The caller of batadv_frag_send_packet() assume that the skb provided to the
function are always consumed. But the pre-check for an empty payload or the
zero fragment size returned an error without any further actions.
A failed pre-check must use the same error handling code as the rest of the
function.
Cc: stable@vger.kernel.org
Fixes: ee75ed8887 ("batman-adv: Fragment and send skbs larger than mtu")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
A TT unicast TVLV contains the number of VLANs stored in it. This number is
an u16 and gets multiplied by the size of the struct
batadv_tvlv_tt_vlan_data (8 bytes). The size can therefore overflow the u16
used to store the tt_vlan_len. All additional safety checks to prevent
out-of-bounds access of the TVLV buffer are invalid due to this overflow.
Using size_t prevents this overflow and ensures that the safety checks
compare against the actual buffer requirements.
Cc: stable@vger.kernel.org
Fixes: 7ea7b4a142 ("batman-adv: make the TT CRC logic VLAN specific")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
batadv_send_tt_request() allocates a tt_req_node when none exists for the
destination originator node. This should prevent that a multiple TT
requests are send at the same time to an originator.
But if allocation of the send buffer failed, this request must be cleaned
up again. But indicator for such a failure is "ret == false". But the
actual implementation is checking for "ret == true".
The check must be inverted to not loose the information about the TT
request directly after it was attempted to be sent out. This should avoid
potential request storms.
Cc: stable@vger.kernel.org
Fixes: 335fbe0f5d ("batman-adv: tvlv - convert tt query packet to use tvlv unicast packets")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
When an mesh interface is registered, it creates an untagged struct
batadv_meshif_vlan on top of it via the NETDEV_REGISTER notifier. But in
this process, another receiver of this notification can veto the
registration. The netdev registration will be aborted because of this veto.
The register_netdevice() call will try to clean up the net_device using
unregister_netdevice_queue() - which only uses the .priv_destructor to
free private resources. In this situation, .dellink will not be called.
The cleanup of the untagged batadv_meshif_vlan must thefore be done in the
destructor to avoid a leak of this object.
Cc: stable@vger.kernel.org
Fixes: 5d2c05b213 ("batman-adv: add per VLAN interface attribute framework")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
On Xe3P, when PSR2 is enabled on a panel that does not support
Early Transport, DC3CO can be entered in the middle of an
active frame. This prevents the pipe from completing the frame
and leaves it in a bad state that does not recover well,
causing visible corruption on screen.
Set CHICKEN_DCPR_4 bit 24 in the PSR2 enable path when Early Transport
is not in use, to notify DMC to prevent DC3CO entry.
v2:
- Remove display from commit header (Suraj Kandpal).
- Add HSD number to intel_display_wa framework (Suraj Kandpal).
- Change register prefix from XE3LPD_ to XE3P_ (Suraj Kandpal).
BSpec: 71483, 75253
Signed-off-by: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260701100650.1689665-2-dibin.moolakadan.subrahmanian@intel.com
Convert `use` imports to vertical layout for better readability and
maintainability.
Signed-off-by: Guru Das Srinagesh <linux@gurudas.dev>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
DSM TRIM currently always emits a single 512-byte page of LBA Range Entries
(ATA_MAX_TRIM_RNUM == 64 ranges), regardless of how many pages the device
can accept in one DATA SET MANAGEMENT command.
The maximum is reported by MAX PAGES PER DSM COMMAND (IDENTIFY DEVICE word
105). Honour it: size the TRIM descriptor as a whole number of 512-byte
pages, bounded by that limit and by the logical sector size (the WRITE SAME
data-out buffer is a single logical block). Build and transfer only as
many pages as the request needs, and set the DSM COUNT field, qc->nbytes
and the maximum WRITE SAME length in the Block Limits VPD page accordingly.
Build the descriptor straight into the WRITE SAME data-out buffer using an
atomic sg_miter mapping, instead of staging it in the shared ata_scsi_rbuf
and copying it out. This removes the global ata_scsi_rbuf_lock and a
memcpy from the TRIM path.
While commit 9379e6b8e0 ("libata: Safely overwrite attached page in WRITE
SAME xlat") replaced direct access to the data-out buffer with an
intermediate step that writes the entries in the ata_scsi_rbuf buffer, this
solution writes to the data-out buffer using sg_miter, which maps each
segment with kmap_atomic (SG_MITER_ATOMIC), so it's highmem- and
multi-segment-safe, and it's usable from the non-sleeping
command-submission path (unlike the page_address() access that
ata_scsi_rbuf originally replaced).
A 512-byte-sector device still uses a single page, so its behaviour is
unchanged.
Add ata_id_dsm_max_pages() to read IDENTIFY DEVICE word 105.
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
ata_scsi_write_same_xlat() translates a SCSI WRITE SAME command with the
UNMAP bit set into an ATA DATA SET MANAGEMENT TRIM command. The TRIM
descriptor is built by ata_format_dsm_trim_descr() into the 2048-byte
ata_scsi_rbuf staging buffer, and the number of bytes copied is compared
against the logical sector size by the caller:
size = ata_format_dsm_trim_descr(scmd, trmax, block, n_block);
if (size != len) /* len == sdp->sector_size */
goto invalid_param_len;
ata_format_dsm_trim_descr() clamps the copy length to ATA_SCSI_RBUF_SIZE
(2048). On a device whose logical sector size exceeds that (e.g. a 4Kn
device, where sector_size == 4096) the function can never return more than
2048, while the caller expects it to return sector_size. The comparison
therefore always fails, so every TRIM is rejected with "Parameter list
length error" and WARN_ON() splats on each attempt. TRIM / discard is
thus completely broken on such devices.
The descriptor was incorrectly sized from the logical sector size. A DSM
TRIM payload is a list of 512-byte pages, each holding up to
ATA_MAX_TRIM_RNUM (64) LBA Range Entries, and is independent of the logical
sector size. The Block Limits VPD page already advertises a single such
page as the maximum WRITE SAME length (65535 * ATA_MAX_TRIM_RNUM logical
blocks), so the block layer never sends a request that needs more than one
page.
Emit exactly one 512-byte page, independent of the logical sector size,
and transfer only that page (COUNT == 1). For a 512-byte-sector device
this is unchanged; devices with larger logical sectors now work instead of
failing every TRIM.
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Fixes: ef2d7392c4 ("libata: SCT Write Same / DSM Trim")
Cc: stable@vger.kernel.org
Signed-off-by: Niklas Cassel <cassel@kernel.org>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Add support to ata_scsi_report_supported_opcodes() for the all command
format indicated with a reporting option of 0. The function
ata_scsi_report_all_supported_opcodes() is introduced to implement this
support. This function operates by testing all commands of the
ata_supported_cmds array and testing them using
ata_scsi_cmd_is_supported(), filling rbuf as it loops through all the
commands that libata can emulate or translate.
With this change, sg_opcodes /dev/sdX is able to list all commands
supported by libata and a device. An example below is shown for a SATA
disk also supporting CDL:
# sg_opcodes /dev/sda
ATA WDC WUH722626AL WZ41
Peripheral device type: disk
Opcode Service CDB RWCDLP, Name
(hex) action(h) size CDLP
-----------------------------------------------
00 6 0,0 Test Unit Ready
01 6 0,0 Rezero Unit
03 6 0,0 Request Sense
08 6 0,0 Read(6)
0a 6 0,0 Write(6)
0b 6 0,0 Seek(6)
12 6 0,0 Inquiry
15 6 0,0 Mode select(6)
1a 6 0,0 Mode sense(6)
1b 6 0,0 Start stop unit
1d 6 0,0 Send diagnostic
25 10 0,0 Read capacity(10)
28 10 0,0 Read(10)
2a 10 0,0 Write(10)
2b 10 0,0 Seek(10)
2f 10 0,0 Verify(10)
35 10 0,0 Synchronize cache(10)
55 10 0,0 Mode select(10)
5a 10 0,0 Mode sense(10)
7f 1ff0 32 0,0 ATA pass-through(32)
85 16 0,0 ATA pass-through(16)
88 16 1,1 Read(16)
8a 16 1,2 Write(16)
8f 16 0,0 Verify(16)
91 16 0,0 Synchronize cache(16)
9e 10 16 0,0 Read capacity(16)
a0 12 0,0 Report luns
a1 12 0,0 ATA pass-through(12)
a3 c 12 0,0 Report supported operation codes
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
Add support for the reporting options value 2 of the in REPORT SUPPORTED
OPERATION CODES command in ata_scsi_report_supported_opcodes(). With this
reporting option, the specified opcode to check must have a service
action. Use ata_scsi_supported_cmd_has_sa() to check for this and fill the
command reply buffer using the one command format based on the command
support.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
ata_scsi_report_supported_opcodes() is ignoring the service action
specified in the SCSI command CDB, but the one command format must take
this field into consideration. With the reporting options field set to 1,
the REPORT SUPPORTED OPERATION CODES command must be failed if the
specified opcode to check has service actions, while reporting option 3
must match supported opcodes together with the specified service action.
