From owner-freebsd-multimedia Fri Mar 24 9:32:20 2000 Delivered-To: freebsd-multimedia@freebsd.org Received: from oberon.dnai.com (oberon.dnai.com [207.181.194.97]) by hub.freebsd.org (Postfix) with ESMTP id 89D5437B800 for ; Fri, 24 Mar 2000 09:32:12 -0800 (PST) (envelope-from kmarx@bigshed.com) Received: from bigshed.com (dnai-216-15-97-193.cust.dnai.com [216.15.97.193]) by oberon.dnai.com (8.9.3/8.9.3) with ESMTP id JAA91734; Fri, 24 Mar 2000 09:31:40 -0800 (PST) Message-ID: <38DBA74B.C2E3DCB6@bigshed.com> Date: Fri, 24 Mar 2000 09:35:07 -0800 From: Ken Marx X-Mailer: Mozilla 4.5 [en] (X11; I; FreeBSD 2.2.8-RELEASE i386) X-Accept-Language: en MIME-Version: 1.0 To: Randall Hopper Cc: freebsd-multimedia@FreeBSD.ORG, Ken Marx Subject: Re: Lost interrupts in snd0 ? References: <38D823E1.87FDA291@bigshed.com> <20000322225322.A3725@ipass.net> <38D9A470.FFC819DD@bigshed.com> <20000323225809.B2570@ipass.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-multimedia@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Randall Hopper wrote: > ... > > Beats me. I'm not sure what you mean by qlen. DMA ring transfer buffers? > The code in dmabuf.c reads: 353 while (!PROCESS_ABORTING (out_sleep_flag[dev]) 354 && audio_devs[dev]->dmap_out->qlen){ 355 int chn; 356 357 out_sleeper[dev] = &chn; 358 DO_SLEEP1(chn, out_sleep_flag[dev], 10 * hz); 359 if (TIMED_OUT (out_sleep_flag[dev]) ) { 360 361 splx(flags); 362 363 return audio_devs[dev]->dmap_out->qlen; 364 365 } 366 } I never see the aborting condition in my debug trace (not shown in above) so either 1) dmap_out->qlen is being set correctly and we miss/never-get an interrupt from the board; or 2) we somehow set/leave qlen incorrectly set which causes us to wait (for the max 10*hz = 1000 tics) for an interrupt that never will/should happen. I'm tending to think #1, but mostly don't understand the code enough to be sure and/or prove it. > ...(Hopefully someone else with Voxware guts knowledge can > help out here :-) (i.e. I don't qualify) Definitely and ditto. I'm going to look into the 3.4-STABLE changes that Conrad mentions. Let you know if I see any difference. Btw, I've been able to start testing on a PIII 500mhz 3.4-RELEASE box. So far don't see any problems, but have yet to really bang on it. Thanks to all, k. -- Ken Marx, kmarx@bigshed.com We are empowered to guard the keys to the kingdom and get our arms around the realistic goals. - http://cgi.bigshed.com/~kmarx/cgi-bin/speak.cgi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-multimedia" in the body of the message