Hans, Thank you for taking time to review these patches. Please see my response inline. I will be able to post the updated patches next week. > -----Original Message----- > From: Hans Verkuil [mailto:hverkuil@xxxxxxxxx] > Sent: Saturday, April 18, 2009 4:58 PM > To: Chaithrika U S > Cc: linux-media@xxxxxxxxxxxxxxx; davinci-linux-open- > source@xxxxxxxxxxxxxxxxxxxx; Manjunath Hadli; Brijesh Jadav > Subject: Re: [PATCH v2 1/4] ARM: DaVinci: DM646x Video: Platform and > board specific setup > > Hi Chaithrika, > > My apologies for the delay, but here are (finally!) my review comments: > > On Wednesday 08 April 2009 13:18:27 Chaithrika U S wrote: > > Platform specific display device setup for DM646x EVM > > > > Add platform device and resource structures. Also define a platform > specific > > clock setup function that can be accessed by the driver to configure > the clock > > and CPLD. > > > > This patch is dependent on a patch submitted earlier, that patch adds > > Pin Mux and clock definitions for Video on DM646x. > > > > Signed-off-by: Manjunath Hadli <mrh@xxxxxx> > > Signed-off-by: Brijesh Jadav <brijesh.j@xxxxxx> > > Signed-off-by: Chaithrika U S <chaithrika@xxxxxx> > > --- > > Applies to DaVinci GIT tree > > > > arch/arm/mach-davinci/board-dm646x-evm.c | 138 > +++++++++++++++++++++++++++ > > arch/arm/mach-davinci/dm646x.c | 63 ++++++++++++ > > arch/arm/mach-davinci/include/mach/dm646x.h | 25 +++++ > > 3 files changed, 226 insertions(+), 0 deletions(-) > > > > diff --git a/arch/arm/mach-davinci/board-dm646x-evm.c > b/arch/arm/mach-davinci/board-dm646x-evm.c > > index bcf11d5..9071a39 100644 > > --- a/arch/arm/mach-davinci/board-dm646x-evm.c > > +++ b/arch/arm/mach-davinci/board-dm646x-evm.c > > @@ -39,6 +39,7 @@ > > #include <mach/serial.h> > > #include <mach/i2c.h> > > #include <mach/mmc.h> > > +#include <mach/mux.h> > > > > #include <linux/platform_device.h> > > #include <linux/i2c.h> > > @@ -49,6 +50,19 @@ > > #define DM646X_EVM_PHY_MASK (0x2) > > #define DM646X_EVM_MDIO_FREQUENCY (2200000) /* PHY bus > frequency */ > > > > +#define VIDCLKCTL_OFFSET (0x38) > > +#define VSCLKDIS_OFFSET (0x6c) > > + > > +#define VCH2CLK_MASK (BIT_MASK(10) | BIT_MASK(9) | > BIT_MASK(8)) > > +#define VCH2CLK_SYSCLK8 (BIT(9)) > > +#define VCH2CLK_AUXCLK (BIT(9) | BIT(8)) > > +#define VCH3CLK_MASK (BIT_MASK(14) | BIT_MASK(13) | > BIT_MASK(12)) > > +#define VCH3CLK_SYSCLK8 (BIT(13)) > > +#define VCH3CLK_AUXCLK (BIT(14) | BIT(13)) > > + > > +#define VIDCH2CLK (BIT(10)) > > +#define VIDCH3CLK (BIT(11)) > > + > > static struct emac_platform_data dm646x_evm_emac_pdata = { > > .phy_mask = DM646X_EVM_PHY_MASK, > > .mdio_max_freq = DM646X_EVM_MDIO_FREQUENCY, > > @@ -103,11 +117,54 @@ int dm646xevm_eeprom_write(void *buf, off_t > off, size_t count) > > } > > EXPORT_SYMBOL(dm646xevm_eeprom_write); > > > > +static struct i2c_client *cpld_client; > > + > > +static int cpld_video_probe(struct i2c_client *client, > > + const struct i2c_device_id *id) > > +{ > > + cpld_client = client; > > + return 0; > > +} > > + > > +static int __devexit cpld_video_remove(struct i2c_client *client) > > +{ > > + cpld_client = NULL; > > + return 0; > > +} > > + > > +static const struct i2c_device_id cpld_video_id[] = { > > + { "cpld_video", 0 }, > > + { } > > +}; > > + > > +static struct i2c_driver cpld_video_driver = { > > + .driver = { > > + .name = "cpld_video", > > + }, > > + .probe = cpld_video_probe, > > + .remove = cpld_video_remove, > > + .id_table = cpld_video_id, > > +}; > > + > > +static void evm_init_cpld(void) > > +{ > > + i2c_add_driver(&cpld_video_driver); > > +} > > + > > static struct i2c_board_info __initdata i2c_info[] = { > > { > > I2C_BOARD_INFO("24c256", 0x50), > > .platform_data = &eeprom_info, > > }, > > + { > > + I2C_BOARD_INFO("adv7343", 0x2A), > > + }, > > + { > > + I2C_BOARD_INFO("ths7303", 0x2C), > > + }, > > Is it necessary to add the v4l2 subdevs here? I think it is better to > use > dm646x_vpif_subdev instead and rely on v4l2_i2c_new_subdev to > initialize > the subdev. I strongly recommend using the v4l2_i2c_new* functions > since > these are guaranteed to load and lock modules correctly. > OK. I will look into this. > > + { > > + I2C_BOARD_INFO("cpld_video", 0x3B), > > + }, > > }; > > > > static struct davinci_i2c_platform_data i2c_pdata = { > > @@ -115,10 +172,90 @@ static struct davinci_i2c_platform_data > i2c_pdata = { > > .bus_delay = 0 /* usec */, > > }; > > > > +static int set_vpif_clock(int mux_mode, int hd) > > +{ > > + int val = 0; > > + int err = 0; > > + unsigned int value; > > + void __iomem *base = IO_ADDRESS(DAVINCI_SYSTEM_MODULE_BASE); > > + > > + /* disable the clock */ > > + value = __raw_readl(base + VSCLKDIS_OFFSET); > > + value |= (VIDCH3CLK | VIDCH2CLK); > > + __raw_writel(value, base + VSCLKDIS_OFFSET); > > + > > + val = i2c_smbus_read_byte(cpld_client); > > + if (val < 0) > > + return val; > > + > > + if (mux_mode == 1) > > + val &= ~0x40; > > + else > > + val |= 0x40; > > + > > + err = i2c_smbus_write_byte(cpld_client, val); > > + if (err) > > + return err; > > + > > + value = __raw_readl(base + VIDCLKCTL_OFFSET); > > + value &= ~(VCH2CLK_MASK); > > + value &= ~(VCH3CLK_MASK); > > + > > + if (hd >= 1) > > + value |= (VCH2CLK_SYSCLK8 | VCH3CLK_SYSCLK8); > > + else > > + value |= (VCH2CLK_AUXCLK | VCH3CLK_AUXCLK); > > + > > + __raw_writel(value, base + VIDCLKCTL_OFFSET); > > + > > + /* enable the clock */ > > + value = __raw_readl(base + VSCLKDIS_OFFSET); > > + value &= ~(VIDCH3CLK | VIDCH2CLK); > > + __raw_writel(value, base + VSCLKDIS_OFFSET); > > + > > + return 0; > > +} > > + > > +static const struct subdev_info dm646x_vpif_subdev[] = { > > + { > > + .addr = 0x2A, > > + .name = "adv7343", > > + }, > > + { > > + .addr = 0x2C, > > + .name = "ths7303", > > + }, > > +}; > > + > > +static struct vpif_output output[] = { > > Should be const. OK > > > + { > > + .id = 0, > > + .name = "Composite" > > + }, > > + { > > + .id = 1, > > + .name = "Component" > > + }, > > + { > > + .id = 2, > > + .name = "S-Video" > > + }, > > +}; > > Is there any need to include the id field? I can't imagine a case where > there are holes in the id sequence. > Will look into this. > > + > > +static struct vpif_config dm646x_vpif_config = { > > + .set_clock = set_vpif_clock, > > + .subdevinfo = (struct subdev_info *)dm646x_vpif_subdev, > > Why is the cast needed here? .subdevinfo should be a const pointer as > well. > OK > > + .subdev_count = ARRAY_SIZE(dm646x_vpif_subdev), > > + .output = output, > > + .output_count = ARRAY_SIZE(output), > > + .card_name = "DM646x EVM", > > +}; > > + > > static void __init evm_init_i2c(void) > > { > > davinci_init_i2c(&i2c_pdata); > > i2c_register_board_info(1, i2c_info, ARRAY_SIZE(i2c_info)); > > + evm_init_cpld(); > > } > > > > static void __init davinci_map_io(void) > > @@ -132,6 +269,7 @@ static __init void evm_init(void) > > evm_init_i2c(); > > davinci_serial_init(&uart_config); > > davinci_init_emac(&dm646x_evm_emac_pdata); > > + dm646x_setup_vpif(&dm646x_vpif_config); > > } > > > > static __init void davinci_dm646x_evm_irq_init(void) > > diff --git a/arch/arm/mach-davinci/dm646x.c b/arch/arm/mach- > davinci/dm646x.c > > index b302f12..a94cb86 100644 > > --- a/arch/arm/mach-davinci/dm646x.c > > +++ b/arch/arm/mach-davinci/dm646x.c > > @@ -12,6 +12,7 @@ > > #include <linux/init.h> > > #include <linux/clk.h> > > #include <linux/platform_device.h> > > +#include <linux/dma-mapping.h> > > > > #include <mach/dm646x.h> > > #include <mach/clock.h> > > @@ -24,6 +25,15 @@ > > #include "clock.h" > > #include "mux.h" > > > > +#define DAVINCI_VPIF_BASE (0x01C12000) > > +#define VDD3P3V_PWDN_OFFSET (0x48) > > +#define VSCLKDIS_OFFSET (0x6C) > > + > > +#define VDD3P3V_VID_MASK (BIT_MASK(7) | BIT_MASK(6) | BIT_MASK(5) > |\ > > + BIT_MASK(4)) > > +#define VSCLKDIS_MASK (BIT_MASK(11) | BIT_MASK(10) | > BIT_MASK(9) |\ > > + BIT_MASK(8)) > > + > > /* > > * Device specific clocks > > */ > > @@ -421,8 +431,61 @@ static struct platform_device dm646x_edma_device > = { > > .resource = edma_resources, > > }; > > > > +static u64 vpif_dma_mask = DMA_32BIT_MASK; > > + > > +static struct resource vpif_resource[] = { > > + { > > + .start = DAVINCI_VPIF_BASE, > > + .end = DAVINCI_VPIF_BASE + 0x03fff, > > + .flags = IORESOURCE_MEM, > > + }, > > + { > > + .start = IRQ_DM646X_VP_VERTINT2, > > + .end = IRQ_DM646X_VP_VERTINT2, > > + .flags = IORESOURCE_IRQ, > > + }, > > + { > > + .start = IRQ_DM646X_VP_VERTINT3, > > + .end = IRQ_DM646X_VP_VERTINT3, > > + .flags = IORESOURCE_IRQ, > > + }, > > +}; > > + > > +static struct platform_device vpif_display_dev = { > > + .name = "vpif_display", > > + .id = -1, > > + .dev = { > > + .dma_mask = &vpif_dma_mask, > > + .coherent_dma_mask = DMA_32BIT_MASK, > > + }, > > + .resource = vpif_resource, > > + .num_resources = ARRAY_SIZE(vpif_resource), > > +}; > > + > > /*------------------------------------------------------------------ > ----*/ > > > > +void dm646x_setup_vpif(struct vpif_config *config) > > +{ > > + unsigned int value; > > + void __iomem *base = IO_ADDRESS(DAVINCI_SYSTEM_MODULE_BASE); > > + > > + value = __raw_readl(base + VSCLKDIS_OFFSET); > > + value &= ~VSCLKDIS_MASK; > > + __raw_writel(value, base + VSCLKDIS_OFFSET); > > + > > + value = __raw_readl(base + VDD3P3V_PWDN_OFFSET); > > + value &= ~VDD3P3V_VID_MASK; > > + __raw_writel(value, base + VDD3P3V_PWDN_OFFSET); > > + > > + davinci_cfg_reg(DM646X_STSOMUX_DISABLE); > > + davinci_cfg_reg(DM646X_STSIMUX_DISABLE); > > + davinci_cfg_reg(DM646X_PTSOMUX_DISABLE); > > + davinci_cfg_reg(DM646X_PTSIMUX_DISABLE); > > + > > + vpif_display_dev.dev.platform_data = config; > > + platform_device_register(&vpif_display_dev); > > +} > > + > > #if defined(CONFIG_TI_DAVINCI_EMAC) || > defined(CONFIG_TI_DAVINCI_EMAC_MODULE) > > > > void dm646x_init_emac(struct emac_platform_data *pdata) > > diff --git a/arch/arm/mach-davinci/include/mach/dm646x.h > b/arch/arm/mach-davinci/include/mach/dm646x.h > > index 02ce872..2f9a1d8 100644 > > --- a/arch/arm/mach-davinci/include/mach/dm646x.h > > +++ b/arch/arm/mach-davinci/include/mach/dm646x.h > > @@ -12,6 +12,7 @@ > > #define __ASM_ARCH_DM646X_H > > > > #include <linux/platform_device.h> > > +#include <linux/i2c.h> > > #include <mach/hardware.h> > > #include <mach/emac.h> > > > > @@ -25,4 +26,28 @@ > > void __init dm646x_init(void); > > void dm646x_init_emac(struct emac_platform_data *pdata); > > > > +void dm646x_video_init(void); > > + > > +struct vpif_output { > > + u16 id; > > + const char *name; > > +}; > > + > > +struct subdev_info { > > + u8 addr; > > + const char *name; > > +}; > > + > > +struct vpif_config { > > + int (*set_clock)(int, int); > > + struct subdev_info *subdevinfo; > > Should be a const pointer. OK > > > + int subdev_count; > > + struct vpif_output *output; > > Ditto. OK > > > + int output_count; > > + const char *card_name; > > +}; > > + > > + > > +void dm646x_setup_vpif(struct vpif_config *config); > > + > > #endif /* __ASM_ARCH_DM646X_H */ > > Regards, > > Hans > > -- > Hans Verkuil - video4linux developer - sponsored by TANDBERG -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