Stop ignoring the service action by passing it to
ata_scsi_cmd_is_supported() and searching for commands in the
array of supported commands (ata_supported_cmds) using both the command
opcode and service action.
Introduce the helper function ata_scsi_supported_cmd_use_sa() to determine
if a particular command has service actions and use this function to fail
a REPORT SUPPORTED OPERATION CODES command if such command is specified
with reporting options 1.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
ata_scsi_report_supported_opcodes() is very limited in functionality as it
lacks support for the all command format and also does not handle
correctly commands that have a service action.
In preparation for adding these missing features, refactor how
ata_scsi_report_supported_opcodes() operates to make modifications and
extensions easier. To do so, introduce the array of supported commands
ata_supported_cmds. This array entries are of type struct ata_scsi_cmd.
This structure stores the operation code, CDB length, and the service
action of a supported SCSI command that libata SAT can translate or
emulate. Since some service actions (e.g. ZI_REPORT_ZONES) can have a
value of 0, the field sa_valid of struct ata_scsi_cmd is used to indicate
if the sa field is valid, or if it should be ignored.
The helper function ata_scsi_get_supported_cmd() is implemented to search
for a particular command by opcode in this array. This function is used in
ata_scsi_cmd_is_supported() together with a struct ata_scsi_cmd_support to
check based on the target device features if the specified command is
supported.
ata_scsi_cmd_is_supported() is used as the main function in
ata_scsi_report_supported_opcodes() to determine if a particular command
is supported and fill the command reply rbuf as needed. In the case of a
command that is not supported, the support field is set to 1 as specified
in SPC, indicating that the command is not supported.
Of note is that the old ata_scsi_report_supported_opcodes() code did not
handle the VARIABLE_LENGTH_CMD/ATA_32 command which is supported and
translated by libata-scsi. The ata_supported_cmds array includes this
command. As commented in the code, WRITE_SAME_16 is left out of the array
of supported command ata_scsi_write_same_xlat() prevents the use of this
command as a passthrough command.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
The helper function ata_dev_is_zac() checks if a device is a ZAC class
device (host managed zoned disk) or if it is a host aware zoned disk, that
is, a regular ATA disk that supports the zoned capabilities. So the name
of this helper function is confusing as it hints at the first case only.
Rename this helper function to ata_dev_is_zoned() to avoid confusions and
better reflect the two cases tested. Use this helper in
ata_scsiop_inq_std(), ata_scsiop_read_cap() and
ata_scsi_report_supported_opcodes() instead of having the same tests open
coded.
While at it, to stay consistent with this renaming, also rename
ata_dev_config_zac() to ata_dev_config_zoned().
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@kernel.org>
The NFSD_CMD_UNLOCK_FILESYSTEM netlink command runs
nfsd4_cancel_copy_by_sb() before nfsd_mutex is held and before
nn->nfsd_serv is confirmed set, the same pre-mutex ordering the procfs
unlock_filesystem path carried. Once nfsd has shut down,
nfs4_state_destroy_net() has freed nn->conf_id_hashtbl but left the
pointer intact, so the cancel helper iterates freed slab memory as an
array of struct list_head and then dereferences a bogus nfs4_client
when it takes clp->async_lock. A local administrator holding
CAP_SYS_ADMIN can reach this use-after-free by stopping the server and
then issuing the command.
Move the async COPY cancel into the nfsd_mutex section, after
nn->nfsd_serv is confirmed, so every NFSv4 state-table walker on this
path observes a running server. Async copies exist only while the
server runs, so gating the cancel on nn->nfsd_serv loses nothing.
Fixes: 327c5168ef ("NFSD: Add NFSD_CMD_UNLOCK_FILESYSTEM netlink command")
Cc: stable@vger.kernel.org
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Link: https://patch.msgid.link/20260613-unlock-filesystem-uaf-v1-2-462b9bec8c84@kernel.org
Signed-off-by: Chuck Lever <cel@kernel.org>
In the `skip_invalidate:` path under `cifs_revalidate_mapping()`, the
sequence of calls:
clear_bit_unlock();
smp_mb__after_atomic();
wake_up_bit();
can be replaced exactly by `clear_and_wake_up_bit()`.
The `clear_and_wake_up_bit()` helper function was introduced in
'commit 8236b0ae31 ("bdi: wake up concurrent wb_shutdown()
callers.")' to replace equivalent instances of this sequence of
operations. This substitution has been applied in multiple subsystems.
Compile-tested with CONFIG_CIFS=y on x86_64, no new warnings present.
Suggested-by: Agatha Isabelle Moreira <code@agatha.dev>
Link: https://kernelnewbies.org/Beginner%20Cleanup%20and%20Refactor%20Tasks%20by%20Agatha%20Isabelle%20Moreira#task_010
Cc: Agatha Isabelle Moreira <code@agatha.dev>
Signed-off-by: Mehdi Hassan <mehdi.h.business@pm.me>
Signed-off-by: Steve French <stfrench@microsoft.com>
Top-level comment is not a kerneldoc, so drop marker to fix W=1 warning:
drivers/iio/adc/ade9000.c:2 This comment starts with '/**', but isn't a kernel-doc comment. Refer to Documentation/doc-guide/kernel-doc.rst
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Maxwell Doose <m32285159@gmail.com>
Reviewed-by: Joshua Crofts <joshua.crofts1@gmail.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
USB controller on Kaanapali SoC is capable of switching role betwen device
and host and also is a wakeup capable device. Add usb-role-switch and
wakeup-source properties to the controller node accordingly.
Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260702184440.442037-1-krishna.kurapati@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
USB controller on SM8750 can be configured as a wakeup capable device.
Add wakeup-source property indicating the same.
Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260702184511.442129-1-krishna.kurapati@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
We know the driver is reporting s3706b, introduce the compatible so we
can more easily introduce quirks for weird touchscreen replacements in
followup series.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: David Heidelberg <david@ixit.cz>
Link: https://lore.kernel.org/r/20260523-synaptics-rmi4-dt-v2-2-0645122babdc@ixit.cz
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Add a default pin configuration for CAM_MCLK3 (GPIO97), which is used
as a master clock for camera sensors on SM8250 boards.
Signed-off-by: Xin Xu <xxsemail@qq.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Link: https://lore.kernel.org/r/tencent_47473BB31C9142449F8C7C1BE69D9071510A@qq.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Add the pm8008 PMIC node on i2c15 with seven LDOs,
using GPIO84 as interrupt and GPIO76 as reset.
Signed-off-by: Xin Xu <xxsemail@qq.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Link: https://lore.kernel.org/r/tencent_D4AB219A942591742C3A6DA4DFB0EBA49505@qq.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Some new futex tests rely on kernel features that may not be supported
on all configurations (e.g., FUTEX2_NUMA or PR_FUTEX_HASH).
Currently, these tests fail with EINVAL when the features are missing.
Add dynamic skip logic to:
- futex_priv_hash: Skip if PR_FUTEX_HASH returns ENOSYS or EINVAL.
- futex_numa_mpol: Skip if futex2 or FUTEX2_NUMA returns ENOSYS or EINVAL.
This allows the tests to pass (as SKIPPED) on kernels without support,
while preserving coverage on configurations that do support them.
[ tglx: Adapt to harness changes, massage change log ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260624070223.1533167-1-wakel@google.com
The futex functional tests cover FUTEX_LOCK_PI timeout semantics
(futex_wait_timeout.c) and robust-list owner death (robust_list.c), but
nothing exercises a non-robust PI owner exiting while holding the lock,
nor the basic ownership / EDEADLK / unlock word semantics of
FUTEX_LOCK_PI.
