Back in 2017, commit 2fd260f03b ("PCI/AER: Remove unused .link_reset()
callback") removed .link_reset() from struct pci_error_handlers, but left
a few code comments behind which still mention it. Remove them.
The code comments in the SolarFlare Ethernet drivers point out that no
.mmio_enabled() callback is needed because the driver's .error_detected()
callback always returns PCI_ERS_RESULT_NEED_RESET, which causes
pcie_do_recovery() to skip .mmio_enabled(). That's not quite correct
because efx_io_error_detected() does return PCI_ERS_RESULT_RECOVERED under
certain conditions and then .mmio_enabled() would indeed be called if it
were implemented. Remove this misleading portion of the code comment as
well.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/1d72a891a7f57115e78a73046e776f7e0c8cd68f.1755008151.git.lukas@wunner.de
After a Fatal Error has been reported by a device and has been recovered
through a Secondary Bus Reset, AER updates the device's error_state to
pci_channel_io_normal before invoking its driver's ->resume() callback.
By contrast, EEH updates the error_state earlier, namely after resetting
the device and before invoking its driver's ->slot_reset() callback.
Commit c58dc575f3 ("powerpc/pseries: Set error_state to
pci_channel_io_normal in eeh_report_reset()") explains in great detail
that the earlier invocation is necessitated by various drivers checking
accessibility of the device with pci_channel_offline() and avoiding
accesses if it returns true. It returns true for any other error_state
than pci_channel_io_normal.
The device should be accessible already after reset, hence the reasoning
is that it's safe to update the error_state immediately afterwards.
This deviation between AER and EEH seems problematic because drivers
behave differently depending on which error recovery mechanism the
platform uses. Three drivers have gone so far as to update the
error_state themselves, presumably to work around AER's behavior.
For consistency, amend AER to update the error_state at the same recovery
steps as EEH. Drop the now unnecessary workaround from the three drivers.
Keep updating the error_state before ->resume() in case ->error_detected()
or ->mmio_enabled() return PCI_ERS_RESULT_RECOVERED, which causes
->slot_reset() to be skipped. There are drivers doing this even for Fatal
Errors, e.g. mhi_pci_error_detected().
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/4517af6359ffb9d66152b827a5d2833459144e3f.1755008151.git.lukas@wunner.de
According to Documentation/PCI/pci-error-recovery.rst, the following shall
occur on failure to recover from a PCIe Uncorrectable Error:
STEP 6: Permanent Failure
-------------------------
A "permanent failure" has occurred, and the platform cannot recover
the device. The platform will call error_detected() with a
pci_channel_state_t value of pci_channel_io_perm_failure.
The device driver should, at this point, assume the worst. It should
cancel all pending I/O, refuse all new I/O, returning -EIO to
higher layers. The device driver should then clean up all of its
memory and remove itself from kernel operations, much as it would
during system shutdown.
Sathya notes that AER does not call error_detected() on failure and thus
deviates from the document (as well as EEH, for which the document was
originally added).
Most drivers do nothing on permanent failure, but the SCSI drivers and a
number of Ethernet drivers do take advantage of the notification to flush
queues and give up resources.
Amend AER to notify such drivers and align with the documentation and EEH.
Link: https://lore.kernel.org/r/f496fc0f-64d7-46a4-8562-dba74e31a956@linux.intel.com/
Suggested-by: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@linux.intel.com>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/ec212d4d4f5c65d29349df33acdc9768ff8279d1.1755008151.git.lukas@wunner.de
Upon failure to recover from a PCIe error through AER, DPC or EDR, a
uevent is sent to inform user space about disconnection of the bridge
whose subordinate devices failed to recover.
However the bridge itself is not disconnected. Instead, a uevent should
be sent for each of the subordinate devices.
Only if the "bridge" happens to be a Root Complex Event Collector or
Integrated Endpoint does it make sense to send a uevent for it (because
there are no subordinate devices).
Right now if there is a mix of subordinate devices with and without
pci_error_handlers, a BEGIN_RECOVERY event is sent for those with
pci_error_handlers but no FAILED_RECOVERY event is ever sent for them
afterwards. Fix it.
Fixes: 856e1eb9bd ("PCI/AER: Add uevents in AER and EEH error/resume")
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org # v4.16+
Link: https://patch.msgid.link/68fc527a380821b5d861dd554d2ce42cb739591c.1755008151.git.lukas@wunner.de
When Advanced Error Reporting was introduced in September 2006 by commit
6c2b374d74 ("PCI-Express AER implemetation: AER core and aerdriver"), it
sought to adhere to the recovery flow and callbacks specified in
Documentation/PCI/pci-error-recovery.rst.
That document had been added in January 2006, when Enhanced Error Handling
(EEH) was introduced for PowerPC with commit 065c635907 ("[PATCH] PCI
Error Recovery: documentation").
However the AER driver deviates from the document in that it never
performs a Secondary Bus Reset on Non-Fatal Errors, but always on Fatal
Errors. By contrast, EEH allows drivers to opt in or out of a Bus Reset
regardless of error severity, by returning PCI_ERS_RESULT_NEED_RESET or
PCI_ERS_RESULT_CAN_RECOVER from their ->error_detected() callback. If all
drivers agree that they can recover without a Bus Reset, EEH skips it.
Should one of them request a Bus Reset, it overrides all other drivers.
This inconsistency between EEH and AER seems problematic because drivers
need to be aware of and cope with it.
The file Documentation/PCI/pcieaer-howto.rst hints at a rationale for
always performing a Bus Reset on Fatal Errors: "Fatal errors [...] cause
the link to be unreliable. [...] This [reset_link] callback is used to
reset the PCIe physical link when a fatal error happens. If an error
message indicates a fatal error, [...] performing link reset at upstream
is necessary."
There's no such rationale provided for never performing a Bus Reset on
Non-Fatal Errors.
The "xe" driver has a need to attempt a reset of local units on graphics
cards upon a Non-Fatal Error. If that is insufficient for recovery, the
driver wants to opt in to a Bus Reset.
Accommodate such use cases and align AER more closely with EEH by
performing a Bus Reset in pcie_do_recovery() if drivers request it and the
faulting device's channel_state is pci_channel_io_normal. The AER driver
sets this channel_state for Non-Fatal Errors. For Fatal Errors, it uses
pci_channel_io_frozen.
This limits the deviation from Documentation/PCI/pci-error-recovery.rst
and EEH to the unconditional Bus Reset on Fatal Errors.
pcie_do_recovery() is also invoked by the Downstream Port Containment and
Error Disconnect Recover drivers. They both set the channel_state to
pci_channel_io_frozen, hence pcie_do_recovery() continues to always invoke
the ->reset_subordinates() callback in their case. That is necessary
because the callback brings the link back up at the containing Downstream
Port.
There are two behavioral changes resulting from this commit:
First, if channel_state is pci_channel_io_normal and one of the affected
drivers returns PCI_ERS_RESULT_NEED_RESET from its ->error_detected()
callback, a Bus Reset will now be performed. There are drivers doing this
and although it would be possible to avoid a behavioral change by letting
them return PCI_ERS_RESULT_CAN_RECOVER instead, the impression I got from
examination of all drivers is that they actually expect or want a Bus
Reset (cxl_error_detected() is a case in point). In any case, if they can
cope with a Bus Reset on Fatal Errors, they shouldn't have issues with a
Bus Reset on Non-Fatal Errors.
Second, if channel_state is pci_channel_io_frozen and all affected drivers
return PCI_ERS_RESULT_CAN_RECOVER from ->error_detected(), their
->mmio_enabled() callback is now invoked prior to performing a Bus Reset,
instead of afterwards. This actually makes sense: For example,
drivers/scsi/sym53c8xx_2/sym_glue.c dumps debug registers in its
->mmio_enabled() callback. Doing so after reset right now captures the
post-reset state instead of the faulting state, which is useless.
