Just like domain_soft_reset() properly zaps runstate area handles, the
secondary time area ones also need discarding to prevent guest memory
corruption once the guest is re-started.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
This is done to shorten line length in the function in preparation for
adding further usages of the vpci_bar data structure.
No functional change.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Teardown of MSIX vPCI related data doesn't currently remove the MSIX
device data from the list of MSIX tables handled by the domain,
leading to a use-after-free of the data in the msix structure.
Remove the structure from the list before freeing in order to solve
it.
Reported-by: Jan Beulich <jbeulich@suse.com>
Fixes: d6281be9d0 ('vpci/msix: add MSI-X handlers')
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
It's possible for a device to be assigned to a domain but have no
vpci structure if vpci_process_pending() failed and called
vpci_remove_device() as a result. The unconditional accesses done by
vpci_{read,write}() and vpci_remove_device() to pdev->vpci would
then trigger a NULL pointer dereference.
Add checks for pdev->vpci presence in the affected functions.
Fixes: 9c244fdef7 ('vpci: add header handlers')
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
All the build jobs exist in two flavors: debug and non-debug, where the
former sets 'debug' variable to 'y' and the latter to 'n'. This variable
is only being recognized by the toolstack, because Xen requires
enabling/disabling debug build via e.g. menuconfig/config file.
As a corollary, we end up building/testing Xen with CONFIG_DEBUG always
set to a default value ('y' for unstable and 'n' for stable branches),
regardless of the type of the build job.
Fix this behavior by setting CONFIG_DEBUG according to the 'debug' value.
Signed-off-by: Michal Orzel <michal.orzel@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
NULL scheduler is not enabled by default on non-debug Xen builds. This
causes the boot time cpupools test to fail on such build jobs. Fix the issue
by explicitly specifying the config options required to enable the NULL
scheduler.
Fixes: 36e3f41587 ("automation: Add a new job for testing boot time cpupools on arm64")
Signed-off-by: Michal Orzel <michal.orzel@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Make may not have copied "_libxl_types_json.h" into $(XEN_INCLUDE)
before starting to build the different objects.
Make sure that the generated headers are copied into $(XEN_INCLUDE)
before using them. This is achieved by telling make about which
headers are needed to use "libxl_internal.h" which use "libxl_json.h"
which uses "_libxl_types_json.h". "libxl_internal.h" also uses
"libxl.h" so add it to the list.
This also prevent `gcc` from using a potentially installed headers
from a previous version of Xen.
Reported-by: Per Bilse <per.bilse@citrix.com>
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
The addition of a call to shadow_blow_tables() from shadow_teardown()
has resulted in the "no vcpus" related assertion becoming triggerable:
If domain_create() fails with at least one page successfully allocated
in the course of shadow_enable(), or if domain_create() succeeds and
the domain is then killed without ever invoking XEN_DOMCTL_max_vcpus.
Note that in-tree tests (test-resource and test-tsx) do exactly the
latter of these two.
The assertion's comment was bogus anyway: Shadow mode has been getting
enabled before allocation of vCPU-s for quite some time. Convert the
assertion to a conditional: As long as there are no vCPU-s, there's
nothing to blow away.
Fixes: e7aa55c0aa ("x86/p2m: free the paging memory pool preemptively")
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com>
A similar assertion/comment pair exists in _shadow_prealloc(); the
comment is similarly bogus, and the assertion could in principle trigger
e.g. when shadow_alloc_p2m_page() is called early enough. Replace those
at the same time by a similar early return, here indicating failure to
the caller (which will generally lead to the domain being crashed in
shadow_prealloc()).
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
When the system is coming up after having been suspended,
restore_vcpu_affinity() is called for each domain in order to adjust
the vcpu's affinity settings in case a cpu didn't come to live again.
The way restore_vcpu_affinity() is doing that is wrong, because the
specific scheduler isn't being informed about a possible migration of
the vcpu to another cpu. Additionally the migration is often even
happening if all cpus are running again, as it is done without check
whether it is really needed.
As cpupool management is already calling cpu_disable_scheduler() for
cpus not having come up again, and cpu_disable_scheduler() is taking
care of eventually needed vcpu migration in the proper way, there is
simply no need for restore_vcpu_affinity().
So just remove restore_vcpu_affinity() completely, together with the
no longer used sched_reset_affinity_broken().
Fixes: 8a04eaa8ea ("xen/sched: move some per-vcpu items to struct sched_unit")
Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Dario Faggioli <dfaggioli@suse.com>
Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
When a domain gets paused the unit runnable state can change to "not
runnable" without the scheduling lock being involved. This means that
a specific scheduler isn't involved in this change of runnable state.
