[RFC PATCH 0/6] fs: Move long delayed work on system_dfl_long_wq

[RFC PATCH 0/6] fs: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Hello,

Currently the code uses the per-cpu workqueue system_long_wq to schedule
long running works.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Another good reason to have this unbound,
is the "queue_delayed_work()" function, used to enqueue the work item.
More details on this will follow in the next section.

Recently, a new unbound workqueue specific for long running work has been
added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

~~~ Details about queue_delayed_work ~~~

system_long_wq is a per-cpu workqueue and it is used as a parameter of
queue_delayed_work(). This function schedule an item that it will later
be enqueued (once the timer will fire). __queue_delayed_work() does the job
receiving as "cpu" WORK_CPU_UNBOUND:

    if (housekeeping_enabled(HK_TYPE_TIMER)) {
    //      [....]
    } else {
            if (likely(cpu == WORK_CPU_UNBOUND))
                    add_timer_global(timer);
            else
                    add_timer_on(timer, cpu);
    }

The timer is global, so can fire everywhere, and the work item will be
enqueued where the timer fired.

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change the
workqueue with the new system_dfl_long_wq, so that the used workqueue is
now unbound and can benefit from scheduler task placement.

Thanks!

Marco Crivellari (6):
  ufs: Move long delayed work on system_dfl_long_wq
  fs/jffs2: Move long delayed work on system_dfl_long_wq
  hfsplus: Move long delayed work on system_dfl_long_wq
  hfs: Move long delayed work on system_dfl_long_wq
  fuse: dax: Move long delayed work on system_dfl_long_wq
  affs: Move long delayed work on system_dfl_long_wq

 fs/affs/super.c    | 2 +-
 fs/fuse/dax.c      | 2 +-
 fs/hfs/super.c     | 2 +-
 fs/hfsplus/super.c | 2 +-
 fs/jffs2/wbuf.c    | 2 +-
 fs/ufs/super.c     | 2 +-
 6 files changed, 6 insertions(+), 6 deletions(-)

-- 
2.53.0

[RFC PATCH 1/6] ufs: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Kees Cook <kees@kernel.org>
Cc: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/ufs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index c4831a8b9b3f..6dcf6d048cce 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -672,7 +672,7 @@ void ufs_mark_sb_dirty(struct super_block *sb)
 	spin_lock(&sbi->work_lock);
 	if (!sbi->work_queued) {
 		delay = msecs_to_jiffies(dirty_writeback_interval * 10);
-		queue_delayed_work(system_long_wq, &sbi->sync_work, delay);
+		queue_delayed_work(system_dfl_long_wq, &sbi->sync_work, delay);
 		sbi->work_queued = 1;
 	}
 	spin_unlock(&sbi->work_lock);
-- 
2.53.0

[RFC PATCH 2/6] fs/jffs2: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: linux-mtd@lists.infradead.org
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/jffs2/wbuf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c
index 8ff7a0b6add2..3b7803c75d58 100644
--- a/fs/jffs2/wbuf.c
+++ b/fs/jffs2/wbuf.c
@@ -1177,7 +1177,7 @@ void jffs2_dirty_trigger(struct jffs2_sb_info *c)
 		return;
 
 	delay = msecs_to_jiffies(dirty_writeback_interval * 10);
-	if (queue_delayed_work(system_long_wq, &c->wbuf_dwork, delay))
+	if (queue_delayed_work(system_dfl_long_wq, &c->wbuf_dwork, delay))
 		jffs2_dbg(1, "%s()\n", __func__);
 }
 
-- 
2.53.0

[RFC PATCH 3/6] hfsplus: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: Viacheslav Dubeyko <slava@dubeyko.com>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Yangtao Li <frank.li@vivo.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/hfsplus/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 40a0feda716b..12ba672e13bc 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -314,7 +314,7 @@ void hfsplus_mark_mdb_dirty(struct super_block *sb)
 	spin_lock(&sbi->work_lock);
 	if (!sbi->work_queued) {
 		delay = msecs_to_jiffies(dirty_writeback_interval * 10);
-		queue_delayed_work(system_long_wq, &sbi->sync_work, delay);
+		queue_delayed_work(system_dfl_long_wq, &sbi->sync_work, delay);
 		sbi->work_queued = 1;
 	}
 	spin_unlock(&sbi->work_lock);
-- 
2.53.0

Re: [RFC PATCH 3/6] hfsplus: Move long delayed work on system_dfl_long_wq

