* [PATCH 7.2 01/10] nfsd: fix all-zero stateid in CB_OFFLOAD for async COPY
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 02/10] dm-integrity: fix NULL pointer dereference when the R flag is used Greg Kroah-Hartman
` (10 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable; +Cc: Greg Kroah-Hartman, patches, Lubos Rendek, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lubos Rendek <l.rendek@proxmox.com>
[ There is no upstream commit: this fixes the 7.2.y backport of
d0beaee498e1. Mainline fixed the problem as part of the async COPY
rework in commit f307b4f7ed17 ("nfsd: split nfsd4_copy into transient
and durable async copy objects"), which is too risky to backport. ]
Since commit 14b978e8d05c ("nfsd: fix stale s2s_cp_stateids IDR entry
for async COPY"), nfsd4_copy() calls dup_copy_fields() before the copy
stateid is created and then writes the new stateid only into the COPY
result. The async copy's own cp_res.cb_stateid stays zero, so
nfsd4_send_cb_offload() sends CB_OFFLOAD with an all-zero stateid. The
client cannot match it to its pending COPY and waits forever in
nfs42_proc_copy().
Also copy the new stateid into async_copy->cp_res.cb_stateid. Calling
dup_copy_fields() after nfs4_init_copy_state() instead is not an
option, because dup_copy_fields() also copies cp_stateid and would
overwrite the stateid that the backport moved to the async copy.
Tested on 7.2.10 with an NFSv4.2 server-side COPY between two
exports: with this change the copy completes and CB_OFFLOAD carries
the same stateid as the COPY reply. Without it, cp(1) hangs on 7.2.4
and 7.2.9; the affected code in nfsd4_copy() is the same in 7.2.10.
Fixes: 14b978e8d05c ("nfsd: fix stale s2s_cp_stateids IDR entry for async COPY")
Closes: https://lore.kernel.org/stable/7cc824c4-632d-49df-ae5e-cfcbb5bf76ad@proxmox.com/
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Lubos Rendek <l.rendek@proxmox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/nfsd/nfs4proc.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 939749709e416..7c29557962b0e 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -2243,6 +2243,8 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
goto out_dec_async_copy_err;
memcpy(&result->cb_stateid, &async_copy->cp_stateid.cs_stid,
sizeof(result->cb_stateid));
+ memcpy(&async_copy->cp_res.cb_stateid, &async_copy->cp_stateid.cs_stid,
+ sizeof(async_copy->cp_res.cb_stateid));
if ((READ_ONCE(copy->nf_dst->nf_file->f_mode) &
FMODE_NOCMTIME) != 0)
async_copy->attr_update = true;
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 02/10] dm-integrity: fix NULL pointer dereference when the R flag is used
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 01/10] nfsd: fix all-zero stateid in CB_OFFLOAD for async COPY Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 03/10] dm-integrity: validate the superblock on resume Greg Kroah-Hartman
` (9 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable; +Cc: Greg Kroah-Hartman, patches, Mikulas Patocka, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mikulas Patocka <mpatocka@redhat.com>
[ Upstream commit 7d4d4f3b668d708d94f62ecdd33ac330a6fd8a84 ]
If the dm-integrity device has the SB_FLAG_DIRTY_BITMAP flag set and the
user activates the device in the 'R' mode, a crash in dm_integrity_resume
happens because the function attempts to read the journal containing the
bitmap.
This patch makes dm-integrity skip any writes to the device in
dm_integrity_resume if the device is activated in the 'R' mode.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Fixes: 468dfca38b1a ("dm integrity: add a bitmap mode")
Cc: stable@vger.kernel.org
Stable-dep-of: fc79aeb86a2c ("dm-integrity: validate the superblock on resume")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/md/dm-integrity.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c
index 4bda7b5dba1a6..cded0607b39c0 100644
--- a/drivers/md/dm-integrity.c
+++ b/drivers/md/dm-integrity.c
@@ -3874,6 +3874,10 @@ static void dm_integrity_resume(struct dm_target *ti)
r = sync_rw_sb(ic, REQ_OP_READ);
if (r)
dm_integrity_io_error(ic, "reading superblock", r);
+
+ if (ic->mode == 'R')
+ goto skip_writes;
+
if ((ic->sb->flags & flags) != flags) {
ic->sb->flags |= flags;
r = sync_rw_sb(ic, REQ_OP_WRITE | REQ_FUA);
@@ -3983,6 +3987,7 @@ static void dm_integrity_resume(struct dm_target *ti)
}
}
+skip_writes:
ic->reboot_notifier.notifier_call = dm_integrity_reboot;
ic->reboot_notifier.next = NULL;
ic->reboot_notifier.priority = INT_MAX - 1; /* be notified after md and before hardware drivers */
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 03/10] dm-integrity: validate the superblock on resume
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 01/10] nfsd: fix all-zero stateid in CB_OFFLOAD for async COPY Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 02/10] dm-integrity: fix NULL pointer dereference when the R flag is used Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 04/10] sched_ext: Fix CPU hotplug hang when a dying CPUs tasks sit in the BPF scheduler Greg Kroah-Hartman
` (8 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, syzbot+675c91651049ad042c5f,
Mikulas Patocka, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mikulas Patocka <mpatocka@redhat.com>
[ Upstream commit fc79aeb86a2c244c1f1894e2d8f2966fe2b60fa2 ]
dm_integrity_resume() re-reads the superblock from the device so that it
picks up the flags and the recalculate position. It performs no
validation on the result, while the constructor validates the superblock
it reads and sizes all the in-memory structures according to it. The user
may modify the superblock on the underlying device while the dm-integrity
device is suspended, so that the two no longer agree.
In particular, access_journal_data() shifts the journal entry index by
ic->sb->log2_sectors_per_block, while ic->journal_section_sectors and the
journal page list were computed with the value that was present at
constructor time. Increasing log2_sectors_per_block makes the index run
past the end of the journal, and dm-integrity then writes 512 bytes
through lowmem_page_address(NULL).
Snapshot the validated superblock in the constructor and refuse to resume
if any of the fields that describe the on-disk geometry changed.
Reported-by: syzbot+675c91651049ad042c5f@syzkaller.appspotmail.com
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org
Fixes: 118ba36e446c ("dm-integrity: fix recalculation in bitmap mode")
Assisted-by: Claude:claude-opus-5
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/md/dm-integrity.c | 50 +++++++++++++++++++++++++++++++++++++++
1 file changed, 50 insertions(+)
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c
index cded0607b39c0..3b36f4b8ec67b 100644
--- a/drivers/md/dm-integrity.c
+++ b/drivers/md/dm-integrity.c
@@ -180,6 +180,7 @@ struct dm_integrity_c {
struct dm_bufio_client *bufio;
struct workqueue_struct *metadata_wq;
struct superblock *sb;
+ struct superblock *sb_copy;
unsigned int journal_pages;
unsigned int n_bitmap_blocks;
@@ -3857,6 +3858,35 @@ static void dm_integrity_postsuspend(struct dm_target *ti)
ic->journal_uptodate = true;
}
+/*
+ * The superblock is re-read from the device on every resume, so that we pick
+ * up the flags and the recalculate position. The geometry described by the
+ * superblock must not change though - the in-memory structures (and the
+ * journal in particular) were sized according to the superblock that was
+ * validated in the constructor. Reject a superblock that was modified behind
+ * our back.
+ *
+ * Only the fields that the driver never rewrites may be tested here. In
+ * particular, "version" is recalculated by sb_set_version on every superblock
+ * write and it depends on SB_FLAG_RECALCULATING and SB_FLAG_DIRTY_BITMAP, and
+ * SB_FLAG_DISCARD_KEYED may be set by dm_integrity_resume itself.
+ */
+static bool superblock_changed(struct dm_integrity_c *ic)
+{
+ const __le32 immutable_flags = cpu_to_le32(SB_FLAG_HAVE_JOURNAL_MAC |
+ SB_FLAG_FIXED_PADDING |
+ SB_FLAG_FIXED_HMAC |
+ SB_FLAG_INLINE);
+
+ return memcmp(ic->sb->magic, ic->sb_copy->magic, sizeof(ic->sb->magic)) != 0 ||
+ ic->sb->log2_interleave_sectors != ic->sb_copy->log2_interleave_sectors ||
+ ic->sb->integrity_tag_size != ic->sb_copy->integrity_tag_size ||
+ ic->sb->journal_sections != ic->sb_copy->journal_sections ||
+ ic->sb->log2_sectors_per_block != ic->sb_copy->log2_sectors_per_block ||
+ ((ic->sb->flags ^ ic->sb_copy->flags) & immutable_flags) != 0 ||
+ memcmp(ic->sb->salt, ic->sb_copy->salt, SALT_SIZE) != 0;
+}
+
static void dm_integrity_resume(struct dm_target *ti)
{
struct dm_integrity_c *ic = ti->private;
@@ -3875,6 +3905,18 @@ static void dm_integrity_resume(struct dm_target *ti)
if (r)
dm_integrity_io_error(ic, "reading superblock", r);
+ if (unlikely(superblock_changed(ic))) {
+ /*
+ * Restore the superblock that we validated in the constructor,
+ * so that the rest of the driver doesn't operate on values
+ * that don't match the in-memory structures.
+ */
+ memcpy(ic->sb, ic->sb_copy, sizeof(struct superblock));
+ DMERR("The superblock was changed while the device was suspended");
+ dm_integrity_io_error(ic, "superblock check", -EINVAL);
+ goto skip_writes;
+ }
+
if (ic->mode == 'R')
goto skip_writes;
@@ -5421,6 +5463,13 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned int argc, char **argv
ic->just_formatted = true;
}
+ ic->sb_copy = kmemdup(ic->sb, sizeof(struct superblock), GFP_KERNEL);
+ if (!ic->sb_copy) {
+ ti->error = "Cannot allocate superblock copy";
+ r = -ENOMEM;
+ goto bad;
+ }
+
if (!ic->meta_dev && ic->mode != 'I') {
r = dm_set_target_max_io_len(ti, 1U << ic->sb->log2_interleave_sectors);
if (r)
@@ -5519,6 +5568,7 @@ static void dm_integrity_dtr(struct dm_target *ti)
kvfree(ic->journal_tree);
if (ic->sb)
free_pages_exact(ic->sb, SB_SECTORS << SECTOR_SHIFT);
+ kfree(ic->sb_copy);
if (ic->internal_shash)
crypto_free_shash(ic->internal_shash);
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 04/10] sched_ext: Fix CPU hotplug hang when a dying CPUs tasks sit in the BPF scheduler
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (2 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 03/10] dm-integrity: validate the superblock on resume Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 05/10] nullfs: refuse file locks Greg Kroah-Hartman
` (7 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Alap Mohan, Joonwoo Park, Tejun Heo,
Andrea Righi, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tejun Heo <tj@kernel.org>
[ Upstream commit d35a535d3e3e71f92a051d94e415f43612c4edfc ]
A CPU going down has to empty its own rq. Its hotplug thread waits in
sched_cpu_wait_empty() until nothing else is left, and the only thing that
wakes it is balance_push(), which runs from __schedule() on the dying CPU
and pushes off the migratable tasks the CPU picks.
sched_ext breaks this. A task sitting on a user DSQ or held by the BPF
scheduler is still counted on its rq, but an inactive CPU no longer calls
ops.dispatch() and can't pull the task back. The dying CPU goes idle with
the task still counted, and one of two things happens:
- Another CPU consumes the task. The rq empties without a __schedule() on
the dying CPU, so the hotplug thread is never woken and cpu_down() hangs
holding cpu_hotplug_lock.
- The task is affine only to the dying CPU. Nothing can move it, and the
offline stalls until the watchdog ejects the BPF scheduler.
When the rq goes offline, re-enqueue every task on it that isn't already on
the local DSQ. An enqueue on an offline rq lands on the local DSQ, so the
dying CPU itself picks the tasks and balance_push() pushes them off, the
same as for the other sched classes. From then on no sched_ext path on
another CPU can pull a task off the rq.
ops.dispatch() currently stops as soon as the CPU goes inactive, and CPU
hotplug then waits for an RCU grace period before taking the rq offline. A
BPF-held task affine only to the dying CPU and preempted inside an RCU
read-side critical section would block that grace period, and the rq would
never go offline. Test SCX_RQ_ONLINE directly so that ops.dispatch() keeps
running until the rq goes offline. Only the dying CPU's own hotplug thread
clears the flag during teardown, and task_can_run_on_remote_rq() still keeps
other rqs' tasks off the CPU.
Fixes: f0e1a0643a59 ("sched_ext: Implement BPF extensible scheduler class")
Fixes: 991ef53a4832 ("sched_ext: Make scx_rq_online() also test cpu_active() in addition to SCX_RQ_ONLINE")
Cc: stable@vger.kernel.org # v6.12+
Reported-by: Alap Mohan <mohaalap@meta.com>
Reported-by: Joonwoo Park <joonwoo@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
[ sashal: Omitted rescue-related changes because the rescue machinery is absent. ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
kernel/sched/ext/ext.c | 29 ++++++++++++++++++++++++++---
kernel/sched/sched.h | 5 +++--
2 files changed, 29 insertions(+), 5 deletions(-)
diff --git a/kernel/sched/ext/ext.c b/kernel/sched/ext/ext.c
index a9f33e2b17ace..8d4992e7e0120 100644
--- a/kernel/sched/ext/ext.c
+++ b/kernel/sched/ext/ext.c
@@ -2031,8 +2031,8 @@ static void set_task_runnable(struct rq *rq, struct task_struct *p)
}
/*
- * list_add_tail() must be used. scx_bypass() depends on tasks being
- * appended to the runnable_list.
+ * list_add_tail() must be used. scx_bypass() and rq_offline_scx()
+ * depend on tasks being appended to the runnable_list.
*/
list_add_tail(&p->scx.runnable_node, &rq->scx.runnable_list);
}
@@ -2868,7 +2868,13 @@ scx_dispatch_sched(struct scx_sched *sch, struct rq *rq,
#endif /* CONFIG_EXT_SUB_SCHED */
}
- if (unlikely(!SCX_HAS_OP(sch, dispatch)) || !scx_rq_online(rq))
+ /*
+ * scx_rq_online() can't be used. Its cpu_active() test goes false
+ * before CPU hotplug waits for an RCU grace period, and
+ * rq_offline_scx() moves this CPU's tasks to the local DSQ only after
+ * the wait. The grace period can depend on those tasks running.
+ */
+ if (unlikely(!SCX_HAS_OP(sch, dispatch)) || !(rq->scx.flags & SCX_RQ_ONLINE))
return SCX_DSP_NONE;
dspc->rq = rq;
@@ -3608,7 +3614,24 @@ static void rq_online_scx(struct rq *rq)
static void rq_offline_scx(struct rq *rq)
{
+ struct task_struct *p, *n;
+
rq->scx.flags &= ~SCX_RQ_ONLINE;
+
+ /* sched domain rebuilds call rq_offline with the CPU staying alive */
+ if (cpu_active(cpu_of(rq)))
+ return;
+
+ /*
+ * An offline CPU no longer calls ops.dispatch(). Re-enqueue its tasks
+ * onto the local DSQ so that they run here and balance_push() moves
+ * them off.
+ */
+ list_for_each_entry_safe_reverse(p, n, &rq->scx.runnable_list, scx.runnable_node) {
+ if (p->scx.dsq == &rq->scx.local_dsq)
+ continue;
+ guard(sched_change)(p, DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK);
+ }
}
static bool check_rq_for_timeouts(struct rq *rq)
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index a6c6f8165eecc..d25defb116caa 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -4185,8 +4185,9 @@ extern void balance_callbacks(struct rq *rq, struct balance_callback *head);
* after which it is enqueued again.
*
* Typically this must be called while holding task_rq_lock, since most/all
- * properties are serialized under those locks. There is currently one
- * exception to this rule in sched/ext which only holds rq->lock.
+ * properties are serialized under those locks. There are currently two
+ * exceptions to this rule in sched/ext which only hold rq->lock: scx_bypass()
+ * and rq_offline_scx().
*/
/*
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 05/10] nullfs: refuse file locks
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (3 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 04/10] sched_ext: Fix CPU hotplug hang when a dying CPUs tasks sit in the BPF scheduler Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 06/10] readdir: take no inode lock on an immutable directory Greg Kroah-Hartman
` (6 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Christian Brauner (Amutable),
Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <brauner@kernel.org>
[ Upstream commit 31a1fb90624d7e275db775a29a0e00a7f8b3df63 ]
Refuse flock() and POSIX locks on nullfs. Its one inode is the root of
every kernel thread and the following patches make it the directory
that stands in for an unmounted mount, so a lock taken through one such
directory would block the locks of every other holder and F_GETLK would
tell them the pid of the holder. Give the directory file operations of
its own: what libfs gives an empty directory plus ->lock and ->flock
that fail with ENOLCK, the way a filesystem without lock support does.
Link: https://patch.msgid.link/20261002-work-mount-fixes-4-v1-17-dd44b89d44ce@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Stable-dep-of: e08f57a7dcb8 ("readdir: take no inode lock on an immutable directory")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/nullfs.c | 29 +++++++++++++++++++++++++++++
1 file changed, 29 insertions(+)
diff --git a/fs/nullfs.c b/fs/nullfs.c
index 20d3aa627a47d..fc79ca42bb178 100644
--- a/fs/nullfs.c
+++ b/fs/nullfs.c
@@ -8,6 +8,34 @@ static const struct super_operations nullfs_super_operations = {
.statfs = simple_statfs,
};
+static loff_t nullfs_dir_llseek(struct file *file, loff_t offset, int whence)
+{
+ /* an empty directory has two entries . and .. at offsets 0 and 1 */
+ return generic_file_llseek_size(file, offset, whence, 2, 2);
+}
+
+static int nullfs_dir_readdir(struct file *file, struct dir_context *ctx)
+{
+ dir_emit_dots(file, ctx);
+ return 0;
+}
+
+/* the one inode of nullfs is shared by every holder, so no locks on it */
+static int nullfs_nolock(struct file *file, int cmd, struct file_lock *fl)
+{
+ return -ENOLCK;
+}
+
+/* what libfs gives an empty directory, plus the refusal of file locks */
+static const struct file_operations nullfs_dir_operations = {
+ .llseek = nullfs_dir_llseek,
+ .read = generic_read_dir,
+ .iterate_shared = nullfs_dir_readdir,
+ .fsync = noop_fsync,
+ .lock = nullfs_nolock,
+ .flock = nullfs_nolock,
+};
+
static int nullfs_fs_fill_super(struct super_block *s, struct fs_context *fc)
{
struct inode *inode;
@@ -28,6 +56,7 @@ static int nullfs_fs_fill_super(struct super_block *s, struct fs_context *fc)
/* nullfs is permanently empty... */
make_empty_dir_inode(inode);
+ inode->i_fop = &nullfs_dir_operations;
simple_inode_init_ts(inode);
inode->i_ino = 1;
/* ... and immutable, reading it leaves no trace either. */
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 06/10] readdir: take no inode lock on an immutable directory
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (4 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 05/10] nullfs: refuse file locks Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 07/10] net: mana: refactor mana_get_strings() and mana_get_sset_count() to use switch Greg Kroah-Hartman
` (5 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Christian Brauner (Amutable),
Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christian Brauner <brauner@kernel.org>
[ Upstream commit e08f57a7dcb83818c058e415988c23beb6c5a47f ]
iterate_dir() takes the directory's i_rwsem shared and holds it across
->iterate_shared(). For a directory that never has an entry and is
never removed the lock keeps nothing still, it only orders every reader
and every writer of that inode behind each other.
For the directory of a nullfs instance that matters. The instance of
the initial mount namespace is the root of every empty mount namespace
and the private instance is the root of every kernel thread, so one
inode is shared across users who have nothing else in common. And a
reader can hold the lock for as long as it likes: back the getdents()
buffer with a mapping of a file on a FUSE mount of your own, let the
copy of "." and ".." fault and let the server wait. Queue an exclusive
taker behind it, a mkdir() in that directory goes through start_dirop()
before the read-only mount is reported, and from then on every lookup
that misses the dcache in that directory, every create and every mount
on it waits until the server answers. One user of an empty mount
namespace stalls all the others.
Add FOP_IMMUTABLE for the file operations of a directory that never
changes and is never removed and let iterate_dir() skip the lock for
it. The flag never changes for a file, ->f_pos is protected by
f_pos_lock since directories are FMODE_ATOMIC_POS, IS_DEADDIR can't be
set on such a directory and neither touch_atime() nor fsnotify take
i_rwsem. Set it on the nullfs directory. The placeholder directories of
libfs never have an entry either but their owners remove them, so they
keep the lock.
Fixes: 9d4e752a24f7 ("namespace: allow creating empty mount namespaces")
Cc: stable@vger.kernel.org # v7.1+
Link: https://patch.msgid.link/20261002-work-mount-fixes-4-v1-19-dd44b89d44ce@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/nullfs.c | 1 +
fs/readdir.c | 13 +++++++++----
include/linux/fs.h | 2 ++
3 files changed, 12 insertions(+), 4 deletions(-)
diff --git a/fs/nullfs.c b/fs/nullfs.c
index fc79ca42bb178..c89181daf208c 100644
--- a/fs/nullfs.c
+++ b/fs/nullfs.c
@@ -34,6 +34,7 @@ static const struct file_operations nullfs_dir_operations = {
.fsync = noop_fsync,
.lock = nullfs_nolock,
.flock = nullfs_nolock,
+ .fop_flags = FOP_IMMUTABLE,
};
static int nullfs_fs_fill_super(struct super_block *s, struct fs_context *fc)
diff --git a/fs/readdir.c b/fs/readdir.c
index 76bb1ae3a4504..f2288841337ad 100644
--- a/fs/readdir.c
+++ b/fs/readdir.c
@@ -87,6 +87,8 @@ EXPORT_SYMBOL(wrap_directory_iterator);
int iterate_dir(struct file *file, struct dir_context *ctx)
{
struct inode *inode = file_inode(file);
+ /* never an entry, never removed: nothing for the lock to keep still */
+ bool locked = !(file->f_op->fop_flags & FOP_IMMUTABLE);
int res = -ENOTDIR;
if (!file->f_op->iterate_shared)
@@ -100,9 +102,11 @@ int iterate_dir(struct file *file, struct dir_context *ctx)
if (res)
goto out;
- res = down_read_killable(&inode->i_rwsem);
- if (res)
- goto out;
+ if (locked) {
+ res = down_read_killable(&inode->i_rwsem);
+ if (res)
+ goto out;
+ }
res = -ENOENT;
if (!IS_DEADDIR(inode)) {
@@ -112,7 +116,8 @@ int iterate_dir(struct file *file, struct dir_context *ctx)
fsnotify_access(file);
file_accessed(file);
}
- inode_unlock_shared(inode);
+ if (locked)
+ inode_unlock_shared(inode);
out:
return res;
}
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 09de7bf6f1d28..a07031fd4ae24 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1980,6 +1980,8 @@ struct file_operations {
#define FOP_ASYNC_LOCK ((__force fop_flags_t)(1 << 6))
/* File system supports uncached read/write buffered IO */
#define FOP_DONTCACHE ((__force fop_flags_t)(1 << 7))
+/* Never changes and is never removed, readdir of a directory takes no lock */
+#define FOP_IMMUTABLE ((__force fop_flags_t)(1 << 8))
/* Wrap a directory iterator that needs exclusive inode access */
int wrap_directory_iterator(struct file *, struct dir_context *,
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 07/10] net: mana: refactor mana_get_strings() and mana_get_sset_count() to use switch
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (5 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 06/10] readdir: take no inode lock on an immutable directory Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 08/10] net: mana: reserve RX buffer headroom to fix forwarding performance Greg Kroah-Hartman
` (4 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Haiyang Zhang, Dipayaan Roy,
Jakub Kicinski, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dipayaan Roy <dipayanroy@linux.microsoft.com>
[ Upstream commit d52f80731055ee43b36bf60088679aca97a484da ]
Refactor mana_get_strings() and mana_get_sset_count() from if/else to
switch statements in preparation for adding ethtool private flags
support which requires handling ETH_SS_PRIV_FLAGS.
No functional change.
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Dipayaan Roy <dipayanroy@linux.microsoft.com>
Link: https://patch.msgid.link/20260729063347.3388035-2-dipayanroy@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 259c4a519100 ("net: mana: reserve RX buffer headroom to fix forwarding performance")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
.../ethernet/microsoft/mana/mana_ethtool.c | 75 ++++++++++++-------
1 file changed, 46 insertions(+), 29 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
index 881df597d7f9c..3531abfdaf433 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
@@ -138,53 +138,70 @@ static int mana_get_sset_count(struct net_device *ndev, int stringset)
struct mana_port_context *apc = netdev_priv(ndev);
unsigned int num_queues = apc->num_queues;
- if (stringset != ETH_SS_STATS)
+ switch (stringset) {
+ case ETH_SS_STATS:
+ return ARRAY_SIZE(mana_eth_stats) +
+ ARRAY_SIZE(mana_phy_stats) +
+ ARRAY_SIZE(mana_hc_stats) +
+ num_queues * (MANA_STATS_RX_COUNT + MANA_STATS_TX_COUNT);
+ default:
return -EINVAL;
-
- return ARRAY_SIZE(mana_eth_stats) + ARRAY_SIZE(mana_phy_stats) + ARRAY_SIZE(mana_hc_stats) +
- num_queues * (MANA_STATS_RX_COUNT + MANA_STATS_TX_COUNT);
+ }
}
-static void mana_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
+static void mana_get_strings_stats(struct mana_port_context *apc, u8 **data)
{
- struct mana_port_context *apc = netdev_priv(ndev);
unsigned int num_queues = apc->num_queues;
int i, j;
- if (stringset != ETH_SS_STATS)
- return;
for (i = 0; i < ARRAY_SIZE(mana_eth_stats); i++)
- ethtool_puts(&data, mana_eth_stats[i].name);
+ ethtool_puts(data, mana_eth_stats[i].name);
for (i = 0; i < ARRAY_SIZE(mana_hc_stats); i++)
- ethtool_puts(&data, mana_hc_stats[i].name);
+ ethtool_puts(data, mana_hc_stats[i].name);
for (i = 0; i < ARRAY_SIZE(mana_phy_stats); i++)
- ethtool_puts(&data, mana_phy_stats[i].name);
+ ethtool_puts(data, mana_phy_stats[i].name);
for (i = 0; i < num_queues; i++) {
- ethtool_sprintf(&data, "rx_%d_packets", i);
- ethtool_sprintf(&data, "rx_%d_bytes", i);
- ethtool_sprintf(&data, "rx_%d_xdp_drop", i);
- ethtool_sprintf(&data, "rx_%d_xdp_tx", i);
- ethtool_sprintf(&data, "rx_%d_xdp_redirect", i);
- ethtool_sprintf(&data, "rx_%d_pkt_len0_err", i);
+ ethtool_sprintf(data, "rx_%d_packets", i);
+ ethtool_sprintf(data, "rx_%d_bytes", i);
+ ethtool_sprintf(data, "rx_%d_xdp_drop", i);
+ ethtool_sprintf(data, "rx_%d_xdp_tx", i);
+ ethtool_sprintf(data, "rx_%d_xdp_redirect", i);
+ ethtool_sprintf(data, "rx_%d_pkt_len0_err", i);
for (j = 0; j < MANA_RXCOMP_OOB_NUM_PPI - 1; j++)
- ethtool_sprintf(&data, "rx_%d_coalesced_cqe_%d", i, j + 2);
+ ethtool_sprintf(data,
+ "rx_%d_coalesced_cqe_%d",
+ i,
+ j + 2);
}
for (i = 0; i < num_queues; i++) {
- ethtool_sprintf(&data, "tx_%d_packets", i);
- ethtool_sprintf(&data, "tx_%d_bytes", i);
- ethtool_sprintf(&data, "tx_%d_xdp_xmit", i);
- ethtool_sprintf(&data, "tx_%d_tso_packets", i);
- ethtool_sprintf(&data, "tx_%d_tso_bytes", i);
- ethtool_sprintf(&data, "tx_%d_tso_inner_packets", i);
- ethtool_sprintf(&data, "tx_%d_tso_inner_bytes", i);
- ethtool_sprintf(&data, "tx_%d_long_pkt_fmt", i);
- ethtool_sprintf(&data, "tx_%d_short_pkt_fmt", i);
- ethtool_sprintf(&data, "tx_%d_csum_partial", i);
- ethtool_sprintf(&data, "tx_%d_mana_map_err", i);
+ ethtool_sprintf(data, "tx_%d_packets", i);
+ ethtool_sprintf(data, "tx_%d_bytes", i);
+ ethtool_sprintf(data, "tx_%d_xdp_xmit", i);
+ ethtool_sprintf(data, "tx_%d_tso_packets", i);
+ ethtool_sprintf(data, "tx_%d_tso_bytes", i);
+ ethtool_sprintf(data, "tx_%d_tso_inner_packets", i);
+ ethtool_sprintf(data, "tx_%d_tso_inner_bytes", i);
+ ethtool_sprintf(data, "tx_%d_long_pkt_fmt", i);
+ ethtool_sprintf(data, "tx_%d_short_pkt_fmt", i);
+ ethtool_sprintf(data, "tx_%d_csum_partial", i);
+ ethtool_sprintf(data, "tx_%d_mana_map_err", i);
+ }
+}
+
+static void mana_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
+{
+ struct mana_port_context *apc = netdev_priv(ndev);
+
+ switch (stringset) {
+ case ETH_SS_STATS:
+ mana_get_strings_stats(apc, &data);
+ break;
+ default:
+ break;
}
}
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 08/10] net: mana: reserve RX buffer headroom to fix forwarding performance
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (6 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 07/10] net: mana: refactor mana_get_strings() and mana_get_sset_count() to use switch Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 09/10] wifi: atk9k: Revert "wifi: ath9k: reject short WMI command responses" Greg Kroah-Hartman
` (3 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Hamza Mahfooz, Simon Horman,
Jakub Kicinski, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hamza Mahfooz <hamzamahfooz@linux.microsoft.com>
[ Upstream commit 259c4a519100e9182f264f8009ee212b32774277 ]
Commit 730ff06d3f5c ("net: mana: Use page pool fragments for RX buffers
instead of full pages to improve memory efficiency.") started handing
out RX buffers with zero headroom so that two buffers fit into one page
at the default MTU.
The MANA TX path, however, stores the per scatter-gather entry DMA
mappings in `struct mana_skb_head` at skb->head, and mana_start_xmit()
therefore calls skb_cow_head(skb, MANA_HEADROOM). The port advertises
this requirement as ndev->needed_headroom = MANA_HEADROOM.
As a result every packet that is received and then forwarded out of a
MANA port fails the skb_cow() in ip_forward() and gets reallocated and
copied by pskb_expand_head(). This is invisible to a plain RX or TX
workload, but it puts a full skb reallocation plus memcpy on the hot
path of every single forwarded packet, which is exactly what a
router/NVA workload does.
Restore the headroom. Note that reserving MANA_HEADROOM (232) is not
enough: ip_forward() asks for LL_RESERVED_SPACE(dev), which rounds
hard_header_len + needed_headroom up to HH_DATA_MOD and is 256 bytes on
ethernet. Use LL_RESERVED_SPACE() directly so the value keeps tracking
both constants. Also, since LL_RESERVED_SPACE() tracks MANA_HEADROOM,
it grows with MAX_SKB_FRAGS and for MAX_SKB_FRAGS >= 19 it is greater
than 256, so we have to account for that by using the headroom the RX
queue actually uses (instead of assuming XDP_PACKET_HEADROOM) and
turning MANA_XDP_MTU_MAX into MANA_XDP_MTU_MAX(ndev) (note that at the
default CONFIG_MAX_SKB_FRAGS=17 they are equivalent).
At the default MTU on a 4K page this means a buffer no longer fits twice
into a page (SKB_DATA_ALIGN(1500 + MANA_RXBUF_PAD + 256) = 2112), so the
frag-vs-single decision is now made by computing the real buffer size
instead of comparing the MTU against PAGE_SIZE / 2. The page_pool
fragment path is still used wherever at least two buffers genuinely fit,
e.g. on 16K and 64K page sizes.
Measured on an Azure VM with a MANA NIC acting as a forwarding NVA (UDP,
1400 byte payload, 4 streams, 8 Gbps offered, only the forwarding
node's kernel differs), 8 runs each, median:
forwarded pps throughput
before 272,830 3.06 Gbps
after 390,560 4.37 Gbps (+43%)
perf on the forwarding node, same workload:
memset_orig __pi_memcpy pskb_expand_head
before 10.07% 3.96% present
after 0.94% 0.64% gone
Cc: stable@vger.kernel.org
Fixes: 730ff06d3f5c ("net: mana: Use page pool fragments for RX buffers instead of full pages to improve memory efficiency.")
Signed-off-by: Hamza Mahfooz <hamzamahfooz@linux.microsoft.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20261003013647.2051416-1-hamzamahfooz@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ sashal: called mana_single_rxbuf_per_page_forced() directly to accommodate missing full-page-rx support. ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
.../net/ethernet/microsoft/mana/mana_bpf.c | 8 +--
drivers/net/ethernet/microsoft/mana/mana_en.c | 60 ++++++++++++++-----
include/net/mana/mana.h | 14 ++++-
3 files changed, 63 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_bpf.c b/drivers/net/ethernet/microsoft/mana/mana_bpf.c
index 5c9961ee9747a..b27fd9b8c4f6d 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_bpf.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_bpf.c
@@ -91,7 +91,7 @@ u32 mana_run_xdp(struct net_device *ndev, struct mana_rxq *rxq,
goto out;
xdp_init_buff(xdp, PAGE_SIZE, &rxq->xdp_rxq);
- xdp_prepare_buff(xdp, buf_va, XDP_PACKET_HEADROOM, pkt_len, true);
+ xdp_prepare_buff(xdp, buf_va, rxq->headroom, pkt_len, true);
act = bpf_prog_run_xdp(prog, xdp);
@@ -183,9 +183,9 @@ static int mana_xdp_set(struct net_device *ndev, struct bpf_prog *prog,
if (!old_prog && !prog)
return 0;
- if (prog && ndev->mtu > MANA_XDP_MTU_MAX) {
+ if (prog && ndev->mtu > MANA_XDP_MTU_MAX(ndev)) {
netdev_err(ndev, "XDP: mtu:%u too large, mtu_max:%lu\n",
- ndev->mtu, MANA_XDP_MTU_MAX);
+ ndev->mtu, MANA_XDP_MTU_MAX(ndev));
NL_SET_ERR_MSG_MOD(extack, "XDP: mtu too large");
return -EOPNOTSUPP;
@@ -238,7 +238,7 @@ static int mana_xdp_set(struct net_device *ndev, struct bpf_prog *prog,
bpf_prog_put(old_prog);
if (prog)
- ndev->max_mtu = min_t(unsigned int, MANA_XDP_MTU_MAX,
+ ndev->max_mtu = min_t(unsigned int, MANA_XDP_MTU_MAX(ndev),
gc->adapter_mtu - ETH_HLEN);
else
ndev->max_mtu = gc->adapter_mtu - ETH_HLEN;
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 8fb29dcddf8e2..af5d4a4f56f16 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -755,26 +755,62 @@ static void *mana_get_rxbuf_pre(struct mana_rxq *rxq, dma_addr_t *da)
return va;
}
-/* Get RX buffer's data size, alloc size, XDP headroom based on MTU */
+/* Reserve enough headroom to satisfy the skb_cow() in ip_forward() and avoid
+ * reallocation: the TX path keeps the SGE DMA mappings in struct mana_skb_head
+ * at skb->head, so the port advertises ndev->needed_headroom = MANA_HEADROOM.
+ */
+static u32 mana_get_rxbuf_headroom(struct mana_port_context *apc)
+{
+ if (mana_xdp_get(apc))
+ return mana_xdp_headroom(apc->ndev);
+
+ return LL_RESERVED_SPACE(apc->ndev);
+}
+
+static u32 mana_get_rxbuf_size(struct mana_port_context *apc, u32 mtu)
+{
+ u32 len = SKB_DATA_ALIGN(mtu + MANA_RXBUF_PAD +
+ mana_get_rxbuf_headroom(apc));
+
+ return ALIGN(len, MANA_RX_FRAG_ALIGNMENT);
+}
+
+/* Returns true when one RX buffer per page is already required by XDP or by
+ * the buffer size implied by the MTU, i.e. regardless of the
+ * MANA_PRIV_FLAG_USE_FULL_PAGE_RXBUF private flag.
+ */
+bool mana_single_rxbuf_per_page_forced(struct mana_port_context *apc, u32 mtu)
+{
+ /* For xdp make sure only one packet fits per page. */
+ if (mana_xdp_get(apc))
+ return true;
+
+ /* Only use the page_pool fragment path when at least two buffers,
+ * including the headroom each of them has to reserve, actually fit
+ * into one page. Otherwise the fragment path degenerates into one
+ * buffer per page while still paying the fragment accounting cost.
+ */
+ return PAGE_SIZE / mana_get_rxbuf_size(apc, mtu) < 2;
+}
+
+/* Get RX buffer's data size, alloc size, headroom and frag count based on MTU */
static void mana_get_rxbuf_cfg(struct mana_port_context *apc,
int mtu, u32 *datasize, u32 *alloc_size,
u32 *headroom, u32 *frag_count)
{
- u32 len, buf_size;
+ u32 buf_size;
/* Calculate datasize first (consistent across all cases) */
*datasize = mtu + ETH_HLEN;
- /* For xdp and jumbo frames make sure only one packet fits per page */
- if (mtu + MANA_RXBUF_PAD > PAGE_SIZE / 2 || mana_xdp_get(apc)) {
- if (mana_xdp_get(apc)) {
- *headroom = XDP_PACKET_HEADROOM;
+ *headroom = mana_get_rxbuf_headroom(apc);
+
+ if (mana_single_rxbuf_per_page_forced(apc, mtu)) {
+ if (mana_xdp_get(apc))
*alloc_size = PAGE_SIZE;
- } else {
- *headroom = 0; /* no support for XDP */
+ else
*alloc_size = SKB_DATA_ALIGN(mtu + MANA_RXBUF_PAD +
*headroom);
- }
*frag_count = 1;
@@ -788,11 +824,7 @@ static void mana_get_rxbuf_cfg(struct mana_port_context *apc,
}
/* Standard MTU case - optimize for multiple packets per page */
- *headroom = 0;
-
- /* Calculate base buffer size needed */
- len = SKB_DATA_ALIGN(mtu + MANA_RXBUF_PAD + *headroom);
- buf_size = ALIGN(len, MANA_RX_FRAG_ALIGNMENT);
+ buf_size = mana_get_rxbuf_size(apc, mtu);
/* Calculate how many packets can fit in a page */
*frag_count = PAGE_SIZE / buf_size;
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 04acb6791dbd9..cd44856fa474c 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -328,7 +328,18 @@ struct mana_recv_buf_oob {
#define MANA_RXBUF_PAD (SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) \
+ ETH_HLEN)
-#define MANA_XDP_MTU_MAX (PAGE_SIZE - MANA_RXBUF_PAD - XDP_PACKET_HEADROOM)
+/* Headroom an RX buffer has to reserve while XDP is attached: the XDP program
+ * needs XDP_PACKET_HEADROOM, and the TX path needs LL_RESERVED_SPACE() (see
+ * mana_get_rxbuf_headroom()). LL_RESERVED_SPACE() grows with MAX_SKB_FRAGS and
+ * exceeds XDP_PACKET_HEADROOM once CONFIG_MAX_SKB_FRAGS is 19 or more.
+ */
+static inline u32 mana_xdp_headroom(struct net_device *ndev)
+{
+ return max_t(u32, LL_RESERVED_SPACE(ndev), XDP_PACKET_HEADROOM);
+}
+
+#define MANA_XDP_MTU_MAX(ndev) \
+ (PAGE_SIZE - MANA_RXBUF_PAD - mana_xdp_headroom(ndev))
struct mana_rxq {
struct gdma_queue *gdma_rq;
@@ -625,6 +636,7 @@ int mana_query_link_cfg(struct mana_port_context *apc);
int mana_set_bw_clamp(struct mana_port_context *apc, u32 speed,
int enable_clamping);
void mana_query_phy_stats(struct mana_port_context *apc);
+bool mana_single_rxbuf_per_page_forced(struct mana_port_context *apc, u32 mtu);
int mana_pre_alloc_rxbufs(struct mana_port_context *apc, int mtu, int num_queues);
void mana_pre_dealloc_rxbufs(struct mana_port_context *apc);
void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc);
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 09/10] wifi: atk9k: Revert "wifi: ath9k: reject short WMI command responses"
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (7 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 08/10] net: mana: reserve RX buffer headroom to fix forwarding performance Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 15:54 ` [PATCH 7.2 10/10] ASoC: cs35l56: Add DAI for SDCA OT25 stream Greg Kroah-Hartman
` (2 subsequent siblings)
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Mehmet Fide,
Toke Høiland-Jørgensen, Linus Torvalds, Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Toke Høiland-Jørgensen <toke@toke.dk>
[ Upstream commit 90201351a862182bdabfc1ad1c97b1e1b6ed5931 ]
This reverts commit 93ce8ea70451e800601bc9444aa33043de15d67b.
Mehmet reported that the offending commit breaks his AR9271 USB adapter
which uses zero-length replies to commands that don't return a value.
Revert the whole thing to restore the functionality.
Cc: stable@vger.kernel.org # the offending commit is in 6.12.113
Reported-by: Mehmet Fide <mehmet.fide@gmail.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/ath/ath9k/wmi.c | 16 ++--------------
drivers/net/wireless/ath/ath9k/wmi.h | 1 -
2 files changed, 2 insertions(+), 15 deletions(-)
diff --git a/drivers/net/wireless/ath/ath9k/wmi.c b/drivers/net/wireless/ath/ath9k/wmi.c
index 4a08babb6d9d6..44e91815a1594 100644
--- a/drivers/net/wireless/ath/ath9k/wmi.c
+++ b/drivers/net/wireless/ath/ath9k/wmi.c
@@ -206,16 +206,9 @@ static void ath9k_wmi_rsp_callback(struct wmi *wmi, struct sk_buff *skb)
{
skb_pull(skb, sizeof(struct wmi_cmd_hdr));
- if (wmi->cmd_rsp_buf && wmi->cmd_rsp_len) {
- if (skb->len < wmi->cmd_rsp_len) {
- wmi->cmd_rsp_status = -EMSGSIZE;
- goto complete;
- }
+ if (wmi->cmd_rsp_buf != NULL && wmi->cmd_rsp_len != 0)
memcpy(wmi->cmd_rsp_buf, skb->data, wmi->cmd_rsp_len);
- }
- wmi->cmd_rsp_status = 0;
-complete:
complete(&wmi->cmd_wait);
}
@@ -307,7 +300,6 @@ static int ath9k_wmi_cmd_issue(struct wmi *wmi,
/* record the rsp buffer and length */
wmi->cmd_rsp_buf = rsp_buf;
wmi->cmd_rsp_len = rsp_len;
- wmi->cmd_rsp_status = 0;
wmi->last_seq_id = wmi->tx_seq_id;
spin_unlock_irqrestore(&wmi->wmi_lock, flags);
@@ -364,13 +356,9 @@ int ath9k_wmi_cmd(struct wmi *wmi, enum wmi_cmd_id cmd_id,
return -ETIMEDOUT;
}
- spin_lock_irqsave(&wmi->wmi_lock, flags);
- ret = wmi->cmd_rsp_status;
- spin_unlock_irqrestore(&wmi->wmi_lock, flags);
-
mutex_unlock(&wmi->op_mutex);
- return ret;
+ return 0;
out:
ath_dbg(common, WMI, "WMI failure for: %s\n", wmi_cmd_to_name(cmd_id));
diff --git a/drivers/net/wireless/ath/ath9k/wmi.h b/drivers/net/wireless/ath/ath9k/wmi.h
index 0bb9d31f1a1cc..5c3b710b8f319 100644
--- a/drivers/net/wireless/ath/ath9k/wmi.h
+++ b/drivers/net/wireless/ath/ath9k/wmi.h
@@ -157,7 +157,6 @@ struct wmi {
u16 tx_seq_id;
u8 *cmd_rsp_buf;
u32 cmd_rsp_len;
- int cmd_rsp_status;
bool stopped;
struct list_head pending_tx_events;
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* [PATCH 7.2 10/10] ASoC: cs35l56: Add DAI for SDCA OT25 stream
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (8 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 09/10] wifi: atk9k: Revert "wifi: ath9k: reject short WMI command responses" Greg Kroah-Hartman
@ 2026-10-17 15:54 ` Greg Kroah-Hartman
2026-10-17 16:47 ` [PATCH 7.2 00/10] 7.2.12-rc1 review Ronald Warsow
2026-10-17 17:19 ` Brett A C Sheffield
11 siblings, 0 replies; 13+ messages in thread
From: Greg Kroah-Hartman @ 2026-10-17 15:54 UTC (permalink / raw)
To: stable
Cc: Greg Kroah-Hartman, patches, Richard Fitzgerald, Mark Brown,
Sasha Levin
7.2-stable review patch. If anyone has any objections, please let me know.
------------------
From: Richard Fitzgerald <rf@opensource.cirrus.com>
[ Upstream commit 86c8360c97f66fa7830fd930e2dc69374007284b ]
Add a DAI for the SDCA OT25 stream on SoundWire DP4.
The SDCA-defined OT25 stream is an amp reference feedback, typically used
for AEC. The firmware outputs this on DP4. The firmware owns all control
registers for this stream so there are no mixer controls or any other
configuration options.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://patch.msgid.link/20261005155235.1386525-2-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
include/sound/cs35l56.h | 1 +
sound/soc/codecs/cs35l56-sdw.c | 9 ++++--
sound/soc/codecs/cs35l56.c | 54 ++++++++++++++++++++++++++++++++++
3 files changed, 62 insertions(+), 2 deletions(-)
diff --git a/include/sound/cs35l56.h b/include/sound/cs35l56.h
index 2490b72c0a7a8..07434e644b89d 100644
--- a/include/sound/cs35l56.h
+++ b/include/sound/cs35l56.h
@@ -312,6 +312,7 @@ struct snd_ctl_elem_value;
#define CS35L56_SDW1_PLAYBACK_PORT 1
#define CS35L56_SDW1_CAPTURE_PORT 3
+#define CS35L56_OT25_CAPTURE_PORT 4
#define CS35L56_NUM_BULK_SUPPLIES 3
#define CS35L56_NUM_DSP_REGIONS 5
diff --git a/sound/soc/codecs/cs35l56-sdw.c b/sound/soc/codecs/cs35l56-sdw.c
index a3812ab4c0227..d52209111fe2e 100644
--- a/sound/soc/codecs/cs35l56-sdw.c
+++ b/sound/soc/codecs/cs35l56-sdw.c
@@ -295,11 +295,11 @@ static int cs35l56_sdw_read_prop(struct sdw_slave *peripheral)
if (ret == 0)
prop->clk_stop_mode1 = !!clock_stop_1;
- ports = devm_kcalloc(cs35l56->base.dev, 2, sizeof(*ports), GFP_KERNEL);
+ ports = devm_kcalloc(cs35l56->base.dev, 3, sizeof(*ports), GFP_KERNEL);
if (!ports)
return -ENOMEM;
- prop->source_ports = BIT(CS35L56_SDW1_CAPTURE_PORT);
+ prop->source_ports = BIT(CS35L56_SDW1_CAPTURE_PORT) | BIT(CS35L56_OT25_CAPTURE_PORT);
prop->sink_ports = BIT(CS35L56_SDW1_PLAYBACK_PORT);
prop->paging_support = true;
prop->quirks = SDW_SLAVE_QUIRKS_INVALID_INITIAL_PARITY;
@@ -315,6 +315,11 @@ static int cs35l56_sdw_read_prop(struct sdw_slave *peripheral)
ports[1].num = CS35L56_SDW1_CAPTURE_PORT;
ports[1].type = SDW_DPN_FULL;
ports[1].ch_prep_timeout = 10;
+
+ /* DP4 - capture */
+ ports[2].num = CS35L56_OT25_CAPTURE_PORT;
