Save/restore records common to all x86 domain types (HVM, PV).
This is only the TSC_INFO record.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Add the context structure used to keep state during the save/restore
process.
Define the set of architecture or domain type specific operations with a
set of callbacks (save_ops, and restore_ops).
Add common functions for writing records.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Provide the C structures matching the binary (wire) format of the new
stream format. All header/record fields are naturally aligned and
explicit padding fields are used to ensure the correct layout (i.e.,
there is no need for any non-standard structure packing pragma or
attribute).
Provide some helper functions for converting types to string for
diagnostic purposes.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
For testing purposes, the environmental variable "XG_MIGRATION_V2" allows the
two save/restore codepaths to coexist, and have a runtime switch.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Progress information is logged via a different logger to regular libxc log
messages, and currently can only express a range. However, not everything
which needs reporting as progress comes with a range. Extend the interface to
allow reporting of a single statement.
The programming interface now looks like:
xc_set_progress_prefix()
set the prefix string to be used
xc_report_progress_single()
report a single action
xc_report_progress_step()
report $X of $Y
The new programming interface is implemented in a compatible way with the
existing caller interface (by reporting a single action as "0 of 0").
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
This implementation of writev_exact() will cope with an iovcnt greater than
IOV_MAX because glibc will actually let this work anyway, and it is very
useful not to have to work about this in the caller of writev_exact(). The
caller is still required to ensure that the sum of iov_len's doesn't overflow
a ssize_t.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
__p2m_lookup should be called with the p2m lock taken. Add an ASSERT in
order to catch wrong caller in debug build.
Signed-off-by: Julien Grall <julien.grall@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Convert all plain ints and strings which are used for xenbus "state"
files to xenbus_state. This makes it easier to find code which deals
with backend/frontend state changes.
Convert usage of libxl__sprintf to GCSPRINTF.
No change in behaviour is expected by this change, beside a small
increase of runtime memory usage in places that used a string constant.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Cc: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
When resuming, the guest needs to check whether the port has changed. HVM
guests use this parameter to get the port number.
(We can't always use xenstore where this value is also written: for example
on Linux the console is resumed very early, before the store is up).
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
This patch firstly enables EPT A/D bits if PML is used, as PML depends on EPT
A/D bits to work. A bit is set for all present p2m types in middle and leaf EPT
entries, and D bit is set for all writable types in EPT leaf entry, except for
log-dirty type with PML.
With PML, for 4K pages, instead of setting EPT entry to read-only, we just need
to clear D bit in order to log that GFN. For superpages, we still need to set it
to read-only as we need to split superpage to 4K pages in EPT violation.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
Using PML, it's possible there are dirty GPAs logged in vcpus' PML buffers
when userspace peek/clear dirty pages, therefore we need to flush them befor
reporting dirty pages to userspace. This applies to both video ram tracking and
paging_log_dirty_op.
This patch adds new p2m layer functions to enable/disable PML and flush PML
buffers. The new functions are named to be generic to cover potential futher
PML-like features for other platforms.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
It's possible domain still remains in log-dirty mode when it is about to be
destroyed, in which case we should manually disable PML for it.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
It's possible domain has already been in log-dirty mode when creating vcpu, in
which case we should enable PML for this vcpu if PML has been enabled for the
domain.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
This patch adds help functions to enable/disable PML, and flush PML buffer for
single vcpu and particular domain for further use.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
A new 4K page pointer is added to arch_vmx_struct as PML buffer for vcpu. And a
new 'status' field is added to vmx_domain to indicate whether PML is enabled for
the domain or not.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
The patch adds PML definition and feature detection. Note PML won't be detected
if PML is disabled from boot parameter. PML is also disabled in construct_vmcs,
as it will only be enabled when domain is switched to log dirty mode.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
PML logs GPA in PML buffer. Original paging_mark_dirty takes MFN as parameter
but it gets guest pfn internally and use guest pfn to as index for looking up
radix log-dirty tree. In flushing PML buffer, calling paging_mark_dirty directly
introduces redundant p2m lookups (gfn->mfn->gfn), therefore we introduce
paging_mark_gfn_dirty which is bulk of paging_mark_dirty but takes guest pfn as
parameter, and in flushing PML buffer we call paging_mark_gfn_dirty directly.
Original paging_mark_dirty then simply is a wrapper of paging_mark_gfn_dirty.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Acked-by: Tim Deegan <tim@xen.org>
A top level EPT parameter "ept=<options>" and a sub boolean "opt_pml_enabled"
are added to control PML. Other booleans can be further added for any other EPT
related features.
The document description for the new parameter is also added.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Tim Deegan <tim@xen.org>
Acked-by: Kevin Tian <kevin.tian@intel.com>
CMPXCHG: in the case of inequality of the rAX and the operand,
need to check CF, PF, AF, SF and OF flags as well.
This adjustment covers the fix of incorrect comparison during
CMPXCHG emulation.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
CMPXCHG sets CF, PF, AF, SF, and OF flags according to the results of the
comparison the rAX with the operand of the instruction.
rAX must be the first argument of the comparison (a minuend), the operand
must be the second one (a subtrahend).
Due to improper order of comparison arguments, CF, PF, AF, SF and OF flags were
set incorrectly in the case of inequality. Need to swap them.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Switching RMID from previous vcpu to next vcpu only needs to write
MSR_IA32_PSR_ASSOC once. Write it with the value of next vcpu is enough,
no need to write '0' first. Idle domain has RMID set to 0 and because MSR
is already updated lazily, so just switch it as it does.
Also move the initialization of per-CPU variable which used for lazy
update from context switch to CPU starting.
