Since commit
f2ba877402 ("arm64: mm: Map the kernel data/bss read-only in the linear map")
the linear alias of the .data and .bss regions is remapped read-only
early during the boot. (Note that a subsequent patch to unmap this
region entirely was reverted just before the v7.2 merge window, and will
be brought back in an improved form for the v7.3 cycle)
Fuad reports that in some cases, the KVM init code may apply relocations
to variables that reside in .data, and does so via the linear map. This
means that remapping .data read-only beforehand is a bad idea, and
results in an early boot crash.
These variables in .data are only present when CONFIG_NVHE_EL2_DEBUG or
CONFIG_NVHE_EL2_TRACING are enabled, which is why it was not spotted in
testing.
So move the remap to mark_rodata_ro(), which is a reasonable place to
put this, and ensures that it happens much later during the boot. It
also means that rodata=off is now taken into account, and so the linear
alias will remain writable in that case.
Fixes: f2ba877402 ("arm64: mm: Map the kernel data/bss read-only in the linear map")
Reviewed-by: Fuad Tabba <fuad.tabba@linux.dev>
Tested-by: Fuad Tabba <fuad.tabba@linux.dev>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kevin Brodsky <kevin.brodsky@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
"fan-supply" already has a type and shouldn't have a type $ref. Drop the
$ref to fix the warning.
Fixes: 39dd85d924 ("dt-bindings: mfd: khadas: Add new compatible for Khadas VIM4 MCU")
Link: https://patch.msgid.link/20260624133643.4146351-1-robh@kernel.org
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
When function tracing or Kprobes is enabled, entering an ACPI Low
Power Idle (LPI) state triggers the following RCU splat:
RCU not on for: acpi_idle_lpi_enter+0x4/0xd8
WARNING: CPU: 8 PID: 0 at include/linux/trace_recursion.h:162 function_trace_call+0x1e8/0x228
The acpi_idle_lpi_enter() function is invoked within the cpuidle
path after RCU has already been disabled for the current local CPU.
Consequently, ftrace's function_trace_call() expects RCU to be
actively watching before recording trace data, emitting a warning
if it is not.
Fix this by annotating acpi_idle_lpi_enter(), the generic __weak
stub, and the RISC-V implementation of acpi_processor_ffh_lpi_enter()
with __cpuidle. This moves these functions into the '.cpuidle.text'
section, implicitly disabling ftrace instrumentation (notrace) along
this sensitive path and preventing trace-induced RCU warnings during
idle entry.
Fixes: a36a7fecfe ("ACPI / processor_idle: Add support for Low Power Idle(LPI) states")
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Acked-by: lihuisong@huawei.com
Link: https://patch.msgid.link/20260616072617.2272-1-lirongqing@baidu.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Add bindings for the regulators found in the Spreadtrum/Unisoc SC2730
PMIC, used e.g. with the UMS512 and UMS9230 SoCs.
Signed-off-by: Otto Pflüger <otto.pflueger@abscue.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Link: https://patch.msgid.link/20260620-sc2730-regulators-v6-1-bbd2db395231@abscue.de
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Commit 97f7d3f9c9 ("ACPICA: Replace strncpy() with strscpy_pad() in
acpi_ut_safe_strncpy()") switched over the ACPICA code in the kernel to
using strscpy_pad() instead of a combination of strncpy() and manual
NUL-termination of the destination string, but it overlooked the fact
that tools also use the code in question and strscpy_pad() is not
defined in those builds.
Address that by using the original ACPICA code in non-kernel builds.
