which would fail if it crossed a page boundary where the guest physical
pages were not contiguous.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Older x86_64 linux kernels use one l4 table per cpu and context switch by
changing an l4 entry pointing to an l3 table. If we're shadowing them
we need to pin l3 shadows to stop them being torn down on every
context switch. (But don't do this for normal 64bit guests).
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Chain hash buckets through the shadow page_info structs instead
of in separately allocated structures. This lets us get rid of
some xenheap allocations and a domain_crash_synchronous() call.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Use this to detect data accesses that straddle a page boundary
and fault on the second page. We cannot handle these properly.
Signed-off-by: Keir Fraser <keir@xensource.com>
This does not work reliably across gcc versions. Workaround this by attaching
the regparm attribute to a variable instead.
Also since guest_to_host_gpr_switch() updates the guest EAX on the stack there
is no need to do so in emulate_privileged_op() as well. Therefore io_emul stub
now returns void.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
This is necessary to be sure that the shadow mode code can always detect
writes to guest page tables.
This should fix Windows restore from hibernation on SVM platforms.
Signed-off-by: Tom Woller <thomas.woller@amd.com>
1. Override ISA IRQs 5,6,10,11 as active-low level-triggered.
These are 'wired' to PCI devices.
2. Limit PCI-ISA link routers to route to 5,6,10,11. Others
are active-high edge-triggered and so invalid to use for
PCI devices. In any case, four ISA IRQs is sufficient to
avoid unnecessary aliasing among four links.
Also remove PCI device 0 IRQ info. It's an embedded host bridge with
no interrupt sources.
Signed-off-by: Keir Fraser <keir@xensource.com>
The Xen IPI spec states that destroy, hard_shutdown, pause, resume, start, suspend, and unpause return void. This patch makes it so and permits proper parsing of the response by clients (libxen) that follow the spec.
Signed-off-by: Jim Fehlig <jfehlig@novell.com>
Thanks to Ewan Mellor, Anthony Liguori, and Daniel P. Berrange for
pointing out problems with and solutions for the problems with the
first version.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
The original /dev/mem (linux/drivers/char/mem.c) is aware of IA64
specific issues. On the other hand paravirtualized /dev/mem
(linux-2.6-xen-sparse/drivers/xen/char/mem.c) is simplifed not to be
aware of IA64. This patch makes it IA64-aware to use X on domain0/IA64.
With this patch and ia64-specific patches, X server can boot on domain0/IA64.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
arch/x86/mm.c and arch/x86/sched.c. Header files were also refactored:
arch specific code got moved to include/x86/arch_mm.h and
include/x86/sched_mm.h.
Signed-off-by: Dietmar Hahn <dietmar.hahn@fujitsu-siemens.com>
Signed-off-by: Grzegorz Milos <gm281@cam.ac.uk>
buffer support and auto translated mode support.
renamed xenoprof_get_buffer::buf_maddr, xenoprof_passive::buf_maddr to
xenoprof_get_buffer::buf_gmaddr, xenoprof_passive::buf_gmaddr
to support auto translated mode. With auto translated mode enabled,
it is gmaddr, not maddr.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Simplify the share function.
Signed-off-by: Keir Fraser <keir@xensource.com>
- compilation fix when CONFIG_OPROFILE=m with minor clean up
- don't export xenoprof_is_primary.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
TODO:
1. Fix IO-APIC ID to not conflict with LAPIC IDS.
2. Fix i8259 device model (seems to work already though!).
3. Add INTSRC overrides in MPBIOS and ACPI tables so
that PCI legacy IRQ routing always ends up at an
IO-APIC input with level trigger. Restricting link
routing to {5,6,10,11} and setting overrides for all
four of those would work.
Signed-off-by: Keir Fraser <keir@xensource.com>
devices' state when a VM configuration file is destroyed. Currently only
the vTPM device's state needs to be explicitly destroyed.
I am also surrounding the saving of the managed domain's configuration
with a try-catch.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
that it logs everything written to stderr by xc_save and xc_restore as
errors whereas in fact the vast majority of this output is
information/debug (and all actual errors are marked by the string ERROR:
at the start of the message) -- this is confusing to folks looking at
the logs and makes automated log analysis tricky.
Fix is to scan for the ERROR: string and log anything without it using
log.info instead.
Signed-off by: Simon Graham <Simon.Graham@stratus.com>
This allows us to use the old pagetables's linear maps in our
remove-writeable-mappings heuristics, fixing the same crash that
cset 12339 did, but still letting us do fast revoke of writeable
mappings of toplevel pagetables.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
