John Martin wrote: > Chris Ridd wrote: >> John Martin wrote: >>> Chris Ridd wrote: >>>> >>>> >>>> Anyway, after trying to create another BE with a downgraded driver, >>>> I rebooted back into "opensolaris-2", aka snv_91, and the graphics >>>> performance is good again. >>>> >>>> Looking at the Xorg.0.log file from the previous snv_91 boot, I do >>>> see some differences: >>>> >>> I don't see the NVIDIA driver version listed in the diffs (173.14.05 >>> versus 169.12). >>> Can you send me both log files? >> >> Here they are. They're both using the 173.14.05 driver from b91. The >> Xorg.0.log file was generated after the power was removed and restored >> to the box. The Xorg.0.log.old was from a "pkg image-update" and "init >> 6", so the power was maintained throughout. > So is the behavior on both your systems, the NVIDIA based desktop and > the ATI based notebook, > that if a simple reboot is done performance is slow but if power is > removed performance improves > (equivalent to b90)?
There's some evidence for that being the case with the NVIDIA desktop, but since I'm almost always power-cycling the laptop I can't tell there. > If so, can you run intrstat when the desktop is slow to see if any > devices are producing > excessive interrupts? Also the output of: > > # echo "::interrupts -d" | mdb -k > > would be useful to see which interrupts are being shared. Next time it happens, sure. Cheers, Chris
