Edgar Toernig wrote: >Manu Abraham wrote: > > >>Edgar Toernig wrote: >> >> >> >>>All good and fine - but I get no errors with the video fifo, >>>only with the audio fifo. Afaics, there are only two differences: >>>the video fifo is two times as deep as the audio fifo (but data rates >>>are 4-8 times as high), and the read-only MAX_LAT register which is >>>10us for video and 64us for audio. >>> >>> >>Yeah, that shouldn't happen. Why don't you try grabbing the entire >>stream as it is from the buffer. This should be almost the same as the >>transfer from the video fifo. At least that could help in identifying >>the problem. >> >> > >The bt878 driver always grabs the entire stream, it's usually between 30 >and 40 Mbits/sec = 4-5 Mbyte/s. A full resolution video (768x576) takes, >depending on color format, between 22 and 44 Mbytes/s. > > > Heh. The driver does indeed ! >>>Not the fifo-depth is changed (that's not possible in the bt878), the >>>trigger point at which access to the PCI bus is requested is lowered. >>> >>> >>Changing the trigger point is indirectly having the same effect as that >>of the buffer size, even though directly it is not handling the buffer size. >> >> > >Nonsense. The fifo can hold 140 bytes - that's about 30us for a full >transponder. If you ask for the PCI bus when the fifo is half full >the PCI arbiter has 15us to grant the bus or the fifo overflows. >If you request the bus when the fifo is only 10% full then the arbiter >has 27us. > > > [..] Manu
