From: Daniel Borkmann > > Maybe the _le has something to do with it. > > Could it be bit-reversing the crc and data bytes all the time? > > The packet will (one would hope) want the crc in the same bit-order > > as the data, so no bit reversal is needed - just the correct logic > > and lookup table. > > Although it's a bit off-topic for this patch, but I'll have a look > into __crc32c_le(); maybe something is just wrong with that code. Yes it generates a nice dependency chain of: xor table_b(,%reg,4),%eax using 8 tables (256*4 bytes each). I bet that with a little bit of pipelining a simple byte by byte loop should manage 1 byte every two clocks (ie 1 memory cycle per clock). David -- To unsubscribe from this list: send the line "unsubscribe linux-sctp" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
