Re: [PATCH 4/10] KVM: fix rcu usage in init_rmode_* functions

[Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>]

On 03/04/2011 07:43 PM, Jan Kiszka wrote:
> On 2011-03-04 11:58, Xiao Guangrong wrote:
>> fix:
>> [ 3494.671786] stack backtrace:
>> [ 3494.671789] Pid: 10527, comm: qemu-system-x86 Not tainted 2.6.38-rc6+ #23
>> [ 3494.671790] Call Trace:
>> [ 3494.671796]  [<ffffffff8107d15f>] ? lockdep_rcu_dereference+0x9d/0xa5
>> [ 3494.671826]  [<ffffffffa03498be>] ? kvm_memslots+0x6b/0x73 [kvm]
>> [ 3494.671834]  [<ffffffffa03499be>] ? gfn_to_memslot+0x16/0x4f [kvm]
>> [ 3494.671843]  [<ffffffffa0349a0d>] ? gfn_to_hva+0x16/0x27 [kvm]
>> [ 3494.671851]  [<ffffffffa034c1c4>] ? kvm_write_guest_page+0x31/0x83 [kvm]
>> [ 3494.671861]  [<ffffffffa034c230>] ? kvm_clear_guest_page+0x1a/0x1c [kvm]
>> [ 3494.671867]  [<ffffffffa015e321>] ? vmx_set_tss_addr+0x83/0x122 [kvm_intel]
>>
>> and:
>> [ 8328.789599] stack backtrace:
>> [ 8328.789601] Pid: 18736, comm: qemu-system-x86 Not tainted 2.6.38-rc6+ #23
>> [ 8328.789603] Call Trace:
>> [ 8328.789609]  [<ffffffff8107d15f>] ? lockdep_rcu_dereference+0x9d/0xa5
>> [ 8328.789621]  [<ffffffffa03988be>] ? kvm_memslots+0x6b/0x73 [kvm]
>> [ 8328.789628]  [<ffffffffa03989be>] ? gfn_to_memslot+0x16/0x4f [kvm]
>> [ 8328.789635]  [<ffffffffa0398a0d>] ? gfn_to_hva+0x16/0x27 [kvm]
>> [ 8328.789643]  [<ffffffffa039b1c4>] ? kvm_write_guest_page+0x31/0x83 [kvm]
>> [ 8328.789699]  [<ffffffffa039b230>] ? kvm_clear_guest_page+0x1a/0x1c [kvm]
>> [ 8328.789713]  [<ffffffffa0206d0e>] ? vmx_create_vcpu+0x316/0x3c8 [kvm_intel]
>>
>> Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
>> ---
>>  arch/x86/kvm/vmx.c |   10 +++++++---
>>  1 files changed, 7 insertions(+), 3 deletions(-)
>>
>> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
>> index e2b8c6b..d871ced 100644
>> --- a/arch/x86/kvm/vmx.c
>> +++ b/arch/x86/kvm/vmx.c
>> @@ -2399,9 +2399,10 @@ static int init_rmode_tss(struct kvm *kvm)
>>  {
>>  	gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT;
> 
> I bet you also wanted to remove this initialization.
> 

Ouch, thank you for reminding me of that!