dmitry@atlantis.dp.ua
2003-Oct-21 00:46 UTC
kern/57174: 4.9-PRERELEASE panic: ata_dmasetup:%
Hello! I'm curious whether this issue would be critical for ANY ata(4)-system. So (to the PR originator): Eugene, could you try your experiment with tagged queueing turned off (hw.ata.tags=0)? Sincerely, Dmitry -- Atlantis ISP, System Administrator e-mail: dmitry@atlantis.dp.ua nic-hdl: LYNX-RIPE
dmitry@atlantis.dp.ua wrote:> > Hello! > > I'm curious whether this issue would be critical for ANY ata(4)-system. > So (to the PR originator): Eugene, could you try your experiment with > tagged queueing turned off (hw.ata.tags=0)?I will tonight. Eugene Grosbein
dmitry@atlantis.dp.ua wrote:> > Hello! > > I'm curious whether this issue would be critical for ANY ata(4)-system. > So (to the PR originator): Eugene, could you try your experiment with > tagged queueing turned off (hw.ata.tags=0)?Btw, I wonder why I did not any feedback from ATA maintainer or RE yet. OTOH, they may be having hard time. Eugene
dmitry@atlantis.dp.ua wrote:> > Hello! > > I'm curious whether this issue would be critical for ANY ata(4)-system. > So (to the PR originator): Eugene, could you try your experiment with > tagged queueing turned off (hw.ata.tags=0)?I've tried. It does not panic with hw.ata.tags=0, glimpseindex took more than a hour to complete. This time CPU utilization was about 50% only and 'systat -vm 3' have shown 100% of hard drive usage. So Tagged Queueing may be guilty (again). Eugene Grosbein