The original commit showed a ~6% regression in a benchmark. The call to
align_timer(firsttick, period) rounds firsttick up to the next mutiple
of the period, if firsttick % period != 0:
align_timer(0, period) -> 0
align_timer(1, period) -> period
align_timer(period, period) -> period
align_timer(period + 1, period) -> 2 * period
So adding the period (sampling_rate) before calling align_timer() will
in most cases incease the expiration to 2 * period (sampling_rate) (the
exception being firsttick % period == 0). This longer timer slows the
reaction time of the algorithm.
This reverts commit a0ed5bcfbe, just without
re-introducing the style issues that were there.
Fixes: a0ed5bcfbe ("xen/cpufreq: fix usages of align_timer() in the on-demand governor")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Today the inline tracking of the stub page is problematic. 0xcc is used to
indicate unused, but it is also a "clear value." A !CONFIG_PV build or
when running with FRED support will not populate the LSTAR/CSTAR stubs at
CPU bringup. If a CPU is then offlined, the stubs page will be freed as
its content will be all 0xcc, regardless of the stubs page still begin
referenced by other CPUs.
The new approach uses a global, CPU-indexed dynamically allocated array of
stub addresses. However, to handle NUMA aware allocations, we cannot
allocate all the memory in advance because of the NUMA dependency. Take
advantage of the fact that Xen will attempt to contiguously pack CPUs on
the same NUMA node (see normalise_cpu_order()), and on CPU bringup use the
same stubs page the previous CPU did if suitable. Note the code would
still function properly even if CPUs from NUMA nodes are not contiguously
packed, it just consumes more memory.
Stubs pages are no longer freed. They remain referenced in the global
CPU-indexed array and are re-used if the CPU is re-onlined.
The stubs array doesn't have an explicit lock. During boot it's accessed
single threaded. During runtime, &cpu_add_remove_lock serializes access.
Fixes: 7a66ac8d16 ("x86: move syscall trampolines off the stack")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Tested-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
wait.c is only used by vm_event.c, so build it when CONFIG_VM_EVENT is
selected.
Provide stubs of functions called from common code. entry.S needs an
ifdef to hide the symbol from the assembly.
Also conditionalize .waitqueue_vcpu in struct vcpu to save space.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Pass the TPM2 ACPI table so that the device can be found by a PVH dom0.
Otherwise dom0 shows:
tpm_tis MSFT0101:00: [Firmware Bug]: failed to get TPM2 ACPI table
tpm_tis MSFT0101:00: probe with driver tpm_tis failed with error -22
TCPA is "Trusted Computing Platform Alliance table", but it is really
the table for a TPM 1.2. Pass that as well.
While doing this, move ACPI_SIG_WPBT to alpabetize the entries.
This exposes TPM event log tables on PVH dom0, bring it in line with a
PV dom0.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Use a define to centralize the common scheduler data in the per-arch
linker scripts. This is in preparation for marking it KEEP(). While
moving, switch to using POINTER_ALIGN.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
On x86, the opening if of call_handlers_*() is indented 4 spaces which
breaks alignment. The alignment is broken by printf(" "), but
that actually aligns ARM's opening "if ( likely() )". Move the indent
to the if that needs it.
This results in x86 hypercall-defs.h changes like:
#define call_handlers_hvm64(num, ret, a1, a2, a3, a4, a5) \
({ \
uint64_t mask = (num) > 63 ? 0 : 1ULL << (num); \
- if ( likely(mask & ((1ULL << __HYPERVISOR_vcpu_op) | (1ULL << __HYPERVISOR_event_channel_op))) ) \
+ if ( likely(mask & ((1ULL << __HYPERVISOR_vcpu_op) | (1ULL << __HYPERVISOR_event_channel_op))) ) \
{ \
if ( (num) == __HYPERVISOR_vcpu_op ) \
ARM's hypercall-defs.h does not change.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
On a system without a control domain, a hardware|xenstore domain
starting xenstored shows:
common/grant_table.c:1041:d2v0 Could not find domain 32756
gnttab: error: mmap failed: Invalid argument
Could not initialize dom32756
setup_structure() does not get called without priv_domid. Subsequent
introduce_domain() calls will fault when calling fire_special_watches().
Default priv_domid to store_domid when there is no control domain.
Fixes: e5b0a94055 ("tools/xenstored: Auto-introduce domains")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
io_apic.c has a lot of ad-hoc for(;;) and while(1) loops for iterating
over irq_pin_list entries. Replace them with a standardized
for loop using pin_list_next() to advance entry.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
removeprefix is only added in Python 3.9.
Instead of the prefix removal, switch to passing in a "depth" parameter,
and incrementing it for each level.
There is a slight change in the generated _libxl_types.c. instances of
KeyedUnion increment depth without outputing any code. The net result
is some cases where jso_sub_1 is followed by jso_sub_3. As an example:
_libxl_types.c
_libxl_types.c
@@ -5535,12 +5535,12 @@
if (!jso_sub_1)
goto out;
if (!libxl__string_is_default(&p->u.pty.path)) {
- json_object *jso_sub_2 = NULL;
- rc = libxl__string_gen_jso(&jso_sub_2, p->u.pty.path);
+ json_object *jso_sub_3 = NULL;
+ rc = libxl__string_gen_jso(&jso_sub_3, p->u.pty.path);
if (rc)
goto out;
- if (json_object_object_add(jso_sub_1, "path", jso_sub_2)) {
- json_object_put(jso_sub_2);
+ if (json_object_object_add(jso_sub_1, "path", jso_sub_3)) {
+ json_object_put(jso_sub_3);
goto out;
}
}
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
io_apic_level_ack_pending() will end up in an infinite loop if
entry->pin == -1. entry does not change, so it will keep reading -1.
Convert to a proper for loop so that continue works. Add a new helper,
pin_list_next(), to handle advancing to the next irq_pin_list entry.
Noticed during code inspection. The infinite loop was not observed.
Fixes: f821102450 ("x86: IRQ Migration logic enhancement.")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
When the fixes commit re-arranged the code, it made edd=off jump to the
mbr code instead of returning from the function. Previously edd_done
was immediately before ret and skipped the MBR check.
Replace edd_done with .Ledd_mbr_sig for the start of the MBR checking,
and replace .Ledd_mbr_sig_skip with .Ledd_done to exit from the function.
edd=off jumps to .Ledd_done to return from the function, and internal
jumps go to .Ledd_mbr_sig to check edd=skipmbr.
Fixes: 5ec164fd61 ("x86/boot: re-arrange how/when we do disk I/O")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-Acked-By: Oleksii Kurochko<oleksii.kurochko@gmail.com>
Split out AMD SVM and AMD IOMMU, and add myself as a reviewer. Jan,
Andrew and Roger are set as maintainers as they were for the X86 entry.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
There is currently an asymmetry between HVM where the page is cleared
and PV where the contents are restored.
Add xc_dom_console_set_disconnected() to only set the connection flag
for PV guests.
xenconsoled is responsible for setting the console connected when it
attaches.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Release-Acked-By: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
Initialize xencons_interface's connection field to
XENCONSOLE_DISCONNECTED. xenconsoled will mark the page as connected
when it establishes the connection.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Release-Acked-By: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
With hyperlaunch, a domU can start before its console ring is connected
by xenconsoled. With nothing emptying the ring, it can quickly fill
during boot. In domU_write_console(), __write_console() returns 0 when
the ring is full. This loops spins until xenconsoled starts emptying
the ring:
while (len) {
ssize_t sent = __write_console(cons, data, len);
if (sent < 0)
return sent;
data += sent;
len -= sent;
if (unlikely(len))
HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
}
The goal of this patch is to add a way for the frontend to know when a
console is connected. This patch adds a new flag to the end of the
console ring structure. It is used for the backend to indicate that it
has connected and started servicing the page.
The two values are
XENCONSOLE_DISCONNECTED 1
XENCONSOLE_CONNECTED 0
XENCONSOLE_DISCONNECTED indicates to the guest that ring is
disconnected, so it will not be serviced. The guest can avoid writing
into it in that case. A domU can use console hypercalls and only
transition to the ring when it is connected and won't fill and block.
Once the backend (xenconsoled) maps and starts servicing the
console, the flag will be set to XENCONSOLE_CONNECTED (0) to indicate
the backend state to the frontend.
The connected value as 0 will be match the default of a zero-ed console
page. Hyperlaunch can set the flag to XENCONSOLE_DISCONNECTED and let
xenconsoled set to XENCONSOLE_CONNECTED.
Old domU hvc_xen drivers won't check the flag.
New domU hvc_xen running on a new xen/xenconsoled will work properly.
New domU hvc_xen on old xen/xenconsoled should only see a 0 for the flag
and behave as if connected.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Release-Acked-By: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
C Xenstored was seen failing to map a domU's event channel. Enhancing
Xen's error messages shows:
common/event_channel.c:427:d0v0 EVTCHNOP failure: d1, port 3267543746 error -22
Port 3267543746 is 0xc2c2c2c2 - SCRUB_PATTERN.
Xenstore expects a non-zero evtchn_port to be valid. Explicitly zero
the page to avoid this.
Fixes: df2c03e2eb ("xen/arm: Alloc XenStore page for Dom0less DomUs from hypervisor")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Acked-by: Michal Orzel <michal.orzel@amd.com>
Update get_implicit_path to return the correct value for a non-dom0
xenstored domain.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
posix.c and minios.c implement the same named functions serving slightly
different purposes.
For xenbus_map()
posix.c maps the local /dev/xen/xsd_kva
minios.c maps dom0 via grant and there is open coding for stub_domid in
map_interface.
Change xenbus_map() to map the local domain's interface. The default
grant table mapping is performed otherwise.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
The dom0_domid variable is misnamed and conflates purposes. If we have
xenstored running in a Linux domain that is not dom0, this variable
controls the lookup of /proc/xen/xsd_kva and the event channel.
Rename to store_domid to better show its purpose.
One implication of this change is that the xenstore domain is not
privileged by virtue of considering store_domid as privileged.
domain_is_unprivileged() removes the dom0_domid/store_domid check, so
xenstore domain is no longer considered privileged.
onearg_domain() is updated to return EINVAL for store_domid or priv_domid
to maintain the ability to call XS_RESUME.
xenbus_master_domid() is removed with store_domid being used instead.
Add a description of the -m/--master-domid options while
doing this.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Usually, priv_domid == dom0_domid == 0, and that is what is expected.
If we rename s/dom0_domid/store_domid/, it seems more likely we want to
actually have the priv_domid as the owner.
That leads to follow on changes to ensure that the priv_domid is created
first.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
With separate control and xenstore domains, the xenstore domain itself
is no longer privileged. Local socket connections do not have
conn->domain assigned, but domain_chk_quota() will go ahead and
dereference it. The !domid_is_unprivileged(conn->id) check usually
means the function would have already returned.
Bail out on !conn->domain to avoid faulting. This adds a similar check
to that in domain_max_chk()
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Replace dom0_init() with init_domains() which uses libxenmanage to
iterate through all existing domains, storing them in a list. The xenstore
domain is introduced first, and then all the other domains are
introduced. The xenstore domain needs to be introduced first to setup
structures needed for firing watches.
dom0_domid is updated with the xenstore domain, since it really
indicates the local domain.
priv_domid is set to the control domain. This makes it limited to a
single domain.
These features let xenstore automatically connect any existing domains,
which means it doesn't need to be done manually from init-dom0less.
For a legacy dom0, the result should be unchanged.
For a late xenstore stubdom it should also be the same, but priv_domid
would be set automatically to control domain (which default to 0
normally).
Always signal the event channel for initial domains. This gets dom0 (a
local xenstored domain) to connect.
Also always write XENSTORE_CONNECTED since we know we are connected at
this point.
To support ARM dom0less domains with xen,enhanced = "no-xenstore" a
failed introduce_domain() becomes non-fatal. Normally,
HVM_PARAM_STORE_EVTCHN is used to identify .
priv_domid from the command line is used, or the first control domain is
used.
dom0_domid will set to the last xenstore found. This will handle dom0
or dom0less, where only 1 xenstore domain can exist, or stubdom, where
dom0 and dom1 exist, and we want to take the stubdom.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Add helpers to lookup the event channel for a domid. This hides some
of the differences between dom0 and stubdom xenstored. Each version
defines its own.
It highlights the different meanings between get_xenbus_evtchn() in a
stubdom, where it looks up dom0's event channel, and dom0, where it
looks up the local event channel.
get_domain_evtchn() replaces get_xenbus_evtchn(), and
get_xenbus_evtchn() is removed from minios.c as it is inlined in the new
function.
The default return 0 will be fine as any other auto-introduced domain
will needs the event channel populated in the grant.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Make introduce_domain() use an event channel from the the xenstore page.
It is only used if non-zero. Otherwise the passed in event channel port
is used.
The is useful for a xenstored stubdom to configure domains autonomously.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Add an additional "caps" argument to the libxenmanage functions to
obtain a domains capabilities - control, hardware, and xenstore.
