Skip to contentSkip to conversation

S3C64XX I2S: declared i2s_v4

PATCH New Jassi 4 participants 5 messages Sep 15, 2009 – Sep 16, 2009

Subsystems (2)
lore
Share & tools
lore

Jassi

author
#0 lore raw

PATCH 2/10 S3C64XX I2S: declared i2s_v4

For now, I2S-v4 is simply enumerated as third I2S controller of S3C6410 rather than a special device.
Also, Declared I2S-v4 device for SMDK6410.

---
 arch/arm/mach-s3c6410/mach-smdk6410.c |    1 +
 arch/arm/plat-s3c/dev-audio.c         |    6 ++++--
 2 files changed, 5 insertions(+), 2 deletions(-)
arch/arm/mach-s3c6410/mach-smdk6410.c +1 -0
index bc9a7de..108ba37 100644
@@ -181,6 +181,7 @@ static struct platform_device *smdk6410_devices[] __initdata = {
 #endif
 	&s3c_device_i2c0,
 	&s3c_device_i2c1,
+	&s3c64xx_device_iisv4,
 	&s3c_device_fb,
 	&s3c_device_usb,
 	&s3c_device_usb_hsotg,
arch/arm/plat-s3c/dev-audio.c +4 -2
index 1322beb..28c030c 100644
@@ -51,6 +51,7 @@ struct platform_device s3c64xx_device_iis1 = {
 };
 EXPORT_SYMBOL(s3c64xx_device_iis1);
 
+#ifdef CONFIG_CPU_S3C6410
 static struct resource s3c64xx_iisv4_resource[] = {
 	[0] = {
 		.start = S3C64XX_PA_IISV4,
@@ -60,9 +61,10 @@ static struct resource s3c64xx_iisv4_resource[] = {
 };
 
 struct platform_device s3c64xx_device_iisv4 = {
-	.name		  = "s3c64xx-iis-v4",
-	.id		  = -1,
+	.name		  = "s3c64xx-iis",
+	.id		  = 2,
 	.num_resources	  = ARRAY_SIZE(s3c64xx_iisv4_resource),
 	.resource	  = s3c64xx_iisv4_resource,
 };
 EXPORT_SYMBOL(s3c64xx_device_iisv4);
+#endif
Signature · 1 line
1.6.2.5
#1 lore raw

PATCH 2/10 S3C64XX I2S: declared i2s_v4

Quoted 35 lines · Jassi

For now, I2S-v4 is simply enumerated as third I2S controller of S3C6410 rather than a special device.
Also, Declared I2S-v4 device for SMDK6410.

---
 arch/arm/mach-s3c6410/mach-smdk6410.c |    1 +
 arch/arm/plat-s3c/dev-audio.c         |    6 ++++--
 2 files changed, 5 insertions(+), 2 deletions(-)
arch/arm/mach-s3c6410/mach-smdk6410.c +1 -0
index bc9a7de..108ba37 100644
@@ -181,6 +181,7 @@ static struct platform_device *smdk6410_devices[] __initdata = {
 #endif
 	&s3c_device_i2c0,
 	&s3c_device_i2c1,
+	&s3c64xx_device_iisv4,
 	&s3c_device_fb,
 	&s3c_device_usb,
 	&s3c_device_usb_hsotg,
arch/arm/plat-s3c/dev-audio.c +1 -0
index 1322beb..28c030c 100644
@@ -51,6 +51,7 @@ struct platform_device s3c64xx_device_iis1 = {
 };
 EXPORT_SYMBOL(s3c64xx_device_iis1);
 
+#ifdef CONFIG_CPU_S3C6410
 static struct resource s3c64xx_iisv4_resource[] = {
 	[0] = {
 		.start = S3C64XX_PA_IISV4,

arch/arm/plat-s3c/dev-audio.c static struct resource s3c64xx_iisv4_resource[] = { · line 65

Quoted 7 lines
arch/arm/plat-s3c/dev-audio.c +2 -2
@@ -60,9 +61,10 @@ static struct resource s3c64xx_iisv4_resource[] = {
 };
 
 struct platform_device s3c64xx_device_iisv4 = {
-	.name		  = "s3c64xx-iis-v4",
-	.id		  = -1,
+	.name		  = "s3c64xx-iis",
+	.id		  = 2,

I'd like to see how much different the two drivers are, and whether they need to be seperated into their own modules.

Changing the name to s3c64xx-iis means that the module auto-load may end up loading the wrong module. Holding off on this change until we se what needs to be done for this block.

Quoted 8 lines
arch/arm/plat-s3c/dev-audio.c +1 -0
 	.num_resources	  = ARRAY_SIZE(s3c64xx_iisv4_resource),
 	.resource	  = s3c64xx_iisv4_resource,
 };
 EXPORT_SYMBOL(s3c64xx_device_iisv4);
+#endif
Signature · 1 line
1.6.2.5
Signature · 5 lines
-- 
Ben

Q:      What's a light-year?
A:      One-third less calories than a regular year.
#2 lore raw

PATCH 2/10 S3C64XX I2S: declared i2s_v4

Quoted 47 lines · Ben Dooks
Quoted 35 lines · Jassi

For now, I2S-v4 is simply enumerated as third I2S controller of S3C6410 rather than a special device.
Also, Declared I2S-v4 device for SMDK6410.

---
?arch/arm/mach-s3c6410/mach-smdk6410.c | ? ?1 + ?arch/arm/plat-s3c/dev-audio.c ? ? ? ? | ? ?6 ++++-- ?2 files changed, 5 insertions(+), 2 deletions(-)

arch/arm/mach-s3c6410/mach-smdk6410.c +0 -0
index bc9a7de..108ba37 100644
@@ -181,6 +181,7 @@ static struct platform_device *smdk6410_devices[] __initdata = {

?#endif
? ? ? &s3c_device_i2c0,
? ? ? &s3c_device_i2c1,
+ ? ? &s3c64xx_device_iisv4,
? ? ? &s3c_device_fb,
? ? ? &s3c_device_usb,
? ? ? &s3c_device_usb_hsotg,

arch/arm/plat-s3c/dev-audio.c +1 -0
index 1322beb..28c030c 100644
@@ -51,6 +51,7 @@ struct platform_device s3c64xx_device_iis1 = {
 ?};
 ?EXPORT_SYMBOL(s3c64xx_device_iis1);
 
+#ifdef CONFIG_CPU_S3C6410
 ?static struct resource s3c64xx_iisv4_resource[] = {
 ? ? ? [0] = {
 ? ? ? ? ? ? ? .start = S3C64XX_PA_IISV4,

arch/arm/plat-s3c/dev-audio.c static struct resource s3c64xx_iisv4_resource[] = { · line 65

Quoted 7 lines
arch/arm/plat-s3c/dev-audio.c +2 -2
@@ -60,9 +61,10 @@ static struct resource s3c64xx_iisv4_resource[] = {
 ?};
 
 ?struct platform_device s3c64xx_device_iisv4 = {
- ? ? .name ? ? ? ? ? ? = "s3c64xx-iis-v4",
- ? ? .id ? ? ? ? ? ? ? = -1,
+ ? ? .name ? ? ? ? ? ? = "s3c64xx-iis",
+ ? ? .id ? ? ? ? ? ? ? = 2,

I'd like to see how much different the two drivers are, and whether they need to be seperated into their own modules.

Quoted 3 lines

Changing the name to s3c64xx-iis means that the module auto-load may end up loading the wrong module. Holding off on this change until we se what needs to be done for this block.

For now, my idea is to first lay ground for I2S-v4 controller. It is easier for implement 6channels support when we already have the controller working for 2channels.

#3 lore raw

PATCH 2/10 S3C64XX I2S: declared i2s_v4

In reply to Ben Dooks
Quoted 8 lines · Ben Dooks

arch/arm/plat-s3c/dev-audio.c static struct resource s3c64xx_iisv4_resource[] = { · line 65

Quoted 3 lines · Jassi
arch/arm/plat-s3c/dev-audio.c +2 -1
-	.id		  = -1,
+	.name		  = "s3c64xx-iis",
+	.id		  = 2,

I'd like to see how much different the two drivers are, and whether they need to be seperated into their own modules.

There's substantial overlap (mostly IISv4 just adds new register bits to control the additional channels), but the trick would be how the driver is supposed to work out that it's handling an IISv4 device and enable the additional functionality.

Quoted 3 lines

Changing the name to s3c64xx-iis means that the module auto-load may end up loading the wrong module. Holding off on this change until we se what needs to be done for this block.

It's certainly not a long-term solution as-is.

#4 lore raw

PATCH 2/10 S3C64XX I2S: declared i2s_v4

In reply to Jassi
Quoted 2 lines · Jassi
 arch/arm/mach-s3c6410/mach-smdk6410.c |    1 +
 arch/arm/plat-s3c/dev-audio.c         |    6 ++++--

This won't apply against the current code, dev-audio.c is now in plat-s3c64xx. Please check that you're working against current code when submitting things otherwise people aren't going to be able to apply your patches. The -next tree is helpful for this.