From owner-freebsd-hackers Fri Feb 15 20:59:13 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from sigbus.com (we-24-126-148-218.we.mediaone.net [24.126.148.218]) by hub.freebsd.org (Postfix) with ESMTP id B0B7437B416; Fri, 15 Feb 2002 20:59:05 -0800 (PST) Received: (from henrich@localhost) by sigbus.com (8.11.1/8.11.1) id g1G4wvq24790; Fri, 15 Feb 2002 20:58:57 -0800 (PST) (envelope-from henrich) Date: Fri, 15 Feb 2002 20:58:57 -0800 From: Charles Henrich To: freebsd-multimedia@freebsd.org, freebsd-hackers@freebsd.org Subject: bktr driver signaling broken Message-ID: <20020215205856.A24721@sigbus.com> Mail-Followup-To: freebsd-multimedia@freebsd.org, freebsd-hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i X-Operating-System: FreeBSD 4.2-RELEASE X-PGP-Fingerprint: 1024/F7 FD C7 3A F5 6A 23 BF 76 C4 B8 C9 6E 41 A4 4F X-GPG-Fingerprint: EA4C AB9B 0C38 17C0 AB3F 11DE 41F6 5883 41E7 4F49 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Okay after bashing my skull against this problem for some time now, I've narrowed this problem down. What Im attempting to do is get a signal from the bktr driver after each frame is posted. The code in the kernel supports this. It even works for some output formats. The problem is the driver never sees a "completed" frame in continous mode, unless your in RGB16 (possibly others, I havent checked). If your in YUV mode, you get signaling something like so: Field is 1 [EVEN:2, ODD:1] Completed field 00311243 00311243 Complete frame? 00111243 & 300000 = 100000 Field is 1 [EVEN:2, ODD:1] Completed field 00111243 00111243 Out of sync, starting over Field is 2 [EVEN:2, ODD:1] Completed field 00311243 00311243 Out of sync, starting over Field is 1 [EVEN:2, ODD:1] Completed field 00311243 00311243 Complete frame? 00111243 & 300000 = 100000 Field is 1 [EVEN:2, ODD:1] Completed field 00111243 00111243 Out of sync, starting over Field is 2 [EVEN:2, ODD:1] Completed field 00311243 00311243 Where for some reason the (card/driver?) is signalling incorrect field ordering (or correct, I dont have the docs). In any case the driver feels since it doesnt have a nice 2 field frame, it never enters the signal the app there is a frame code. Any of the authors of this driver, or anyone with brooktree experience got any ideas? -Crh Charles Henrich Eon Entertainment henrich@msu.edu http://www.sigbus.com:81/~henrich To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message