Skip to content

Conversation

@bardliao
Copy link
Collaborator

Initial upstream merge for 6.11
tag sof-dev-rebase-20240715 pushed to https://github.com/bardliao/linux/releases/tag/sof-dev-rebase-20240730

liuhangbin and others added 30 commits July 24, 2024 12:50
…learning limit

If the testing kernel doesn't support setting fdb_max_learned or show
fdb_n_learned, just skip it. Or we will get errors like

./bridge_fdb_learning_limit.sh: line 218: [: null: integer expression expected
./bridge_fdb_learning_limit.sh: line 225: [: null: integer expression expected

Fixes: 6f84090 ("selftests: forwarding: bridge_fdb_learning_limit: Add a new selftest")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Johannes Nixdorf <jnixdorf-oss@avm.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit 7caa976 ("ftrace: riscv: move from REGS to ARGS"),
kprobe on ftrace is not supported by riscv, because riscv's support for
FTRACE_WITH_REGS has been replaced with support for FTRACE_WITH_ARGS, and
KPROBES_ON_FTRACE will be supplanted by FPROBES. So remove the deprecated
kprobe on ftrace support, which is misunderstood.

Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20240613111347.1745379-1-ruanjinjie@huawei.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The "targets:" is duplicated in another line, remove the one with less
targets.

Signed-off-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Emil Renner Berthing <emil.renner.berthing@canonical.com>
Link: https://lore.kernel.org/r/20240613153053.3835-1-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
io_uring_cancel_generic() should retry if any state changes like a
request is completed, however in case of a task exit it only goes for
another loop and avoids schedule() if any tracked (i.e. REQ_F_INFLIGHT)
request got completed.

Let's assume we have a non-tracked request executing in iowq and a
tracked request linked to it. Let's also assume
io_uring_cancel_generic() fails to find and cancel the request, i.e.
via io_run_local_work(), which may happen as io-wq has gaps.
Next, the request logically completes, io-wq still hold a ref but queues
it for completion via tw, which happens in
io_uring_try_cancel_requests(). After, right before prepare_to_wait()
io-wq puts the request, grabs the linked one and tries executes it, e.g.
arms polling. Finally the cancellation loop calls prepare_to_wait(),
there are no tw to run, no tracked request was completed, so the
tctx_inflight() check passes and the task is put to indefinite sleep.

Cc: stable@vger.kernel.org
Fixes: 3f48cf1 ("io_uring: unify files and task cancel")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/acac7311f4e02ce3c43293f8f1fda9c705d158f1.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IORING_SETUP_IOPOLL rings don't have any netpoll handling, let's fail
attempts to register netpolling in this case, there might be people who
will mix up IOPOLL and netpoll.

Cc: stable@vger.kernel.org
Fixes: ef1186c ("io_uring: add register/unregister napi function")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/1e7553aee0a8ae4edec6742cd6dd0c1e6914fba8.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The __must_hold annotation in io_match_task() uses a non existing
parameter "req", fix it.

Fixes: 6af3f48 ("io_uring: fix link traversal locking")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3e65ee7709e96507cef3d93291746f2c489f2307.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We don't have to return error code from an op handler back to core
io_uring, so once io_uring_cmd() sets the results and handles errors we
can juts return IOU_OK and simplify the code.

Note, only valid with e0b23d9 ("io_uring: optimise ltimeout for
inline execution"), there was a problem with iopoll before.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8eae2be5b2a49236cd5f1dadbd1aa5730e9e2d4f.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We removed the reliance on the flag by the cancellation path and now
it's unused.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e57afe566bbe4fefeb44daffb08900f2a4756577.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is a difference in how io_queue_sqe and io_wq_submit_work treat
error codes they get from io_issue_sqe. The first one fails anything
unknown but latter only fails when the code is negative.

It doesn't make sense to have this discrepancy, align them to the
io_queue_sqe behaviour.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c550e152bf4a290187f91a4322ddcb5d6d1f2c73.1721819383.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The perfect_match parameter of the update_vlan_hash operation is __le16,
and is correctly converted from host byte-order in the lone caller,
stmmac_vlan_update().

However, the implementations of this caller, dwxgmac2_update_vlan_hash()
and dwxgmac2_update_vlan_hash(), both treat this parameter as host byte
order, using the following pattern:

	u32 value = ...
	...
	writel(value | perfect_match, ...);

This is not correct because both:
1) value is host byte order; and
2) writel expects a host byte order value as it's first argument

I believe that this will break on big endian systems. And I expect it
has gone unnoticed by only being exercised on little endian systems.

The approach taken by this patch is to update the callback, and it's
caller to simply use a host byte order value.

Flagged by Sparse.
Compile tested only.

Fixes: c7ab0b8 ("net: stmmac: Fallback to VLAN Perfect filtering if HASH is not available")
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct nexthop_grp contains two reserved fields that are not initialized by
nla_put_nh_group(), and carry garbage. This can be observed e.g. with
strace (edited for clarity):

    # ip nexthop add id 1 dev lo
    # ip nexthop add id 101 group 1
    # strace -e recvmsg ip nexthop get id 101
    ...
    recvmsg(... [{nla_len=12, nla_type=NHA_GROUP},
                 [{id=1, weight=0, resvd1=0x69, resvd2=0x67}]] ...) = 52

The fields are reserved and therefore not currently used. But as they are, they
leak kernel memory, and the fact they are not just zero complicates repurposing
of the fields for new ends. Initialize the full structure.

Fixes: 430a049 ("nexthop: Add support for nexthop groups")
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update my email address, clarify support status, and delete the
web site that hasn't been used in a long time.

Signed-off-by: Jay Vosburgh <j.vosburgh@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
netconsole has no maintainer, and Breno has been working on
improving it consistently for some time. So I think we found
the maintainer :)

Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Breno Leitao <leitao@debian.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ACPI SPCR code has been used to enable console output for ARM64 and
X86. The same code can be reused for RISC-V. Furthermore, SPCR table is
mandated for headless system as outlined in the RISC-V BRS
Specification, chapter 6.

Signed-off-by: Sia Jee Heng <jeeheng.sia@starfivetech.com>
Reviewed-by: Sunil V L <sunilvl@ventanamicro.com>
Link: https://lore.kernel.org/r/20240502073751.102093-2-jeeheng.sia@starfivetech.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
…f_init()

ci_leaf_init() is a declared static function. The implementation of the
function body and the caller do not use the parameter (struct device_node
*node) input parameter, so remove it.

Fixes: 6a24915 ("Revert "riscv: Set more data to cacheinfo"")
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
Reviewed-by: Jeremy Linton <jeremy.linton@arm.com>
Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
Link: https://lore.kernel.org/r/20240617131425.7526-1-cuiyunhui@bytedance.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Before cacheinfo can be built correctly, we need to initialize level
and type. Since RISC-V currently does not have a register group that
describes cache-related attributes like ARM64, we cannot obtain them
directly, so now we obtain cache leaves from the ACPI PPTT table
(acpi_get_cache_info()) and set the cache type through split_levels.

Suggested-by: Jeremy Linton <jeremy.linton@arm.com>
Suggested-by: Sudeep Holla <sudeep.holla@arm.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Jeremy Linton <jeremy.linton@arm.com>
Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
Link: https://lore.kernel.org/r/20240617131425.7526-2-cuiyunhui@bytedance.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
After adding ACPI support to populate_cache_leaves(), RISC-V can build
cacheinfo through the ACPI PPTT table, thus enabling the ACPI_PPTT
configuration.

Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
Reviewed-by: Jeremy Linton <jeremy.linton@arm.com>
Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
Reviewed-by: Sunil V L <sunilvl@ventanamicro.com>
Link: https://lore.kernel.org/r/20240617131425.7526-3-cuiyunhui@bytedance.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
When a process' cred struct is replaced, this _almost_ always invokes
the cred_prepare LSM hook; but in one special case (when
KEYCTL_SESSION_TO_PARENT updates the parent's credentials), the
cred_transfer LSM hook is used instead.  Landlock only implements the
cred_prepare hook, not cred_transfer, so KEYCTL_SESSION_TO_PARENT causes
all information on Landlock restrictions to be lost.

This basically means that a process with the ability to use the fork()
and keyctl() syscalls can get rid of all Landlock restrictions on
itself.

Fix it by adding a cred_transfer hook that does the same thing as the
existing cred_prepare hook. (Implemented by having hook_cred_prepare()
call hook_cred_transfer() so that the two functions are less likely to
accidentally diverge in the future.)

Cc: stable@kernel.org
Fixes: 385975d ("landlock: Set up the security framework and manage credentials")
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20240724-landlock-houdini-fix-v1-1-df89a4560ca3@google.com
Signed-off-by: Mickaël Salaün <mic@digikod.net>
Check that keyctl(KEYCTL_SESSION_TO_PARENT) preserves the parent's
restrictions.

Fixes: e119981 ("selftests/landlock: Add user space tests")
Co-developed-by: Jann Horn <jannh@google.com>
Signed-off-by: Jann Horn <jannh@google.com>
Link: https://lore.kernel.org/r/20240724.Ood5aige9she@digikod.net
Signed-off-by: Mickaël Salaün <mic@digikod.net>
Our test report the following hung task:

[ 2538.459400] INFO: task "kworker/0:0":7 blocked for more than 188 seconds.
[ 2538.459427] Call trace:
[ 2538.459430]  __switch_to+0x174/0x338
[ 2538.459436]  __schedule+0x628/0x9c4
[ 2538.459442]  schedule+0x7c/0xe8
[ 2538.459447]  schedule_preempt_disabled+0x24/0x40
[ 2538.459453]  __mutex_lock+0x3ec/0xf04
[ 2538.459456]  __mutex_lock_slowpath+0x14/0x24
[ 2538.459459]  mutex_lock+0x30/0xd8
[ 2538.459462]  del_gendisk+0xdc/0x350
[ 2538.459466]  sd_remove+0x30/0x60
[ 2538.459470]  device_release_driver_internal+0x1c4/0x2c4
[ 2538.459474]  device_release_driver+0x18/0x28
[ 2538.459478]  bus_remove_device+0x15c/0x174
[ 2538.459483]  device_del+0x1d0/0x358
[ 2538.459488]  __scsi_remove_device+0xa8/0x198
[ 2538.459493]  scsi_forget_host+0x50/0x70
[ 2538.459497]  scsi_remove_host+0x80/0x180
[ 2538.459502]  usb_stor_disconnect+0x68/0xf4
[ 2538.459506]  usb_unbind_interface+0xd4/0x280
[ 2538.459510]  device_release_driver_internal+0x1c4/0x2c4
[ 2538.459514]  device_release_driver+0x18/0x28
[ 2538.459518]  bus_remove_device+0x15c/0x174
[ 2538.459523]  device_del+0x1d0/0x358
[ 2538.459528]  usb_disable_device+0x84/0x194
[ 2538.459532]  usb_disconnect+0xec/0x300
[ 2538.459537]  hub_event+0xb80/0x1870
[ 2538.459541]  process_scheduled_works+0x248/0x4dc
[ 2538.459545]  worker_thread+0x244/0x334
[ 2538.459549]  kthread+0x114/0x1bc

[ 2538.461001] INFO: task "fsck.":15415 blocked for more than 188 seconds.
[ 2538.461014] Call trace:
[ 2538.461016]  __switch_to+0x174/0x338
[ 2538.461021]  __schedule+0x628/0x9c4
[ 2538.461025]  schedule+0x7c/0xe8
[ 2538.461030]  blk_queue_enter+0xc4/0x160
[ 2538.461034]  blk_mq_alloc_request+0x120/0x1d4
[ 2538.461037]  scsi_execute_cmd+0x7c/0x23c
[ 2538.461040]  ioctl_internal_command+0x5c/0x164
[ 2538.461046]  scsi_set_medium_removal+0x5c/0xb0
[ 2538.461051]  sd_release+0x50/0x94
[ 2538.461054]  blkdev_put+0x190/0x28c
[ 2538.461058]  blkdev_release+0x28/0x40
[ 2538.461063]  __fput+0xf8/0x2a8
[ 2538.461066]  __fput_sync+0x28/0x5c
[ 2538.461070]  __arm64_sys_close+0x84/0xe8
[ 2538.461073]  invoke_syscall+0x58/0x114
[ 2538.461078]  el0_svc_common+0xac/0xe0
[ 2538.461082]  do_el0_svc+0x1c/0x28
[ 2538.461087]  el0_svc+0x38/0x68
[ 2538.461090]  el0t_64_sync_handler+0x68/0xbc
[ 2538.461093]  el0t_64_sync+0x1a8/0x1ac

  T1:				T2:
  sd_remove
  del_gendisk
  __blk_mark_disk_dead
  blk_freeze_queue_start
  ++q->mq_freeze_depth
  				bdev_release
 				mutex_lock(&disk->open_mutex)
  				sd_release
 				scsi_execute_cmd
 				blk_queue_enter
 				wait_event(!q->mq_freeze_depth)
  mutex_lock(&disk->open_mutex)

SCSI does not set GD_OWNS_QUEUE, so QUEUE_FLAG_DYING is not set in
this scenario. This is a classic ABBA deadlock. To fix the deadlock,
make sure we don't try to acquire disk->open_mutex after freezing
the queue.

Cc: stable@vger.kernel.org
Fixes: eec1be4 ("block: delete partitions later in del_gendisk")
Signed-off-by: Yang Yang <yang.yang@vivo.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Fixes: and Cc: stable tags are missing. Otherwise this patch looks fine
Link: https://lore.kernel.org/r/20240724070412.22521-1-yang.yang@vivo.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In ublk_ctrl_uring_cmd(), ioctl command NR should be used for
matching _IOC_NR(cmd_op).

Fix it by adding one private macro, and this way is clean.

Fixes: 13fe8e6 ("ublk: add UBLK_CMD_DEL_DEV_ASYNC")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20240724143311.2646330-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Adjust quirks for 0x3a20, 0x3a30, 0x3a50 to match the 0x3a60. This
set has now been confirmed to work with this patch.

Signed-off-by: Luke D. Jones <luke@ljones.dev>
Fixes: 811dd42 ("ALSA: hda/realtek: Add quirks for Asus ROG 2024 laptops using CS35L41")
Cc: <stable@vger.kernel.org>
Link: https://patch.msgid.link/20240723011224.115579-1-luke@ljones.dev
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Commit e3ec0fe ("hostfs: Convert hostfs_read_folio() to use a
folio") simplified hostfs_read_folio(), but in the process of converting
to using folios natively also mis-used the folio_zero_tail() function
due to the very confusing API of that function.

Very arguably it's folio_zero_tail() API itself that is buggy, since it
would make more sense (and the documentation kind of implies) that the
third argument would be the pointer to the beginning of the folio
buffer.

But no, the third argument to folio_zero_tail() is where we should start
zeroing the tail (even if we already also pass in the offset separately
as the second argument).

So fix the hostfs caller, and we can leave any folio_zero_tail() sanity
cleanup for later.

Reported-and-tested-by: Maciej Żenczykowski <maze@google.com>
Fixes: e3ec0fe ("hostfs: Convert hostfs_read_folio() to use a folio")
Link: https://lore.kernel.org/all/CANP3RGceNzwdb7w=vPf5=7BCid5HVQDmz1K5kC9JG42+HVAh_g@mail.gmail.com/
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Christian Brauner <brauner@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
…x/kernel/git/vfs/vfs

Pull vfs fixes from Christian Brauner:
 "VFS:

   - The new 64bit mount ids start after the old mount id, i.e., at the
     first non-32 bit value. However, we started counting one id too
     late and thus lost 4294967296 as the first valid id. Fix that.

   - Update a few comments on some vfs_*() creation helpers.

   - Move copying of the xattr name out from the locks required to start
     a filesystem write.

   - Extend the filelock lock UAF fix to the compat code as well.

   - Now that we added the ability to look up an inode under RCU it's
     possible that lockless hash lookup can find and lock an inode after
     it gets I_FREEING set. It then waits until inode teardown in
     evict() is finished.

     The flag however is still set after evict() has woken up all
     waiters. If the inode lock is taken late enough on the waiting side
     after hash removal and wakeup happened the waiting thread will
     never be woken.

     Before RCU based lookup this was synchronized via the
     inode_hash_lock. But since unhashing requires the inode lock as
     well we can check whether the inode is unhashed while holding inode
     lock even without holding inode_hash_lock.

  pidfd:

   - The nsproxy structure contains nearly all of the namespaces
     associated with a task. When a namespace type isn't supported
     nsproxy might contain a NULL pointer or always point to the initial
     namespace type. The logic isn't consistent. So when deriving
     namespace fds we need to ensure that the namespace type is
     supported.

     First, so that we don't risk dereferncing NULL pointers. The
     correct bigger fix would be to change all namespaces to always set
     a valid namespace pointer in struct nsproxy independent of whether
     or not it is compiled in. But that requires quite a few changes.

     Second, so that we don't allow deriving namespace fds when the
     namespace type doesn't exist and thus when they couldn't also be
     derived via /proc/self/ns/.

   - Add missing selftests for the new pidfd ioctls to derive namespace
     fds. This simply extends the already existing testsuite.

  netfs:

   - Fix debug logging and fix kconfig variable name so it actually
     works.

   - Fix writeback that goes both to the server and cache. The streams
     are only activated once a subreq is added. When a server write
     happens the subreq doesn't need to have finished by the time the
     cache write is started. If the server write has already finished by
     the time the cache write is about to start the cache write will
     operate on a folio that might already have been reused. Fix this by
     preactivating the cache write.

   - Limit cachefiles subreq size for cache writes to MAX_RW_COUNT"

* tag 'vfs-6.11-rc1.fixes.2' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
  inode: clarify what's locked
  vfs: Fix potential circular locking through setxattr() and removexattr()
  filelock: Fix fcntl/close race recovery compat path
  fs: use all available ids
  cachefiles: Set the max subreq size for cache writes to MAX_RW_COUNT
  netfs: Fix writeback that needs to go to both server and cache
  pidfs: add selftests for new namespace ioctls
  pidfs: handle kernels without namespaces cleanly
  pidfs: when time ns disabled add check for ioctl
  vfs: correct the comments of vfs_*() helpers
  vfs: handle __wait_on_freeing_inode() and evict() race
  netfs: Rename CONFIG_FSCACHE_DEBUG to CONFIG_NETFS_DEBUG
  netfs: Revert "netfs: Switch debug logging to pr_debug()"
…/linux/kernel/git/crng/random

Pull random number generator updates from Jason Donenfeld:
 "This adds getrandom() support to the vDSO.

  First, it adds a new kind of mapping to mmap(2), MAP_DROPPABLE, which
  lets the kernel zero out pages anytime under memory pressure, which
  enables allocating memory that never gets swapped to disk but also
  doesn't count as being mlocked.

  Then, the vDSO implementation of getrandom() is introduced in a
  generic manner and hooked into random.c.

  Next, this is implemented on x86. (Also, though it's not ready for
  this pull, somebody has begun an arm64 implementation already)

  Finally, two vDSO selftests are added.

  There are also two housekeeping cleanup commits"

* tag 'random-6.11-rc1-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:
  MAINTAINERS: add random.h headers to RNG subsection
  random: note that RNDGETPOOL was removed in 2.6.9-rc2
  selftests/vDSO: add tests for vgetrandom
  x86: vdso: Wire up getrandom() vDSO implementation
  random: introduce generic vDSO getrandom() implementation
  mm: add MAP_DROPPABLE for designating always lazily freeable mappings
A "nosymfollow" flag was added in commit
dab741e ("Add a "nosymfollow" mount option.")

While we don't need to implement any special logic on
the AppArmor kernel side to handle it, we should provide
user with a correct list of mount flags in audit logs.

Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Reviewed-by: Georgia Garcia <georgia.garcia@canonical.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
Fix the 'make W=1' warning:
WARNING: modpost: missing MODULE_DESCRIPTION() in security/apparmor/apparmor_policy_unpack_test.o

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
If the label is not stale (which is the common case), the fact that the
passed file object holds a reference can be leverged to avoid the
ref/unref cycle. Doing so reduces performance impact of apparmor on
parallel open() invocations.

When benchmarking on a 24-core vm using will-it-scale's open1_process
("Separate file open"), the results are (ops/s):
before: 6092196
after:  8309726 (+36%)

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
Due to a bug in earlier userspaces, a transition table may be present
even when the dfa is not. Commit 7572fea
("apparmor: convert fperm lookup to use accept as an index") made the
verification check more rigourous regressing old userspaces with
the bug. For compatibility reasons allow the orphaned transition table
during unpack and discard.

Fixes: 7572fea ("apparmor: convert fperm lookup to use accept as an index")
Signed-off-by: Georgia Garcia <georgia.garcia@canonical.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
const qualify the struct ctl_table argument in the proc_handler function
signatures. This is a prerequisite to moving the static ctl_table
structs into .rodata data which will ensure that proc_handler function
pointers cannot be modified.

This patch has been generated by the following coccinelle script:

```
  virtual patch

  @r1@
  identifier ctl, write, buffer, lenp, ppos;
  identifier func !~ "appldata_(timer|interval)_handler|sched_(rt|rr)_handler|rds_tcp_skbuf_handler|proc_sctp_do_(hmac_alg|rto_min|rto_max|udp_port|alpha_beta|auth|probe_interval)";
  @@

  int func(
  - struct ctl_table *ctl
  + const struct ctl_table *ctl
    ,int write, void *buffer, size_t *lenp, loff_t *ppos);

  @r2@
  identifier func, ctl, write, buffer, lenp, ppos;
  @@

  int func(
  - struct ctl_table *ctl
  + const struct ctl_table *ctl
    ,int write, void *buffer, size_t *lenp, loff_t *ppos)
  { ... }

  @r3@
  identifier func;
  @@

  int func(
  - struct ctl_table *
  + const struct ctl_table *
    ,int , void *, size_t *, loff_t *);

  @r4@
  identifier func, ctl;
  @@

  int func(
  - struct ctl_table *ctl
  + const struct ctl_table *ctl
    ,int , void *, size_t *, loff_t *);

  @R5@
  identifier func, write, buffer, lenp, ppos;
  @@

  int func(
  - struct ctl_table *
  + const struct ctl_table *
    ,int write, void *buffer, size_t *lenp, loff_t *ppos);

```

* Code formatting was adjusted in xfs_sysctl.c to comply with code
  conventions. The xfs_stats_clear_proc_handler,
  xfs_panic_mask_proc_handler and xfs_deprecated_dointvec_minmax where
  adjusted.

* The ctl_table argument in proc_watchdog_common was const qualified.
  This is called from a proc_handler itself and is calling back into
  another proc_handler, making it necessary to change it as part of the
  proc_handler migration.

Co-developed-by: Thomas Weißschuh <linux@weissschuh.net>
Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Co-developed-by: Joel Granados <j.granados@samsung.com>
Signed-off-by: Joel Granados <j.granados@samsung.com>
@plbossart
Copy link
Member

@bardliao do you mind cancelling the tests and applying the fix from Takashi?

One snd_power_unref() was forgotten and left at __snd_ctl_elem_info()
in the previous change for reorganizing the locking order.

Fixes: fcc62b1 ("ALSA: control: Take power_ref lock primarily")
Link: thesofproject#5127
Signed-off-by: Takashi Iwai <tiwai@suse.de>
@bardliao bardliao force-pushed the merge/sound-upstream-20240730 branch from 57eb1d6 to a9151e4 Compare August 1, 2024 07:28
@bardliao
Copy link
Collaborator Author

bardliao commented Aug 1, 2024

@bardliao do you mind cancelling the tests and applying the fix from Takashi?

Yes, pushed the fixed patch.

@plbossart
Copy link
Member

LGTM @bardliao we have the known issue with the Xe driver but other than that no new issues reported. We should merge this before the week-end.

@bardliao
Copy link
Collaborator Author

bardliao commented Aug 2, 2024

Let's merge and check the daily test result.

@bardliao bardliao merged commit baba2a0 into thesofproject:topic/sof-dev Aug 2, 2024
@marc-hb
Copy link
Collaborator

marc-hb commented Aug 3, 2024

Let's merge and check the daily test result.

There was no need to wait for daily test results, the PR test results were already all broken:

[    5.365011] kernel: sysctl table check failed: dev/xe/(null) procname is null
[    5.365030] kernel: sysctl table check failed: dev/xe/(null) No proc_handler

@bardliao
Copy link
Collaborator Author

bardliao commented Aug 3, 2024

Let's merge and check the daily test result.

There was no need to wait for daily test results, the PR test results were already all broken:

[    5.365011] kernel: sysctl table check failed: dev/xe/(null) procname is null
[    5.365030] kernel: sysctl table check failed: dev/xe/(null) No proc_handler

I believe that is a xe driver issue and not relate to audio.

@marc-hb
Copy link
Collaborator

marc-hb commented Aug 5, 2024

I believe that is a xe driver issue and not relate to audio.

Enabling xe is one of our current objectives: for audio.

Even if it were unrelated to audio, it's generally better to ignore the error first because now every PR is red.

@kv2019i
Copy link
Collaborator

kv2019i commented Aug 5, 2024

@bardliao @marc-hb xe issues tracked at https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/2419

@msatwood
Copy link

msatwood commented Aug 5, 2024

from the dmesg of tgl, i915 is doing the guc/huc load so xe isnt even loaded. Did some paths in the SOF driver hardcode xe dependencies?

@marc-hb
Copy link
Collaborator

marc-hb commented Aug 5, 2024

from the dmesg of tgl, i915 is doing the guc/huc load so xe isnt even loaded.

xe may not run and not print anything (not a big surprise on TGL and MTL) but it's definitely loaded! I don't think it used to?

root@jf-mtlp-rvp-nocodec-3:~# journalctl -k -p 3
Aug 05 17:03:36 jf-mtlp-rvp-nocodec-3 kernel: i8042: Can't read CTR while initializing i8042
Aug 05 17:03:36 jf-mtlp-rvp-nocodec-3 kernel: i8042 i8042: probe with driver i8042 failed with error -5
Aug 05 17:03:37 jf-mtlp-rvp-nocodec-3 kernel: sysctl table check failed: dev/xe/(null) procname is null
Aug 05 17:03:37 jf-mtlp-rvp-nocodec-3 kernel: sysctl table check failed: dev/xe/(null) No proc_handler

root@jf-mtlp-rvp-nocodec-3:~# lsmod | grep xe
xe                   2940928  0
drm_ttm_helper         16384  1 xe
gpu_sched              65536  1 xe
drm_suballoc_helper    12288  1 xe
drm_gpuvm              45056  1 xe
drm_exec               12288  2 drm_gpuvm,xe
i2c_algo_bit           12288  2 xe,i915
drm_buddy              20480  2 xe,i915
ttm                   102400  3 drm_ttm_helper,xe,i915
drm_display_helper    204800  2 xe,i915
drm_kms_helper        241664  4 drm_display_helper,drm_ttm_helper,xe,i915
video                  73728  2 xe,i915
drm                   741376  15 gpu_sched,drm_kms_helper,drm_exec,drm_gpuvm,drm_suballoc_helper,drm_display_helper,drm_buddy,drm_ttm_helper,xe,i915,ttm

@marc-hb
Copy link
Collaborator

marc-hb commented Aug 5, 2024

Fix submitted: https://patchwork.freedesktop.org/series/136874/

drm/xe/observation: Drop empty sysctl table entry

An empty sysctl table entry was inadvertently left behind for observation
sysctl. The breaks on 6.11 with the following errors:

--- a/drivers/gpu/drm/xe/xe_observation.c
+++ b/drivers/gpu/drm/xe/xe_observation.c
@@ -66,7 +66,6 @@  static struct ctl_table observation_ctl_table[] = {
 	 .extra1 = SYSCTL_ZERO,
 	 .extra2 = SYSCTL_ONE,
 	 },
-	{}
 };
 
 /**

@marc-hb
Copy link
Collaborator

marc-hb commented Aug 5, 2024

Fix successfully tested in cherry-pick #5133

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.