... and, if it is not created, make the top level 'device' node in
xenstore writable by the guest instead.
The purpose and semantics of the suspend event channel node are explained
in xenstore-paths.pandoc [1]. It was originally introduced in xend by
commit 17636f47a4 "Teach xc_save to use event-channel-based domain
suspend if available.". Note that, because, the top-level frontend
'device' node was created writable by the guest in xend, there was no
need to explicitly create the 'suspend-event-channel' node as a writable
node.
However, libxl creates the 'device' node as read-only by the guest and so
explicit creation of the 'suspend-event-channel' node is necessary to make
it usable. This unfortunately has the side-effect of making some old
Windows PV drivers [2] cease to function. This is because they scan the top
level 'device' node, find the 'suspend' node and expect it to contain the
usual sub-nodes describing a PV frontend. When this is found not to be the
case, enumeration ceases and (because the 'suspend' node is observed before
the 'vbd' node) no system disk is enumerated. Windows will then crash with
bugcheck code 0x7B (missing system disk).
This patch adds a boolean 'xend_suspend_evtchn_compat' field into
libxl_create_info and a similarly named option in xl.cfg to set it.
If the value is true then the xenstore node is not created. Instead the
old xend behaviour of making top level device node writable by the guest is
re-instated. If the value is false (the default) then the current libxl
behaviour persists.
xenstore-paths.pandoc is also modified to say that the suspend event
channel node may not exist and, if it does not exist, then the guest may
create it. A note is also added concerning the writability of the top
level device node.
[1] https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/misc/xenstore-paths.pandoc;hb=HEAD#l177
[2] https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/5/html/para-virtualized_windows_drivers_guide/sect-para-virtualized_windows_drivers_guide-installing_and_configuring_the_para_virtualized_drivers-installing_the_para_virtualized_drivers
NOTE: While adding the new LIBXL_HAVE_CREATEINFO_... definition into
libxl.h, this patch corrects the previous stanza which erroneously
implies libxl_domain_create_info is a function.
Signed-off-by: Paul Durrant <paul@xen.org>
Reviewed-by: Ian Jackson <ian.jackson@eu.citrix.com>
Provide a #define for other cases; it didn't seem worthwhile to me to
introduce an IOMMU_INTREMAP Kconfig option at this point.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Paul Durrant <paul@xen.org>
There is now only one version of PVH implementation in Xen. Drop "v2" to
avoid confusion.
Signed-off-by: Wei Liu <wl@xen.org>
Acked-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
The written ABI states that %es will be set up, but libxc doesn't do so. In
practice, it breaks `rep movs` inside guests before they reload %es.
The written ABI doesn't mention %ss, but libxc does set it up. Having %ds
different to %ss is obnoxous to work with, as different registers have
different implicit segments.
Modify the spec to state that %ss is set up as a flat read/write segment.
This a) matches the Multiboot 1 spec, b) matches what is set up in practice,
and c) is the more sane behaviour for guests to use.
Fixes: 68e1183411 ('libxc: introduce a xc_dom_arch for hvm-3.0-x86_32 guests')
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Wei Liu <wl@xen.org>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
The names of the serial parameters use hyphens, not underscores.
Signed-off-by: Sarah Newman <srn@prgmr.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
This patch updates Kconfig to a more recent version of Kconfig, found
in Linux v5.4.0, 219d54332a09 ("Linux 5.4").
With the updated version of Kconfig, other changes are necessary to
avoid breaking the build.
Kconfig files:
- fix Kconfig files that where using option env=*:
Since Linux commit 104daea149c4 ("kconfig: reference environment
variables directly and remove 'option env='"), we can access the
environment directly via $() and "option env=" as been removed.
- CONFIG_EXPERT='y' will now appear in .config file if
XEN_CONFIG_EXPERT=y in the environment. The alternative is to change
"EXPERT" to "$(XEN_CONFIG_EXPERT)" in all Kconfig files.
Makefile:
- silentoldconfig target as been removed from Kconfig. To update
include/generated/autoconf.h, we need to use syncconfig target
instead.
Makefile.kconfig:
- Import newer needed code from Linux's Makefile.lib and
Kbuild.include and Makefile.build.
- Set Q to empty, Xen build system doesn't silence commands. Having Q
empty mean we can import stuff from Linux without having to remove the
leading $(Q) from build commands. And quiet='' means commands will be
echoed.
