that OFFLINE is only sent if the ONLINE event has been sent. Secondly, this
fixes one possible cause of the reopened bug #549 as it prevents be->netif
being dereferenced inside netback_uevent. Early reports suggest that this is
not the only cause of #549, so that bug remains open for now.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
the intention appears to have been 500msec, but struct timeval when used with
a select() is in usec, not msec. The timeout has been set to 250msec (500msec
seems a little long, in any case).
Signed-off-by: Ewan Mellor <ewan@xensource.com>
guest guest physical address, even when the guest paging is disabled.
The gdbserver-xen tries to access guest pdes & ptes to map memory of
hvm guest being debugged; and it gets a seg-fault because guest has not
setup it's paging yet. The attached patch adds guest paging state check,
so that the map_domain_va() can get the correct guest physical address
from guest va.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
only for x86. The x86 implementation of smp_call_function has been simplified
and is now based on on_slected_cpus().
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
Now that sysfs is mounted in the RAM disk of the test environment, the
test code does not need to do this anymore. This patch also simplifies
previous tests and the APIs used there. If prerequisites are not met
(vtpm manager not running) the patch will 'skip'.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
The default is preserved but can be overridden by either passing it with -r to runtest.sh or by setting the xmtest_repo environment variable.
Signed-off-by: Paul Larson <pl@us.ibm.com>
1. subdirs-y -> subdir-y (match Linux name, and also obj-y
is singular so this makes sense).
2. subdirs can be declared with or without trailing slash
3. As with Linux, subdirs can be declared in the obj-y list
but they must be distinguished by a trailing slash
Signed-off-by: Keir Fraser <keir@xensource.com>
genapic structure that Xen does not require (mostly because we do
not support Visual Workstation nor IBM/Sequent NUMA-Q).
Signed-off-by: Keir Fraser <keir@xensource.com>
Also change v_end type from unsigned long to unsigned long long, for we
will have PAE VMX guest with more than 4G memory on PAE xen.
Signed-off-by: Xin Li <xin.b.li@intel.com>
Also introduces some cleanups to the Xen build system.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
make "halt -p" and "shutdown -h now" work. Also added apm 0x05 CPU
idle, 0x08 Enable PM and 0x0f apm engage/disengage subfunctions.
Signed-off-by: Winston Wang <winston.l.wang@intel.com>
There are some below code segments in guest OS
1. Rsm psr.dt
...
2. itc.d r18
...
3. rfi
After executing instruction 1, domain is in metaphysical mode.
When executing instruction 2, VMM gets control to emulate this
instruction. Firstly VMM will try to get opcode, which may
trigger a tlb miss. At this time domain is in metaphysical mode
and the fault address is in region 5. vcpu_translate handles this
as normal guest metaphysical mode.
It's not correct; sometimes this will make dom0 hang.
cpu_translate should handle this situation as if
guest is not in metaphysical mode.
Signed-off-by: Anthony Xu <anthony.xu@intel.com>
