>>> On 03.05.17 at 14:00, wrote:
> On 03/05/17 11:44, Razvan Cojocaru wrote:
>> On 05/03/17 12:30, Jan Beulich wrote:
>> On 03.05.17 at 11:21, wrote:
At 10:15 +0100 on 03 May (1493806508), Tim Deegan wrote:
> At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
>> +else
On 03/05/17 11:44, Razvan Cojocaru wrote:
> On 05/03/17 12:30, Jan Beulich wrote:
> On 03.05.17 at 11:21, wrote:
>>> At 10:15 +0100 on 03 May (1493806508), Tim Deegan wrote:
At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
> +else if ( ctxt.cur > sizeof(*desc) )
>
On 05/03/17 12:30, Jan Beulich wrote:
On 03.05.17 at 11:21, wrote:
>> At 10:15 +0100 on 03 May (1493806508), Tim Deegan wrote:
>>> At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
+else if ( ctxt.cur > sizeof(*desc) )
{
uint32_t off;
-con
Hi Jan,
On 02/05/17 16:43, Jan Beulich wrote:
On 02.05.17 at 17:21, wrote:
hvm_save_cpu_ctxt() returns success without writing any data into
hvm_domain_context_t when all VCPUs are offline. This can then crash
the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
"off < (ctxt.cur -
>>> On 03.05.17 at 11:21, wrote:
> At 10:15 +0100 on 03 May (1493806508), Tim Deegan wrote:
>> At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
>> > +else if ( ctxt.cur > sizeof(*desc) )
>> > {
>> > uint32_t off;
>> > -const struct hvm_save_descriptor *desc;
>> >
At 10:15 +0100 on 03 May (1493806508), Tim Deegan wrote:
> At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
> > +else if ( ctxt.cur > sizeof(*desc) )
> > {
> > uint32_t off;
> > -const struct hvm_save_descriptor *desc;
> >
> > -rv = -ENOENT;
> >
On 05/03/17 12:15, Tim Deegan wrote:
> At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
>> Hmm, with both of you being of that opinion, I've taken another
>> look. I think I see now why you think that way (this being data
>> from an internal producer, overflow/underflow are not a primary
>
At 00:31 -0600 on 03 May (1493771502), Jan Beulich wrote:
> Hmm, with both of you being of that opinion, I've taken another
> look. I think I see now why you think that way (this being data
> from an internal producer, overflow/underflow are not a primary
> concern), so I'll withdraw my objection t
>>> On 02.05.17 at 18:11, wrote:
> On 02/05/17 17:02, Tim Deegan wrote:
>> At 18:21 +0300 on 02 May (1493749307), Razvan Cojocaru wrote:
>>> hvm_save_cpu_ctxt() returns success without writing any data into
>>> hvm_domain_context_t when all VCPUs are offline. This can then crash
>>> the hypervisor
On 05/02/2017 07:11 PM, Andrew Cooper wrote:
> On 02/05/17 17:02, Tim Deegan wrote:
>> At 18:21 +0300 on 02 May (1493749307), Razvan Cojocaru wrote:
>>> hvm_save_cpu_ctxt() returns success without writing any data into
>>> hvm_domain_context_t when all VCPUs are offline. This can then crash
>>> the
On 02/05/17 17:02, Tim Deegan wrote:
> At 18:21 +0300 on 02 May (1493749307), Razvan Cojocaru wrote:
>> hvm_save_cpu_ctxt() returns success without writing any data into
>> hvm_domain_context_t when all VCPUs are offline. This can then crash
>> the hypervisor (with FATAL PAGE FAULT) in hvm_save_one
On 05/02/2017 06:41 PM, Jan Beulich wrote:
On 02.05.17 at 17:21, wrote:
>> hvm_save_cpu_ctxt() returns success without writing any data into
>> hvm_domain_context_t when all VCPUs are offline. This can then crash
>> the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
>> "off < (c
At 18:21 +0300 on 02 May (1493749307), Razvan Cojocaru wrote:
> hvm_save_cpu_ctxt() returns success without writing any data into
> hvm_domain_context_t when all VCPUs are offline. This can then crash
> the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
> "off < (ctxt.cur - sizeof(*de
>>> On 02.05.17 at 17:21, wrote:
> hvm_save_cpu_ctxt() returns success without writing any data into
> hvm_domain_context_t when all VCPUs are offline. This can then crash
> the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
> "off < (ctxt.cur - sizeof(*desc))" for() test, where ctxt
>>> On 02.05.17 at 17:21, wrote:
> hvm_save_cpu_ctxt() returns success without writing any data into
> hvm_domain_context_t when all VCPUs are offline. This can then crash
> the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
> "off < (ctxt.cur - sizeof(*desc))" for() test, where ctxt
hvm_save_cpu_ctxt() returns success without writing any data into
hvm_domain_context_t when all VCPUs are offline. This can then crash
the hypervisor (with FATAL PAGE FAULT) in hvm_save_one() via the
"off < (ctxt.cur - sizeof(*desc))" for() test, where ctxt.cur remains 0,
causing an underflow which
16 matches
Mail list logo
