* kvm-arm64/vgic-lr-overflow: (50 commits)
: Support for VGIC LR overflows, courtesy of Marc Zyngier
:
: Address deficiencies in KVM's GIC emulation when a vCPU has more active
: IRQs than can be represented in the VGIC list registers. Sort the AP
: list to prioritize inactive and pending IRQs, potentially spilling
: active IRQs outside of the LRs.
:
: Handle deactivation of IRQs outside of the LRs for both EOImode=0/1,
: which involves special consideration for SPIs being deactivated from a
: different vCPU than the one that acked it.
KVM: arm64: Convert ICH_HCR_EL2_TDIR cap to EARLY_LOCAL_CPU_FEATURE
KVM: arm64: selftests: vgic_irq: Add timer deactivation test
KVM: arm64: selftests: vgic_irq: Add Group-0 enable test
KVM: arm64: selftests: vgic_irq: Add asymmetric SPI deaectivation test
KVM: arm64: selftests: vgic_irq: Perform EOImode==1 deactivation in ack order
KVM: arm64: selftests: vgic_irq: Remove LR-bound limitation
KVM: arm64: selftests: vgic_irq: Exclude timer-controlled interrupts
KVM: arm64: selftests: vgic_irq: Change configuration before enabling interrupt
KVM: arm64: selftests: vgic_irq: Fix GUEST_ASSERT_IAR_EMPTY() helper
KVM: arm64: selftests: gic_v3: Disable Group-0 interrupts by default
KVM: arm64: selftests: gic_v3: Add irq group setting helper
KVM: arm64: GICv2: Always trap GICV_DIR register
KVM: arm64: GICv2: Handle deactivation via GICV_DIR traps
KVM: arm64: GICv2: Handle LR overflow when EOImode==0
KVM: arm64: GICv3: Force exit to sync ICH_HCR_EL2.En
KVM: arm64: GICv3: nv: Plug L1 LR sync into deactivation primitive
KVM: arm64: GICv3: nv: Resync LRs/VMCR/HCR early for better MI emulation
KVM: arm64: GICv3: Avoid broadcast kick on CPUs lacking TDIR
KVM: arm64: GICv3: Handle in-LR deactivation when possible
KVM: arm64: GICv3: Add SPI tracking to handle asymmetric deactivation
...
Signed-off-by: Oliver Upton <oupton@kernel.org>
* kvm-arm64/sea-user:
: Userspace handling of SEAs, courtesy of Jiaqi Yan
:
: Add support for processing external aborts in userspace in situations
: where the host has failed to do so, allowing the VMM to potentially
: reinject an external abort into the VM.
Documentation: kvm: new UAPI for handling SEA
KVM: selftests: Test for KVM_EXIT_ARM_SEA
KVM: arm64: VM exit to userspace to handle SEA
Signed-off-by: Oliver Upton <oupton@kernel.org>
* kvm-arm64/misc:
: Miscellaneous fixes/cleanups for KVM/arm64
:
: - Fix for need_resched warnings on non-preemptible kernels when
: tearing down a VM's stage-2
:
: - Improvements to KVM struct allocation, getting rid of pointless
: __GFP_HIGHMEM and switching to kvzalloc()
:
: - SYNC ITS configuration before injecting LPIs in vgic_lpi_stress
: selftest
KVM: arm64: Reschedule as needed when destroying the stage-2 page-tables
KVM: arm64: Split kvm_pgtable_stage2_destroy()
KVM: arm64: Only drop references on empty tables in stage2_free_walker
KVM: selftests: SYNC after guest ITS setup in vgic_lpi_stress
KVM: selftests: Assert GICR_TYPER.Processor_Number matches selftest CPU number
KVM: arm64: Use kvzalloc() for kvm struct allocation
KVM: arm64: Drop useless __GFP_HIGHMEM from kvm struct allocation
Signed-off-by: Oliver Upton <oupton@kernel.org>
In addition to generic VFIO PCI functionality, the driver implements
VFIO migration uAPI, allowing userspace to enable migration for Intel
Graphics SR-IOV Virtual Functions.
The driver binds to VF device and uses API exposed by Xe driver to
transfer the VF migration data under the control of PF device.
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Alex Williamson <alex@shazbot.org>
Link: https://patch.msgid.link/20251127093934.1462188-5-michal.winiarski@intel.com
Link: https://lore.kernel.org/all/20251128125322.34edbeaf.alex@shazbot.org/
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 2e38c50ae4)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
A guest can write 1 to TCR_ELx.HA, making the KVM software walker update
the access flag in a table descriptor even if FEAT_HAFDBS is not present.
Avoid this by making wi->ha depend on FEAT_HAFDBS being enabled in the VM,
similar to how the software walker treats FEAT_HPDS.
This is not needed for VTCR_EL2.HA, since a guest will always write to
the in-memory copy of the register, where the HA bit is masked (set to
0) by KVM if the VM doesn't have FEAT_HAFDBS.
Fixes: c59ca4b5b0c3 ("KVM: arm64: Implement HW access flag management in stage-1 SW PTW")
Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com>
Link: https://msgid.link/20251128100946.74210-5-alexandru.elisei@arm.com
Signed-off-by: Oliver Upton <oupton@kernel.org>
According to ARM DDI 0487L.b, the HA bit in TCR_EL2 when the translation
regime is EL2 (or !ELIsInHost(EL2)) is bit 21, not 39.
Fixes: c59ca4b5b0c3 ("KVM: arm64: Implement HW access flag management in stage-1 SW PTW")
Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com>
Link: https://msgid.link/20251128100946.74210-3-alexandru.elisei@arm.com
Signed-off-by: Oliver Upton <oupton@kernel.org>
Commit 2608563b46 ("KVM: arm64: Add support for FEAT_XNX stage-2
permissions") added the KVM_PGTABLE_PROX_{UX,PX} permissions to stage 2 and
to EL2 translation regimes, but left them undocumented. Let's fix that.
Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com>
Link: https://msgid.link/20251128100946.74210-2-alexandru.elisei@arm.com
Signed-off-by: Oliver Upton <oupton@kernel.org>
Clang warns (or errors with CONFIG_WERROR=y / W=e):
arch/arm64/kvm/hyp/pgtable.c:757:2: error: label at end of compound statement is a C23 extension [-Werror,-Wc23-extensions]
757 | }
| ^
With older versions of clang (15 and older) and GCC (at least the minimum
supported, 8.1), this is an unconditional hard error:
arch/arm64/kvm/hyp/pgtable.c: In function 'kvm_pgtable_stage2_pte_prot':
arch/arm64/kvm/hyp/pgtable.c:756:2: error: label at end of compound statement
default:
^~~~~~~
arch/arm64/kvm/hyp/pgtable.c:756:10: error: label at end of compound statement: expected statement
default:
^
;
Add a break statement to this default case to clear up the error/warning.
Fixes: 2608563b46 ("KVM: arm64: Add support for FEAT_XNX stage-2 permissions")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251125-arm64-kvm-hyp-pgtable-fix-c23-ext-warn-v1-1-98b506ddefbf@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Add a basic test for AT emulation in the EL2&0 and EL1&0 translation
regimes.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-16-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Everything is in place to update the access flag at S1 and S2. Expose
support for the access flag flavor of FEAT_HAFDBS to NV guests.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-15-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Give the stage-2 walk similar treatment to stage-1: update the access
flag during the table walk and do so for any walk context.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-14-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Atomically update the Access flag at stage-1 when the guest has
configured the MMU to do so. Make the implementation choice (and liberal
interpretation of speculation) that any access type updates the Access
flag, including AT and CMO instructions.
Restart the entire walk by returning to the exception-generating
instruction in the case of a failed Access flag update.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-13-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
KVM's software PTW will soon support 'hardware' updates to the access
flag. Similar to fault handling, races to update the descriptor will be
handled by restarting the instruction. Prepare for this by propagating
errors up to the AT emulation, only retiring the instruction if the walk
succeeds.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-12-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Implementing FEAT_HAFDBS in KVM's software PTWs requires the ability to
CAS a descriptor to update the in-memory value. Add an accessor to do
exactly that, coping with the fact that guest descriptors are in user
memory (duh).
While FEAT_LSE required on any system that implements NV, KVM now uses
the stage-1 PTW for non-nested use cases meaning an LL/SC implementation
is necessary as well.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-11-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Use the existing page table definitions instead of magic numbers for the
stage-2 table walk.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-10-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Implementing FEAT_HAFDBS means adding another descriptor accessor that
needs to deal with the guest-configured endianness. Prepare by moving
the endianness handling into the read accessor and out of the main body
of the S1/S2 PTWs.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-9-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
The stage-2 table walker passes down the vCPU as a void pointer. That
might've made sense if the walker was generic although at this point it
is clear this will only ever be used in the context of a vCPU.
Suggested-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-8-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Going through a function pointer doesn't serve much purpose when there's
only one implementation.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-7-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Everything is in place to support FEAT_XNX, advertise support.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-6-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Although KVM doesn't make direct use of the feature, guest hypervisors
can use FEAT_XNX which influences the permissions of the shadow stage-2.
Update ptdump to separately print the privileged and unprivileged
execute permissions.
Reviewed-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://msgid.link/20251124190158.177318-5-oupton@kernel.org
Signed-off-by: Oliver Upton <oupton@kernel.org>
Device specific VFIO driver variant for Xe will implement VF migration.
Export everything that's needed for migration ops.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-4-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 17f22465c5)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
In certain scenarios (such as VF migration), VF driver needs to interact
with PF driver.
Add a helper to allow VF driver access to PF xe_device.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-3-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 8b3cce3ad9)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
All of the necessary building blocks are now in place to support SR-IOV
VF migration.
Flip the enable/disable logic to match VF code and disable the feature
only for platforms that don't meet the necessary prerequisites.
To allow more testing and experiments, on DEBUG builds any missing
prerequisites will be ignored.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-2-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 01c724aa7b)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Add a scope-based helpers for runtime PM that may be used to simplify
cleanup logic and potentially avoid goto-based cleanup.
For example, using
guard(xe_pm_runtime)(xe);
will get runtime PM and cause a corresponding put to occur automatically
when the current scope is exited. 'xe_pm_runtime_noresume' can be used
as a guard replacement for the corresponding 'noresume' variant.
There's also an xe_pm_runtime_ioctl conditional guard that can be used
as a replacement for xe_runtime_ioctl():
ACQUIRE(xe_pm_runtime_ioctl, pm)(xe);
if ((ret = ACQUIRE_ERR(xe_pm_runtime_ioctl, &pm)) < 0)
/* failed */
In a few rare cases (such as gt_reset_worker()) we need to ensure that
runtime PM is dropped when the function is exited by any means
(including error paths), but the function does not need to acquire
runtime PM because that has already been done earlier by a different
function. For these special cases, an 'xe_pm_runtime_release_only'
guard can be used to handle the release without doing an acquisition.
These guards will be used in future patches to eliminate some of our
goto-based cleanup.
v2:
- Specify success condition for xe_pm runtime_ioctl as _RET >= 0 so
that positive values will be properly identified as success and
trigger destructor cleanup properly.
v3:
- Add comments to the kerneldoc for the existing 'get' functions
indicating that scope-based handling should be preferred where
possible. (Gustavo)
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patch.msgid.link/20251118164338.3572146-31-matthew.d.roper@intel.com
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
(cherry picked from commit 59e7528dbf)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
There wasn't much of reviewing activity from Kishon of PHY subsystem
during last few years [1] and last maintainer commit is from 2020, so
move Kishon to Credits to indicate that PHY subsystem has only one
active maintainer. Thank you Kishon for working on the PHY subsystem
and for all the maintenance effort.
Link: https://lore.kernel.org/all/?q=f%3A%22Kishon+Vijay+Abraham+I%22 [1]
Cc: Kishon Vijay Abraham I <kishon@kernel.org>
Cc: Kishon Vijay Abraham I <kvijayab@amd.com>
Cc: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://patch.msgid.link/20251120175537.171340-2-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Call drm_gem_fb_begin_cpu_access() and drm_gem_fb_end_cpu_access()
around cursor image updates. Imported buffers might have to be
synchronized for CPU access before they can be used.
Ignore errors from drm_gem_fb_begin_cpu_access(). These errors can
often be transitory. The cursor image will be updated on the next
frame. Meanwhile display a white square where the cursor would be.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>>
Link: https://patch.msgid.link/20251126094626.41985-4-tzimmermann@suse.de
Copy the ARGB4444 cursor buffer to system memory if it is located in
I/O memory. While this cannot happen with ast's native GEM objects, an
imported buffer object might be on the external device's I/O memory.
If the cursor buffer is located in system memory continue to use it
directly.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>>
Link: https://patch.msgid.link/20251126094626.41985-3-tzimmermann@suse.de
Move the format conversion of the cursor framebuffer into the new
helper ast_cursor_plane_get_argb4444(). It returns a buffer in system
memory, which the atomic_update handler copies to video memory.
The returned buffer is either the GEM buffer itself, or a temporary
copy within the plane in ARGB4444 format.
As a small change, list supported formats explicitly in the switch
statement. Do not assume ARGB8888 input by default. The cursor
framebuffer knows its format, so should we.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://patch.msgid.link/20251126094626.41985-2-tzimmermann@suse.de
Switch to the permissive MIT license to make the EROFS on-disk format
more interoperable across various use cases.
It was previously recommended by the Composefs folks, for example:
https://github.com/composefs/composefs/pull/216#discussion_r1356409501
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Acked-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Acked-by: Yue Hu <zbestahu@gmail.com>
Acked-by: Chao Yu <chao@kernel.org>
Acked-by: Jianan Huang <jnhuang95@gmail.com>
Acked-by: Chunhai Guo <guochunhai@vivo.com>
Acked-by: Hongbo Li <lihongbo22@huawei.com>
Modify kernel-doc comments in local_lock.h to prevent warnings:
Warning: include/linux/local_lock.h:9 function parameter 'lock' not described in 'local_lock_init'
Warning: include/linux/local_lock.h:56 function parameter 'lock' not described in 'local_trylock_init'
Warning: include/linux/local_lock.h:56 expecting prototype for local_lock_init(). Prototype was for local_trylock_init() instead
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://patch.msgid.link/20251128065925.917917-1-rdunlap@infradead.org
The Linux kernel coding style advises to avoid common variable
names in function-like macros to reduce the risk of namespace
collisions.
Throughout local_lock_internal.h, several macros use the rather common
variable names 'l' and 'tl'. This already resulted in an actual
collision: the __local_lock_acquire() function like macro is currently
shadowing the parameter 'l' of the:
class_##_name##_t class_##_name##_constructor(_type *l)
function factory from <linux/cleanup.h>.
Rename the variable 'l' to '__l' and the variable 'tl' to '__tl'
throughout the file to fix the current namespace collision and
to prevent future ones.
[ bigeasy: Rebase, update all l and tl instances in macros ]
Signed-off-by: Vincent Mailhol <mailhol@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Waiman Long <longman@redhat.com>
Link: https://patch.msgid.link/20251127144140.215722-3-bigeasy@linutronix.de
The local_lock_t was never added to the MAINTAINERS file since its
inclusion.
Add local_lock_t to the locking primitives section.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Waiman Long <longman@redhat.com>
Link: https://patch.msgid.link/20251127144140.215722-2-bigeasy@linutronix.de
mutex_init() invokes __mutex_init() providing the name of the lock and
a pointer to a the lock class. With LOCKDEP enabled this information is
useful but without LOCKDEP it not used at all. Passing the pointer
information of the lock class might be considered negligible but the
name of the lock is passed as well and the string is stored. This
information is wasting storage.
Split __mutex_init() into a _genereic() variant doing the initialisation
of the lock and a _lockdep() version which does _genereic() plus the
lockdep bits. Restrict the lockdep version to lockdep enabled builds
allowing the compiler to remove the unused parameter.
This results in the following size reduction:
text data bss dec filename
| 30237599 8161430 1176624 39575653 vmlinux.defconfig
| 30233269 8149142 1176560 39558971 vmlinux.defconfig.patched
-4.2KiB -12KiB
| 32455099 8471098 12934684 53860881 vmlinux.defconfig.lockdep
| 32455100 8471098 12934684 53860882 vmlinux.defconfig.patched.lockdep
| 27152407 7191822 2068040 36412269 vmlinux.defconfig.preempt_rt
| 27145937 7183630 2067976 36397543 vmlinux.defconfig.patched.preempt_rt
-6.3KiB -8KiB
| 29382020 7505742 13784608 50672370 vmlinux.defconfig.preempt_rt.lockdep
| 29376229 7505742 13784544 50666515 vmlinux.defconfig.patched.preempt_rt.lockdep
-5.6KiB
[peterz: folded fix from boqun]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Waiman Long <longman@redhat.com>
Link: https://lkml.kernel.org/r/20251125145425.68319-1-boqun.feng@gmail.com
Link: https://patch.msgid.link/20251105142350.Tfeevs2N@linutronix.de
Use netfs_alloc/free_folioq_buffer() rather than doing its own version.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Shyam Prasad N <sprasad@microsoft.com>
cc: Tom Talpey <tom@talpey.com> (RDMA, smbdirect)
cc: linux-cifs@vger.kernel.org
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Show cached directory smb lease key in /proc/fs/cifs/open_dirs
for debugging purposes.
Signed-off-by: Bharath SM <bharathsm@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Add the SMB lease key in /proc/fs/cifs/open_files for
debugging purposes.
Signed-off-by: Bharath SM <bharathsm@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
ipc_msg_send_request() waits for a generic netlink reply using an
ipc_msg_table_entry on the stack. The generic netlink handler
(handle_generic_event()/handle_response()) fills entry->response under
ipc_msg_table_lock, but ipc_msg_send_request() used to validate and free
entry->response without holding the same lock.
Under high concurrency this allows a race where handle_response() is
copying data into entry->response while ipc_msg_send_request() has just
freed it, leading to a slab-use-after-free reported by KASAN in
handle_generic_event():
BUG: KASAN: slab-use-after-free in handle_generic_event+0x3c4/0x5f0 [ksmbd]
Write of size 12 at addr ffff888198ee6e20 by task pool/109349
...
Freed by task:
kvfree
ipc_msg_send_request [ksmbd]
ksmbd_rpc_open -> ksmbd_session_rpc_open [ksmbd]
Fix by:
- Taking ipc_msg_table_lock in ipc_msg_send_request() while validating
entry->response, freeing it when invalid, and removing the entry from
ipc_msg_table.
- Returning the final entry->response pointer to the caller only after
the hash entry is removed under the lock.
- Returning NULL in the error path, preserving the original API
semantics.
This makes all accesses to entry->response consistent with
handle_response(), which already updates and fills the response buffer
under ipc_msg_table_lock, and closes the race that allowed the UAF.
Cc: stable@vger.kernel.org
Reported-by: Qianchang Zhao <pioooooooooip@gmail.com>
Reported-by: Zhitong Liu <liuzhitong1993@gmail.com>
Signed-off-by: Qianchang Zhao <pioooooooooip@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
sc->first_error might already be set and sc->status
is thus unexpected, so this should avoid the WARN[_ON]_ONCE()
if sc->first_error is already set and have a usable error path.
While there set sc->first_error as soon as possible.
This is done based on a problem seen in similar places on
the server. And there it was already very useful in order
to find the problem when we have a meaningful WARN_ONCE()
that prints details about the connection.
This is much more useful:
[ 309.560973] expected[NEGOTIATE_NEEDED] != RDMA_CONNECT_RUNNING
first_error=0 local=192.168.0.200:445 remote=192.168.0.100:60445
[ 309.561034] WARNING: CPU: 2 PID: 78 at transport_rdma.c:643
recv_done+0x2fa/0x3d0 [ksmbd]
than what we had before (only):
[ 894.140316] WARNING: CPU: 1 PID: 116 at
fs/smb/server/transport_rdma.c:642 recv_done+0x308/0x360 [ksmbd]
Fixes: 58dfba8a2d ("smb: client/smbdirect: replace SMBDIRECT_SOCKET_CONNECTING with more detailed states")
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Paulo Alcantara <pc@manguebit.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
These will be used in various places in order to assert
the current status mostly during the connect and negotiation
phase. It will replace the WARN_ON_ONCE(sc->status != ...)
calls, which are very useless in order to identify the
problem that happened.
As a start client and server will need to define their own
__SMBDIRECT_SOCKET_DISCONNECT(__sc) macro in order to use
SMBDIRECT_CHECK_STATUS_DISCONNECT().
