On Wed, Feb 14, 2007 at 07:00:37PM -0800, Jay Lan wrote: > Zou, Nanhai wrote: > > <snip> > > >>>> put_kernel_page: page at 0xe000003018b3c000 not in reserved memory > >>>> > >>>> and followed by tons of "Bad page state" and "Bad pte" errors. > >>> It took a long long time, but the kdump kernel eventually came up. > >>> > >>> What essential IA64 patches do i need (besides the zero-size-vmcore) > >>> for kexec/kdump to work in 2.6.20? > >> You might want to just check out Linus' git tree and work from there, > >> a couple of kexec changes have gone in there recently. > >> > >> I've attached 3 that may be useful to you, though to be honest > >> I doubt any of them fix problems that you are seeing. > >> > > > > Oops, I think the first patch [IA64] register memory ranges in a > > consistent manner is important..., > > I posted the almost same patch 2 months before, but unfortunately that > > patch was missed in 2.6.20 release. > > Jay, I guess that patch will fix your problem. > > That patch fixed the "Bad page state" errors and i do not see > "put_kernel_page: page at 0xe0... not in reserved memory" messages > any more, but the "Bad pte" error still haunts me. The other > two patches do not look like related. > > I will try Linus' git tree tomorrow as Horms suggested.
I wonder if there is any chance that SGI could make an SN machine available (remotely) for me to do some testing. It might make it easier for us to move forward on making sure the kernel and kexec-tools-testing work (and keep working) on SN. -- Horms H: http://www.vergenet.net/~horms/ W: http://www.valinux.co.jp/en/ _______________________________________________ fastboot mailing list [email protected] https://lists.osdl.org/mailman/listinfo/fastboot