Update the xenstored callers at the same time.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
Include the event channel in struct xenstore_domain_interface. This way
the toolstack or xen can communicate the event channel to xenstored in
memory xenstored already needs to access.
xenstored maps the grant with the well known GNTTAB_RESERVED_XENSTORE
index, so no further information is needed.
Suggested-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Expose a domain's capabilities - control, hardware or xenstore - through
stable get domain state hypercall.
The xenstore domain can use this information to assign appropriate
permissions on connections.
Repurpose the 16bit pad field for this purpose.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@amd.com>
Only access the HVM union b_info->u.hvm on HVM guests. The union
access is not guarded, so this reads and sets the default even on
non-HVM guests. Usually this doesn't matter as PV and PVH unions are
smaller and zero-initialized, but the zero default will be re-written as
a -1 boolean. Generally, it could incorrectly set b_info->altp2m
through aliased data.
Fixes: 0291089f6e ("xen: enable altp2m at create domain domctl")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
is_periodic_irq() was removed in the Fixes commit, but the prototype
remained. Drop it now.
Fixes: ddc35d1cc9 ("[HVM] Enable more than one platform timer...")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
The format string lacks a space, so mfn and type run together:
(XEN) d0v0 0000:06:00.7: existing mapping (mfn: 753037type: 0) at 0x1 clobbers MSIX MMIO area
Add a space. Additionally, move the format string to a single long line
to improve grep-ability.
Fixes: 677053fac1 ("vpci/msix: carve p2m hole for MSIX MMIO regions")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
With a tab stop of 8, the alignment is off. Replace the hard tabs with
spaces to match the file.
Fixes: 284d5633be ("Tools: add online/offline hotplug user interfaces")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Denis Mukhin <dmukhin@ford.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
The PCI device irq is read with fscanf(%u). Switch the irq variable to
unsigned int to match.
Linux driver/pci/pci-sysfs.c:irq_show() uses %u to print the value.
However, unsigned int irq doesn't compile because of:
error: pointer targets in passing argument 4 of 'xc_physdev_map_pirq' differ in signedness [-Werror=pointer-sign]
Add int pirq to provide the desired type instead of re-using irq.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
A PCI device's irq field is an 8-bit number. A value of 0xff indicates
that the device IRQ is not connected. Additionally, the Linux ACPI code
can convert these 0xff values to IRQ_NOTCONNECTED(0x80000000) because
"0x80000000 is guaranteed to be outside the available range of
interrupts and easy to distinguish from other possible incorrect
values." When the hypercall to assign that IRQ fails, device
passthrough as a whole fails.
Add checking for a valid IRQ and skip the IRQ handling for PCI devices
outside that range. This allows for passthrough of devices without
legacy IRQs.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
Use of setjmp/longjmp as removed in 2006, but the include remained.
Remove it now.
Fixes: 1bac3b49cd ("Import the current version of talloc from the Samba 3 source base")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Add a capabilities property to dom0less to allow building a
disaggregated system. Only a single hardware domain and single xenstore
domain can be specified. Multiple control domains are possible.
Introduce bootfdt.h to contain these constants.
When using the hardware or xenstore capabilities, adjust the grant and
event channel limits similar to dom0.
For a hardware domain, disallow specifying "vpl011", "nr_spis",
"multiboot,device-tree" and "passthrough" nodes. Also, require an IOMMU
when not direct-mapped,
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Store the result of finding a "multiboot,device-tree" node. This will
simplity adding hardware domain checks.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
The hardware domain is unable to seed a control domain, but we want the
control domain to use xenstore. Rely on the hypervisor to seed dom0less
grant table entries for Xenstore, so this seeding is unnecessary.
However, that only works for the new xenstore late init. The legacy
protocol which uses init-dom0less to populate the page still needs to
seed the grant.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
xenstored maps other domains' xenstore pages. Currently this relies on
init-dom0less or xl to seed the grants from Dom0. With split
hardware/control/xenstore domains, this is problematic since we don't
want the hardware domain to be able to map other domains' resources
without their permission. Instead have the hypervisor seed the grant
table entry for every dom0less domain. The grant is then accessible as
normal.
C xenstored uses grants, so it can map the xenstore pages from a
non-dom0 xenstore domain. OCaml xenstored uses foreign mappings, so it
can only run from a privileged domain (dom0).
