Date: Sun, 11 Jun 1995 12:21:33 +0200 (MET DST) From: J Wunsch <j@uriah.heep.sax.de> To: freebsd-hackers@freebsd.org (FreeBSD hackers) Subject: Re: VM86 calls and BIOS disk driver Message-ID: <199506111021.MAA04899@uriah.heep.sax.de> In-Reply-To: <199506110228.CAA07664@netfl15a.devetir.qld.gov.au> from "Stephen Hocking" at Jun 11, 95 02:28:09 am
next in thread | previous in thread | raw e-mail | index | archive | help
As Stephen Hocking wrote: > > > Is anyone actively working on the VM86 stuff with an eye to > implementing a generic BIOS disk driver, so as to support the hardware > or controllers that are not yet supported by a specific driver? Why I say > that is because this TMC-885 controller does not work wuth -current's > seagate.c, and only on the 2.0 seagate.c with BLIND_TRANSFER disabled. If I > could use the onboard BIOS, I'm fairly sure that it'd do better than the > huge amount of CPU it eats now polling the damn thing. Not voting against VM86 work, but what makes you believe the latter statement? If the driver uses PIO (either a FreeBSD driver or the BIOS driver), it will eat up CPU. The BIOS driver, while arguably less work, will be worse, since it suffers from additional overhead. If the driver uses bus-master DMA, it might have better performance, but you'll have a hard time telling the BIOS driver its correct (physical) addresses to transfer (unless you're going to always use bounce buffers). I'd still anticipate troubles from the BIOS not knowing how to behave in a `more advanced' environment. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199506111021.MAA04899>