I have no plans to attend HPCA at the moment. There are folks here who are using my original version of this patch and would like to check it into our internal version of gem5, but I think we should use a version of it that doesn't break anything else in the process. Unfortunately I don't have an SVM capable machine around to test on so I can't vet things myself, but if you guys have a patch you're confident doesn't break either Intel or AMD KVM I can pass that along. Please let me know if there's anything I can do to help.
Gabe On Tue, Jan 31, 2017 at 8:05 AM, Jason Lowe-Power <ja...@lowepower.com> wrote: > > > > On Jan. 30, 2017, 10:31 p.m., Gabe Black wrote: > > > Hi folks. I put together this patch a long time ago, there was some > discussion about it, and then I lost track of what was going on with it. > Have the issues it causes with AMD cpus been tracked down? Glancing through > this again, I think there are a couple things which are genuinely wrong > which should fix things regardless of the underlying CPU (like reversed > bits, things that shouldn't be set but are), but I wouldn't want to push it > if it would break some other things which are currently working. Please > update me on the current state of things. > > > > mike upton wrote: > > I developed another patch on top that had both AMD and intel > working. I cant remember the number, and it should certainly be retested > before merging. > > I will need to see if I can fins a current AMD box. > > > > So the answer to your question is: yes there were issues, the issues > were tracked down and there is a additional patch that gets both intel and > AMD SE virtualization working. The code has not been tested in 2 years > though. > > I've been using some version of these patches in my local repository for a > while now. I think I may have some patches on top of these to fix other > issues, too. > > I've added this as a possible project for a Sprint this Sunday ( > http://gem5.org/Sprint_Ideas#Push_in_fixes_to_x86_KVM). Any chance you're > going to be at HPCA, and want to work on this, Mike or Gabe? > > > - Jason > > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/2557/#review9350 > ----------------------------------------------------------- > > > On Dec. 10, 2014, 10:11 a.m., Gabe Black wrote: > > > > ----------------------------------------------------------- > > This is an automatically generated e-mail. To reply, visit: > > http://reviews.gem5.org/r/2557/ > > ----------------------------------------------------------- > > > > (Updated Dec. 10, 2014, 10:11 a.m.) > > > > > > Review request for Default. > > > > > > Repository: gem5 > > > > > > Description > > ------- > > > > Changeset 10606:aa3eb7453246 > > --------------------------- > > x86: kvm: Fix the KVM CPU in SE and FS on Intel CPUs. > > > > There were a number of problems with how things were initialized which > prevent > > VMX from running the simulation as a guest. > > > > > > Diffs > > ----- > > > > src/arch/x86/process.cc 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/arch/x86/regs/misc.hh 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/arch/x86/system.hh 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/arch/x86/system.cc 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/arch/x86/utility.hh 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/arch/x86/utility.cc 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > src/cpu/kvm/x86_cpu.cc 8fc6e7a835d1d313e139c9095251105f904ac1b4 > > > > Diff: http://reviews.gem5.org/r/2557/diff/ > > > > > > Testing > > ------- > > > > > > Thanks, > > > > Gabe Black > > > > > > _______________________________________________ > gem5-dev mailing list > gem5-dev@gem5.org > http://m5sim.org/mailman/listinfo/gem5-dev > _______________________________________________ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev
