Date: Tue, 1 Jun 1999 08:05:16 -0700 From: Daryll Strauss <daryll@harlot.rb.ca.us> To: Stephen Hocking-Senior Programmer PGS Tensor Perth <shocking@prth.pgs.com> Cc: dfr@freebsd.org, multimedia@freebsd.org Subject: Re: Encouraging progress with FreeBSD 3dfx device driver. Message-ID: <19990601080516.B5797@harlot.rb.ca.us> In-Reply-To: <199906011229.UAA10598@ariadne.tensor.pgs.com>; from Stephen Hocking-Senior Programmer PGS Tensor Perth on Tue, Jun 01, 1999 at 08:29:19PM %2B0800 References: <199906011229.UAA10598@ariadne.tensor.pgs.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jun 01, 1999 at 08:29:19PM +0800, Stephen Hocking-Senior Programmer PGS Tensor Perth wrote: > I've made some huge steps during the past couple of days with the driver. For > a long time I was having problems with the structure being passed to the ioctl > call being corrupted. It turned out that I wasn't setting up the call properly > in the linux emulator. Now the main problem lies with mmaping the card memory. > Watching the system calls whizz by, it works up until a couple of ioctls's > past the mmap. There it hangs hard. A softreset leaves the card in a bizzarre > state, where even any calls that touch it later hang. It takes a power cyle to > make it sensible again. I'm going to be reviewing carefully the mmap and > address setup. Sending bad data to the card can place it in bad states. The memory map should be the entire 16MB of the card. There are also two MTRRs setup (on PIIs) on for the entire region as write combining and another hole poked in that for the control registers that has to be marked uncached. I'm not sure if this helps. If there's other questions on the kernel interface, I can probably answer them. I just can't go into the details of glide. - |Daryll To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-multimedia" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990601080516.B5797>