> Otbert Froehlich wrote: > > I had comparable problems with my FSC ESPRIMO P5615 > (motherboard D2461-A1). > > Problems went away after either > > uncommenting: AllowMSI=0 in > /kernel/drv/nvidia.conf > > or > > executing: NVIDIA-Solaris-x86-177.80-xVM.run . > > > > But always I have to execute '/usr/sbin/psradm -n > -v -F 1' because as byproduct of both methods > > command 'psrinfo' is showing after boot for > processor 1 state 'FAILED'. > > > > Last time I have seen a failing processor 1 was > with the last release of SXDE. There are no > > problems with processor 1 if I boot either build 97 > or build 101 (after deleting lines 'nvidia > > "pci10de,24."' in /etc/driver_aliases). > > I attached output of command > nvidia-SunOS-bug-report.sh and excerpts of file > /var/adm/messages > > of both build 97 and the unmodified build 101. > Do you still have the failed CPU 1 start even with > the updated NVIDIA > driver?
Yes, I am using now the driver from NVIDIA-Solaris-x86-177.80-xVM.run and after the graphical login I have to restart CPU 1. > The log file with the original driver is shows: > > ... NOTICE: apic: local nmi: 0 1 1 1 > ... NOTICE: apic: local nmi: 1 1 1 1 > ... pcplusmp: vector 0x9 ioapic 0x2 intin 0x9 is > bound to cpu 1 > ... NOTICE: amd_iommu: No AMD IOMMU ACPI IVRS > table > > _______________________________________________ > xwin-discuss mailing list > xwin-discuss at opensolaris.org -- This message posted from opensolaris.org
