Marco Gittler wrote: >>> +/* I2C */ >>> + >>> >> >> Looks like i2c is using the 8-bit addresses instead of 7-bit >> addresses, just >> like how m920x, gl861 and au6610 did at first... This needs to be fixed. >> > this seem very difficult. the device uses 16 bit (or better 2 byte for > addressing something) > the big problem ist that f.e. 0xb880 must READ data (remote-code) and > another example > voltage switch: the adress changes from 0xb600 to 0xb601. (both write to > device) > > so is not using standard i2c addresses and must be complete other handled. > i mean a wrapper for the address stuff must be ni to make if full i2c > comaptible. > > the device is not the best i every saw ;), so the design is also not so > nice. > > any help how the convert from the addr to i2c addr can be done is welcome.
OK. We will have time to revisit this issue -- just try to keep it in the back of your mind. >> >>> +static int opera1_usb_i2c_msgxfer(struct dvb_usb_device *dev, u16 addr, >>> + u8 * buf, u16 len, int flag) >>> +{ >>> + int ret = 0; >>> + u8 request; >>> + u16 value; >>> + >>> + if (!dev) { >>> + info("no usb_device"); >>> + return -EINVAL; >>> + } >>> + if (mutex_lock_interruptible(&dev->usb_mutex) < 0) >>> + return -EAGAIN; >>> + >>> + request = (addr & 0xff00) >> 8; >>> + if (!request) >>> + request = 0xb1; >>> + value = (addr & 0xff); >>> + if (flag & OPERA_READ_MSG) { >>> + value |= 0x01; >>> + } >>> + if (request == 0xa0) >>> + value = 0xe600; >>> + ret = opera1_xilinx_rw(dev->udev, request, value, buf, len, flag); >>> + >>> + mutex_unlock(&dev->usb_mutex); >>> + return ret; >>> +} >>> + [snip] >>> +static int opera1_tuner_attach(struct dvb_usb_adapter *adap) >>> +{ >>> + adap->pll_addr = 0xc0; >>> + adap->pll_desc = &dvb_pll_opera1; >>> + adap->fe->ops.tuner_ops.set_params = dvb_usb_tuner_set_params_i2c; >>> >> >> We should be using "dvb_attach(dvb_pll_attach, adap->fe, 0x60, >> &adap->dev->i2c_adap, &dvb_pll_opera1)" here, instead. >> >> also, please note the usage of 8-bit i2c addr instead of standard 7-bit. >> >> >> > done, except the i2c addr (reason is like above) fine. Once the i2c handling is converted, we'll change this as appropriate. This will be OK in the meantime. >>> + return 0; >>> +} >>> + [snip] >>> +static struct dvb_usb_device_properties opera1_properties = { >>> + .caps = DVB_USB_IS_AN_I2C_ADAPTER, >>> + .usb_ctrl = CYPRESS_FX2, >>> + .firmware = "opera.fw", >>> >> >> Please see the other firmware names... we should conform to the >> standard, and >> name it: >> >> "dvb-usb-opera-01.fw" >> > done > firmware extract utility is coming soon OK. > > ------------------------------------------------------------------------ > Thanks for the patch.... Please remember that your sign-off is required with every patch, in order for us to apply it to the kernel source. For THIS patch below, I submit my own: Signed-off-by: Michael Krufky <[EMAIL PROTECTED]> Please respond to this email with yours. ...there is still one trivial whitespace problem, though. see below: > diff -r c8b73ec18b42 linux/drivers/media/dvb/dvb-usb/dvb-usb-ids.h > --- a/linux/drivers/media/dvb/dvb-usb/dvb-usb-ids.h Thu Apr 19 12:04:50 > 2007 -0300 > +++ b/linux/drivers/media/dvb/dvb-usb/dvb-usb-ids.h Thu Apr 19 19:04:37 > 2007 +0200 > @@ -32,13 +32,13 @@ > #define USB_VID_LITEON 0x04ca > #define USB_VID_MEDION 0x1660 > #define USB_VID_MSI 0x0db0 > +#define USB_VID_OPERA1 0x695c > #define USB_VID_PINNACLE 0x2304 > #define USB_VID_VISIONPLUS 0x13d3 > #define USB_VID_TWINHAN 0x1822 > #define USB_VID_ULTIMA_ELECTRONIC 0x05d8 > #define USB_VID_UNIWILL 0x1584 > #define USB_VID_WIDEVIEW 0x14aa > -#define USB_VID_OPERA1 0x695c > > /* Product IDs */ > #define USB_PID_ADSTECH_USB2_COLD 0xa333 > diff -r c8b73ec18b42 linux/drivers/media/dvb/dvb-usb/opera1.c > --- a/linux/drivers/media/dvb/dvb-usb/opera1.c Thu Apr 19 12:04:50 > 2007 -0300 > +++ b/linux/drivers/media/dvb/dvb-usb/opera1.c Thu Apr 19 19:53:36 > 2007 +0200 > @@ -256,9 +256,8 @@ static int opera1_frontend_attach(struct > > static int opera1_tuner_attach(struct dvb_usb_adapter *adap) > { > - adap->pll_addr = 0xc0; > - adap->pll_desc = &dvb_pll_opera1; > - adap->fe->ops.tuner_ops.set_params = dvb_usb_tuner_set_params_i2c; > + dvb_attach(dvb_pll_attach, adap->fe, 0xc0, > + &adap->dev->i2c_adap, &dvb_pll_opera1); the 1st char of "&adap->dev ....... " should line up directly below the 1st char of "dvb_pll_attach". not a big problem, but should be fixed before commit. > return 0; > } > > @@ -478,7 +477,7 @@ static struct dvb_usb_device_properties > static struct dvb_usb_device_properties opera1_properties = { > .caps = DVB_USB_IS_AN_I2C_ADAPTER, > .usb_ctrl = CYPRESS_FX2, > - .firmware = "opera.fw", > + .firmware = "dvb-usb-opera-01.fw", > .size_of_priv = sizeof(struct opera1_state), > > .power_ctrl = opera1_power_ctrl, > @@ -533,7 +532,7 @@ static int opera1_probe(struct usb_inter > if (udev->descriptor.idProduct == USB_PID_OPERA1_WARM && > udev->descriptor.idVendor == USB_VID_OPERA1 && > (d == NULL > - || opera1_xilinx_load_firmware(udev, "opera1-fpga.fw") > != 0) > + || opera1_xilinx_load_firmware(udev, > "dvb-usb-opera1-fpga.fw") != 0) > ) { > return -EINVAL; > } > > > ------------------------------------------------------------------------ > > _______________________________________________ > linux-dvb mailing list > linux-dvb@linuxtv.org > http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb -- Michael Krufky _______________________________________________ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb