On Tue, Sep 22, 2009 at 7:08 AM, hermann pitton <hermann-pitton@xxxxxxxx> wrote: > > All fine. > > But you do cut off parts of my messages. > > I would like to ask you to exercise a breakage, not only for stuff six > years back, but on recent S2 API. > at this time there are definitely no DVB applications available which check the flags of the encoded IOCTL values for the S2-API definitions, there are just very few applications available which use this API at all. if (IOCTL_WRITE(cmd)) copy data into buffer for ioctl -- from the driver side -- if (IOCTL_WRITE(cmd)) copy data from the other side process_ioctl() At least the API specification should be valid however the rest is implemented is another story. It works as it is but it's still wrong. Since the S2-API is rather new it might be better to fix it up. It would be a little bit weird too that coding style has a higher priority in Linux than correct code. it breaks the macros in asm-generic/ioctl.h #define IOC_INOUT ((_IOC_WRITE|_IOC_READ) << _IOC_DIRSHIFT) #define IOC_IN (_IOC_WRITE << _IOC_DIRSHIFT) #define IOC_OUT (_IOC_READ << _IOC_DIRSHIFT) #define _IOC_DIR(nr) (((nr) >> _IOC_DIRSHIFT) & _IOC_DIRMASK) Basically I don't mind too much since we already maintain our own version which also supports other unix systems and is not limited to linux only... Markus -- 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