There is only one other driver which implements ->mmio_enabled() and
returns PCI_ERS_RESULT_CAN_RECOVER from ->error_detected() for
channel_state pci_channel_io_frozen, drivers/scsi/ipr.c (IBM Power RAID).
It appears to only be used on EEH platforms. So the second behavioral
change is limited to these two drivers.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/28fd805043bb57af390168d05abb30898cf4fc58.1755008151.git.lukas@wunner.de
Since commit 6b730a4050 “ext4: hoist ext4_block_write_begin and
replace the __block_write_begin”, the comment should be updated
accordingly from "__block_write_begin" to "ext4_block_write_begin".
Fixes: 6b730a4050 (“ext4: hoist ext4_block_write_begin and replace...")
Signed-off-by: Baolin Liu <liubaolin@kylinos.cn>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Link: https://patch.msgid.link/20250812021709.1120716-1-liubaolin12138@163.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The CS_ONLINE flag was introduced prior to the CSS_ONLINE flag in the
cpuset subsystem. Currently, the flag setting sequence is as follows:
1. cpuset_css_online() sets CS_ONLINE
2. css->flags gets CSS_ONLINE set
...
3. cgroup->kill_css sets CSS_DYING
4. cpuset_css_offline() clears CS_ONLINE
5. css->flags clears CSS_ONLINE
The is_cpuset_online() check currently occurs between steps 1 and 3.
However, it would be equally safe to perform this check between steps 2
and 3, as CSS_ONLINE provides the same synchronization guarantee as
CS_ONLINE.
Since CS_ONLINE is redundant with CSS_ONLINE and provides no additional
synchronization benefits, we can safely remove it to simplify the code.
Signed-off-by: Chen Ridong <chenridong@huawei.com>
Acked-by: Waiman Long <longman@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Convert the Silicon Labs SI570 binding to DT schema format. It's a
straight-forward conversion.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250804222010.4082782-1-robh@kernel.org
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Convert the Silicon Labs SI5341 binding to DT schema format. It's a
straight-forward conversion.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250804222034.4083410-1-robh@kernel.org
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Convert the Silicon Labs SI514 and SI544 bindings to DT schema format.
Combine the bindings into a single schema as they are the same.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250804222042.4083656-1-robh@kernel.org
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
SM7225 is a variation of SM6350, and also needs an entry in the table.
Fixes: 1924272b9c ("soc: qcom: Add UBWC config provider")
Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Bjorn Andersson <andersson@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
After the commit 45a2974157 ("drm/msm: Use the central UBWC config
database") the MDSS driver errors out if UBWC database didn't provide it
with the UBWC configuration. Make UBWC database return zero data for
MSM8916 / APQ8016, MSM8974 / APQ8074, MSM8226 and MSM8939.
Fixes: 1924272b9c ("soc: qcom: Add UBWC config provider")
Acked-by: Bjorn Andersson <andersson@kernel.org>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/665313/
Link: https://lore.kernel.org/r/20250723-ubwc-no-ubwc-v3-1-81bdb75685bf@oss.qualcomm.com
After a recent change in clang to expose uninitialized warnings from
const variables and pointers [1], there is a warning around crtc_state
in dpu_plane_virtual_atomic_check():
drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c:1145:6: error: variable 'crtc_state' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
1145 | if (plane_state->crtc)
| ^~~~~~~~~~~~~~~~~
drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c:1149:58: note: uninitialized use occurs here
1149 | ret = dpu_plane_atomic_check_nosspp(plane, plane_state, crtc_state);
| ^~~~~~~~~~
drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c:1145:2: note: remove the 'if' if its condition is always true
1145 | if (plane_state->crtc)
| ^~~~~~~~~~~~~~~~~~~~~~
1146 | crtc_state = drm_atomic_get_new_crtc_state(state,
drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c:1139:35: note: initialize the variable 'crtc_state' to silence this warning
1139 | struct drm_crtc_state *crtc_state;
| ^
| = NULL
Initialize crtc_state to NULL like other places in the driver do, so
that it is consistently initialized.