- Add $(PHONY) to .PHONY. Like it is intended by Kbuild.
Makefile.host is also updated and copied from Linux.
Dependency change:
- Now depends on flex/bison, maybe we could _shipped those files like
before. Linux doesn't do that anymore.
The .gitignore in kconfig/ has more entries, compared to upstream, for
file generated by Makefile.host.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
The CONTROL command (former DEBUG command) isn't specified in the
xenstore protocol doc. Add it.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Backport: 4.9+
DIRECTORY_PART was missing in docs/misc/xenstore.txt. Add it.
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Wei Liu <wl@xen.org>
Backport: 4.9+
Jürgen Groß <jgross@suse.com> points out that this is entirely wrong.
Adding the "Backport" tag so we find this revert too.
This reverts commit d34dc88098.
Backport: 4.9+
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
... specification.
This was added by commit 0ca64ed8 "xenstore: add support for reading
directory with many children" but not added to the specification at that
point. A version of xenstored supporting the command was first released
in Xen 4.9.
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Backport: 4.9+
Specify applicability and the default value. Also state that, in case of
EFI, the microcode update blob specified in the EFI cfg takes precedence
over `ucode=scan`, if the latter is specified on Xen commend line.
No functional changes.
Signed-off-by: Eslam Elnikety <elnikety@amazon.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Due to AMD and Hygon being unable to selectively trap CR4 bit modifications
running 32-bit PV guest inside PV shim comes with significant performance
hit. Moreover, for SMEP in particular every time CR4.SMEP changes on context
switch to/from 32-bit PV guest, it gets trapped by L0 Xen which then
tries to perform global TLB invalidation for PV shim domain. This usually
results in eventual hang of a PV shim with at least several vCPUs.
Since the overall security risk is generally lower for shim Xen as it being
there more of a defense-in-depth mechanism, choose to disable SMEP/SMAP in
it by default on AMD and Hygon unless a user chose otherwise.
Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
pandoc is currently failing to generate the pdf with the following
error:
! Undefined control sequence.
l.1048 metadata string format is: key=value\0
In this case, we want to print \0 so we need to backslash-escape the
first character.
Interestingly pandoc will not complain when creating html and will just
ignore \0 completely.
Fixes: 5083e0ff93 ("livepatch: Add metadata runtime retrieval mechanism")
Signed-off-by: Julien Grall <julien@xen.org>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Reviewed-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
At the moment, the diagram is only indented with 2 spaces. So pandoc
will try to badly interpret it and not display it correctly.
Fix it by indenting all the block by 4 spaces (i.e an extra 2 spaces).
Fixes: d661611d08 ("docs/markdown: Switch to using pandoc, and fix underscore escaping")
Signed-off-by: Julien Grall <julien@xen.org>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
Cluster mode can only be used with interrupt remapping support, since
the top 16bits of the APIC ID are filled with the cluster ID, and
hence on systems where the physical ID is still smaller than 255 the
cluster ID is not. Force x2APIC to use physical mode if there's no
interrupt remapping support.
Note that this requires a further patch in order to enable x2APIC
without interrupt remapping support.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
c/s 5de961d9c0 "x86: do not enable global pages when virtualized on AMD or
Hygon hardware" in fact does. Fix the calculation in pge_init().
While fixing this, adjust the command line documenation, first to use the
newer style, and to expand the description to discuss cases where the option
might be useful to use, but Xen can't account for by default.
Fixes: 5de961d9c0 ('x86: do not enable global pages when virtualized on AMD or Hygon hardware')
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
And remove all mention of it in docs. It hasn't been used since
9ead9afcb9 ("Add configure --with-sysconfig-leaf-dir=SUBDIR to set
CONFIG_LEAF_DIR").
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Extend the livepatch list operation to fetch also payloads' metadata.
This is achieved by extending the sysctl list interface with 2 extra
guest handles:
* metadata - an array of arbitrary size strings
* metadata_len - an array of metadata strings' lengths (uin32_t each)
Payloads' metadata is a string of arbitrary size and does not have an
upper bound limit. It may also vary in size between payloads.
In order to let the userland allocate enough space for the incoming
data add a metadata total size field to the list sysctl operation and
fill it with total size of all payloads' metadata.