In the RTDS scheduler this can result in an inconsistency in case a
unit is losing its "runnable" capability while the RTDS scheduler's
scheduling function is active. RTDS will remove the unit from the run
queue, but doesn't do so for the replenish queue, leading to hitting
an ASSERT() in replq_insert() later when the domain is unpaused again.
Fix that by removing the unit from the replenish queue as well in this
case.
Fixes: 7c7b407e77 ("xen/sched: introduce unit_runnable_state()")
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Dario Faggioli <dfaggioli@suse.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
efi_init_memory() in both relevant places is treating EFI_MEMORY_RUNTIME
higher priority than the type of the range. To avoid accessing memory at
runtime which was re-used for other purposes, make
efi_arch_process_memory_map() follow suit. While in theory the same would
apply to EfiACPIReclaimMemory, we don't actually "reclaim" or clobber
that memory (converted to E820_ACPI on x86) there (and it would be a bug
if the Dom0 kernel tried to reclaim the range, bypassing Xen's memory
management, plus it would be at least bogus if it clobbered that space),
hence that type's handling can be left alone.
Fixes: bf6501a62e ("x86-64: EFI boot code")
Fixes: facac0af87 ("x86-64: EFI runtime code")
Fixes: 6d70ea10d4 ("Add ARM EFI boot support")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Currently, the fields 'root_order' and 'sl0' of the pa_range_info for
the 52-bit pa range have the values 3 and 3, respectively.
This configuration does not match any of the valid root table configurations
for 4KB granule and t0sz 12, described in ARM DDI 0487I.a D8.2.7.
More specifically, according to ARM DDI 0487I.a D8.2.7, in order to support
the 52-bit pa size with 4KB granule, the p2m root table needs to be configured
either as a single table at level -1 or as 16 concatenated tables at level 0.
Since, currently there is no support for level -1, set the 'root_order' and
'sl0' fields of the 52-bit pa_range_info according to the second approach.
Note that the values of those fields are not used so far. This patch updates
their values only for the sake of correctness.
Fixes: 407b13a71e ("xen/arm: p2m don't fall over on FEAT_LPA enabled hw")
Signed-off-by: Xenia Ragiadakou <burzalodowa@gmail.com>
Reviewed-by: Michal Orzel <michal.orzel@amd.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
All functions in domain_build.c should be marked __init. This was
spotted when building the hypervisor with -Og.
Fixes: 1050a7b91c ("xen/arm: add pci-domain for disabled devices")
Signed-off-by: Stewart Hildebrand <stewart.hildebrand@amd.com>
Acked-by: Julien Grall <jgrall@amazon.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
During live update we will load the /tool/xenstored path from the previous binary,
and then try to mkdir /tool again which will fail with EEXIST.
Check for existence of the path before creating it.
The write call to /tool/xenstored should not need any changes
(and we do want to overwrite any previous path, in case it changed).
Prior to 7110192b1d live update would work only if the binary path was
specified, and with 7110192b1d and this live update also works when
no binary path is specified in `xenstore-control live-update`.
Fixes: 7110192b1d ("tools/oxenstored: Fix Oxenstored Live Update")
Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
The xendomains script uses the output of `xl list -l` to collect the
id and name of each domain, which is used in the shutdown logic, amongst
other purposes.
The linked commit added a "domid" field to libxl_domain_create_info.
This causes the output of `xl list -l` to contain two "domid"s per
domain, which may not be equal. This in turn causes `xendomains stop` to
issue two shutdown commands per domain, one of which is to a duplicate
and/or invalid domid.
To work around this, make the LIST_GREP pattern more restrictive for
domid, so it only detects the domid at the top level and not the domid
inside c_info.
Fixes: 4a3a25678d ("libxl: allow creation of domains with a specified or random domid")
Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
CONFIG_HAS_PCI is not defined for the tools build, and as a result the
vpci harness would never get build. Fix this by building it
unconditionally, there's nothing arch specific in it.
Reported-by: Andrew Cooper <Andrew.Cooper3@citrix.com>
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Instead of pci_get_pdev_by_domain(), which is no longer present in the
hypervisor.
While there add parentheses around the define value.
Fixes: a37f9ea7a6 ('PCI: fold pci_get_pdev{,_by_domain}()')
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Some vpci functions got the cfcheck attribute added, but that's not
defined in the user-space test harness, so add a dummy define in order
for the harness to build.
