soc_dai *dai,
* Returns 0 for success or negative error code.
*/
static int rt5677_pll_calc(const unsigned int freq_in,
- const unsigned int freq_out, struct rt5677_pll_code *pll_code)
+ const unsigned int freq_out, struct rl6231_pll_code *pll_code)
{
- int max_n = RT5677_PLL_N_MAX, max_m = RT5677_PLL_M_MAX;
- int k, red, n_t, pll_out, in_t;
- int n = 0, m = 0, m_t = 0;
- int out_t, red_t = abs(freq_out - freq_in);
- bool m_bp = false, k_bp = false;
-
- if (RT5677_PLL_INP_MAX < freq_in || RT5677_PLL_INP_MIN > freq_in)
+ if (RT5677_PLL_INP_MIN > freq_in)
return -EINVAL;
- k = 100000000 / freq_out - 2;
- if (k > RT5677_PLL_K_MAX)
- k = RT5677_PLL_K_MAX;
- for (n_t = 0; n_t <= max_n; n_t++) {
- in_t = freq_in / (k + 2);
- pll_out = freq_out / (n_t + 2);
- if (in_t < 0)
- continue;
- if (in_t == pll_out) {
- m_bp = true;
- n = n_t;
- goto code_find;
- }
- red = abs(in_t - pll_out);
- if (red < red_t) {
- m_bp = true;
- n = n_t;
- m = m_t;
- if (red == 0)
- goto code_find;
- red_t = red;
- }
- for (m_t = 0; m_t <= max_m; m_t++) {
- out_t = in_t / (m_t + 2);
- red = abs(out_t - pll_out);
- if (red < red_t) {
- m_bp = false;
- n = n_t;
- m = m_t;
- if (red == 0)
- goto code_find;
- red_t = red;
- }
- }
- }
- pr_debug("Only get approximation about PLL\n");
-
-code_find:
-
- pll_code->m_bp = m_bp;
- pll_code->k_bp = k_bp;
- pll_code->m_code = m;
- pll_code->n_code = n;
- pll_code->k_code = k;
- return 0;
+ return rl6231_pll_calc(freq_in, freq_out, pll_code);
}
static int rt5677_set_dai_pll(struct snd_soc_dai *dai, int pll_id, int source,
@@ -3051,7 +3001,7 @@ static int rt5677_set_dai_pll(struct snd_soc_dai *dai, int pll_id, int source,
{
struct snd_soc_codec *codec = dai->codec;
struct rt5677_priv *rt5677 = snd_soc_codec_get_drvdata(codec);
- struct rt5677_pll_code pll_code;
+ struct rl6231_pll_code pll_code;
int ret;
if (source == rt5677->pll_src && freq_in == rt5677->pll_in &&
@@ -3109,15 +3059,12 @@ static int rt5677_set_dai_pll(struct snd_soc_dai *dai, int pll_id, int source,
return ret;
}
- dev_dbg(codec->dev, "m_bypass=%d k_bypass=%d m=%d n=%d k=%d\n",
- pll_code.m_bp, pll_code.k_bp,
- (pll_code.m_bp ? 0 : pll_code.m_code), pll_code.n_code,
- (pll_code.k_bp ? 0 : pll_code.k_code));
+ dev_dbg(codec->dev, "m_bypass=%d m=%d n=%d k=%d\n",
+ pll_code.m_bp, (pll_code.m_bp ? 0 : pll_code.m_code),
+ pll_code.n_code, pll_code.k_code);
regmap_write(rt5677->regmap, RT5677_PLL1_CTRL1,
- pll_code.n_code << RT5677_PLL_N_SFT |
- pll_code.k_bp << RT5677_PLL_K_BP_SFT |
- (pll_code.k_bp ? 0 : pll_code.k_code));
+ pll_code.n_code << RT5677_PLL_N_SFT | pll_code.k_code);
regmap_write(rt5677->regmap, RT5677_PLL1_CTRL2,
(pll_code.m_bp ? 0 : pll_code.m_code) << RT5677_PLL_M_SFT |
pll_code.m_bp << RT5677_PLL_M_BP_SFT);
diff --git a/sound/soc/codecs/rt5677.h b/sound/soc/codecs/rt5677.h
index af4e9c7..08252e7 100644
--- a/sound/soc/codecs/rt5677.h
+++ b/sound/soc/codecs/rt5677.h
@@ -1425,14 +1425,6 @@ enum {
RT5677_AIFS,
};
-struct rt5677_pll_code {
- bool m_bp; /* Indicates bypass m code or not. */
- bool k_bp; /* Indicates bypass k code or not. */
- int m_code;
- int n_code;
- int k_code;
-};
-
struct rt5677_priv {
struct snd_soc_codec *codec;
struct rt5677_platform_data pdata;
--
1.8.3.2
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc
2014-06-17 4:41 [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc Axel Lin
@ 2014-06-23 5:32 ` Oder Chiou
2014-06-30 18:42 ` Mark Brown
1 sibling, 0 replies; 3+ messages in thread
From: Oder Chiou @ 2014-06-23 5:32 UTC (permalink / raw)
To: Axel Lin, Mark Brown
Cc: Oder Chiou, alsa-devel@alsa-project.org, Liam Girdwood
> -----Original Message-----
> From: Axel Lin [mailto:axel.lin@ingics.com]
> Sent: Tuesday, June 17, 2014 12:42 PM
> To: Mark Brown
> Cc: Oder Chiou; Liam Girdwood; alsa-devel@alsa-project.org
> Subject: [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc
>
> The implementation of rt5677_pll_calc() has the same logic of rl6231_pll_calc().
> The only difference is the lower boundary checking for freq_in.
>
> This patch calls rl6231_pll_calc() instead of open-coded.
> The k_bp of struct rt5677_pll_code is always false, thus also remove the
> code to check pll_code.k_bp.
>
> Signed-off-by: Axel Lin <axel.lin@ingics.com>
Tested-by: Oder Chiou <oder_chiou@realtek.com>
> ---
> Hi Oder,
> This patch removes the usage of k_bp from struct rt5677_pll_code because it is
> always false. I think if k_bp is necessay in the further development, maybe
> it's better to add it to rl6231_pll_code. How do you think?
>
It is the better way, thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc
2014-06-17 4:41 [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc Axel Lin
2014-06-23 5:32 ` Oder Chiou
@ 2014-06-30 18:42 ` Mark Brown
1 sibling, 0 replies; 3+ messages in thread
From: Mark Brown @ 2014-06-30 18:42 UTC (permalink / raw)
To: Axel Lin; +Cc: Oder Chiou, alsa-devel, Liam Girdwood
[-- Attachment #1.1: Type: text/plain, Size: 403 bytes --]
On Tue, Jun 17, 2014 at 12:41:31PM +0800, Axel Lin wrote:
> The implementation of rt5677_pll_calc() has the same logic of rl6231_pll_calc().
> The only difference is the lower boundary checking for freq_in.
>
> This patch calls rl6231_pll_calc() instead of open-coded.
> The k_bp of struct rt5677_pll_code is always false, thus also remove the
> code to check pll_code.k_bp.
Applied, thanks.
[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
[-- Attachment #2: Type: text/plain, Size: 0 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2014-06-30 18:43 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-06-17 4:41 [PATCH RFC/RFT] ASoC: rt5677: Convert to use rl6231_pll_calc Axel Lin
2014-06-23 5:32 ` Oder Chiou
2014-06-30 18:42 ` Mark Brown
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).