Add futex_lock_pi_exiting.c with three tests:
- lock_unlock_basic: an uncontended FUTEX_LOCK_PI puts the owner TID in
the futex word, a recursive lock by the owner returns EDEADLK, and
FUTEX_UNLOCK_PI clears the word.
- owner_dies_with_blocked_waiter: a thread acquires a PI futex and exits
while holding it; a blocked FUTEX_LOCK_PI waiter must come out cleanly
(0, EOWNERDEAD or ESRCH) and own the lock when it acquires.
- stress_owner_exits: repeatedly drive the same exiting-owner path.
The exiting-owner retry path is where commit 210d36d892 ("futex: Clear
stale exiting pointer in futex_lock_pi() retry path") fixed a stale task
pointer that tripped WARN_ON_ONCE() in wait_for_owner_exiting(). That
warning does not change the syscall return value, so this test does not
assert on it directly; the stress loop exists to give a kernel booted
with panic_on_warn=1 (as fuzzers and CI commonly run) a chance to trip
on it.
Signed-off-by: Valery Borovsky <vebohr@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260607183249.246037-1-vebohr@gmail.com
Migrate robust_list test to the kselftest harness framework, removing mixed
legacy ksft_* API usages and passing test metadata to cloned child
functions via child_args struct.
Also, fix an out-of-bounds waitpid bug during multithreaded tests.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-8-wakel@google.com
Migrate futex_priv_hash test to the kselftest harness framework, removing
mixed legacy ksft_* API usages and passing test metadata to all helper
functions.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-7-wakel@google.com
Migrate futex_numa_mpol test to the kselftest harness framework, removing
mixed legacy ksft_* API usages and passing test metadata to helper
functions.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-6-wakel@google.com
Migrate futex_requeue_pi_signal_restart test to the kselftest harness
framework, removing mixed legacy ksft_* API usages and ensuring proper
thread joining.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-5-wakel@google.com
Migrate futex_requeue_pi_mismatched_ops test to the kselftest harness
framework, removing mixed legacy ksft_* API usages and passing test
metadata to helper thread.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-4-wakel@google.com
Migrate futex_requeue_pi test to the kselftest harness framework, removing
mixed legacy ksft_* API usages and passing test metadata to all helper
threads via thread arguments.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-3-wakel@google.com
Migrate futex_requeue test to the kselftest harness framework, removing
mixed legacy ksft_* API usages and ensuring proper thread joining.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260601065126.3623867-2-wakel@google.com
Migrate futex_wait_uninitialized_heap test to the kselftest harness
framework, removing mixed legacy ksft_* API usages and ensuring proper
thread joining.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260525092002.3762888-4-wakel@google.com
Migrate futex_wait_private_mapped_file test to the kselftest harness
framework, removing mixed legacy ksft_* API usages and passing test
metadata to helper thread.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260525092002.3762888-3-wakel@google.com
Migrate futex_wait test to the kselftest harness framework, removing mixed
legacy ksft_* API usages and ensuring proper thread joining.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260525092002.3762888-2-wakel@google.com
In futex_waitv negative tests (invalid_flag, unaligned_address, etc.), test
results are evaluated as:
if (res == EINVAL)
Since sys_futex_waitv returns -1 on error and sets errno, direct positive
comparisons against res are always false, causing tests to silently pass
regardless of real errors.
Correct these validations to assert EXPECT_EQ(res, -1) and compare errno
directly against expected constants.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260526010635.23980-3-wakel@google.com
Currently, multiple futex functional tests (wait_timeout, waitv,
wait_wouldblock) mix low-level ksft_* logging and result APIs with the
kselftest_harness.h framework. On older kernels where system calls like
futex_waitv are missing (returning -ENOSYS), this mixed usage triggers
framework inconsistencies, causing the test to fail with: "Illegal usage of
low-level ksft APIs in harness test".
Address this by completely refactoring these tests to exclusively use
the high-level kselftest_harness.h framework (gtest-like API), mapping
all low-level calls to native harness macros:
- Non-fatal assertions: Replace ksft_test_result_fail() with EXPECT_EQ()
and EXPECT_NE().
- Fatal assertions: Replace ksft_exit_fail_msg() with ASSERT_EQ() and
ASSERT_TRUE() to immediately terminate execution upon setup failure.