Fixes: 4ed7d5525f ('xen/vpci: CFI hardening')
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Hardware using GICv2 needs to create a P2M mapping of 8KB GICv2 area
when the domain is created. Considering the worst case of page tables
which requires 6 P2M pages as the two pages will be consecutive but not
necessarily in the same L3 page table and keep a buffer, populate 16
pages as the default value to the P2M pages pool in p2m_init() at the
domain creation stage to satisfy the GICv2 requirement. For GICv3, the
above-mentioned P2M mapping is not necessary, but since the allocated
16 pages here would not be lost, hence populate these pages
unconditionally.
With the default 16 P2M pages populated, there would be a case that
failures would happen in the domain creation with P2M pages already in
use. To properly free the P2M for this case, firstly support the
optionally preemption of p2m_teardown(), then call p2m_teardown() and
p2m_set_allocation(d, 0, NULL) non-preemptively in p2m_final_teardown().
As non-preemptive p2m_teardown() should only return 0, use a
BUG_ON to confirm that.
Since p2m_final_teardown() is called either after
domain_relinquish_resources() where relinquish_p2m_mapping() has been
called, or from failure path of domain_create()/arch_domain_create()
where mappings that require p2m_put_l3_page() should never be created,
relinquish_p2m_mapping() is not added in p2m_final_teardown(), add
in-code comments to refer this.
Fixes: cbea5a1149 ("xen/arm: Allocate and free P2M pages from the P2M pool")
Suggested-by: Julien Grall <jgrall@amazon.com>
Signed-off-by: Henry Wang <Henry.Wang@arm.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Reviewed-by: Bertrand Marquis <bertrand.marquis@arm.com>
Release-acked-by: George Dunlap <george.dunlap@citrix.com>
p2m_init() is mostly trivial initialisation, but has two fallible operations
which are on either side of the backpointer trigger for teardown to take
actions.
p2m_free_vmid() is idempotent with a failed p2m_alloc_vmid(), so rearrange
p2m_init() to perform all trivial setup, then set the backpointer, then
perform all fallible setup.
This will simplify a future bugfix which needs to add a third fallible
operation.
No practical change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Reviewed-by: Bertrand Marquis <bertrand.marquis@arm.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
qemu-trad build system, when built from xen.git, will make use of
Rules.mk (setup via qemu-trad.git/xen-setup). This mean that changes
to Rules.mk will have an impact our ability to build qemu-trad.
Recent commit e4f5949c44 ("tools: Add -Werror by default to all
tools/") have added "-Werror" to the CFLAGS and qemu-trad start to use
it. But this fails and there's lots of warning that are now turned
into error.
We should teach qemu-trad and xen.git to not have to use Rules.mk when
building qemu-trad, but for now, avoid adding -Werror to CFLAGS when
building qemu-trad.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Using a full path to the C libraries when preparing one of the ocaml
binding for those libraries make the binding unusable by external
project. The full path is somehow embedded and reused by the external
project when linking against the binding.
Instead, we will use the proper way to link a library, by using '-l'.
For in-tree build, we also need to provide the search directory via
'-L'.
(The search path -L are still be embedded, but at least that doesn't
prevent the ocaml binding from been used.)
Related-to: xen-project/xen#96
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
Tested-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
gengotypes.py creates both "types.gen.go" and "helpers.gen.go", but
make can start gengotypes.py twice. Rework the rules so that
gengotypes.py is executed only once.
Also, add the ability to provide a path to tell gengotypes.py where to
put the files. This doesn't matter yet but it will when for example
the script will be run from tools/ to generate the targets.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: George Dunlap <george.dunlap@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Instead of moving the public "_libxl_*.h" headers, we make a copy to
the destination so that make doesn't try to remake the targets
"_libxl_*.h" in libs/light/ again.
A new .PRECIOUS target is added to tell make to not deletes the
intermediate targets generated by "gentypes.py".
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Currently, a rebuild of libxl will always rebuild "build.o". This is because
the target depends on "acpi" which never exist. So instead we will have
"build.o" have as prerequisites targets that are actually generated by "acpi",
that is $(DSDT_FILES-y).
While "dsdt_*.c" isn't really a dependency for "build.o", a side
effect of building that dsdt_*.c is to also generate the "ssdt_*.h"
that "build.o" needs, but I don't want to list all the headers needed
by "build.o" and duplicate the information available in
"libacpi/Makefile" at this time.
Also avoid duplicating the "acpi" target for Arm, and unique one for
both architecture. And move the "acpi" target to be with other targets
rather than in the middle of the source listing. For the same reason,
move the prerequisites listing for both $(DSDT_FILES-y) and "build.o".
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Rework "xen-xsm" rules to not have to change directory to run
mkflask.sh, and store mkflask.sh path in a var, and use a full path
for FLASK_H_DEPEND, and output directory is made relative.