Extend the libxc to handle the metadata back-to-back data transfers
as well as metadata length array data transfers.
The xen-livepatch userland tool is extended to always display the
metadata for each received module. The metadata is received with the
following format: key=value\0key=value\0...key=value\0. The format is
modified to the following one: key=value;key=value;...key=value.
The new format allows to easily parse the metadata for a given module
by a machine.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Reviewed-by: Norbert Manthey <nmanthey@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
The payloads' name strings can be of arbitrary size (typically small
with an upper bound of XEN_LIVEPATCH_NAME_SIZE).
Current implementation of the list operation interface allows to copy
names in the XEN_LIVEPATCH_NAME_SIZE chunks regardless of its actual
size and enforces space allocation requirements on userland tools.
To unify and simplify the interface, handle the name strings of
arbitrary size by copying them in adhering chunks to the userland.
In order to let the userland allocate enough space for the incoming
data add an auxiliary interface xc_livepatch_list_get_sizes() that
provides the current number of payload entries and the total size of
all name strings. This is achieved by extending the sysctl list
interface with an extra fields: name_total_size.
The xc_livepatch_list_get_sizes() issues the livepatch sysctl list
operation with the nr field set to 0. In this mode the operation
returns the number of payload entries and calculates the total sizes
for all payloads' names.
When the sysctl operation is issued with a non-zero nr field (for
instance with a value obtained earlier with the prior call to the
xc_livepatch_list_get_sizes()) the new field name_total_size provides
the total size of actually copied data.
Extend the libxc to handle the name back-to-back data transfers.
The xen-livepatch tool is modified to start the list operation with a
call to the xc_livepatch_list_get_sizes() to obtain the actual number
of payloads as well as the necessary space for names.
The tool now always requests the actual number of entries and leaves
the preemption handling to the libxc routine. The libxc still returns
'done' and 'left' parameters with the same semantic allowing the tool
to detect anomalies and react to them. At the moment it is expected
that the tool receives the exact number of entries as requested.
The xen-livepatch tool has been also modified to handle the name
back-to-back transfers correctly.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Bjoern Doebel <doebel@amazon.de>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
This is the initial implementation of the expectations enhancement
to improve inline asm livepatching.
Expectations are designed as optional feature, since the main use of
them is planned for inline asm livepatching. The flag enabled allows
to control the expectation state.
Each expectation has data and len fields that describe the data
that is expected to be found at a given patching (old_addr) location.
The len must not exceed the data array size. The data array size
follows the size of the opaque array, since the opaque array holds
the original data and therefore must match what is specified in the
expectation (if enabled).
The payload structure is modified as each expectation structure is
part of the livepatch_func structure and hence extends the payload.
Each expectation is checked prior to the apply action (i.e. as late
as possible to check against the most current state of the code).
For the replace action a new payload's expectations are checked AFTER
all applied payloads are successfully reverted, but BEFORE new payload
is applied. That breaks the replace action's atomicity and in case of
an expectation check failure would leave a system with all payloads
reverted. That is obviously insecure. Use it with caution and act
upon replace errors!
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Reviewed-by: Norbert Manthey <nmanthey@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Livepatch only tracks an entire payload applied/reverted state. But,
with an option to supply the apply_payload() and/or revert_payload()
functions as optional hooks, it becomes possible to intermix the
execution of the original apply_payload()/revert_payload() functions
with their dynamically supplied counterparts.
It is important then to track the current state of every function
being patched and prevent situations of unintentional double-apply
or unapplied revert.
To support that, it is necessary to extend public interface of the
livepatch. The struct livepatch_func gets additional field holding
the applied/reverted state marker.
To reflect the livepatch payload ABI change, bump the version flag
LIVEPATCH_PAYLOAD_VERSION up to 2.
[And also update the top of the design document]
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Bjoern Doebel <doebel@amazon.de>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Julien Grall <julien.grall@arm.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
By default, in the quiescing zone, a livepatch payload is applied with
apply_payload() and reverted with revert_payload() functions. Both of
the functions receive the payload struct pointer as a parameter. The
functions are also a place where standard 'load' and 'unload' module
hooks are executed.
To increase livepatching system's agility and provide more flexible
long-term livepatch solution, allow to overwrite the default apply
and revert action functions with hook-like supplied alternatives.
The alternative functions are optional and the default functions are
used by default.
Since the alternative functions have direct access to the livepatch
payload structure, they can better control context of the 'load' and
'unload' hooks execution as well as exact instructions replacement
workflows. They can be also easily extended to support extra features
in the future.
To simplify the alternative function generation move code responsible
for payload and livepatch region registration outside of the function.
That way it is guaranteed that the registration step occurs even for
newly supplied functions.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Petre Eftime <epetre@amazon.com>
Reviewed-by: Martin Pohlack <mpohlack@amazon.com>
Reviewed-by: Norbert Manthey <nmanthey@amazon.com>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Bjoern Doebel <doebel@amazon.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
This is an implementation of 4 new livepatch module vetoing hooks,
that can be optionally supplied along with modules.
Hooks that currently exists in the livepatch mechanism aren't agile
enough and have various limitations:
* run only from within a quiescing zone
* cannot conditionally prevent applying or reverting
* do not have access to the module context
To address these limitations the following has been implemented:
1) pre-apply hook
runs before the apply action is scheduled for execution. Its main
purpose is to prevent from applying a livepatch when certain
expected conditions aren't met or when mutating actions implemented
in the hook fail or cannot be executed.
2) post-apply hook
runs after the apply action has been executed and quiescing zone
exited. Its main purpose is to provide an ability to follow-up on
actions performed by the pre- hook, when module application was
successful or undo certain preparation steps of the pre- hook in
case of a failure. The success/failure error code is provided to
the post- hooks via the rc field of the payload structure.
3) pre-revert hook
runs before the revert action is scheduled for execution. Its main
purpose is to prevent from reverting a livepatch when certain
expected conditions aren't met or when mutating actions implemented
in the hook fail or cannot be executed.
4) post-revert hook
runs after the revert action has been executed and quiescing zone
exited. Its main purpose is to perform cleanup of all previously
executed mutating actions in order to restore the original system
state from before the current module application.
The success/failure error code is provided to the post- hooks via
the rc field of the payload structure.
The replace action performs atomically the following actions:
- revert all applied modules
- apply a single replacement module.
With the vetoing hooks in place various inter-hook dependencies may
arise. Also, during the revert part of the operation certain vetoing
hooks may detect failing conditions that previously were satisfied.
That could in turn lead to situation when the revert part must be
rolled back with all the pre- and post- hooks re-applied, which again
can't be guaranteed to always succeed.
The simplest response to this complication is to disallow the replace
action completely on modules with vetoing hooks.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Petre Eftime <epetre@amazon.com>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Reviewed-by: Norbert Manthey <nmanthey@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
By default Livepatch enforces the following buildid-based dependency
chain between livepatch modules:
1) first module depends on given hypervisor buildid
2) every consecutive module depends on previous module's buildid
This way proper livepatch stack order is maintained and enforced.
While it is important for production livepatches it limits agility and
blocks usage of testing or debug livepatches. These kinds of livepatch
modules are typically expected to be loaded at any time irrespective
of current state of the modules stack.
To enable testing and debug livepatches allow user dynamically ignore
the inter-modules dependency. In this case only hypervisor buildid
match is verified and enforced.
To allow userland pass additional paremeters for livepatch actions
add support for action flags.
Each of the apply, revert, unload and revert action gets additional
32-bit parameter 'flags' where extra flags can be applied in a mask
form.
Initially only one flag '--nodeps' is added for the apply action.
This flag modifies the default buildid dependency check as described
above.
The global sysctl interface input flag parameter is defined with a
single corresponding flag macro:
LIVEPATCH_ACTION_APPLY_NODEPS (1 << 0)
The userland xen-livepatch tool is modified to support the '--nodeps'
flag for apply and load commands. A general mechanism for specifying
more flags in the future for apply and other action is however added.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Eslam Elnikety <elnikety@amazon.de>
Reviewed-by: Petre Eftime <epetre@amazon.com>
Reviewed-by: Leonard Foerster <foersleo@amazon.de>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Reviewed-by: Norbert Manthey <nmanthey@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
This change is part of a independant stacked livepatch modules
feature. This feature allows to bypass dependencies between modules
upon loading, but still verifies Xen build ID matching.
In order to prevent (up)loading any livepatches built for different
hypervisor version as indicated by the Xen Build ID, add checking for
the payload's vs Xen's build id match.
To achieve that embed into every livepatch another section with a
dedicated hypervisor build id in it. After the payload is loaded and
the .livepatch.xen_depends section becomes available, perform the
check and reject the payload if there is no match.
Signed-off-by: Pawel Wieczorkiewicz <wipawel@amazon.de>
Reviewed-by: Andra-Irina Paraschiv <andraprs@amazon.com>
Reviewed-by: Bjoern Doebel <doebel@amazon.de>
Reviewed-by: Eslam Elnikety <elnikety@amazon.de>
Reviewed-by: Martin Pohlack <mpohlack@amazon.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Ross Lagerwall <ross.lagerwall@citrix.com>
When using global pages a full tlb flush can only be performed by
toggling the PGE bit in CR4, which is usually quite expensive in terms
of performance when running virtualized. This is specially relevant on
AMD or Hygon hardware, which doesn't have the ability to do selective
CR4 trapping, but can also be relevant on e.g. Intel if the underlying
hypervisor also traps accesses to the PGE CR4 bit.
In order to avoid this performance penalty, do not use global pages
when running virtualized on AMD or Hygon hardware. A command line option
'global-pages' is provided in order to allow the user to select whether
global pages will be enabled for PV guests.
The above figures are from a PV shim running on AMD hardware with
32 vCPUs:
PGE enabled, x2APIC mode:
(XEN) Global lock flush_lock: addr=ffff82d0804b01c0, lockval=1adb1adb, not locked
(XEN) lock:1841883(1375128998543), block:1658716(10193054890781)
Average lock time: 746588ns
Average block time: 6145147ns
PGE disabled, x2APIC mode:
(XEN) Global lock flush_lock: addr=ffff82d0804af1c0, lockval=a8bfa8bf, not locked
(XEN) lock:2730175(657505389886), block:2039716(2963768247738)
Average lock time: 240829ns
Average block time: 1453029ns
As seen from the above figures the lock and block time of the flush
lock is reduced to approximately 1/3 of the original value.
Note that XEN_MINIMAL_CR4 and mmu_cr4_features are not modified, and
thus global pages are left enabled for the hypervisor. This is not an
issue because the code to switch the control registers (cr3 and cr4)
already takes into account such situation and performs the necessary
flushes. The same already happens when using XPTI or PCIDE, as the
guest cr4 doesn't have global pages enabled in that case either.
Also note that the suspend and resume code is correct in writing
mmu_cr4_features into cr4 on resume, since that's the cr4 used by the
idle vCPU which is the context used by the suspend and resume routine.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
XSA-302 relies on the use of libxl's "assignable-add" feature to prepare
devices to be assigned to untrusted guests.
Unfortunately, this is not considered a strictly required step for
device assignment. The PCI passthrough documentation on the wiki
describes alternate ways of preparing devices for assignment, and
libvirt uses its own ways as well. Hosts where these alternate methods
are used will still leave the system in a vulnerable state after the
device comes back from a guest.
Default to always quarantining PCI devices, but provide a command line
option to revert back to prior behavior (such that people who both
sufficiently trust their guests and want to be able to use devices in
Dom0 again after they had been in use by a guest wouldn't need to
"manually" move such devices back from DomIO to Dom0).
This is XSA-306.
Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Wei Liu <wl@xen.org>
See patch documentation and comments.
This is part of XSA-305 / CVE-2019-11135
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
To protect against the TSX Async Abort speculative vulnerability, Intel have
released new microcode for affected parts which introduce the MSR_TSX_CTRL
control, which allows TSX to be turned off. This will be architectural on
future parts.
Introduce tsx= to provide a global on/off for TSX, including its enumeration
via CPUID. Provide stub virtualisation of this MSR, as it is not exposed to
guests at the moment.
VMs may have booted before microcode is loaded, or before hosts have rebooted,
and they still want to migrate freely. A VM which booted seeing TSX can
migrate safely to hosts with TSX disabled - TSX will start unconditionally
aborting, but still behave in a manner compatible with the ABI.
The guest-visible behaviour is equivalent to late loading the microcode and
setting the RTM_DISABLE bit in the course of live patching.
This is part of XSA-305 / CVE-2019-11135
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
CVE-2018-12207 covers a set of errata on various Intel processors, whereby a
machine check exception can be generated in a corner case when an executable
mapping changes size or cacheability without TLB invalidation. HVM guest
kernels can trigger this to DoS the host.
To mitigate, in affected hardware, all EPT superpages are marked NX. When an
instruction fetch violation is observed against the superpage, the superpage
is shattered to 4k and has execute permissions restored. This prevents the
guest kernel from being able to create the necessary preconditions in the iTLB
to exploit the vulnerability.
This does come with a workload-dependent performance overhead, caused by
increased TLB pressure. Performance can be restored, if guest kernels are
trusted not to mount an attack, by specifying ept=exec-sp on the command line.
This is part of XSA-304 / CVE-2018-12207
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: George Dunlap <george.dunlap@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
The current documentation is misleading for a few reasons:
1) The synchronization happens on all exit/entry from/to the guest.
This includes from EL0 (i.e userspace).
2) Trusted guest can also generate SErrors (e.g. memory failure)
3) Without RAS support, SErrors are IMP DEFINED. Unless you have a
complete TRM in hand, you can't really make a decision.
4) The documentation is written around performance when this is not
the first concern.
The documentation is now reworked to focus on the consequences of using
serrors="panic" and avoid to go in details on the exact implementation.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Juergen Gross <jgross@suse.com>
Per the Arm ARM (D4.5 in ARM DDI 0487E.a), SError may be precise or
imprecise.
Imprecise means the state presented to the exception handler is not
guaranteed to be consistent with any point in the excution stream from
which the exception was taken. In other words, they are likely to be
fatal as you can't return safely from them.
Without the RAS extension, the Arm architecture does not provide a way
to differentiate between imprecise and precise SError. Furthermore Xen
has no support for RAS yet. So from a software POV, there is not much
we can do.
More generally, forwarding blindly SErrors to the guest is likely to be
the wrong thing to do. Indeed, Xen is not able to know what is the
content of the SError. This may be a critical device used by the
hypervisor that is about to fail.
In a nutshell, the option serrors=forward is not safe to use in any
environment with the current state of Xen. Therefore the option and any
code related to it are completely removed.
Take the opportunity to rework the comment in do_trap_data_abort() as
all SErrors/External Abort generated by the hypervisor will result in
a crash of the system no matter what the user passed on the command
line.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Juergen Gross <jgross@suse.com>
When serrors=diverse is selected by the user, we will only synchronize
the pending SErrors on entry to hypervisor from guest context and exit
from guest to hypervisor context.
We don't need synchronize SErrors between guest context switch as they
would be categorized to Hypervisor generated SErrors in any case.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Juergen Gross <jgross@suse.com>
l1tf-barrier is an inappropriate name, and came about because of restrictions
on could be discussed publicly when the patches were proposed.
In practice, it is for general Spectre v1 mitigations, and is necessary in all
cases. An adversary which can control speculation in Xen can leak data in
cross-core (BCBS, etc) or remote (NetSpectre) scenarios - the problem is not
limited to just L1TF with HT active.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Juergen Gross <jgross@suse.com>
The binding for the dom0less module does not match Xen implementation.
Any module should contain the compatible "multiboot,module" to be
recognized.
This was clearly an oversight as other examples with Xen code base
provide the compatible "multiboot,module".
So fix the binding and the example associated to it.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Juergen Gross <jgross@suse.com>
When one core is loading ucode, handling NMI on sibling threads or
on other cores in the system might be problematic. By rendezvousing
all CPUs in NMI handler, it prevents NMI acceptance during ucode
loading.
Basically, some work previously done in stop_machine context is
moved to NMI handler. Primary threads call in and load ucode in
NMI handler. Secondary threads wait for the completion of ucode
loading on all CPU cores. An option is introduced to disable this
behavior.
Control thread doesn't rendezvous in NMI handler by calling self_nmi()
(in case of unknown_nmi_error() being triggered). The side effect is
control thread might be handling an NMI while other threads are loading
ucode. If an ucode is to update something shared by a whole socket,
control thread may be accessing things that are being updating by the
ucode loading on other cores. It is not safe. Update ucode on the
control thread first to mitigate this issue.
Signed-off-by: Sergey Dyasli <sergey.dyasli@citrix.com>
Signed-off-by: Chao Gao <chao.gao@intel.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Both Cortex-A57 (erratum 1319537) and Cortex-A72 (erratum 1319367) can
end with corrupted TLBs if they speculate an AT instruction while S1/S2
system registers in inconsistent state.