- Test skipping: Replace ksft_exit_skip() with the graceful SKIP() macro
combined with early runtime availability checks for sys_futex_waitv.
- Debug logging: Replace ksft_print_dbg_msg() with TH_LOG() to inherit
automatic file/line context.
- Success reporting: Remove explicit ksft_test_result_pass() calls,
deferring to automated harness completion reporting.
Additionally:
- Fix a critical SIGSEGV crash in early syscall probing logic caused by
passing a NULL pointer to inline user-space timespec conversions.
- Introduce TEST_TIMEOUT() and GET_ABS_TIMEOUT() macros in wait_timeout
to encapsulate assertions while preserving source line attribution.
- Pass test metadata (_metadata) into secondary threads to allow native
harness assertions during concurrent execution.
[ tglx: Fixup coding style ]
Signed-off-by: Wake Liu <wakel@google.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260526010635.23980-2-wakel@google.com
'head' is defined as 'static struct robust_list_head' that stores the
local variable of 'struct lock_struct a' raising the Wdangling-pointer
warning.
robust_list.c: In function ‘child_circular_list’:
robust_list.c:522:24: warning: storing the address of local variable
‘a’ in ‘head.list.next’ [-Wdangling-pointer=]
522 | head.list.next = &a.list;
| ~~~~~~~~~~~~~~~^~~~~~~~~
robust_list.c:513:28: note: ‘a’ declared here
513 | struct lock_struct a, b, c;
| ^
robust_list.c:512:40: note: ‘head’ declared here
512 | static struct robust_list_head head;
| ^~~~
Since 'head' doesn't need static storge duration, removing the static
keyword of it to fix this.
Signed-off-by: Ankit Khushwaha <ankitkhushwaha.linux@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260420173544.55288-1-ankitkhushwaha.linux@gmail.com
After removing all extra entries from the UBWC database it is easy to
define generic entries, common for all machine classes.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-28-72f2749bc807@oss.qualcomm.com
[bjorn: Dropped unused "ubwc_3_0_hbb16", per compiler warning]
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Sort out the remaining UBWC swizzle values, using flags to control
whether level 2 and level 3 swizzling are enabled or not.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-27-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
UBWC devices before 4.0 use standard UBWC swizzle levels. As all the
drivers now use the qcom_ubwc_swizzle() helper, move those values to the
helper, leaving UBWC 4.0+ intact for now.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-26-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
All the users have been migrated to using qcom_ubwc_macrotile_mode()
instead of reading the raw value from the config structure. Drop the
field from struct qcom_ubwc_cfg_data and replace it with the calculated
value. Split single UBWC_3_0 into UBWC_3_0 (no macrotile mode) and
UBWC_3_1 (with macrotile mode).
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-25-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
According to the documentation, UBWC bank spreading should be enabled
for all targets. It's just not all targets have separate bit to control
it. Drop the bit from the database and make the helper always return
true. If we need to change it later, the helper can be adjusted
according to the programming guides.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-24-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
The ubwc_dec_version field has been inherited from the MDSS driver and
it is equal to the version of the UBWC decoder in the display block
only. Other IP Cores can have different UBWC decoders and so the version
would vary between blocks.
As the value is no longer used as is not relevant to other UBWC database
consumers, drop it from the UBWC database.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-23-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
According to the documentation, the MAL should be set for all UBWC 1.0
targets, no matter what is the version of the UBWC decoders are present
on the device. The helper comes from DPU / GPU world, where there was no
separate bit to control MAL before UBWC 2.0. As the helper is now being
used by other drivers too, correct the helper to return the correct MAL
value (Iris doesn't support UBWC 1.0 devices for now, so there is no
changes of the behaviour).
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260520-ubwc-rework-v5-22-72f2749bc807@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
checkpatch.pl reports an improper SPDX comment style for this file:
WARNING: Improper SPDX comment style for 'drivers/soc/qcom/cmd-db.c',
please use '//' instead
Per Documentation/process/license-rules.rst, C source files must use the
'// SPDX-License-Identifier: ...' form instead of the '/* ... */' form.
Fix it.
Signed-off-by: Mayur Kumar <kmayur809@gmail.com>
Reviewed-by: Sneh Mankad <sneh.mankad@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260510193840.59156-1-kmayur809@gmail.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