Rename "all-y" target to a more descriptive "xen/lib/x86/all".
Removed the "dist" target which was the only one existing in tools/.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
$(SHLIB_LDFLAGS) is used by more targets that the single targets that
except it (libxenfoo.so.X.Y). There is also some dynamic libraries in
stats/ that uses $(SHLIB_LDFLAGS) (even if those are never built), and
there's libxenlight_test.so which doesn't needs a version script.
Also, libxenlight_test.so might failed to build if the version script
doesn't exist yet.
For these reasons, avoid changing the generic $(SHLIB_LDFLAGS) flags,
and add the flag directly on the command line.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
"$DIR" might not be a full path and it might not have `pwd` as ".."
directory. So use `cd -` to undo the first `cd` command.
Also, use `basename` to make a symbolic link with a relative path.
This doesn't matter yet but it will when for example the commands to
clone OVMF is been run from tools/ rather than tools/firmware/.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
This doesn't matter yet but it will when for example the script will
be run from tools/ to generate files tools/libs/light/.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Remove "build" targets.
Use "$(TARGETS)" to list binary to be built.
Cleanup "clean" rule.
Also drop conditional install of $(BIN) and $(LIBBIN) as those two
variables are now always populated.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
gdbsx/:
- Make use of subdir facility for the "clean" target.
- No need to remove the *.a, they aren't in this dir.
- Avoid calling "distclean" in subdirs as "distclean" targets do only
call "clean", and the "clean" also runs "clean" in subdirs.
- Avoid the need to make "gx_all.a" and "xg_all.a" in the "all"
recipe by forcing make to check for update of "xg/xg_all.a" and
"gx/gx_all.a" by having "FORCE" as prerequisite. Now, when making
"gdbsx", make will recurse even when both *.a already exist.
- List target in $(TARGETS).
gdbsx/*/:
- Fix dependency on *.h.
- Remove some dead code.
- List targets in $(TARGETS).
- Remove "build" target.
- Cleanup "clean" targets.
- remove comments about the choice of "ar" instead of "ld"
- Use "$(AR)" instead of plain "ar".
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Reviewed-by: Luca Fancellu <luca.fancellu@arm.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
I observed this ASSERT_UNREACHABLE in partner_rings_remove consistently
trip. It was in OpenXT with the viptables patch applied.
dom10 shuts down.
dom7 is REJECTED sending to dom10.
dom7 shuts down and this ASSERT trips for dom10.
The argo_send_info has a domid, but there is no refcount taken on
the domain. Therefore it's not appropriate to ASSERT that the domain
can be looked up via domid. Replace with a debug message.
Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
Reviewed-by: Christopher Clark <christopher.w.clark@gmail.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
This patch partially reverts the changes made by the following commit:
da0e58c038e6 intel_idle: add 'preferred_cstates' module argument
As that commit describes, on early Sapphire Rapids Xeon platforms the C1 and
C1E states were mutually exclusive, so that users could only have either C1 and
C6, or C1E and C6.
However, Intel firmware engineers managed to remove this limitation and make C1
and C1E to be completely independent, just like on previous Xeon platforms.
Therefore, this patch:
* Removes commentary describing the old, and now non-existing SPR C1E
limitation.
* Marks SPR C1E as available by default.
* Removes the 'preferred_cstates' parameter handling for SPR. Both C1 and
C1E will be available regardless of 'preferred_cstates' value.
We expect that all SPR systems are shipping with new firmware, which includes
the C1/C1E improvement.
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Origin: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 1548fac47a11
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Having IBRS enabled while the SMT sibling is idle unnecessarily slows
down the running sibling. OTOH, disabling IBRS around idle takes two
MSR writes, which will increase the idle latency.
Therefore, only disable IBRS around deeper idle states. Shallow idle
states are bounded by the tick in duration, since NOHZ is not allowed
for them by virtue of their short target residency.
Only do this for mwait-driven idle, since that keeps interrupts disabled
across idle, which makes disabling IBRS vs IRQ-entry a non-issue.
Note: C6 is a random threshold, most importantly C1 probably shouldn't
disable IBRS, benchmarking needed.
Suggested-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Origin: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git bf5835bcdb96
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Similar to SPR, the C1 and C1E states on ADL are mutually exclusive.
Only one of them can be enabled at a time.
But contrast to SPR, which usually has a strong latency requirement
as a Xeon processor, C1E is preferred on ADL for better energy
efficiency.
