|
Bugzilla – Full Text Bug Listing |
| Summary: | SMP kernel installed even on non-SMP systems | ||
|---|---|---|---|
| Product: | [openSUSE] SUSE Linux 10.1 | Reporter: | Michael Stather <kontakt> |
| Component: | Installation | Assignee: | Lukas Ocilka <locilka> |
| Status: | RESOLVED DUPLICATE | QA Contact: | Klaus Kämpf <kkaempf> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | suse-beta |
| Version: | Beta 8 | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | Other | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Michael Stather
2006-03-13 18:12:19 UTC
I'd say it makes more sense installing the correct kernel from the beginning. Please attach `hwinfo --cpu' here. I didn´t know this was supported. What about switching to a multicore CPU, is this supported too? r098088:/home/michael # hwinfo --cpu 01: None 00.0: 10103 CPU [Created at cpu.290] Unique ID: rdCR.j8NaKXDZtZ6 Hardware Class: cpu Arch: Intel Vendor: "AuthenticAMD" Model: 15.47.2 "AMD Athlon(tm) 64 Processor 3200+" Features: fpu,vme,de,pse,tsc,msr,pae,mce,cx8,sep,mtrr,pge,mca,cmov,pat,pse36,clflush,mmx,fxsr,sse,sse2,syscall,nx,mmxext,fxsr_opt,lm,3dnowext,3dnow,pni,lahf_lm Clock: 2010 MHz Cache: 512 kb Config Status: cfg=new, avail=yes, need=no, active=unknown AFAIK this is not supported, but we are dealing with a feature request here anyway ;) The correct kernel could be identified by the installer quite easily. Reassigning to Jiri. OK, cool :) Detecting the correct kernel during installation is a great idea but my suggestion also is to create a little YaST2 module where you can change this after installing. For example when you update to a multicore system, or perhaps you could even offer optimized kernels, then. And if the optimized one doesn´t work the user can switch back to the regular one. I see the same problem -- SMP kernel gets installed on a IBM ThinkPad T40p and Dell Lattitude D600. Both are Pentium M based systems. T42p: 01: None 00.0: 10103 CPU [Created at cpu.290] Unique ID: rdCR.j8NaKXDZtZ6 Hardware Class: cpu Arch: Intel Vendor: "GenuineIntel" Model: 6.9.5 "Intel(R) Pentium(R) M processor 1600MHz" Features: <snip> Clock: 1600 MHz Cache: 1024 kb Config Status: cfg=new, avail=yes, need=no, active=unknown Dell D600: 01: None 00.0: 10103 CPU [Created at cpu.290] Unique ID: rdCR.j8NaKXDZtZ6 Hardware Class: cpu Arch: Intel Vendor: "GenuineIntel" Model: 6.9.5 "Intel(R) Pentium(R) M processor 1700MHz" Features: <snip> Clock: 1700 MHz Cache: 1024 kb Config Status: cfg=new, avail=yes, need=no, active=unknown Moving to Beta 8. Default installation actually installs both kernels kernel-default-2.6.16_rc6_git1-4 and kernel-smp-2.6.16_rc6_git1-4. /boot/vmlinuz points to the smp kernel. /boot/vmlinuz.previous exists and points to the default kernel; non-smp. I guess thats becaus both kernels get installed, first the non-smp and second the smp kernel which replaces the non-smp as defualt. Same here with Pentium M processor. It was a default KDE installation without any removed or added selections. Same here with Pentium 4 Mobile and another with Pentium M - also default KDE installtion. Please, attach YaST logs. http://en.opensuse.org/Bug_Reporting_FAQ#YaST Thanks Olaf (od), do you do anything with this in the Bootloader? I also see this on NLD. This is a blocker IMO -> setting to critical for now and adding NLD guy. I see this on two machines here one AMD64 and a AMD64 Turion. However, not all seem to be affected (neighbours laptop has -default with 10.1 Beta8). Will attach y2logs of that both. You find three y2logs here: ~trenn/Export/default_smp_mixup_on_Beta8_y2logs - One is a working config (10.1 Beta8) (default kernel is correctly used even CPU is nearly the same as the next) - One is a not working AMD64 Turion (NLD10 Beta8) (smp instead of default installed). - One is a not working AMD64 Turion (NLD10 Beta8) (smp instead of default installed). I reassign this to you Ocilka for now, please reassign if this is wrong. The Needinfo flag on michaelstather was wrong anyway. Isn't this simply a duplicate of 156513 ?? Klaus, I hope so ;) Thanks *** This bug has been marked as a duplicate of 156513 *** |