>>> On 24.05.11 at 22:09, Keir Fraser <keir.xen@xxxxxxxxx> wrote:
> It depends on c/s 23142, and I think it isn't even needed in the absence of
> that other patch.
Correct - no checking on the context was done up to 23142, so as
long as that one doesn't get backported the fix to it obviously doesn't
need to be either.
It's odd in the first place that you're asking, as it sort of implies you've
been seeing -EOPNOTSUPP problems on 4.1, which you'd need to
explain/investigate independently of the problem addressed here.
Jan
> -- Keir
>
> On 24/05/2011 20:24, "AP Xen" <apxeng@xxxxxxxxx> wrote:
>
>> Is is possible for this (23348:3e8e1800d472) to be included in Xen 4.1.1?
>>
>> Thanks,
>> AP
>>
>> On Mon, May 16, 2011 at 6:48 PM, Shriram Rajagopalan <rshriram@xxxxxxxxx>
>> wrote:
>>> On Mon, May 16, 2011 at 8:02 AM, Jan Beulich <JBeulich@xxxxxxxxxx> wrote:
>>>>
>>>>>>> On 15.05.11 at 00:15, Shriram Rajagopalan <rshriram@xxxxxxxxx> wrote:
>>>>> This one works only for 64-bit domUs. 32bit domU (on 64bit dom0) fails
>>>>> with
>>>>> usual EOPNOTSUPP.
>>>>
>>>> Next (hopefully final) try below.
>>>>
>>>> Jan
>>>>
>>>> --- a/xen/arch/x86/domain.c
>>>> +++ b/xen/arch/x86/domain.c
>>>> @@ -856,6 +856,15 @@ int arch_set_info_guest(
>>>> goto out;
>>>> }
>>>>
>>>> + init_int80_direct_trap(v);
>>>> +
>>>> + /* IOPL privileges are virtualised. */
>>>> + v->arch.pv_vcpu.iopl = (v->arch.user_regs.eflags >> 12) & 3;
>>>> + v->arch.user_regs.eflags &= ~X86_EFLAGS_IOPL;
>>>> +
>>>> + /* Ensure real hardware interrupts are enabled. */
>>>> + v->arch.user_regs.eflags |= X86_EFLAGS_IF;
>>>> +
>>>> if ( !v->is_initialised )
>>>> {
>>>> v->arch.pv_vcpu.ldt_base = c(ldt_base);
>>>> @@ -863,11 +872,27 @@ int arch_set_info_guest(
>>>> }
>>>> else
>>>> {
>>>> - bool_t fail = v->arch.pv_vcpu.ctrlreg[3] != c(ctrlreg[3]);
>>>> + unsigned long pfn = pagetable_get_pfn(v->arch.guest_table);
>>>> + bool_t fail;
>>>>
>>>> + if ( !compat )
>>>> + {
>>>> + fail = xen_pfn_to_cr3(pfn) != c.nat->ctrlreg[3];
>>>> #ifdef CONFIG_X86_64
>>>> - fail |= v->arch.pv_vcpu.ctrlreg[1] != c(ctrlreg[1]);
>>>> + if ( pagetable_is_null(v->arch.guest_table_user) )
>>>> + fail |= c.nat->ctrlreg[1] || !(flags & VGCF_in_kernel);
>>>> + else
>>>> + {
>>>> + pfn = pagetable_get_pfn(v->arch.guest_table_user);
>>>> + fail |= xen_pfn_to_cr3(pfn) != c.nat->ctrlreg[1];
>>>> + }
>>>> + } else {
>>>> + l4_pgentry_t *l4tab = __va(pfn_to_paddr(pfn));
>>>> +
>>>> + pfn = l4e_get_pfn(*l4tab);
>>>> + fail = compat_pfn_to_cr3(pfn) != c.cmp->ctrlreg[3];
>>>> #endif
>>>> + }
>>>>
>>>> for ( i = 0; i < ARRAY_SIZE(v->arch.pv_vcpu.gdt_frames); ++i )
>>>> fail |= v->arch.pv_vcpu.gdt_frames[i] != c(gdt_frames[i]);
>>>> @@ -907,15 +932,6 @@ int arch_set_info_guest(
>>>> v->arch.pv_vcpu.ctrlreg[0] &= X86_CR0_TS;
>>>> v->arch.pv_vcpu.ctrlreg[0] |= read_cr0() & ~X86_CR0_TS;
>>>>
>>>> - init_int80_direct_trap(v);
>>>> -
>>>> - /* IOPL privileges are virtualised. */
>>>> - v->arch.pv_vcpu.iopl = (v->arch.user_regs.eflags >> 12) & 3;
>>>> - v->arch.user_regs.eflags &= ~X86_EFLAGS_IOPL;
>>>> -
>>>> - /* Ensure real hardware interrupts are enabled. */
>>>> - v->arch.user_regs.eflags |= X86_EFLAGS_IF;
>>>> -
>>>> cr4 = v->arch.pv_vcpu.ctrlreg[4];
>>>> v->arch.pv_vcpu.ctrlreg[4] = cr4 ? pv_guest_cr4_fixup(v, cr4) :
>>>> real_cr4_to_pv_guest_cr4(mmu_cr4_features);
>>>>
>>>>
>>> ok. this one works :). I ve tested with remus too :P. Can you please please
>>> spin out a patch to be pushed into the repo ? thanks for the help.
>>>
>>> shriram
>>>
>>> _______________________________________________
>>> Xen-devel mailing list
>>> Xen-devel@xxxxxxxxxxxxxxxxxxx
>>> http://lists.xensource.com/xen-devel
>>>
>>>
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home