Add a define to indicate the late alloc xsentore PFN, to better indicate
what is being checked. Use UINT64_MAX instead of ~0ULL as the HVM_PARAM
field is a uint64_t. UINT64_MAX is not defined, so add it.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
To allocate the xenstore event channel and initialize the grant table
entry, the xenstore domid is neeed. A dom0 is created before the domUs,
so it is normally available through hardware_domain. With capabilities
and dom0less, the xenstore domain may not be created first.
Keep the population of the page and HVM_PARAM_STORE_PFN in the normal
domain construction, but delay event channel creation and grant seeding
to after all domUs are created. HVM_PARAM_STORE_PFN now serves as
indication to setup xenstore since the device tree is no longer
immediately available. 0 means no xenstore. ~0ULL means legacy so only
the event channel needs setup, and any other value means to seed the
page.
dom0 needs to set xs_domid when it is serving as the xenstore domain.
The domain running xenstored needs to be the handler for VIRQ_DOM_EXC,
so set that as well - it otherwise defaults to hardware domain.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
When creating a hardware domain, have the dom0less code call
construct_hwdom() which is shared with the dom0 code. The hardware
domain requires building that best matches the dom0 build path. Re-use
it to keep them in sync.
The device tree node of the dom0less config is now passed into
construct_hwdom(). dom0 uses /chosen for process_shm while a hwdom will
use the value from its dom0less device tree node.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
A PCI device may not have a legacy IRQ. In that case, we don't need to
do anything, so don't fail in libxl__arch_hvm_map_gsi() and
libxl__arch_hvm_unmap_gsi().
Requires an updated pciback to return -ENOENT.
Fixes: f97f885c71 ("tools: Add new function to do PIRQ (un)map on PVH dom0")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
It is valid for a PCI device to not have a legacy IRQ. In that case, do
not print an error to keep the logs clean.
This relies on pciback being updated to return -ENOENT for a missing
GSI.
Fixes: b93e5981d2 ("tools: Add new function to get gsi from dev")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
Factor out construct_hwdom() from construct_dom0(). This will be
re-used by the dom0less code when building a domain with the hardware
capability.
iommu_hwdom_init(d) is moved into construct_hwdom() which moves it after
kernel_probe(). kernel_probe() doesn't seem to depend on its setting.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
The xl subcommands are generic, but only built when
LIBXL_HAVE_NO_SUSPEND_RESUME is false. Currently that is only x86, so
list them there.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Oleksii Kurochko<oleksii.kurochko@gmail.com>
When a VM transitioned to LIBXL_SHUTDOWN_REASON_SUSPEND, the xl daemon
was exiting as 0 = DOMAIN_RESTART_NONE "No domain restart".
Later, when the VM actually shutdown, the missing xl daemon meant the
domain wasn't cleaned up properly.
Add a new DOMAIN_RESTART_SUSPENDED to handle the case. The xl daemon
keeps running to react to future shutdown events.
The domain death event needs to be re-enabled to catch subsequent
events. The libxl_evgen_domain_death is moved from death_list to
death_reported, and then it isn't found on subsequent iterations through
death_list. We enable the new event before disabling the old event, to
keep the xenstore watch active. If it is unregistered and
re-registered, it'll fire immediately for our suspended domain which
will end up continuously re-triggering.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
xl/libxl memory is leaking into a PVH guest through uninitialized
portions of the ACPI tables.
Use libxl_zalloc() to obtain zero-ed memory to avoid this issue.
This is XSA-464 / CVE-2024-45819.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Fixes: 14c0d328da ("libxl/acpi: Build ACPI tables for HVMlite guests")
Reviewed-by: Jan Beulich <jbeulich@suse.com>
"$dev" needs to be set correctly for backendtype=phy as well as
backendtype=tap. Move the setting into the conditional, so it can be
handled properly for each.
(dev could be captured during tap-ctl allocate for blktap module, but it
would not be set properly for the find_device case. The backendtype=tap
case would need to be handled regardless.)
Fixes: f16ac12bd4 ("hotplug: Restore block-tap phy compatibility")
Fixes: 76a484193d ("hotplug: Update block-tap")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
backendtype=phy using the blktap kernel module needs to use write_dev,
but tapback can't support that. tapback should perform better, but make
the script compatible with the old kernel module again.