Cc: stable@vger.kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/2106
Fixes: 774bcfb731 ("drm/msm/dpu: add support for virtual planes")
Link: 2464313eef [1]
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Currently, the high bitfield of certain DSI registers
do not align with the configuration of the SWI registers
description. This can lead to wrong programming these DSI
registers, for example for 4k resloution where H_TOTAL is
taking 13 bits but software is programming only 12 bits
because of the incorrect bitmask for H_TOTAL bitfeild,
this is causing DSI FIFO errors. To resolve this issue,
increase the high bitfield of the DSI registers from 12 bits
to 16 bits in dsi.xml to match the SWI register configuration.
Signed-off-by: Ayushi Makhija <quic_amakhija@quicinc.com>
Fixes: 4f52f5e63b ("drm/msm: import XML display registers database")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/666229/
Link: https://lore.kernel.org/r/20250730123938.1038640-1-quic_amakhija@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Fix a regression introduced by commit b41642c877 ("rcu: Fix
rcu_read_unlock() deadloop due to IRQ work") which results in boot hang
as reported by kernel test bot at [1]. This issue happens because RCU
re-initializes the deferred QS IRQ work everytime it is queued. With
commit b41642c877, the IRQ work re-initialization can happen while it
is already queued. This results in IRQ work being requeued to itself.
When IRQ work finally fires, as it is requeued to itself, it is
repeatedly executed and results in hang. Fix this with initializing the
IRQ work only once before the CPU boots.
[1] https://lore.kernel.org/rcu/202508071303.c1134cce-lkp@intel.com/
-----BEGIN PGP SIGNATURE-----
iHQEABYIAB0WIQSi2tPIQIc2VEtjarIAHS7/6Z0wpQUCaJlhIQAKCRAAHS7/6Z0w
pbjqAQCfjAYzQiFDEFoiOKvbMsaikg1fmj+Td1g+UsC3utAiBAD1HcQJsCAV21Ft
0xw6B+xBLbUC+dyIhfIffM+NQmrcCg==
=Y0kO
-----END PGP SIGNATURE-----
Merge tag 'rcu.fixes.6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux
Pull RCU fix from Neeraj Upadhyay:
"Fix a regression introduced by commit b41642c877 ("rcu: Fix
rcu_read_unlock() deadloop due to IRQ work") which results in boot
hang as reported by kernel test bot at [1].
This issue happens because RCU re-initializes the deferred QS IRQ work
everytime it is queued. With commit b41642c877, the IRQ work
re-initialization can happen while it is already queued. This results
in IRQ work being requeued to itself. When IRQ work finally fires, as
it is requeued to itself, it is repeatedly executed and results in
hang.
Fix this with initializing the IRQ work only once before the CPU
boots"
Link: https://lore.kernel.org/rcu/202508071303.c1134cce-lkp@intel.com/ [1]
* tag 'rcu.fixes.6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux:
rcu: Fix racy re-initialization of irq_work causing hangs
To configure and enable the DSI PHY PLL clocks, the MDSS AHB clock must
be active for MMIO operations. Typically, this AHB clock is enabled as
part of the DSI PHY interface enabling (dsi_phy_enable_resource).
However, since these PLL clocks are registered as clock entities, they
can be enabled independently of the DSI PHY interface, leading to
enabling failures and subsequent warnings:
```
msm_dsi_phy 5e94400.phy: [drm:dsi_pll_14nm_vco_prepare] *ERROR* DSI PLL lock failed
------------[ cut here ]------------
dsi0pllbyte already disabled
WARNING: CPU: 3 PID: 1 at drivers/clk/clk.c:1194 clk_core_disable+0xa4/0xac
CPU: 3 UID: 0 PID: 1 Comm: swapper/0 Tainted:
Tainted: [W]=WARN
Hardware name: Qualcomm Technologies, Inc. Robotics RB1 (DT)
pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[...]
