* SEP regression in kvm-20?
@ 2007-04-23 4:56 Rusty Russell
[not found] ` <1177304168.17026.49.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
0 siblings, 1 reply; 4+ messages in thread
From: Rusty Russell @ 2007-04-23 4:56 UTC (permalink / raw)
To: kvm-devel
Hi all,
I note that Linux under kvm 20 doesn't use sysenter, while it did under
kvm 19. There's no "sep" in /proc/cpuinfo and the vendor id looks kinda
random too: this looks like a cpuid problem?
Cheers,
Rusty.
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
^ permalink raw reply [flat|nested] 4+ messages in thread[parent not found: <1177304168.17026.49.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>]
* Re: SEP regression in kvm-20? [not found] ` <1177304168.17026.49.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org> @ 2007-04-23 7:17 ` Avi Kivity [not found] ` <462C5DA7.1040407-atKUWr5tajBWk0Htik3J/w@public.gmane.org> 0 siblings, 1 reply; 4+ messages in thread From: Avi Kivity @ 2007-04-23 7:17 UTC (permalink / raw) To: Rusty Russell; +Cc: kvm-devel Rusty Russell wrote: > Hi all, > > I note that Linux under kvm 20 doesn't use sysenter, while it did under > kvm 19. There's no "sep" in /proc/cpuinfo x86info does report sep, but arch/i386/kernel/cpu/intel.c has this to say: /* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */ if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633) clear_bit(X86_FEATURE_SEP, c->x86_capability); whereas our cpu is cpu family : 6 model : 2 so, no SEP for you. I imagine we can bump the model and mask to 3. > and the vendor id looks kinda > random too: this looks like a cpuid problem? > I see a vendor_id : GenuineIntel ? -- error compiling committee.c: too many arguments to function ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ^ permalink raw reply [flat|nested] 4+ messages in thread
[parent not found: <462C5DA7.1040407-atKUWr5tajBWk0Htik3J/w@public.gmane.org>]
* Re: SEP regression in kvm-20? [not found] ` <462C5DA7.1040407-atKUWr5tajBWk0Htik3J/w@public.gmane.org> @ 2007-04-23 11:18 ` Rusty Russell [not found] ` <1177327093.17026.67.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org> 0 siblings, 1 reply; 4+ messages in thread From: Rusty Russell @ 2007-04-23 11:18 UTC (permalink / raw) To: Avi Kivity; +Cc: kvm-devel On Mon, 2007-04-23 at 10:17 +0300, Avi Kivity wrote: > Rusty Russell wrote: > > Hi all, > > > > I note that Linux under kvm 20 doesn't use sysenter, while it did under > > kvm 19. There's no "sep" in /proc/cpuinfo > > x86info does report sep, but arch/i386/kernel/cpu/intel.c has this to say: > > /* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it > until model 3 mask 3 */ > if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633) > clear_bit(X86_FEATURE_SEP, c->x86_capability); > > whereas our cpu is > > cpu family : 6 > model : 2 > > so, no SEP for you. I imagine we can bump the model and mask to 3. Hi Avi! Seems strange that this changed between kvm 19 and kvm 20. kvm-20/qemu/target-i386/helper2.c seems to offer 6/3/3. > > and the vendor id looks kinda > > random too: this looks like a cpuid problem? > > > > I see a > > vendor_id : GenuineIntel vendor_id here is a char 223: processor : 0 vendor_id : � cpu family : 0 model : 2 model name : 00/02 stepping : 8 cpu MHz : 2835.485 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 65535 wp : yes flags : fpu vme tsc mce up pni monitor ds_cpl vmx smx est tm2 ssse3 cid cx16 xtpr bogomips : 4207.41 clflush size : 32 FYI, Host is: processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz stepping : 6 cpu MHz : 2128.059 cache size : 2048 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 2 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr lahf_lm bogomips : 4258.53 clflush size : 64 Cheers, Rusty. ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ kvm-devel mailing list kvm-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/kvm-devel ^ permalink raw reply [flat|nested] 4+ messages in thread
[parent not found: <1177327093.17026.67.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>]
* Re: SEP regression in kvm-20? [not found] ` <1177327093.17026.67.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org> @ 2007-04-23 13:15 ` Avi Kivity 0 siblings, 0 replies; 4+ messages in thread From: Avi Kivity @ 2007-04-23 13:15 UTC (permalink / raw) To: Rusty Russell; +Cc: kvm-devel Rusty Russell wrote: > vendor_id here is a char 223: > > processor : 0 > vendor_id : � > I fixed an assembly constraint problem (& pushed to kernel.org) which may help with this. -- error compiling committee.c: too many arguments to function ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ kvm-devel mailing list kvm-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/kvm-devel ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2007-04-23 13:15 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-04-23 4:56 SEP regression in kvm-20? Rusty Russell
[not found] ` <1177304168.17026.49.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2007-04-23 7:17 ` Avi Kivity
[not found] ` <462C5DA7.1040407-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-04-23 11:18 ` Rusty Russell
[not found] ` <1177327093.17026.67.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2007-04-23 13:15 ` Avi Kivity
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox