In theory, after Guest executes "cover", incomplete frame changes to complete
frame. So Guest should set INCOMPL to 0 just after "cover". At least before
guest psr.ic and psr.i are turned on.
Previously, only when Guest executes "rfi", INCOMPL is set to 0. The window
between "cover" and "rfi" causes trouble in below scenario.
1. Application A calls system call.
2. In OS breaks handler entry, INCOMPL is 0. Due to its system call,
Linux kernel doesn't execute "cover".
3. Before returning to Application A, schedule happens, Application B begins
to run.
4. A TLB miss happens on the context of B, this may make INCOMPL 1, before
Returning to B, (that means "rfi" is not executed, and INCOMPL is still 1)
schedule happens again. A resumes to run with INCOMPL 1 (this is incorrect now).
5. As mentioned before, this is system call, "cover" is executed in
ia64_leave_kernel path. Because INCOMPL is 1, this "cover" is not actually
executed, but this "cover" should be executed.
5. Thus application A's frame is destroyed. Issue appears.
From: Anthony Xu <anthony.xu@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Fix all remaining warnings generated by gas.
Enforce the no gas warnings by using --fatal-warnings gas flag.
Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
This patch fixed vm_summary info and provide correct max_itr_entry,
max_dtr_entry,impl_va_msb, rid_size and so on.
Signed-off-by: Zhang xiantao <xiantao.zhang@intel.com>
linux default implementations of dma_declare_coheremnt_memory() and its
families which is defined in include/linux/dma-mapping.h.
So those in pci-dma-xen.c are unnecessary. #ifdef out them.
PATCHNAME: pci-dma-xen-common
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
This small patch intends to provide correct total memory info for
control panel and fixed hardcode for that. The total memory info doesn't
include the memory FW used.
Signed-off-by : Zhang Xiantao <xiantao.zhang@intel.com>
Our patch fix domain0 boot panic on large memory system.
(e.g. amount of installed memory is 16GB)
Memory of domain0 is not reserved now.
Our patch can make this memory reserved.
And we clean up initrd_start of domain0.
Signed-off-by: Kouya Shimura <kouya@jp.fujitsu.com>
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
Move contents of switch_to macro from xensystem.h to context_switch function.
Initialize FPU on all processors. FPH is always enabled in Xen.
Speed up context-switch (a little bit!) by not enabling/disabling FPH.
Cleanup (unused function/variablesi/fields, debug printf...)
vmx_ia64_switch_to removed (was unused).
Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
Use sal_cache_flush to emulate SAL_CACHE_FLUSH instead of fc.
Only prints one vcpu_set_itc message, do not print set itv message.
Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
1. emulate ldpf8 in mmio
2. handle floating point register rotation in functions setfpreg and getfpreg
Signed-off-by: Anthony Xu <anthony.xu@intel.com>
This small patch output message of elf_sanity_check as x86.
Currently this panic is the only following message.
(XEN) *** LOADING DOMAIN 0 ***
(XEN)
(XEN) ****************************************
(XEN) Panic on CPU 0:
(XEN) Could not set up DOM0 guest OS
(XEN) ****************************************
(XEN)
(XEN) Reboot in five seconds...
(XEN) machine_restart called: spinning....
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
space handlers were making calls into ACPI code which uses semaphores.
Since semaphores can not be locked while atomic (because they could
sleep), I changed the way the PCI backend responds to requests from the
frontend. Previously, the virtual configuration space handlers ran in
the same context as the event channel interrupt handler (which was often
atomic if not always atomic). Now the interrupt handler schedules a
callback function (a bottom half) in the system work queue (keventd)
that will get called in process context at a slightly later time. This
allows the handlers in the virtual configuration space to run in process
context and to call any core PCI function regardless of whether it will
sleep or not.
Signed-off-by: Ryan Wilson <hap9@epoch.ncsc.mil>
controlled by the outcome of the table scan, if apic= was given on the
command line its effect should not be overridden here.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Since part of the VMX guest initialization is running in the context of
dom0, so when we want to make sure the operation of shadow ops is
always done by vcpu0 of the guest VMX domain, we should make sure
the current vcpu0 is running in guest domain and not in the dom0.
Signed-off-by: Xiaohui Xin xiaohui.xin@intel.com
Signed-off-by: Xin Li <xin.b.li@intel.com>
