I tried latest xen and linux-xen staging tree, but failed to run HVM domain on x86-64 environment. domU creation is OK. However the weird thing is not HVM domain itself. Instead system crashed on dom0 context. I saw once with some stack dump that xen''s page fault handler is executed on a dom0''s stack which then causes nested page fault due to unable to fetch vcpu pointer. But I''m curious why stack switch doesn''t happen? Does anybody else observe same failure? I''m still debugging it and send out early in case it''s my environment issue. Thanks, Kevin _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Cui, Dexuan
2007-Jul-19 06:45 UTC
RE: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
It seems some c/s between 15516 and 15604 breaks x86_64 Xen. IA32pae Xen c/s 15605 is OK. -- Dexuan -----Original Message----- From: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin Sent: 2007年7月19日 14:32 To: xen-devel@lists.xensource.com Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen I tried latest xen and linux-xen staging tree, but failed to run HVM domain on x86-64 environment. domU creation is OK. However the weird thing is not HVM domain itself. Instead system crashed on dom0 context. I saw once with some stack dump that xen''s page fault handler is executed on a dom0''s stack which then causes nested page fault due to unable to fetch vcpu pointer. But I''m curious why stack switch doesn''t happen? Does anybody else observe same failure? I''m still debugging it and send out early in case it''s my environment issue. Thanks, Kevin _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Keir Fraser
2007-Jul-19 08:20 UTC
Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
I definitely am able to boot a 32-bit HVM guest on 64-bit Xen with current unstable. -- Keir On 19/7/07 07:45, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> It seems some c/s between 15516 and 15604 breaks x86_64 Xen. > IA32pae Xen c/s 15605 is OK. > > -- Dexuan > > -----Original Message----- > From: xen-devel-bounces@lists.xensource.com > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin > Sent: 2007年7月19日 14:32 > To: xen-devel@lists.xensource.com > Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen > > I tried latest xen and linux-xen staging tree, but failed to run HVM > domain on x86-64 environment. domU creation is OK. > > However the weird thing is not HVM domain itself. Instead system > crashed on dom0 context. I saw once with some stack dump that > xen''s page fault handler is executed on a dom0''s stack which then > causes nested page fault due to unable to fetch vcpu pointer. But > I''m curious why stack switch doesn''t happen? > > Does anybody else observe same failure? I''m still debugging it and > send out early in case it''s my environment issue. > > Thanks, > Kevin > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Cui, Dexuan
2007-Jul-19 08:26 UTC
RE: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
I''m using a 64-bit Linux HVM guest on the latest 64-bit Xen. -- Dexuan -----Original Message----- From: Keir Fraser [mailto:keir@xensource.com] Sent: 2007年7月19日 16:21 To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen I definitely am able to boot a 32-bit HVM guest on 64-bit Xen with current unstable. -- Keir On 19/7/07 07:45, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> It seems some c/s between 15516 and 15604 breaks x86_64 Xen. > IA32pae Xen c/s 15605 is OK. > > -- Dexuan > > -----Original Message----- > From: xen-devel-bounces@lists.xensource.com > [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin > Sent: 2007年7月19日 14:32 > To: xen-devel@lists.xensource.com > Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen > > I tried latest xen and linux-xen staging tree, but failed to run HVM > domain on x86-64 environment. domU creation is OK. > > However the weird thing is not HVM domain itself. Instead system > crashed on dom0 context. I saw once with some stack dump that > xen''s page fault handler is executed on a dom0''s stack which then > causes nested page fault due to unable to fetch vcpu pointer. But > I''m curious why stack switch doesn''t happen? > > Does anybody else observe same failure? I''m still debugging it and > send out early in case it''s my environment issue. > > Thanks, > Kevin > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Keir Fraser
2007-Jul-19 08:31 UTC
Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
How far does it get? K. On 19/7/07 09:26, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> I''m using a 64-bit Linux HVM guest on the latest 64-bit Xen. > > -- Dexuan > > -----Original Message----- > From: Keir Fraser [mailto:keir@xensource.com] > Sent: 2007年7月19日 16:21 > To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com > Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen > > I definitely am able to boot a 32-bit HVM guest on 64-bit Xen with current > unstable. > > -- Keir > > On 19/7/07 07:45, "Cui, Dexuan" <dexuan.cui@intel.com> wrote: > >> It seems some c/s between 15516 and 15604 breaks x86_64 Xen. >> IA32pae Xen c/s 15605 is OK. >> >> -- Dexuan >> >> -----Original Message----- >> From: xen-devel-bounces@lists.xensource.com >> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin >> Sent: 2007年7月19日 14:32 >> To: xen-devel@lists.xensource.com >> Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen >> >> I tried latest xen and linux-xen staging tree, but failed to run HVM >> domain on x86-64 environment. domU creation is OK. >> >> However the weird thing is not HVM domain itself. Instead system >> crashed on dom0 context. I saw once with some stack dump that >> xen''s page fault handler is executed on a dom0''s stack which then >> causes nested page fault due to unable to fetch vcpu pointer. But >> I''m curious why stack switch doesn''t happen? >> >> Does anybody else observe same failure? I''m still debugging it and >> send out early in case it''s my environment issue. >> >> Thanks, >> Kevin >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Cui, Dexuan
