Re: [alsa-devel] [PATCH] ASoC: Intel: cht_bsw_rt5645.c: Remove buffer and snprintf calls

From: Hans de Goede
Date: Mon May 06 2019 - 11:49:04 EST


On 06-05-19 17:24, Pierre-Louis Bossart wrote:
On 5/4/19 10:16 AM, Nariman wrote:
From: Damian van Soelen <dj.vsoelen@xxxxxxxxx>

The snprintf calls filling cht_rt5645_cpu_dai_name / cht_rt5645_codec_aif_name
always fill them with the same string ("ssp0-port" resp "rt5645-aif2") so
instead of keeping these buffers around and making the cpu_dai_name /
codec_aif_name point to this, simply update the foo_dai_name and foo_aif_name pointers to
directly point to a string constant containing the desired string.

Signed-off-by: Damian van Soelen <dj.vsoelen@xxxxxxxxx>

Need Nariman's Signoff-of-by tag here.

Yes, Nariman can you resend with your S-o-b added once the other part
of this discussion is resolved?

Note you need to pass

--subject-prefix="PATCH v2"

To git send-email when sending out the new version with your S-o-b added
to the 3 patches which you are not the author of.

 sound/soc/intel/boards/cht_bsw_rt5645.c | 26 ++++---------------------
 1 file changed, 4 insertions(+), 22 deletions(-)

diff --git a/sound/soc/intel/boards/cht_bsw_rt5645.c b/sound/soc/intel/boards/cht_bsw_rt5645.c
index cbc2d458483f..b15459e56665 100644
--- a/sound/soc/intel/boards/cht_bsw_rt5645.c
+++ b/sound/soc/intel/boards/cht_bsw_rt5645.c
@@ -506,8 +506,6 @@ static struct cht_acpi_card snd_soc_cards[] = {
 static char cht_rt5645_codec_name[SND_ACPI_I2C_ID_LEN];
-static char cht_rt5645_codec_aif_name[12]; /*Â = "rt5645-aif[1|2]" */
-static char cht_rt5645_cpu_dai_name[10]; /*Â = "ssp[0|2]-port" */
 static bool is_valleyview(void)
@@ -641,28 +639,12 @@ static int snd_cht_mc_probe(struct platform_device *pdev)
ÂÂÂÂÂ log_quirks(&pdev->dev);
ÂÂÂÂÂ if ((cht_rt5645_quirk & CHT_RT5645_SSP2_AIF2) ||
-ÂÂÂÂÂÂÂ (cht_rt5645_quirk & CHT_RT5645_SSP0_AIF2)) {
-ÂÂÂÂÂÂÂ /* fixup codec aif name */
-ÂÂÂÂÂÂÂ snprintf(cht_rt5645_codec_aif_name,
-ÂÂÂÂÂÂÂÂÂÂÂ sizeof(cht_rt5645_codec_aif_name),
-ÂÂÂÂÂÂÂÂÂÂÂ "%s", "rt5645-aif2");
-ÂÂÂÂÂÂÂ cht_dailink[dai_index].codec_dai_name =
-ÂÂÂÂÂÂÂÂÂÂÂ cht_rt5645_codec_aif_name;
-ÂÂÂ }
+ÂÂÂÂÂÂÂ (cht_rt5645_quirk & CHT_RT5645_SSP0_AIF2))
+ÂÂÂÂÂÂÂÂÂÂÂ cht_dailink[dai_index].codec_dai_name = "rt5645-aif2";

same, not equivalent. SSP2_AIF2 is not handled.

Same remark from me, and also the same for the 3th patch you've
responded to.



ÂÂÂÂÂ if ((cht_rt5645_quirk & CHT_RT5645_SSP0_AIF1) ||
-ÂÂÂÂÂÂÂ (cht_rt5645_quirk & CHT_RT5645_SSP0_AIF2)) {
-ÂÂÂÂÂÂÂ /* fixup cpu dai name name */
-ÂÂÂÂÂÂÂ snprintf(cht_rt5645_cpu_dai_name,
-ÂÂÂÂÂÂÂÂÂÂÂ sizeof(cht_rt5645_cpu_dai_name),
-ÂÂÂÂÂÂÂÂÂÂÂ "%s", "ssp0-port");
-ÂÂÂÂÂÂÂ cht_dailink[dai_index].cpu_dai_name =
-ÂÂÂÂÂÂÂÂÂÂÂ cht_rt5645_cpu_dai_name;
-ÂÂÂ }
+ÂÂÂÂÂÂÂ (cht_rt5645_quirk & CHT_RT5645_SSP0_AIF2))
+ÂÂÂÂÂÂÂÂÂÂÂ cht_dailink[dai_index].cpu_dai_name = "ssp0-port";

and same here, SSP0_AIF1 will no longer work.

ÂÂÂÂÂ /* override plaform name, if required */
ÂÂÂÂÂ platform_name = mach->mach_params.platform;