The workaround is the same as for Cortex A-76 implemented by commit
a18be06aca "xen/arm: Implement workaround for Cortex-A76 erratum 1165522",
so it is only necessary to plumb in the cpuerrata framework.
Signed-off-by: Julien Grall <julien.grall@arm.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
The domain builder no longer uses local CPUID instructions for policy
decisions. This resolves a key issue for PVH dom0's. However, as PV dom0's
have never had faulting enforced, leave a command line option to restore the
old behaviour.
Advertise virtualised faulting support to control domains unless the opt-out
has been used.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Juergen Gross <jgross@suse.com>
debugtrace is normally writing trace entries into a single trace
buffer. There are cases where this is not optimal, e.g. when hunting
a bug which requires writing lots of trace entries and one cpu is
stuck. This will result in other cpus filling the trace buffer and
finally overwriting the interesting trace entries of the hanging cpu.
In order to be able to debug such situations add the capability to use
per-cpu trace buffers. This can be selected by specifying the
debugtrace boot parameter with the modifier "cpu:", like:
debugtrace=cpu:16
At the same time switch the parsing function to accept size modifiers
(e.g. 4M or 1G).
Printing out the trace entries is done for each buffer in order to
minimize the effort needed during printing. As each entry is prefixed
with its sequence number sorting the entries can easily be done when
analyzing them.
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Allow limiting the max C-state sub-state by appending to the max_cstate
command-line parameter. E.g. max_cstate=1,0
The limit only applies to the highest legal C-state. For example:
max_cstate = 1, max_csubstate = 0 ==> C0, C1 okay, but not C1E
max_cstate = 1, max_csubstate = 1 ==> C0, C1 and C1E okay, but not C2
max_cstate = 2, max_csubstate = 0 ==> C0, C1, C1E, C2 okay, but not C3
max_cstate = 2, max_csubstate = 1 ==> C0, C1, C1E, C2 okay, but not C3
Signed-off-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
While the MWAIT idle driver already takes it to mean an actual C state,
the ACPI idle driver so far used it as a list index. The list index,
however, is an implementation detail of Xen and affected by firmware
settings (i.e. not necessarily uniform for a particular system).
While touching this code also avoid invoking menu_get_trace_data()
when tracing is not active. For consistency do this also for the
MWAIT driver.
Note that I'm intentionally not adding any sorting logic to set_cx():
Before and after this patch we assume entries to arrive in order, so
this would be an orthogonal change.
Take the opportunity and add minimal documentation for the command line
option.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
We currently have an asymmetric setup where CONFIG_VERBOSE_DEBUG controls
extra diagnostics for a PV dom0, and opt_dom0_verbose controls extra
diagnostics for a PVH dom0.
Default opt_dom0_verbose to CONFIG_VERBOSE_DEBUG and use opt_dom0_verbose
consistently.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
This would have been very handy when debugging some pv-l1tf issues. As there
is no cost to supporting it, wire it up.
Due to the way dom0 is constructed, switching into shadow mode must be done
after the pagetables are written, and because of partially being in dom0
context, shadow_enable() doesn't like the state it finds.
Reuse the pv_l1tf tasklet for convenience, which will switch dom0 into shadow
mode just before it starts executing.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Extend early prink code to be able to handle other SCIF(X)
compatible interfaces as well. These interfaces have lot in common,
but mostly differ in offsets and bits for some registers.
Introduce "EARLY_PRINTK_VERSION" config option to choose which
interface version should be used (to properly apply register offsets).
Please note, nothing has been technically changed for Renesas "Lager"
and other supported boards (SCIF).
The "EARLY_PRINTK_VERSION" option for that board should be empty:
CONFIG_EARLY_PRINTK=scif,0xe6e60000
Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
Acked-by: Julien Grall <julien.grall@arm.com>
Paul Durrant
Jan Beulich
Wei Liu
Andrew Cooper
Sarah Newman
Anthony PERARD
Juergen Gross
Ian Jackson
Eslam Elnikety
Igor Druzhinin
Julien Grall
Roger Pau Monné
Konrad Rzeszutek Wilk
Pawel Wieczorkiewicz
Julien Grall
Stefano Stabellini
Sergey Dyasli
Ross Lagerwall
Oleksandr Tyshchenko