On Sat, 2015-01-31 at 09:43 +0100, Paolo Bonzini wrote:
On 31/01/2015 00:55, Alex Williamson wrote:
Commit d8d95814609e replaced a number of memory_region_destroy()
calls with object_unparent() calls. The logic appears to be that
subregions need to be unparented, but the base region is
On 31/01/2015 00:55, Alex Williamson wrote:
Commit d8d95814609e replaced a number of memory_region_destroy()
calls with object_unparent() calls. The logic appears to be that
subregions need to be unparented, but the base region is destroyed
with the device object. Doing hotplug testing
On 31/01/2015 16:10, Alex Williamson wrote:
Explicit object_unparent() is only needed if you recreate the memory
region during the lifetime of the object. This is rarely needed, and it
is simple to spot if it's needed. If you do memory_region_init* outside
the realize function, most likely
Commit d8d95814609e replaced a number of memory_region_destroy()
calls with object_unparent() calls. The logic appears to be that
subregions need to be unparented, but the base region is destroyed
with the device object. Doing hotplug testing with vfio-pci I
occasionally get a segfault from
