Re: [Qemu-devel] Re: [PATCH] savevm: fix corruption in vmstate_subsection_load().

2011-01-24 Thread Yoshiaki Tamura
2011/1/20 Yoshiaki Tamura : > 2011/1/20 Paolo Bonzini : >> On 01/20/2011 09:57 AM, Yoshiaki Tamura wrote: >>> >>> 2011/1/20 Paolo Bonzini: On 12/14/2010 10:07 AM, Yoshiaki Tamura wrote: > > Although it's rare to happen in live migration, when the head of a > byte stream contai

Re: [Qemu-devel] Re: [PATCH] savevm: fix corruption in vmstate_subsection_load().

2011-01-20 Thread Yoshiaki Tamura
2011/1/20 Paolo Bonzini : > On 01/20/2011 09:57 AM, Yoshiaki Tamura wrote: >> >> 2011/1/20 Paolo Bonzini: >>> >>> On 12/14/2010 10:07 AM, Yoshiaki Tamura wrote: Although it's rare to happen in live migration, when the head of a byte stream contains 0x05 >>> >>> IIUC, this happens if

Re: [Qemu-devel] Re: [PATCH] savevm: fix corruption in vmstate_subsection_load().

2011-01-20 Thread Paolo Bonzini
On 01/20/2011 09:57 AM, Yoshiaki Tamura wrote: 2011/1/20 Paolo Bonzini: On 12/14/2010 10:07 AM, Yoshiaki Tamura wrote: Although it's rare to happen in live migration, when the head of a byte stream contains 0x05 IIUC, this happens if you have VMS_STRUCT and the field after the VMS_STRUCT sta

Re: [Qemu-devel] Re: [PATCH] savevm: fix corruption in vmstate_subsection_load().

2011-01-20 Thread Yoshiaki Tamura
2011/1/20 Paolo Bonzini : > On 12/14/2010 10:07 AM, Yoshiaki Tamura wrote: >> >> Although it's rare to happen in live migration, when the head of a >> byte stream contains 0x05 > > IIUC, this happens if you have VMS_STRUCT and the field after the VMS_STRUCT > starts with 0x5. > > I think you should

[Qemu-devel] Re: [PATCH] savevm: fix corruption in vmstate_subsection_load().

2011-01-20 Thread Paolo Bonzini
On 12/14/2010 10:07 AM, Yoshiaki Tamura wrote: Although it's rare to happen in live migration, when the head of a byte stream contains 0x05 IIUC, this happens if you have VMS_STRUCT and the field after the VMS_STRUCT starts with 0x5. I think you should also add this in vmstate_subsection_loa