I can attest to that. I had to port a Windows driver to Linux for an internal test card (the card & code never left the building). The Windows code was ~50k lines long, and did the exact same thing as my final Linux driver of ~7500 lines. Part of the problem was all of the Windows API's that needed to be included. Another problem was the Windows driver was written in C++ with heavy OO code.
Tobin
On Thu, 2007-08-23 at 00:42 +0200, Takashi Iwai wrote:
At Thu, 23 Aug 2007 00:09:55 +0200, Rene Herman wrote:
On 08/22/2007 11:50 PM, Takashi Iwai wrote:
As James mentioned, the datasheet is vial. "SOME SORT OF" datasheet is. Yes, this can be any document.
Where existing Windows driver source would, I assume, count as "some sort".
Well, as my personal preference, Windows driver source code isn't suitable as the primary information source. It's good as a reference, for example, for debugging a bug in your driver. But, the proprietary driver code is often hard to disclose publicly than the other technical documents. And, above all, when you write a driver based on windows code, it tends to result in a bad code ;)
Takashi _______________________________________________ Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel