tech-kern: Re: bus_dmamap_sync() in ti(4) or the lack thereof

Subject: Re: bus_dmamap_sync() in ti(4) or the lack thereof
To: None <tnn+nbsd@nygren.pp.se>
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
List: tech-kern
Date: 03/03/2007 10:41:52

tnn+nbsd@nygren.pp.se wrote:

> Yeah, I've noticed it still doesn't work right. It's gone from 
> non-functional
> to usable, but tcpdump still shows bogus packets with invalid checksums
> and performance is bad.

Hmm. I'm afraid no one tried ti(4) on big endian machines.
ti(4) seems to have some BYTESWAP functions, but I have
seen few boards which handle byteswap properly by hardware
and I have had to add htole32()/le32toh() etc. around all
DMA descriptor accesses.
---
Izumi Tsutsui