```
This issue is particularly prevalent at boot time during the disabling of
unused clocks (clk_disable_unused()) which includes enabling the parent
clock(s) when CLK_OPS_PARENT_ENABLE flag is set (this is the case for the
14nm DSI PHY PLL consumers).
To resolve this issue, we move the AHB clock as a PM dependency of the DSI
PHY device (via pm_clk). Since the DSI PHY device is the parent of the PLL
clocks, this resolves the PLL/AHB dependency. Now the AHB clock is enabled
prior the PLL clk_prepare callback, as part of the runtime-resume chain.
We also eliminate dsi_phy_[enable|disable]_resource functions, which are
superseded by runtime PM.
Note that it breaks compatibility with kernels before 6.0, as we do not
support anymore the legacy `iface_clk` name.
Signed-off-by: Loic Poulain <loic.poulain@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/663239/
Link: https://lore.kernel.org/r/20250709140836.124143-1-loic.poulain@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Remove redundant "_kunit" suffix from test module name.
The naming is now consistent with other Rust components as the test
context is already implied by the #[kunit_tests] macro and test module
location.
Co-developed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Hui Zhu <zhuhui@kylinos.cn>
Reviewed-by: David Gow <davidgow@google.com>
Link: https://lore.kernel.org/r/4eb554c3bf03dd4f9e6dea659497938baab61dba.1753929369.git.zhuhui@kylinos.cn
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Add a practical usage example to the documentation of KVec::as_slice()
showing how to:
Create a new KVec.
Push elements into it.
Convert to a slice via as_slice().
Co-developed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Hui Zhu <zhuhui@kylinos.cn>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Kunwu Chan <chentao@kylinos.cn>
Reviewed-by: David Gow <davidgow@google.com>
Link: https://lore.kernel.org/r/4e7f396f38ed8a780f863384bfc3d7de135ef3ea.1753929369.git.zhuhui@kylinos.cn
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Commit 34331d7bee ("smb: client: fix first command failure during
re-negotiation") addressed a race condition by updating lstrp before
entering negotiate state. However, this approach may have some unintended
side effects.
The lstrp field is documented as "when we got last response from this
server", and updating it before actually receiving a server response
could potentially affect other mechanisms that rely on this timestamp.
For example, the SMB echo detection logic also uses lstrp as a reference
point. In scenarios with frequent user operations during reconnect states,
the repeated calls to cifs_negotiate_protocol() might continuously
update lstrp, which could interfere with the echo detection timing.
Additionally, commit 266b5d02e1 ("smb: client: fix race condition in
negotiate timeout by using more precise timing") introduced a dedicated
neg_start field specifically for tracking negotiate start time. This
provides a more precise solution for the original race condition while
preserving the intended semantics of lstrp.
Since the race condition is now properly handled by the neg_start
mechanism, the lstrp update in cifs_negotiate_protocol() is no longer
necessary and can be safely removed.
Fixes: 266b5d02e1 ("smb: client: fix race condition in negotiate timeout by using more precise timing")
Cc: stable@vger.kernel.org
Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
Signed-off-by: Wang Zhaolong <wangzhaolong@huaweicloud.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
We already called ib_drain_qp() before and that makes sure
send_done() was called with IB_WC_WR_FLUSH_ERR, but
didn't called atomic_dec_and_test(&sc->send_io.pending.count)
So we may never reach the info->send_pending == 0 condition.
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Fixes: 5349ae5e05 ("smb: client: let send_done() cleanup before calling smbd_disconnect_rdma_connection()")
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
This is step 4/4 of a patch series to fix mid_q_entry memory leaks
caused by race conditions in callback execution.
In compound_send_recv(), when wait_for_response() is interrupted by
signals, the code attempts to cancel pending requests by changing
their callbacks to cifs_cancelled_callback. However, there's a race
condition between signal interruption and network response processing
that causes both mid_q_entry and server buffer leaks:
```
User foreground process cifsd
cifs_readdir
open_cached_dir
cifs_send_recv
compound_send_recv
smb2_setup_request
smb2_mid_entry_alloc
smb2_get_mid_entry
smb2_mid_entry_alloc
mempool_alloc // alloc mid
kref_init(&temp->refcount); // refcount = 1
mid[0]->callback = cifs_compound_callback;
mid[1]->callback = cifs_compound_last_callback;
smb_send_rqst
rc = wait_for_response
wait_event_state TASK_KILLABLE
cifs_demultiplex_thread
allocate_buffers
server->bigbuf = cifs_buf_get()
standard_receive3
->find_mid()
smb2_find_mid
__smb2_find_mid
kref_get(&mid->refcount) // +1
cifs_handle_standard
handle_mid
/* bigbuf will also leak */
mid->resp_buf = server->bigbuf
server->bigbuf = NULL;
dequeue_mid
/* in for loop */
mids[0]->callback
cifs_compound_callback
/* Signal interrupts wait: rc = -ERESTARTSYS */
/* if (... || midQ[i]->mid_state == MID_RESPONSE_RECEIVED) *?
midQ[0]->callback = cifs_cancelled_callback;
cancelled_mid[i] = true;
/* The change comes too late */
mid->mid_state = MID_RESPONSE_READY
release_mid // -1
/* cancelled_mid[i] == true causes mid won't be released
in compound_send_recv cleanup */
/* cifs_cancelled_callback won't executed to release mid */
```
The root cause is that there's a race between callback assignment and
execution.
Fix this by introducing per-mid locking:
- Add spinlock_t mid_lock to struct mid_q_entry
- Add mid_execute_callback() for atomic callback execution
- Use mid_lock in cancellation paths to ensure atomicity
This ensures that either the original callback or the cancellation
callback executes atomically, preventing reference count leaks when
requests are interrupted by signals.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=220404
Fixes: ee258d7915 ("CIFS: Move credit processing to mid callbacks for SMB3")
Signed-off-by: Wang Zhaolong <wangzhaolong@huaweicloud.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Update nova to import `ARef` from `sync::aref` instead of `types`.
This aligns with the ongoing effort to move `ARef` and
`AlwaysRefCounted` to sync.
Suggested-by: Benno Lossin <lossin@kernel.org>
Link: https://github.com/Rust-for-Linux/linux/issues/1173
Signed-off-by: Shankari Anand <shankari.ak0208@gmail.com>
Link: https://lore.kernel.org/r/20250716090941.811418-1-shankari.ak0208@gmail.com
[ Alter subject and commit message to be nova specific. - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
While nothing was really needed for virtualenv to work on most
distros, we had an issue with OpenMandriva.
While checking for it, it was noticed that there was no check if
python-virtualenv was installed.
This didn't solve the issues we faced there: at least with
the half-broken OpenMandriva Lx 4.0 docker container we used,
ensurepip was not available anywhere, causing venv to fail.
Add a distro-specific note about that.
Note: at least at the time we did our tests, OpenMandriva Lx 4.0
docker was shipped with wrong dnf repositories. Also, there
was no repos available for it anymore. So, we had to do some
hacks to upgrade to 4.3 before being able to run any tests.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/e3a0e5eccd50eb506846e3e8487a2d9124ef83e2.1754992972.git.mchehab+huawei@kernel.org
This program is somewhat complex. Add some docstring documentation,
explaining what each function and class is supposed to do.
Most of the focus here were to describe the ancillary functions used
to detect dependency needs.
The main SphinxDependencyChecker still requires a lot of care,
and probably need to be reorganized to clearly split the 4 types
of output it produces:
- Need to upgrade Python binary;
- System install needs;
- Virtual env install needs;
- Python install needs via system packages, to run Sphinx
natively.
Yet, for now, I'm happy of having it a lot better documented
than its Perl version.
-
While here, rename a parameter to have its usage better
documented.
No functional changes.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/0cadab2cab3f78ae6d9f378e92a45125fbc5188f.1754992972.git.mchehab+huawei@kernel.org
The code at get_system_release() is actually a helper function,
independent from the actual Sphinx verification checker. Move
it to MissingCheckers class, where other checkers are present.
With that, the entire distro-specific handler logic, with
all its complexity is confined at SphinxDependencyChecker
class.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/4b42a85bbb6575bb34a58cf66019038c4afa1d5b.1754992972.git.mchehab+huawei@kernel.org
Better organize the code by moving the more generic methods
to MissingCheckers. Such class contain only binary and package
dependent missing checkers, but no distro-specific data or code.
All distro-specific data/code remains at SphinxDependencyChecker
class.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/11a252fe816bd7c85583d26ade0666eb2b481bf0.1754992972.git.mchehab+huawei@kernel.org
Better implement support for RHEL-based distros. While here,
get rid of a Fedora 28 support which cause troubles with
server distros. Also, get rid of yum, as RHEL8 already
suppords dnf, and this is not the minimal version we may
still support.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/4d1b27d3a381f011e150bb50176babba83af9e1a.1754992972.git.mchehab+huawei@kernel.org
When is_optional was added in Perl, it was a boolean. With
time, it ended becoming a sort of enum, which makes the
module harder to maintain.
Convert it to a enum-like class and add more options to it.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/42290a24f3b1dbea9ebe19747cf5622bb2f2cf5c.1754992972.git.mchehab+huawei@kernel.org
Currently, if Python < 3.7, package install will fail. That happens
with OpenSuse Leap and RHEL-based ver 8 distros.
OpenSuse allows installing Sphinx with Python 3.11, but RHEL-based
distros don't.
Prepare to recomend only venv on such cases. For now, just split
the recomendation on a new function that will check for a
paramtere to be called.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/4fb2181c960e89774309a833f80209a1a3ab10d2.1754992972.git.mchehab+huawei@kernel.org
It took me a lot of time, but I guess understand now what it
takes to install a package on Gentoo.
Handling dependencies is a nightmare, as Gentoo refuses to emerge
some packages if there's no package.use file describing them.
To make it worse, compilation flags shall also be present there
for some packages. If USE is not perfect, error/warning messages
like those are shown:
gnome-base/librsvg dev-texlive/texlive-xetex media-fonts/dejavu dev-python/pyyaml
...
!!! The following binary packages have been ignored due to non matching USE:
=media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_13 qt6 svg
=media-gfx/graphviz-12.2.1-r1 X pdf python_single_target_python3_12 -python_single_target_python3_13 qt6 svg
=media-gfx/graphviz-12.2.1-r1 X pdf qt6 svg
=media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_10 qt6 svg
=media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_10 python_single_target_python3_12 -python_single_target_python3_13 qt6 svg
=media-fonts/noto-cjk-20190416 X
=app-text/texlive-core-2024-r1 X cjk -xetex
=app-text/texlive-core-2024-r1 X -xetex
=app-text/texlive-core-2024-r1 -xetex
=dev-libs/zziplib-0.13.79-r1 sdl
If emerge is allowed, it will simply ignore the above packages,
creating an incomplete installation, which will later fail when
one tries to build docs with images or build PDFs.
After the fix, command line commands to produce the needed USE
chain will be emitted, if they don't exist yet.
sudo su -c 'echo "media-gfx/graphviz" > /etc/portage/package.use/graphviz'
sudo su -c 'echo "media-gfx/imagemagick" > /etc/portage/package.use/imagemagick'
sudo su -c 'echo "media-libs/harfbuzz icu" > /etc/portage/package.use/media-libs'
sudo su -c 'echo "media-fonts/noto-cjk" > /etc/portage/package.use/media-fonts'
sudo su -c 'echo "app-text/texlive-core xetex" > /etc/portage/package.use/texlive'
sudo su -c 'echo "dev-libs/zziplib sdl" > /etc/portage/package.use/zziblib'
The new logic tries to be smart enough to detect for missing files
and missing arguments. Yet, as Gentoo seems to require users to
manage those package.use files by hand, the logic isn't perfect:
users may still need to verify for conflicts on different use
files.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/365fe5e7d568da932dcffde65f48f2c1256cb773.1754992972.git.mchehab+huawei@kernel.org