2007-Jul-19 08:33 UTC
RE: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
Sometimes immediately after grub, and sometimes goes further. Anyway, once the HVM domain is created, it''s very likely for xen to crash on dom0''s context. At that time, because crashing cpu does infinitely page fault on guest stack, another cpu then block at smp_function_call and then whole system halts. Weird! -- Dexuan -----Original Message----- From: Keir Fraser [mailto:keir@xensource.com] Sent: 2007年7月19日 16:32 To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen How far does it get? K. On 19/7/07 09:26, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> I''m using a 64-bit Linux HVM guest on the latest 64-bit Xen. > > -- Dexuan > > -----Original Message----- > From: Keir Fraser [mailto:keir@xensource.com] > Sent: 2007年7月19日 16:21 > To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com > Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen > > I definitely am able to boot a 32-bit HVM guest on 64-bit Xen with current > unstable. > > -- Keir > > On 19/7/07 07:45, "Cui, Dexuan" <dexuan.cui@intel.com> wrote: > >> It seems some c/s between 15516 and 15604 breaks x86_64 Xen. >> IA32pae Xen c/s 15605 is OK. >> >> -- Dexuan >> >> -----Original Message----- >> From: xen-devel-bounces@lists.xensource.com >> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin >> Sent: 2007年7月19日 14:32 >> To: xen-devel@lists.xensource.com >> Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen >> >> I tried latest xen and linux-xen staging tree, but failed to run HVM >> domain on x86-64 environment. domU creation is OK. >> >> However the weird thing is not HVM domain itself. Instead system >> crashed on dom0 context. I saw once with some stack dump that >> xen''s page fault handler is executed on a dom0''s stack which then >> causes nested page fault due to unable to fetch vcpu pointer. But >> I''m curious why stack switch doesn''t happen? >> >> Does anybody else observe same failure? I''m still debugging it and >> send out early in case it''s my environment issue. >> >> Thanks, >> Kevin >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Keir Fraser
2007-Jul-19 09:52 UTC
Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen
I don''t actually have any 64-bit guests to hand to test with, but looking at the code it looks like I broke start_vmx() by causing vmx_save_host_msrs() to be called only on cpu0. Oops. -- Keir On 19/7/07 09:33, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> Sometimes immediately after grub, and sometimes goes further. Anyway, once the > HVM domain is created, it''s very likely for xen to crash on dom0''s context. At > that time, because crashing cpu does infinitely page fault on guest stack, > another cpu then block at smp_function_call and then whole system halts. > Weird! > > -- Dexuan > > > -----Original Message----- > From: Keir Fraser [mailto:keir@xensource.com] > Sent: 2007年7月19日 16:32 > To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com > Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen > > How far does it get? > > K. > > > On 19/7/07 09:26, "Cui, Dexuan" <dexuan.cui@intel.com> wrote: > >> I''m using a 64-bit Linux HVM guest on the latest 64-bit Xen. >> >> -- Dexuan >> >> -----Original Message----- >> From: Keir Fraser [mailto:keir@xensource.com] >> Sent: 2007年7月19日 16:21 >> To: Cui, Dexuan; Tian, Kevin; xen-devel@lists.xensource.com >> Subject: Re: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen >> >> I definitely am able to boot a 32-bit HVM guest on 64-bit Xen with current >> unstable. >> >> -- Keir >> >> On 19/7/07 07:45, "Cui, Dexuan" <dexuan.cui@intel.com> wrote: >> >>> It seems some c/s between 15516 and 15604 breaks x86_64 Xen. >>> IA32pae Xen c/s 15605 is OK. >>> >>> -- Dexuan >>> >>> -----Original Message----- >>> From: xen-devel-bounces@lists.xensource.com >>> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of Tian, Kevin >>> Sent: 2007年7月19日 14:32 >>> To: xen-devel@lists.xensource.com >>> Subject: [Xen-devel] Anyone succeeds HVM on latest x86-64 xen >>> >>> I tried latest xen and linux-xen staging tree, but failed to run HVM >>> domain on x86-64 environment. domU creation is OK. >>> >>> However the weird thing is not HVM domain itself. Instead system >>> crashed on dom0 context. I saw once with some stack dump that >>> xen''s page fault handler is executed on a dom0''s stack which then >>> causes nested page fault due to unable to fetch vcpu pointer. But >>> I''m curious why stack switch doesn''t happen? >>> >>> Does anybody else observe same failure? I''m still debugging it and >>> send out early in case it''s my environment issue. >>> >>> Thanks, >>> Kevin >>> >>> _______________________________________________ >>> Xen-devel mailing list >>> Xen-devel@lists.xensource.com >>> http://lists.xensource.com/xen-devel >>> >>> _______________________________________________ >>> Xen-devel mailing list >>> Xen-devel@lists.xensource.com >>> http://lists.xensource.com/xen-devel >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Reasonably Related Threads
- [Patch] the interface of invalidating qemu mapcache
- [Patch] Add NMI Injection and Pending Support in VMX
- A race condition introduced by changeset 15175: Re-init hypercall stubs page after HVM save/restore
- It seems the "machine check exception handling" breaks HVM guest
- [Patch] Fix the slow wall clock time issue in x64 SMP Vista