Signed-off-by: Chao Peng <chao.p.peng@linux.intel.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Dario Faggioli <dario.faggioli@citrix.com>
Prior to 1a09c5113a ("libxlu: rework internal representation of
setting") empty lists in config files did get accepted. Restore that
behavior.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Currently, unless a (non-default) ioreq server is explicitly disabled before
being destroyed, its gmfns will not be placed back into the p2m but still
released back into the ioreq_gmfn mask. This is somewhat counter-intuitive
and easily remedied by this small patch.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
hvm_free_ioreq_gmfn has the sense of the ioreq_gmfn mask inverted; it
needs to set a bit to release the gmfn, not clear it.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
hwdom_shutdown() operates with hardware domains, use the proper wording.
Eliminate pointless braces from switch cases.
Use hardware_domain->domain_id instead of hardware_domid to print the actual
domain ID as in some cases it can differ (e.g. Dom0 dies before the actual HW
domain got created, kexec for the HW domain is being performed,...).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
1aeb1156fa ("x86 don't change affinity with interrupt unmasked")
introducing RTE reads prior to the respective interrupt having got
enabled for the first time uncovered a bug in 2ca9fbd739 ("AMD IOMMU:
allocate IRTE entries instead of using a static mapping"): We obviously
shouldn't be translating RTEs for which remapping didn't get set up
yet.
Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
update_intremap_entry_from_msi() doesn't write to its data pointer on
some error paths, so we copying that variable into the msg would count
as undefined behaviour.
Signed-off-by: Tim Deegan <tim@xen.org>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Modifiers to the pointer passed into list_next_entry() are also being
applied to the macro's return type, and hence if the input pointer is
const-qualified a variable the result gets assigned to would also need
to be. As that doesn't seem desirable here, drop the const qualifier
on the input pointer instead.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Tim Deegan <tim@xen.org>
g++ 4.1.x dies with "cc1plus: error: output filename specified twice"
on the currently used construct. That's apparently due to it converting
the manually specified "c++" into "c++-header", and mis-handling that
(which, when using "c++-header" explicitly btw gets mis-handled even
with 4.9.x and also, using "c-header", by the plain C compiler).
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Tim Deegan <tim@xen.org>
Older gcc warns (and due to -Werror fails) on this ASSERT() now that
"node" is of unsigned type. Make it more useful at once.
Coverity-ID: 1055630
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Tim Deegan <tim@xen.org>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
None of these structures currently contain a hole. However, there is a risk
that a change to the structure might introduce a hole, and thus create a
hypervisor stack leak to the toolstack.
Mitigate this risk by preemptively zeroing these structures. These are not
hotpaths, so the slight overhead is not an issue.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Just like the other messages in this function this one should be issued
through plain printk() - the current vCPU is irrelevant here. (Noticed
while backporting to older trees, which don't have gprintk().)
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Kevin Tian <kevin.tian@intel.com>
It was confusing GCC's uninitialized-variable detection.
Signed-off-by: Tim Deegan <tim@xen.org>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Always recheck with poll() right before making the callback.
All sorts of things may have happened since poll() originally signaled
the fd. We would like the main functional libxl code not to have to
worry about spurious wakeups.
In particular, this fixes a bug in the save/restore callout: the save
helper message reader operates with the fd in blocking mode. In a
multithreaded program one thread might have eaten all the messages out
of the fd while another one is busy returning from poll and reacquiring
the libxl lock, possibly resulting in a deadlock.
(Also, we abolish the anomalous direct caller of efd->func.)
Reported-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reported-by: Jim Fehlig <jfehlig@suse.com>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Jim Fehlig <jfehlig@suse.com>
CC: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Jim Fehlig <jfehlig@suse.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
No functional change, only code motion.
Currently, contrary to this function's name, there are two sites where
efd->func() is called so one of them doesn't go through here just yet.
That will be dealt with in the next commit.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Jim Fehlig <jfehlig@suse.com>
CC: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Jim Fehlig <jfehlig@suse.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
For top-level targets which use more than a single program to produce content
(txt already, pdf once pandoc is supported), these current checks are
unsuitable.
By moving the the install checks to the rules which actually use the programs,
it is now possible to build a subset of a top-level target depending on the
installed programs.
As a bonus, it removes the need to recurse for txt, man-pages and pdf targets.
A side effect of this is that every individual source which cannot be
generated will have a specific message logged, giving the file and program.
As such, these message are updated to consistently report the target file
which was not generated.
Finally, update "ifdef foo" to "ifneq($(foo),)" to be more resilient to errors
caused by having foo defined as an empty string.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
On creation time, tap and xnb interfaces are created with an mtu of
1500 bytes, assuming the bridge will have the same value.
Instead, check the bridge mtu and configure the new xnb or
tap interface with the same value.
The tools used are sed and ifconfig, both included on base. No need
to install additional ports (no new dependences).
Signed-off-by: Gustau Perez <gustau.perez@gmail.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Campbell <ian.cmapbell@citrix.com>
[ ijc -- clarified title ]
New functions to provide logical and and or of two bitmaps. These are
generically useful utility functions added to the public API for the
benefit of libxl's users.
In the future they may also be useful internally, e.g. in the
vNUMA configuration check function.
Signed-off-by: Linda Jacobson <lindaj@jma3.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- rewrote commit message and fixed typo ]
Andrew Cooper
Pranavkumar Sawargaonkar
Julien Grall
Olaf Hering
Boris Ostrovsky
Kai Huang
Eugene Korenevsky
Chao Peng
Jan Beulich
Paul Durrant
Vitaly Kuznetsov
Tim Deegan
Ian Jackson
Gustau Perez
Giuseppe Mazzotta
Wei Liu
Linda Jacobson
Tamas K Lengyel