+ ports[2].type = SDW_DPN_FULL;
+ ports[2].ch_prep_timeout = 10;
prop->src_dpn_prop = &ports[1];
dev_dbg(&peripheral->dev, "clock stop mode 1 supported: %s\n",
diff --git a/sound/soc/codecs/cs35l56.c b/sound/soc/codecs/cs35l56.c
index 8beef93ac4676..79f0a39d12b96 100644
--- a/sound/soc/codecs/cs35l56.c
+++ b/sound/soc/codecs/cs35l56.c
@@ -32,6 +32,7 @@
#include <sound/cs-amp-lib.h>
#include <sound/pcm.h>
#include <sound/pcm_params.h>
+#include <sound/sdw.h>
#include <sound/soc.h>
#include <sound/soc-dapm.h>
#include <sound/tlv.h>
@@ -323,6 +324,7 @@ static const struct snd_soc_dapm_widget cs35l56_dapm_widgets[] = {
SND_SOC_DAPM_SIGGEN("VDDBMON ADC"),
SND_SOC_DAPM_SIGGEN("VBSTMON ADC"),
SND_SOC_DAPM_SIGGEN("TEMPMON ADC"),
+ SND_SOC_DAPM_SIGGEN("OT25 Reference"),
SND_SOC_DAPM_INPUT("Calibrate"),
};
@@ -387,6 +389,8 @@ static const struct snd_soc_dapm_route cs35l56_audio_map[] = {
{ "SDW1 Capture", NULL, "SDW1 TX2 Source" },
{ "SDW1 Capture", NULL, "SDW1 TX3 Source" },
{ "SDW1 Capture", NULL, "SDW1 TX4 Source" },
+
+ { "OT25", NULL, "OT25 Reference" },
};
static int cs35l56_dsp_event(struct snd_soc_dapm_widget *w,
@@ -711,6 +715,37 @@ static int cs35l56_sdw_dai_set_stream(struct snd_soc_dai *dai,
return 0;
}
+static int cs35l56_ot25_dai_hw_params(struct snd_pcm_substream *substream,
+ struct snd_pcm_hw_params *params,
+ struct snd_soc_dai *dai)
+{
+ struct cs35l56_private *cs35l56 = snd_soc_component_get_drvdata(dai->component);
+ struct sdw_stream_runtime *sdw_stream = snd_soc_dai_get_dma_data(dai, substream);
+ struct sdw_stream_config sconfig = { };
+ struct sdw_port_config pconfig = { };
+ int ret;
+
+ dev_dbg(cs35l56->base.dev, "%s: rate %d\n", __func__, params_rate(params));
+
+ if (!cs35l56->base.init_done)
+ return -ENODEV;
+
+ if (!sdw_stream)
+ return -EINVAL;
+
+ snd_sdw_params_to_config(substream, params, &sconfig, &pconfig);
+ pconfig.num = CS35L56_OT25_CAPTURE_PORT;
+
+ ret = sdw_stream_add_slave(cs35l56->sdw_peripheral, &sconfig, &pconfig,
+ 1, sdw_stream);
+ if (ret) {
+ dev_err(dai->dev, "Failed to add OT25 stream: %d\n", ret);
+ return ret;
+ }
+
+ return 0;
+}
+
static const struct snd_soc_dai_ops cs35l56_sdw_dai_ops = {
.set_tdm_slot = cs35l56_sdw_dai_set_tdm_slot,
.shutdown = cs35l56_sdw_dai_shutdown,
@@ -719,6 +754,13 @@ static const struct snd_soc_dai_ops cs35l56_sdw_dai_ops = {
.set_stream = cs35l56_sdw_dai_set_stream,
};
+static const struct snd_soc_dai_ops cs35l56_ot25_dai_ops = {
+ .shutdown = cs35l56_sdw_dai_shutdown,
+ .hw_params = cs35l56_ot25_dai_hw_params,
+ .hw_free = cs35l56_sdw_dai_hw_free,
+ .set_stream = cs35l56_sdw_dai_set_stream,
+};
+
static struct snd_soc_dai_driver cs35l56_dai[] = {
{
.name = "cs35l56-asp1",
@@ -767,6 +809,18 @@ static struct snd_soc_dai_driver cs35l56_dai[] = {
.symmetric_rate = 1,
.ops = &cs35l56_sdw_dai_ops,
},
+ {
+ .name = "cs35l56-ot25",
+ .id = 3,
+ .capture = {
+ .stream_name = "OT25",
+ .channels_min = 1,
+ .channels_max = 2,
+ .rates = CS35L56_RATES,
+ .formats = CS35L56_TX_FORMATS,
+ },
+ .ops = &cs35l56_ot25_dai_ops,
+ },
};
static int cs35l56_write_cal(struct cs35l56_private *cs35l56)
--
2.53.0
^ permalink raw reply related [flat|nested] 13+ messages in thread* Re: [PATCH 7.2 00/10] 7.2.12-rc1 review
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (9 preceding siblings ...)
2026-10-17 15:54 ` [PATCH 7.2 10/10] ASoC: cs35l56: Add DAI for SDCA OT25 stream Greg Kroah-Hartman
@ 2026-10-17 16:47 ` Ronald Warsow
2026-10-17 17:19 ` Brett A C Sheffield
11 siblings, 0 replies; 13+ messages in thread
From: Ronald Warsow @ 2026-10-17 16:47 UTC (permalink / raw)
To: Greg Kroah-Hartman, stable
Cc: patches, linux-kernel, torvalds, akpm, linux, shuah, patches,
lkft-triage, pavel, jonathanh, f.fainelli, sudipm.mukherjee,
conor, hargar, broonie, achill, sr
Hi
kernel build / boot test on x86_64 (Intel).
No regressions here.
Thanks
Tested-by: Ronald Warsow <rwarsow@gmx.de>
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [PATCH 7.2 00/10] 7.2.12-rc1 review
2026-10-17 15:54 [PATCH 7.2 00/10] 7.2.12-rc1 review Greg Kroah-Hartman
` (10 preceding siblings ...)
2026-10-17 16:47 ` [PATCH 7.2 00/10] 7.2.12-rc1 review Ronald Warsow
@ 2026-10-17 17:19 ` Brett A C Sheffield
11 siblings, 0 replies; 13+ messages in thread
From: Brett A C Sheffield @ 2026-10-17 17:19 UTC (permalink / raw)
To: gregkh
Cc: stable, patches, linux-kernel, torvalds, akpm, linux, shuah,
patches, lkft-triage, pavel, jonathanh, f.fainelli,
sudipm.mukherjee, rwarsow, conor, hargar, broonie, achill, sr,
Brett A C Sheffield
# Librecast Test Results
044/044 [ OK ] liblcrq
010/010 [ OK ] libmld
120/120 [ OK ] liblibrecast
CPU/kernel: Linux auntie 7.2.12-rc1-g590446e5a2fe #2 SMP PREEMPT_DYNAMIC Sat Oct 17 16:46:51 -00 2026 x86_64 AMD Ryzen 9 9950X 16-Core Processor AuthenticAMD GNU/Linux
Tested-by: Brett A C Sheffield <bacs@librecast.net>
^ permalink raw reply [flat|nested] 13+ messages in thread