On Sat, Aug 11, 2012 at 6:37 AM, Antti Palosaari <crope@xxxxxx> wrote: > On 08/11/2012 03:31 AM, Manu Abraham wrote: >> >> On Sat, Aug 11, 2012 at 5:44 AM, Antti Palosaari <crope@xxxxxx> wrote: >>> >>> On 08/11/2012 01:12 AM, Mauro Carvalho Chehab wrote: >>>> >>>> >>>> Em 13-07-2012 20:15, CrazyCat escreveu: >>> >>> >>> >>>>> #define DTV_ISDBS_TS_ID 42 >>>>> >>>>> #define DTV_DVBT2_PLP_ID 43 >>>>> +#define DTV_DVBS2_MIS_ID 43 >>>> >>>> >>>> >>>> It would be better to define it as: >>>> >>>> #define DTV_DVBS2_MIS_ID DTV_DVBT2_PLP_ID >>>> >>>> Even better, we should instead find a better name that would cover both >>>> DVB-T2 and DVB-S2 program ID fields, like: >>>> >>>> #define DTV_DVB_MULT 43 >>>> #define DTV_DVBT2_PLP_ID DTV_DVB_MULT >>>> >>>> And use the new symbol for both DVB-S2 and DVB-T2, deprecating the >>>> legacy symbol. >>> >>> >>> >>> Also DTV_ISDBS_TS_ID means same. All these three DTV_ISDBS_TS_ID, >>> DTV_DVBT2_PLP_ID and DTV_DVBS2_MIS_ID are same thing - just named >>> differently between standards. I vote for common name TS ID (I have said >>> that already enough many times...). >> >> >> I agree, but a still more generic term like STREAM_ID would be more >> appropriate, > > > Ack. Since this stream could be something else than MPEG2-TS better to give > more generic name. > > >> as it happens at different layers for different delivery >> systems.DVB-S2 additionally >> provides BBHEADER at Physical Layer. In any case setting PLP_ID for DVB-S2 >> is completely confusing. >> >> Anyway, the demuxer part is also missing .. > > > Demuxer for MIS? I am not any familiar with MIS but I know there is "raw" > demux payload used already for ATSC-M/H. It just passes all the data coming > from demod "TS". Just grabbing and sending the DMA'd data to userspace is not nice. With ATSC-M/H you don't simply have a TS. With an ATSC-M/H capable demod, which supports ATSC (standard) outputs a TS. the M/H part is not a TS at all. Even with ATSC-M/H passing raw data causes all the IP (network) packets to be parsed in userspace, which should have been properly done in kernel space as a filter. If we were to do everything similarly, then we don't need any API at all, just a simple read and let each application do whatever it wants. Regards, Manu -- 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