Fixes: 97f7d3f9c9 ("ACPICA: Replace strncpy() with strscpy_pad() in acpi_ut_safe_strncpy()")
Reported-by: Jiri Slaby <jirislaby@kernel.org>
Closes: https://lore.kernel.org/all/79e9e913-0fb1-4110-804b-c3b5d0edafe4@kernel
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Kees Cook <kees@kernel.org>
[ rjw: Fixed up the number of added code lines ]
Link: https://patch.msgid.link/12923581.O9o76ZdvQC@rafael.j.wysocki
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The example DTS is tested in a wrapped node with address/size-cells=1,
thus reg had incorrect entry leading to dt_binding_check fails:
thermal/amlogic,thermal.example.dtb: temperature-sensor@20000 (amlogic,t7-thermal): reg: [[0, 131072], [0, 80]] is too long
Fixes: b1c8ccdbd4 ("dt-bindings: thermal: amlogic: Add support for T7")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://patch.msgid.link/20260622100231.438435-4-krzysztof.kozlowski@oss.qualcomm.com
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
The @filter_np parameter descriptions in of_map_id() and of_map_msi_id()
contained the text '*filter_np' in prose. Docutils interprets a leading
'*' as the start of RST emphasis (italic), but finds no closing '*',
triggering:
Documentation/devicetree/kernel-api:11: ./drivers/of/base.c:2134:
WARNING: Inline emphasis start-string without end-string. [docutils]
Documentation/devicetree/kernel-api:11: ./drivers/of/base.c:2260:
WARNING: Inline emphasis start-string without end-string. [docutils]
Fix by wrapping '*filter_np' in double backticks (*filter_np) to
render it as an RST inline code literal, which is also the correct
kernel-doc convention for pointer expressions.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202606130111.ldC96rqf-lkp@intel.com/
Signed-off-by: Vijayanand Jitta <vijayanand.jitta@oss.qualcomm.com>
Link: https://patch.msgid.link/20260619-iommu_map_kdoc_fix-v1-1-9573e1cf30b3@oss.qualcomm.com
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
fwnode_property_get_reference_args() should return -ENOENT when an out
of bound index is passed. An issue arised with the OF backend because
the OF API use signed indexes while the fwnode API use unsigned ones.
When an index value greater the INT_MAX was passed to the OF backend
it got casted to a negative value and it returned -EINVAL instead of
-ENOENT. This patch add a check to of_fwnode_get_reference_args() to
catch negative index before they are passed to the OF API and return
-ENOENT right away.
This issue appeared when the following pattern was used in the LED
subsystem:
index = fwnode_property_match_string(fwnode, "led-names", name)
led_node = fwnode_find_reference(fwnode, "leds", index);
Unlike the same pattern with the OF API, this pattern implicitly cast
the signed return value of fwnode_property_match_string() to an
unsigned index leading to the above issue with the OF backend. It can
be argued that the return value of fwnode_property_match_string()
should be checked separately, but I think there is value in supporting
such simple and straight to the point patterns.
Link: https://lore.kernel.org/linux-leds/aimVRwJPhlGxsIUj@tom-desktop/T/#mc43cbf7e0599991b56dd0d9680714d28d145fbc8
Cc: Tommaso Merciai <tommaso.merciai.xr@bp.renesas.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Alban Bedel <alban.bedel@lht.dlh.de>
Link: https://patch.msgid.link/20260618152035.1600436-1-alban.bedel@lht.dlh.de
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Otherwise zone append commands will miss their integrity data. While
this works "fine" for auto-PI, it break file system PI and non-PI
metadata.
With this XFS on ZNS namespace with non-PI metadata and 512 byte sectors
with PI work, while PI 4k sector formats with PI work only when Caleb's
"block: fix integrity offset/length conversions" is applied as well.
Note that unlike regular writes, zone append does need remapping as
partitions are not supported on zoned block devices.
Fixes: df3c485e0e ("block: switch on bio operation in bio_integrity_prep")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://patch.msgid.link/20260624080014.1998650-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
bio_integrity_alloc_buf usage of GFP_ flags is messed up. For one it
mixes GFP_NOFS and GFP_NOIO for neighbouring allocations, but it also
makes the allocations fail more often than needed. That code was copied
from bio_alloc_bioset which needs to do that so that it can punt to the
rescuer workqueue, but none of that is needed for the integrity
allocations that either sits in the file system or at the very bottom
of the I/O stack. Failing early means we'll do a fully waiting
allocation from the mempool ->alloc callback which is usually much
larger than required.
Fix this by passing a gfp_t so that the file system path can pass
GFP_NOFS and the auto-integrity code can pass GFP_NOIO, and don't
modify the allocation type except for disabling warnings.