Add custom C-state tables for ADL with both C1 and C1E, and
1. Enable the "C1E promotion" bit in MSR_IA32_POWER_CTL and mark C1
with the CPUIDLE_FLAG_UNUSABLE flag, so C1 is not available by
default.
2. Add support for the "preferred_cstates" module parameter, so that
users can choose to use C1 instead of C1E by booting with
"intel_idle.preferred_cstates=2".
Separate custom C-state tables are introduced for the ADL mobile and
desktop processors, because of the exit latency differences between
these two variants, especially with respect to PC10.
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
[ rjw: Changelog edits, code rearrangement ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Origin: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git d1cf8bbfed1e
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
From: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Add a Sapphire Rapids Xeon C6 optimization, similar to what we have for Sky Lake
Xeon: if package C6 is disabled, adjust C6 exit latency and target residency to
match core C6 values, instead of using the default package C6 values.
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Origin: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 3a9cf77b60dc
Make sure a contradictory "preferred-cstates" wouldn't cause bypassing
of the added logic.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
On Sapphire Rapids Xeon (SPR) the C1 and C1E states are basically mutually
exclusive - only one of them can be enabled. By default, 'intel_idle' driver
enables C1 and disables C1E. However, some users prefer to use C1E instead of
C1, because it saves more energy.
This patch adds a new module parameter ('preferred_cstates') for enabling C1E
and disabling C1. Here is the idea behind it.
1. This option has effect only for "mutually exclusive" C-states like C1 and
C1E on SPR.
2. It does not have any effect on independent C-states, which do not require
other C-states to be disabled (most states on most platforms as of today).
3. For mutually exclusive C-states, the 'intel_idle' driver always has a
reasonable default, such as enabling C1 on SPR by default. On other
platforms, the default may be different.
4. Users can override the default using the 'preferred_cstates' parameter.
5. The parameter accepts the preferred C-states bit-mask, similarly to the
existing 'states_off' parameter.
6. This parameter is not limited to C1/C1E, and leaves room for supporting
other mutually exclusive C-states, if they come in the future.
Today 'intel_idle' can only be compiled-in, which means that on SPR, in order
to disable C1 and enable C1E, users should boot with the following kernel
argument: intel_idle.preferred_cstates=4
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Origin: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git da0e58c038e6
Enable C1E (if requested) not only on the BSP's socket / package. Alter
command line option to fit our model, and extend it to also accept
string form arguments.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
The current bindings function, but the preexisting
type physinfo_arch_cap_flag =
| X86 of x86_physinfo_arch_cap_flag
is a special case in the Ocaml type system with an unusual indirection, and
will break when a second option, e.g. `| ARM of ...` is added.
Also, the position the list is logically wrong. Currently, the types express
a list of elements which might be an x86 flag or an arm flag (and can
intermix), whereas what we actually want is either a list of x86 flags, or a
list of ARM flags (that cannot intermix).
Rework the Ocaml types to avoid the ABI special case and move the list
primitive, and adjust the C bindings to match.
Fixes: 2ce11ce249 ("x86/HVM: allow per-domain usage of hardware virtualized APIC")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
Reposition the defines to match the outer style and to make the logic
half-legible.
No functional change.
Fixes: 0570d7f276 ("x86/msr: introduce an option for compatible MSR behavior selection")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
With the addition of vmx_add_msr() calls to construct_vmcs() there are
now cases where simply freeing the VMCS isn't enough: The MSR bitmap
page as well as one of the MSR area ones (if it's the 2nd vmx_add_msr()
which fails) may also need freeing. Switch to using vmx_destroy_vmcs()
instead.
Fixes: 3bd36952da ("x86/spec-ctrl: Introduce an option to control L1D_FLUSH for HVM HAP guests")
Fixes: 53a570b285 ("x86/spec-ctrl: Support IBPB-on-entry")
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
protmode_load_seg() would better adhere to that "feature" of clearing
base (and limit) during NULL selector loads.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Henry Wang <Henry.Wang@arm.com>
While the comment next to the lock dropping in preparation of
recursively calling acquire_grant_for_copy() mistakenly talks about the
rd == td case (excluded a few lines further up), the same concerns apply
to the calling of release_grant_for_copy() on a subsequent error path.
This is CVE-2022-33748 / XSA-411.
Fixes: ad48fb963d ("gnttab: fix transitive grant handling")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Jan Beulich
Volodymyr Babchuk
Andrew Cooper
Roger Pau Monné
Michal Orzel
Anthony PERARD
Juergen Gross
Xenia Ragiadakou
Stewart Hildebrand
Edwin Török
Peter Hoyes
Henry Wang
Jason Andryuk
Artem Bityutskiy
Peter Zijlstra
Zhang Rui