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Paulo Alcantara <pc@manguebit.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
This can be used like this:
int err = somefunc();
pr_warn("err=%1pe\n", SMBDIRECT_DEBUG_ERR_PTR(err));
This will be used in the following fixes in order
to be prepared to identify real world problems
more easily.
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Paulo Alcantara <pc@manguebit.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
ksmbd maintains delete-on-close and pending-delete state in
ksmbd_inode->m_flags. In vfs_cache.c this field is accessed under
inconsistent locking: some paths read and modify m_flags under
ci->m_lock while others do so without taking the lock at all.
Examples:
- ksmbd_query_inode_status() and __ksmbd_inode_close() use
ci->m_lock when checking or updating m_flags.
- ksmbd_inode_pending_delete(), ksmbd_set_inode_pending_delete(),
ksmbd_clear_inode_pending_delete() and ksmbd_fd_set_delete_on_close()
used to read and modify m_flags without ci->m_lock.
This creates a potential data race on m_flags when multiple threads
open, close and delete the same file concurrently. In the worst case
delete-on-close and pending-delete bits can be lost or observed in an
inconsistent state, leading to confusing delete semantics (files that
stay on disk after delete-on-close, or files that disappear while still
in use).
Fix it by:
- Making ksmbd_query_inode_status() look at m_flags under ci->m_lock
after dropping inode_hash_lock.
- Adding ci->m_lock protection to all helpers that read or modify
m_flags (ksmbd_inode_pending_delete(), ksmbd_set_inode_pending_delete(),
ksmbd_clear_inode_pending_delete(), ksmbd_fd_set_delete_on_close()).
- Keeping the existing ci->m_lock protection in __ksmbd_inode_close(),
and moving the actual unlink/xattr removal outside the lock.
This unifies the locking around m_flags and removes the data race while
preserving the existing delete-on-close behaviour.
Reported-by: Qianchang Zhao <pioooooooooip@gmail.com>
Reported-by: Zhitong Liu <liuzhitong1993@gmail.com>
Signed-off-by: Qianchang Zhao <pioooooooooip@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
strcpy() is deprecated [1] and using strcat() is discouraged. Replace
them by assigning the prefix directly and by using memcpy() to copy the
pathname. Using memcpy() is safe because we already know the length of
the source string and that it is guaranteed to be NUL-terminated.
Allocate only as many bytes as needed and replace kzalloc() with
kmalloc() since memcpy() overwrites the entire buffer anyway.
Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy [1]
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Add STATUS_INFO_LENGTH_MISMATCH mapping to EMSGSIZE.
Currently, STATUS_INFO_LENGTH_MISMATCH has no mapping to any error code,
making it difficult to distinguish between invalid parameters and length
mismatch.
Map STATUS_INFO_LENGTH_MISMATCH to EMSGSIZE while keeping the EINVAL for
invalid parameters. Although the buf_len check only checks for buf_size
being less than required, there was no error code for lower buf_size.
Hence, EMSGSIZE is used.
Signed-off-by: Aaditya Kansal <aadityakansal390@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Under high concurrency, A tree-connection object (tcon) is freed on
a disconnect path while another path still holds a reference and later
executes *_put()/write on it.
Reported-by: Qianchang Zhao <pioooooooooip@gmail.com>
Reported-by: Zhitong Liu <liuzhitong1993@gmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
The ksmbd listener thread was using busy waiting on a listening socket by
calling kernel_accept() with SOCK_NONBLOCK and retrying every 100ms on
-EAGAIN. Since this thread is dedicated to accepting new connections,
there is no need for non-blocking mode.
Switch to a blocking accept() call instead, allowing the thread to sleep
until a new connection arrives. This avoids unnecessary wakeups and CPU
usage. During teardown, call shutdown() on the listening socket so that
accept() returns -EINVAL and the thread exits cleanly.
The socket release mutex is redundant because kthread_stop() blocks until
the listener thread returns, guaranteeing safe teardown ordering.
Also remove sk_rcvtimeo and sk_sndtimeo assignments, which only caused
accept() to return -EAGAIN prematurely.
Signed-off-by: Qingfang Deng <dqfext@gmail.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