Fixes: ec7f31b2a2 ("block: make bio auto-integrity deadlock safe")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://patch.msgid.link/20260624080014.1998650-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Take a css reference under RCU, drop RCU, and then associate the bio with
the blkg. This avoids nesting queue_lock under RCU and prepares to protect
blkcg with blkcg_mutex instead of queue_lock.
Use css_tryget() instead of css_tryget_online() so swap writeback for
pages charged to a dying memcg still passes the dying css to
bio_associate_blkg_from_css(). That preserves the existing closest-live
ancestor fallback instead of charging those bios to the root blkg.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/c910d2c39d3ec97f67de68af636a52394342d55f.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The correct lock order is q->queue_lock before blkcg->lock, and in order
to prevent deadlock from blkcg_destroy_blkgs(), trylock is used for
q->queue_lock while blkcg->lock is already held, this is hacky.
Refactor blkcg_destroy_blkgs() to hold blkcg->lock only long enough to
get the first blkg and then release it. Then take q->queue_lock and
blkcg->lock in the correct order to destroy the blkg. This is a very cold
path, so the extra lock/unlock cycles are acceptable.
Also prepare to convert protecting blkcg with blkcg_mutex instead of
queue_lock.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/00b03cf74a9937cb4d6dd67a189ddc00a3de0451.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If a bio is already associated with a blkg, the blkcg is already pinned
until the bio is done, so there is no need for RCU protection. Otherwise,
protect blkcg_css() with RCU independently. Prepare to protect blkcg with
blkcg_mutex instead of queue_lock.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/8496fa234b21d4b31b7f068766906d0bffcac8e6.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Change this in two steps:
1) hold rcu lock and do blkg_lookup() from fast path;
2) hold queue_lock directly from slow path, and don't nest it under rcu
lock;
Prepare to convert protecting blkcg with blkcg_mutex instead of
queue_lock.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/93f33cc9e5a39dddb78dcd934d0c1d04b564fb00.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
With previous modification to delay freeing policy data after an RCU grace
period, prfill() can run under RCU instead of taking queue_lock. However,
policy teardown can still clear blkg->pd[plid] after blkcg_print_blkgs()
observes the policy enabled bit.
Load policy data once with READ_ONCE() and skip the blkg if teardown
already cleared it. Do the same in recursive stat walks for descendant
blkgs. Remove the stale BFQ debug queue_lock assertion because
blkcg_print_blkgs() no longer calls prfill() with queue_lock held. This
also lets ioc_qos_prfill() and ioc_cost_model_prfill() use IRQ-safe
ioc->lock locking without re-enabling IRQs while queue_lock is still held.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/db7633d5e263dd1c2bf9b901762545a84b7d714e.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently blkcg_print_blkgs() must hold RCU to iterate blkgs from a
blkcg, and prfill() must hold queue_lock to prevent policy data from
being freed by policy deactivation. As a consequence, queue_lock has to
be nested under RCU from blkcg_print_blkgs().
Delay freeing policy data until after an RCU grace period so prfill() can
be protected by RCU alone.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/e20e5d984b41a026d61851966bed35eb094c4bff.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blkcg_print_one_stat() will be called for each blkg:
- access blkg->iostat, which is freed from rcu callback
blkg_free_workfn();
- access policy data from pd_stat_fn(), which is freed from
pd_free_fn(), while pd_free_fn() can be called by removing blkcg or
deactivating policy;
Take blkcg->lock while iterating so the blkgs stay online and both
blkg->iostat and policy data for activated policies stay valid. Use
irq-safe locking because blkcg->lock can be nested under q->queue_lock,
which is used from IRQ completion paths.
Prepare to convert protecting blkgs from request_queue with mutex.
Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/05799877e720dcd300e2ddd4625e8e162959d7cc.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If device_link_add() fails in snd_soc_bind_card() just skip that
driver pair and carry on.
This means that ASoC must now keep track of which components it was
able to device_link to the card->dev. The new card_device_link member
of struct snd_soc_component is non-NULL if a device_link exists.
The intent of the device link is to ensure that the machine driver
system-suspends before the component drivers, to prevent ASoC
suspend attempting to reconfigure a driver that has already suspended.
It isn't possible to create this device link if the machine driver is
a parent of the component driver or already has a device_link in the
opposite direction. In this case skip the link. A warn is placed in
kernel log since this might indicate a genuine design problem with
those two drivers (this can be downgraded to dbg in future when
people are happy that all these special drivers correctly handle their
reversed shutdown order).
Fixes: 0f54ce994b ("ASoC: soc-core: Create device_link to ensure correct suspend order")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Closes: https://lore.kernel.org/linux-sound/61bd38e7-5eb9-4448-a93f-afa2ccbd1c9d@opensource.cirrus.com/T/#m496fe5a11b0a3649afd2e85da5e1cea82bb16d8a
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://patch.msgid.link/20260623135821.4125543-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently that header is only included via:
<sound/dmaengine_pcm.h> ->
<sound/soc.h> ->
<linux/platform_device.h>
which doesn't look reliable, still more in the presence of the comment:
/* For the current users of sound/soc.h to avoid build issues */
in <sound/soc.h>.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
Reviewed-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
Link: https://patch.msgid.link/20260624083708.254517-2-u.kleine-koenig@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>
da9121_i2c_probe() stores i2c_get_match_data() in chip->subvariant_id
and da9121_assign_chip_model() switches on DA9121_SUBTYPE_* values. The
OF table provides those subvariant values, but the I2C id table
currently provides DA9121_TYPE_* values.
Make the I2C id table use the same subvariant namespace as the OF table
so non-DT I2C matches feed the expected data type into the model
assignment code.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260624060024.61300-1-pengpeng@iscas.ac.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
Use cpu_physical_id() instead of default_cpu_present_to_apicid() when
getting the APIC ID of the pCPU on which a vCPU is running/loaded, as the
kernel has gone way off the rails if a vCPU is loaded on a pCPU that has
been physically removed from the system. Even if the impossible were to
happen, the absolutely worst case scenario is that hardware will ring the
AIVC doorbell on the wrong pCPU, i.e. a severely broken system will
experience mild performance issues.
Kill off KVM's superfluous kvm_cpu_get_apicid() wrapper along with the
for-KVM export of default_cpu_present_to_apicid(), as they existed purely
for the wonky AVIC usage.
Cc: Kai Huang <kai.huang@intel.com>
Cc: Yosry Ahmed <yosry@kernel.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Acked-by: Naveen N Rao (AMD) <naveen@kernel.org>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Reviewed-by: Yosry Ahmed <yosry@kernel.org>
Message-ID: <20260612185459.591892-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Turn arch.n_used_mmu_pages into a stat, mmu_shadow_pages, as the number of
live shadow pages is arguably _the_ most critical datapoint when it comes
to analyzing the shadow MMU. Before the TDP MMU came along, i.e. when the
shadow MMU was the only MMU, explicitly tracking the number of shadow pages
wasn't as interesting, because the same information could more or less be
gleaned from the pages_{1g,2m,4k} stats. But with the TDP MMU, where the
shadow MMU is only used for nested TDP, it becomes extremely difficult, if
not impossible, to determine which SPTEs are coming from the TDP MMU, and
which are coming from the shadow MMU.
E.g. when triaging/debugging shadow MMU performance issues due to "too many
shadow pages", being able to observe that 99%+ of all shadow pages are
unsync is critical to being able to deduce that KVM is effectively leaking
shadow pages.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20260612133727.411902-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-----BEGIN PGP SIGNATURE-----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=OiBj
-----END PGP SIGNATURE-----
Merge tag 'kvm-s390-next-7.2-2' of https://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD
* Fix S390_USER_OPEREXEC so it can now be enabled regardless of other
unrelated capabilities
* Fix handling of the _PAGE_UNUSED pte bit that could lead to guest
memory corruption in some scenarios
* A bunch of misc gmap fixes (locking, behaviour under memory pressure)
* Fix CMMA dirty tracking
The GOP (and even Bspec on some platforms) is a bit inconsistent
on what the CDCLK_FREQ_DECIMAL divider should be. Currently any
mismatch there causes a full CDCLK PLL disable+re-enable, which
we really don't want to do if any displays are currently active.
Let's instead just reprogram CDCLK_FREQ_DECIMAL when that is the
only thing amiss. For any other (more serious) mismatch we still
punt to the full PLL reprogramming.
We also need to tweak the bxt_cdclk_cd2x_pipe() stuff a bit to
consistently select pipe==NONE since we have no idea which pipes
are enabled at this point. Since we're not actually changing the
CDCLK frequency here we don't need to sync the update to any
pipe.
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/work_items/16209
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260612173653.7830-2-ville.syrjala@linux.intel.com
Reviewed-by: Michał Grzelak <michal.grzelak@intel.com>
(cherry picked from commit 3f9de66f8acbf8ff45a91b4920605ed10c6b7c06)
Fixes: ba91b9eecb ("drm/i915/cdclk: Decouple cdclk from state->modeset")
Fixes: d66a21947e ("drm/i915/bxt: Sanitize CDCLK to fix breakage during S4 resume")
Fixes: c73666f394 ("drm/i915/skl: If needed sanitize bios programmed cdclk")
Cc: <stable@vger.kernel.org> # v4.5+
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
The TPR_THRESHOLD field in the VMCS is used by VMX to induce VM exits
when the guest's virtual TPR falls under the specified threshold,
allowing KVM to inject previously masked interrupts.
KVM handles these VM exits in handle_tpr_below_threshold().
Commit eb90f3417a ("KVM: vmx: speed up TPR below threshold vmexits")
optimized this function by calling apic_update_ppr() instead of raising
KVM_REQ_EVENT. apic_update_ppr() then raises KVM_REQ_EVENT if there is
a pending, deliverable interrupt.
However, if there are no new interrupts pending, apic_update_ppr() does
not issue the request. Thus, kvm_lapic_update_cr8_intercept() and
vmx_update_cr8_intercept() are not called before VM entry, which results
in a high, stale TPR_THRESHOLD. This is problematic due to the following
sentence in 28.2.1.1 "VM-Execution Control Fields" in the SDM:
The following check is performed if the “use TPR shadow” VM-execution
control is 1 and the “virtualize APIC accesses” and “virtual-interrupt
delivery” VM-execution controls are both 0: the value of bits 3:0 of
the TPR threshold VM-execution control field should not be greater
than the value of bits 7:4 of VTPR.
This error condition is typically not observed when KVM runs on a bare
metal system because modern processors support APICv, which enables
virtual-interrupt delivery, and which KVM uses when possible. This
causes the processor to no longer generate TPR-below-threshold exits
and to no longer check TPR_THRESHOLD on entry. However, when running
on older platforms, or under nested virtualization on a hypervisor that
does not support virtual-interrupt delivery and enforces this check
(like Hyper-V) this can cause a VM entry failure with hardware error
0x7, as seen in [1].
Call kvm_lapic_update_cr8_intercept() if apic_update_ppr() does not
find a deliverable interrupt (and thus does not raise KVM_REQ_EVENT).
Remove calls to kvm_lapic_update_cr8_intercept() on paths that end up in
apic_update_ppr(), as they now become redundant. This ensures that any
path that updates the guest's PPR also figures out if KVM needs to wait
for a TPR change (using TPR_THRESHOLD on VMX or CR8 intercepts on SVM).
