Ivy Town processors have slightly different properties
than Ivy Bridge processors, particuarly as socket count grows.
Here we add dedicated tables covering 1-2 socket,
3-4 socket, and > 4 socket IVT configurations.
This reduces the frequency of deep transitions on those systems,
which can impact throughput.
Signed-off-by: Len Brown <len.brown@intel.com>
Socket count determination needs to be done differently for us: The
relevant code runs in a pre-SMP initcall, and hence can't use
cpu_to_socket() (as only the boot CPU's cpu_data[].phys_proc_id got set
up by that time). As a replacement the patch introduces
apicid_to_socket(), thus estimating the socket count based on the boot
CPU's internal topology information (if all CPUs in a system are the
same, wich ought to be the common case, this estimate will be precise).
Note that apicid_to_socket() handles Intel topology determination only
for now, as the mwait-idle driver currently only supports Intel CPUs.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Add CPU ID for Atom N2600/N2800 processors. Datasheets indicate support
for this, detailed information about potential quirks or limitations are
missing, though. So we just reuse the definition for the previous ATOM
series. Tests on N2800 systems showed that this addition is fine an can
reduce power consumption by about 0.25 W (personally confirmed on Intel
DN2800MT).
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Bay Trail (BYT) is a family of Silvermont-core Atom Processor SOCs,
including the Intel Atom Processor Z36xxx and Z37xxx Series.
Although it shares the Silvermont core with Avoton,
BYT is optimized for mobile, and thus it supports
different power saving CPU idle states.
Note that not all versions of Bay Trail HW support all
of the states listed in the driver.
Signed-off-by: Len Brown <len.brown@intel.com>
Tested-by: Aubrey Li <aubrey.li@linux.intel.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Like acpi_idle, mwait-idle compared sub-state numbers
to the number of supported sub-states -- discarding
sub-states numbers that were numbered >= the number of states.
But some Bay Trail SOCs use sparse sub-state numbers,
so we can't make such a comparison if we are going
to access those states.
So now we simply check that _some_ sub-states are
supported for the given state, and assume that the
sub-state number in our driver is valid.
In practice, the driver is correct, and even if it were not,
the hardware clips invalid sub-state requests to valid ones.
No entries in the driver require this change,
but Bay Trail will need it.
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
So far for non-Intel CPUs struct cpuinfo_x86's apicid field didn't get
set, despite MCE code consuming it.
Do some formatting/ordering adjustment to the touched code at once.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
As identified in e5545fb6, likely() and unlikely() must convert their
expressions to booleans before comparing to 1 or 0, to avoid truncation
issues.
While editing this file, add inclusion guards.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
gic.h contains defines of vgic. Move these to vgic
header file. Also move inline functions and defines/macros
in vgic.c to vgic.h. These vgic functions and macros
can be used by other GIC HW specific VGIC drivers,
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@caviumnetworks.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Julien Grall <julien.grall@linaro.org>
vgic_irq_rank structure contains gic specific data elements.
Move this out of domain.h to new vgic header file vgic.h
Allocate memory dynamically in vgic driver.
This patch reduces the size of domain struct and helps to
keep domain struct within PAGE_SIZE when future GIC hw versions
are added
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@caviumnetworks.com>
Acked-by: Julien Grall <julien.grall@linaro.org>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
arch_domain in domain.h defines all the GIC registers
that needs to be saved/restored directly.
These GIC registers are GIC HW version specific. The number
of registers and size of registers varies with GIC version.
So move these registers to gic.h file and make a union of these
registers. This helps to define GIC HW version specific
structure to this union for later GIC versions.
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@caviumnetworks.com>
Acked-by: Julien Grall <julien.grall@linaro.org>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
GIC driver contains both generic and hardware specific low
level functionality in gic.c file.
With this patch, low level functionality is moved to separate
file gic-v2.c and generic code is kept in gic.c file
Callbacks are registered by low level driver with generic driver
and are called wherever required.
The locking mechanism is not changed.
This helps to separate generic and hardware functionality
and implement future hardware version drivers.
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@caviumnetworks.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Julien Grall <julien.grall@linaro.org>
Tested-by: Julien Grall <julien.grall@linaro.org>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
gic-v2 driver uses fixmap to map the registers.
Instead use ioremap to access mmio registers.
With this patch, gic-v2 register definitions are updated
to use absolute offset address instead of dividing the
register offset by 4.
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@caviumnetworks.com>
Acked-by: Julien Grall <julien.grall@linaro.org>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
This wasn't done upon the initial addition of the file since the
section processing needs some adjustment for this: We can't mark the
two arrays needing relocations as either __initdata or __initconst, as
both have the potential of creating section conflicts (with -fPIC the
compiler may want to mark the section writable due to the necessary
relocations, colliding with other __initconst uses, and if the compiler
chooses to use the provided section name, __initdata on a constant
object might collide with other __initdata uses). However, local data
known to be referenced only by init code/data can also safely be moved
into .init.*. Hence the respective logic is being tweaked such that it
won't complain on non-empty .*.local sections.
While at it also drop the non-local section names from
SPECIAL_DATA_SECTIONS - they can't be safely converted.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
While this was intended to only do cleanup (replace the two bogus
"ret |= " constructs, and a simple formatting correction), this now
also
- fixes the bit manipulations for size_order > 0
a) correct an off-by-one in the use of size_order for shifting (till
now double the requested size got invalidated)
b) in fact setting bit 12 and up if necessary (without which too
small a region might have got invalidated)
c) making them capable of dealing with regions of 4Gb size and up
- corrects the return value handling, such that a later iteration's
success won't clear an earlier iteration's error indication
- uses PCI_BDF2() instead of open coding it
- bail immediately on bad passed in invalidation type, rather than
repeatedly printing the same message for each ATS-capable device, at
once also no longer hiding that failure from the caller
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Yang Zhang <yang.z.zhang@intel.com>
If the "generation_id" option is set in the domain configuration,
generate and set a new random VM generation ID every time a domain is
created or restored.
xl lacks the infrastructure to fully track the lifecycle of VM images
as they are snapshotted and cloned (etc) so always using a new ID is
the safe option and ensures that a new one will be used where it matters.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Toolstacks may specify a VM generation ID using the u.hvm.ms_vm_genid
field in the libxl_domain_build_info structure, when creating a
domain.
The toolstack is responsible for providing the correct generation ID
according to the Microsoft specification (e.g., generating new random
ones with libxl_ms_vm_genid_generate() as appropriate when restoring).
Although the specification requires that a ACPI Notify event is raised
if the generation ID is changed, the generation ID is never changed
when the domain is in a state to receive such an event (it's either
newly created or suspended).
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Cc: David Scott <dave.scott@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Make the set of MCEs that xen-mceinj can inject by data-driven instead
of a whole pile of cut-and-paste functions.
There is no functional change.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Christoph Egger <chegger@amazon.de>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Correctly name --page option and make it clear that it takes a
parameter. Use consistent formatting for help message.
Remove unused options.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
err() calls perror() which appends a newline.
Make Lprintf() also append a newline so err() and Lprintf() are
consistent.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Christoph Egger <chegger@amazon.de>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
On FreeBSD ZVOLs/disks/partitions are char devices, so libxl must
allow it's usage. This is already taken care in the OS-specific bits
of libxl by libxl__try_phy_backend, so remove the generic gate.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Cc: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
blktap2 will be enabled by default on Linux systems only, also give
the oportunity to disable it from configure.
Remove the gate in the Makefile that only allows enabling blktap{1,2}
on Linux, since this is now done in a more flexible way on configure.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Cc: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- reran autogen.sh ]
XSA-93 adds a couple of new functions to trap coprocessor registers. They
unconditionally inject an undefined instruction to guest.