Fixes: 76a484193d ("hotplug: Update block-tap")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Anthony PERARD <anthony.perard@vates.tech>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Python 3.12.2 warns:
xen/tools/gen-cpuid.py:50: SyntaxWarning: invalid escape sequence '\s'
"\s+([\s\d]+\*[\s\d]+\+[\s\d]+)\)"
xen/tools/gen-cpuid.py:51: SyntaxWarning: invalid escape sequence '\s'
"\s+/\*([\w!]*) .*$")
Specify the strings as raw strings so '\s' is read as literal '\' + 's'.
This avoids escaping all the '\'s in the strings.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
An XSM hook for get_dom0_console is currently missing. Using XSM with
a PVH dom0 shows:
(XEN) FLASK: Denying unknown platform_op: 64.
Wire up the hook, and allow it for dom0.
Fixes: 4dd160583c ("x86/platform: introduce hypercall to get initial video console settings")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Daniel P. Smith <dpsmith@apertussolutions.com>
Switch to unsigned int for the dom0 e820 index. This eliminates the
potential for array underflows, and the compiler might be able to
generate better code.
Requested-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Xen tries to load a PVH dom0 kernel at the fixed guest physical address
from the elf headers. For Linux, this defaults to 0x1000000 (16MB), but
it can be configured.
Unfortunately there exist firmwares that have reserved regions at this
address, so Xen fails to load the dom0 kernel since it's not RAM.
The PVH entry code is not relocatable - it loads from absolute
addresses, which fail when the kernel is loaded at a different address.
With a suitably modified kernel, a reloctable entry point is possible.
Add XEN_ELFNOTE_PHYS32_RELOC which specifies optional alignment,
minimum, and maximum addresses needed for the kernel. The presence of
the NOTE indicates the kernel supports a relocatable entry path.
Change the loading to check for an acceptable load address. If the
kernel is relocatable, support finding an alternate load address.
The primary motivation for an explicit align field is that Linux has a
configurable CONFIG_PHYSICAL_ALIGN field. This value is present in the
bzImage setup header, but not the ELF program headers p_align, which
report 2MB even when CONFIG_PHYSICAL_ALIGN is greater. Since a kernel
is only considered relocatable if the PHYS32_RELOC elf note is present,
the alignment contraints can just be specified within the note instead
of searching for an alignment value via a heuristic.
Load alignment uses the PHYS32_RELOC note value if specified.
Otherwise, the maxmum ELF PHDR p_align value is selected if greater than
or equal to PAGE_SIZE. Finally, the fallback default is 2MB.
libelf-private.h includes common-macros.h to satisfy the fuzzer build.
Link: https://gitlab.com/xen-project/xen/-/issues/180
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
While parsing the PHDRs, store the maximum p_align value. This may be
consulted for moving a PVH image's load address.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Consolidate to a single set of common macros for tools.
MB() will gain another use in libelf, so this movement makes it
available.
Requested-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
The local MB() & GB() macros will be replaced with a common
implementation, but those only work with numeric values. Introduce a
static inline mb_to_bytes() in place of the MB() macro to convert the
variable values, at the same time addressing the lack of
parenthesization of the prior macro's parameter.
Fixes: 134d53f577 ("tools/init-xenstore-domain: fix memory map for PVH stubdom")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Anthony PERARD <anthony.perard@citrix.com>
The Xen ELF Notes are only used with x86. libelf's elf_xen_note_check()
exits early for ARM binaries with "ELF: Not bothering with notes on
ARM".
Add a comment to the top of elfnote.h specifying that Notes are only used
with x86 binaries. This is to avoid adding disclaimers for individual
notes.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
The XEN_ELFNOTE_L1_MFN_VALID is an array of pairs of values, but it is
printed as:
(XEN) ELF: note: L1_MFN_VALID = 0
This is a limitation of only printing either string or numeric values.
Switch from the boolean to an enum which allows more flexibility in
printing the values. Introduce ELFNOTE_NAME to only print the name
without a value like:
(XEN) ELF: note: L1_MFN_VALID
Details can optionally be printed for specific notes.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
A new ELF note will specify the alignment for a relocatable PVH kernel.
ELF notes are suitable for vmlinux and other ELF files, so this
Linux-specific bzImage parsing in unnecessary.
This reverts commit c44cac2290.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Expand bzimage_parse() to return kernel_alignment from the setup_header.
This will be needed if loading a PVH kernel at a physical offset to
compensate for a reserved E820 region.
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Acked-by: Jan Beulich <jbeulich@suse.com>
Replace hard tabs to match the rest of the file.
Fixes: 48a3fd1432 ("domain: expose newly introduced hypercalls as XENFEAT")
Signed-off-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Jason Andryuk