Link: https://github.com/coconut-svsm/svsm/issues/1081 [1]
Tested-by: Stefano Garzarella <sgarzare@redhat.com>
Cc: stable@vger.kernel.org
Fixes: eb90f3417a ("KVM: vmx: speed up TPR below threshold vmexits")
Signed-off-by: Carlos López <clopez@suse.de>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20260618174347.1981064-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
WARN once if KVM's tracking for pending EOIs for Real-Time Clock IRQs goes
off the rails, as there's no reason to bug the host or risk a DoS due to
spamming dmesg with endless WARNs. Absolute worst case scenario, guest
time will go awry.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Message-ID: <20260618174527.1982333-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
WARN and return an error up the stack if the PIC or I/O APIC encounters an
invalid vector when injecting an IRQ, as there is no danger to the host and
thus no justification for potentially panicking the kernel. Don't bug the
VM either, as the risk of corrupting the guest is minuscule, and the guest
might even be completely tolerant of a lost interrupt.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Message-ID: <20260618185213.2019937-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Bug the VM, not the host kernel, if KVM's ISR count {under,over}flows when
tracking in-flight ISRs. There is zero danger to the host if KVM messes up
its IRQ tracking.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Message-ID: <20260618185350.2020845-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Instead of bugging the host kernel, WARN and terminate the VM if KVM
attempts to write-protect at a level that cannot use leaf SPTEs.
There is no reason to bring down the entire host; even termininating
the VM is likely overkill, but in theory a missed write could corrupt
guest memory, so play it safe.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Message-ID: <20260618185641.2022368-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If KVM attempts to translate what it thinks is an L2 GPA with a non-nested
MMU, simply WARN and return the GPA, i.e. trust the MMU more than the
caller, as there is zero reason to potentially panic the host kernel just
because KVM misused an API.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Message-ID: <20260618185746.2023283-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Drop a BUG_ON() that has been reachable since it was first added, way back
in 2009, and instead use get_unaligned() to perform potentially-unaligned
accesses.
For a given store, KVM x86's emulator tracks the entire value in the
destination operand, x86_emulate_ctxt.dst. If the destination is memory,
and the target splits multiple pages and/or is emulated MMIO, then KVM
handles each fragment independently. E.g. on a page split starting at page
offset 0xffc, KVM writes 4 bytes to the first page, then the remaining
bytes to the second page, using ctxt->dst as the source for both (with
appropriate offsets).
If the destination splits a page *and* hits emulated MMIO on the second
page, then KVM will complete the write to the first page, then emulate the
MMIO access to the second page. If there is a datamatch-enabled ioeventfd
at offset 0 of the second page, then KVM will process the remainder of the
store as a potential ioeventfd signal.
Putting it all together, if the guest emits a store that splits a page
starting at page offset N, and the second page has a datamatch-enabled
ioeventfd at offset 0, then KVM will check for datamatch using
&dst.valptr[N] as the source. Due to dst (and thus dst.valptr) being
32-byte aligned, if N is not aligned to @len, the BUG_ON() fires.
E.g. with a 16-byte store at page offset 0xffc, to an ioeventfd of len 8,
all initial checks in ioeventfd_in_range() will succeed, and the BUG_ON()
fires due to @val being 4-byte aligned, but not 8-byte aligned.
------------[ cut here ]------------
kernel BUG at arch/x86/kvm/../../../virt/kvm/eventfd.c:783!
Oops: invalid opcode: 0000 [#1] SMP
CPU: 0 UID: 1000 PID: 615 Comm: repro Not tainted 7.1.0-rc2-ff238429d1ea #365 PREEMPT
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
RIP: 0010:ioeventfd_write+0x6c/0x70 [kvm]
Call Trace:
<TASK>
__kvm_io_bus_write+0x85/0xb0 [kvm]
kvm_io_bus_write+0x53/0x80 [kvm]
vcpu_mmio_write+0x66/0xf0 [kvm]
emulator_read_write_onepage+0x12a/0x540 [kvm]
emulator_read_write+0x109/0x2b0 [kvm]
x86_emulate_insn+0x4f8/0xfb0 [kvm]
x86_emulate_instruction+0x181/0x790 [kvm]
kvm_mmu_page_fault+0x313/0x630 [kvm]
vmx_handle_exit+0x18a/0x590 [kvm_intel]
kvm_arch_vcpu_ioctl_run+0xc81/0x1c90 [kvm]
kvm_vcpu_ioctl+0x2d5/0x970 [kvm]
__x64_sys_ioctl+0x8a/0xd0
do_syscall_64+0xb7/0x890
entry_SYSCALL_64_after_hwframe+0x4b/0x53
RIP: 0033:0x7f19c931a9bf
</TASK>
Modules linked in: kvm_intel kvm irqbypass
---[ end trace 0000000000000000 ]---
In a perfect world, the fix would be to simply delete the BUG_ON(), as KVM
x86 doesn't perform alignment checks on "normal" memory accesses at CPL0.
Sadly, C99 ruins all the fun; while the x86 architecture plays nice,
dereferencing an unaligned pointer directly is undefined behavior in C,
e.g. triggers splats when running with CONFIG_UBSAN_ALIGNMENT=y.
Fixes: d34e6b175e ("KVM: add ioeventfd support")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20260612225241.678509-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
snd_seq_event_dup() copies an incoming event into a pool cell and, in
the UMP-enabled build, clears the trailing cell->ump.raw.extra word that
the memcpy() did not cover. The guard deciding whether to clear it
compares the copied size against sizeof(cell->event):
memcpy(&cell->ump, event, size);
if (size < sizeof(cell->event))
cell->ump.raw.extra = 0;
For a legacy (non-UMP) event, size == sizeof(struct snd_seq_event) ==
sizeof(cell->event), so the condition is false and the extra word keeps
stale data. The cell pool is allocated with kvmalloc() (not zeroed) and
cells are reused via a free list, so that word holds uninitialised heap
or leftover event data.
When such a cell is delivered to a UMP client (client->midi_version > 0)
that set SNDRV_SEQ_FILTER_NO_CONVERT -- so the legacy event reaches it
unconverted -- snd_seq_read() reads it out as the larger struct
snd_seq_ump_event and copies the stale word to user space, a 4-byte
kernel heap infoleak to an unprivileged /dev/snd/seq client.
Compare against sizeof(cell->ump) instead, so the trailing word is zeroed
for every event shorter than the UMP cell.
Fixes: 46397622a3 ("ALSA: seq: Add UMP support")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: HyeongJun An <sammiee5311@gmail.com>
Link: https://patch.msgid.link/20260623233841.853326-1-sammiee5311@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
If the allocation of the bits array failed, kvm_s390_set_cmma_bits()
would return 0 instead of an error code.
Rework the function to use the __free() macros and thus simplify the
code flow; when the above mentioned allocation fails, simply return
-ENOMEM.
Fixes: e38c884df9 ("KVM: s390: Switch to new gmap")
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-10-imbrenda@linux.ibm.com>
The existing cmma selftest depended on the host allocating page tables
for all present memslots. Since the gmap rewrite, memory that is not
accessed by the guest might not have page tables allocated yet.
This caused the test to fail due to a mismatch in the assertion.
Fix by having the guest access also the second half of the test
memslot, thus guaranteeing that its page tables are present.
Fixes: e38c884df9 ("KVM: s390: Switch to new gmap")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-9-imbrenda@linux.ibm.com>
It is possible that some guest memory areas have not been touched yet
when starting migration mode, and thus have no ptes allocated. Only
existing and allocated ptes should count toward the total of dirty cmma
entries.
When starting migration mode, enable the migration_mode flag
immediately, so that any subsequent ESSA will trap in the host and
cause cmma_dirty_pages to be increased as needed.
Subsequently, set the cmma_d bit on all existing cmma-clean PGSTEs,
increasing cmma_dirty_pages as needed. Skipping cmma-dirty pages
prevents double counting.
Conversely, when disabling migration mode, set cmma_dirty_pages to 0
and clear the cmma_d bit in all existing PGSTEs.
The invariant is that when migration mode is off, no PGSTE has its
cmma_d bit set, and cmma_dirty_pages is 0. kvm->slots_lock protects
kvm_s390_vm_start_migration() and kvm_s390_vm_stop_migration() from
each other and from kvm_s390_get_cmma_bits().
Also fix dat_get_cmma() to properly wrap around if the first attempt
reached the end of guest memory without finding cmma-dirty pages.
[ imbrenda: Moved kvm_s390_sync_request_broadcast() before gmap_set_cmma_all_dirty() ]
Fixes: e38c884df9 ("KVM: s390: Switch to new gmap")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-8-imbrenda@linux.ibm.com>
Add the missing locking around dat_reset_cmma().