When debugging an OS at early stage, it may be hard to know why the guest
received an UNDEFINED. Add some debug message to help the developer when Xen
is built in debug mode.
Signed-off-by: Julien Grall <julien.grall@linaro.org>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
The function init_traps sets up the handler taken when Xen hits a
BUG_ON/ASSERT.
If an error happen before init_traps is called, we lose the backtrace.
When a trap is taken from HYP mode, we need to:
- get the processor ID, percpu has to be initialized and the
smp_processor_id correctly set.
- initialize current, even though it's not used for the time being. Move
the debug sanity earlier.
Signed-off-by: Julien Grall <julien.grall@linaro.org>
Tested-by: Naresh Bhat <naresh.bhat@linaro.org>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- fixed typos and rewrapped commit log ]
The {GET,SET,MEMCPY,MEMSET_ARRAY}_FIELD() macros previously required
'dinfo->guest_width' to exist in scope, which is somewhat antisocial, and
makes them awkward to use in other contexts.
Update these macros to take guest width as a parameter, and in all cases pass
dinfo->guest_width.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
CC: Ian Campbell <Ian.Campbell@citrix.com>
CC: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
The VM generation ID support in libxc/libxl was based on a draft
specification which subsequently changed considerably. Remove much of
the code in anticipation of introducing something simpler that
conforms to the current specification from Microsoft.
Switch to using a HVM param (HVM_PARAM_VM_GENERATION_ID_ADDR) instead
of the hvmloader/generation-id-address XenStore key. This simplifies
save/restore since it only needs to transfer the value of this param.
There are no changes to the migration stream format or the public
libxl API.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- updated xc_nomigrate.c for interface change ]
HVM_PARAM_VM_GENERATION_ID_ADDR is the guest physical address of the
VM Generation ID. This parameter will be written by hvmloader and read
by the toolstack when updating the gen. ID (e.g., after restoring from a
snapshot).
A HVM parameter is easier for the save/restore code to work with (than
a XenStore key).
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
The random bytes are obtained from /dev/urandom and are suitable for
almost all uses (except for generating long-lived secure keys).
Documentation suggests that /dev/urandom is widely available on Unix-like
systems (such FreeBSD and NetBSD).
A public libxl_random_bytes() (or similar) could be trivially added,
if this required in the future.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
HVMOP_get_param and HVMOP_set_param take a uint32_t for the parameter
index and a uint64_t for the value. So add xc_hvm_param_get() and
xc_hvm_param_set() that take the same types.
The existing xc_get_hvm_param() and xc_set_hvm_param() are retained
for compatibility but are deprecated.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- dropped one hunk against xen-access.c which didn't apply due to
intervening changes ]
Testing whether the result of local_irq_save is zero doesn't work on
ARM.
Suggested by Julien Grall.
Signed-off-by: Thomas Leonard <talex5@gmail.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
The old code said "Xenbus tests disabled, because of a Xend bug".
Since xend doesn't exist any longer, we can enable these again.
Signed-off-by: Thomas Leonard <talex5@gmail.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
This allows closing any ports opened by the arch-specific code.
ARM will use it for the debug port.
Signed-off-by: Thomas Leonard <talex5@gmail.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Moves some x86-specific code into arch/x86.
Signed-off-by: Karim Allah Ahmed <karim.allah.ahmed@gmail.com>
[talex5@gmail.com: split into multiple patches]
Signed-off-by: Thomas Leonard <talex5@gmail.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Karim Allah Ahmed <karim.allah.ahmed@gmail.com>
[talex5@gmail.com: separated from big ARM commit]
[talex5@gmail.com: restored comment, moved prototypes to headers]
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
[talex5@gmail.com: restored stack address printk on x86]
[talex5@gmail.com: moved first printk's after start_info setup on x86]
Signed-off-by: Thomas Leonard <talex5@gmail.com>
Jan Beulich