Fixes: e38c884df9 ("KVM: s390: Switch to new gmap")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-7-imbrenda@linux.ibm.com>
Under heavy memory pressure, handle_sske() and handle_pfmf() might
cause an endless loop if the mmu cache runs empty, the atomic
allocations fail, and the top-up function also fails. While quite
unlikely, that scenario is not impossible.
Fix the issue by not ignoring the return value of
kvm_s390_mmu_cache_topup(), and appropriately returning an error code
in case of failure.
Fixes: e38c884df9 ("KVM: s390: Switch to new gmap")
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-6-imbrenda@linux.ibm.com>
The correct length to pass to kvm_s390_get_guest_pages() is asce.tl + 1,
not asce.dt + 1. It was a typo, which, due to fortuitous circumstances,
did not cause bugs. It should nonetheless be fixed.
Fixes: e5f98a6899 ("KVM: s390: Add some helper functions needed for vSIE")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-5-imbrenda@linux.ibm.com>
Special pages / folios should not be set dirty. This also applies to
large pages.
Add a missing check in gmap_clear_young_crste() to prevent setting the
large page dirty if it is a special page.
Fixes: a2c17f9270 ("KVM: s390: New gmap code")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-4-imbrenda@linux.ibm.com>
If userspace passes a start address that is out of bounds,
_dat_walk_gfn_range() will fail with -EFAULT, but state.end will not be
touched and will stay 0. This will cause *count to underflow and report
a very high number, and the function will end up erroneously reporting
success.
Fix by only setting *count if the end address is not smaller than the
starting address. This way invalid starting addresses will correctly
return -EFAULT and *count will correctly indicate that no values have
been returned.
Fixes: 7b368470e1 ("KVM: s390: KVM page table management functions: CMMA")
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-3-imbrenda@linux.ibm.com>
The _PAGE_UNUSED softbit should not really be lying around. Its sole
purpose is to signal to try_to_unmap_one() and try_to_migrate_one()
that the page can be discarded instead of being moved / swapped.
KVM has no way to know why a page is being unmapped, so it sets the bit
on userspace ptes corresponding to unused guest pages every time they
get unmapped. KVM has no reasonable way to clear the bit once the page
is in use again.
While set_ptes() checks and clears the bit, other paths that set new
ptes did not. This led to used pages being thrown out as if they were
unused, causing guest corruption.
Fix the issue by clearing the _PAGE_UNUSED bit for present ptes in
set_pte(), i.e. whenever a present pte is getting set. The check in
set_ptes() is then redundant and can be removed.
Also fix gmap_helper_try_set_pte_unused() to only set the bit if the
pte is present; the _PAGE_UNUSED bit is only defined for present ptes
and thus should not be set for non-present ptes.
Fixes: c98175b791 ("KVM: s390: Add gmap_helper_set_unused()")
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260623153331.233784-2-imbrenda@linux.ibm.com>
There is a possibility that the user_operexec capability
only works if facility bit 74 is enabled. This is now fixed,
but add a selftest to demonstrate that.
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Acked-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260507200836.3500368-3-farman@linux.ibm.com>
The KVM_CAP_S390_USER_OPEREXEC capability allows operation exceptions
to be forwarded to userspace. But the actual enablement at the hardware
level occurs in kvm_arch_vcpu_postcreate(), and only if STFLE.74 or
user_instr0 are enabled. The latter is associated with a separate
capability (KVM_CAP_S390_USER_INSTR0), so the only way this happens
for the USER_OPEREXEC capability is if STFLE.74 is enabled. KVM
unconditionally enables this bit in kvm_arch_init_vm(), but the guest
could disable it from the CPU model and thus ignore this capability.
Add USER_OPEREXEC to the check in kvm_arch_vcpu_postcreate(), such that
either capability would enable this type of exception.
Fixes: 8e8678e740 ("KVM: s390: Add capability that forwards operation exceptions")
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
[Fixed patch title, as recommended by frankja@linux.ibm.com]
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260507200836.3500368-2-farman@linux.ibm.com>