[Viacheslav Dubeyko]

On Fri, 2026-05-08 at 15:45 +0200, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_long_wq. This workqueue should be used when long works are
> expected and it is a per-cpu workqueue.
> 
> The function(s) end up calling __queue_delayed_work(), which set a global
> timer that could fire anywhere, enqueuing the work where the timer fired.
> 
> Unbound works could benefit from scheduler task placement, to optimize
> performance and power consumption. Long work shouldn't stick to a single
> CPU.
> 
> Recently, a new unbound workqueue specific for long running work has
> been added:
> 
>     c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
> 
> Since the workqueue work doesn't rely on per-cpu variables, there is no
> obvious reason that justify the use of a per-cpu workqueue. So change
> system_long_wq with system_dfl_long_wq so that the work may benefit from
> scheduler task placement.
> 
> Cc: Viacheslav Dubeyko <slava@dubeyko.com>
> Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
> Cc: Yangtao Li <frank.li@vivo.com>
> Cc: linux-fsdevel@vger.kernel.org
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
> ---
>  fs/hfsplus/super.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
> index 40a0feda716b..12ba672e13bc 100644
> --- a/fs/hfsplus/super.c
> +++ b/fs/hfsplus/super.c
> @@ -314,7 +314,7 @@ void hfsplus_mark_mdb_dirty(struct super_block *sb)
>  	spin_lock(&sbi->work_lock);
>  	if (!sbi->work_queued) {
>  		delay = msecs_to_jiffies(dirty_writeback_interval * 10);
> -		queue_delayed_work(system_long_wq, &sbi->sync_work, delay);
> +		queue_delayed_work(system_dfl_long_wq, &sbi->sync_work, delay);
>  		sbi->work_queued = 1;
>  	}
>  	spin_unlock(&sbi->work_lock);

Looks good.

Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>

Thanks,
Slava.

[RFC PATCH 4/6] hfs: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: Viacheslav Dubeyko <slava@dubeyko.com>
Cc: John Paul Adrian Glaubitz
Cc: Yangtao Li <frank.li@vivo.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/hfs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index a4f2a2bfa6d3..cf4e7a6112e7 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -82,7 +82,7 @@ void hfs_mark_mdb_dirty(struct super_block *sb)
 	spin_lock(&sbi->work_lock);
 	if (!sbi->work_queued) {
 		delay = msecs_to_jiffies(dirty_writeback_interval * 10);
-		queue_delayed_work(system_long_wq, &sbi->mdb_work, delay);
+		queue_delayed_work(system_dfl_long_wq, &sbi->mdb_work, delay);
 		sbi->work_queued = 1;
 	}
 	spin_unlock(&sbi->work_lock);
-- 
2.53.0

Re: [RFC PATCH 4/6] hfs: Move long delayed work on system_dfl_long_wq

[Viacheslav Dubeyko]

On Fri, 2026-05-08 at 15:45 +0200, Marco Crivellari wrote:
> Currently the code enqueue work items using {queue|mod}_delayed_work(),
> using system_long_wq. This workqueue should be used when long works are
> expected and it is a per-cpu workqueue.
> 
> The function(s) end up calling __queue_delayed_work(), which set a global
> timer that could fire anywhere, enqueuing the work where the timer fired.
> 
> Unbound works could benefit from scheduler task placement, to optimize
> performance and power consumption. Long work shouldn't stick to a single
> CPU.
> 
> Recently, a new unbound workqueue specific for long running work has
> been added:
> 
>     c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")
> 
> Since the workqueue work doesn't rely on per-cpu variables, there is no
> obvious reason that justify the use of a per-cpu workqueue. So change
> system_long_wq with system_dfl_long_wq so that the work may benefit from
> scheduler task placement.
> 
> Cc: Viacheslav Dubeyko <slava@dubeyko.com>
> Cc: John Paul Adrian Glaubitz
> Cc: Yangtao Li <frank.li@vivo.com>
> Cc: linux-fsdevel@vger.kernel.org
> Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
> ---
>  fs/hfs/super.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/hfs/super.c b/fs/hfs/super.c
> index a4f2a2bfa6d3..cf4e7a6112e7 100644
> --- a/fs/hfs/super.c
> +++ b/fs/hfs/super.c
> @@ -82,7 +82,7 @@ void hfs_mark_mdb_dirty(struct super_block *sb)
>  	spin_lock(&sbi->work_lock);
>  	if (!sbi->work_queued) {
>  		delay = msecs_to_jiffies(dirty_writeback_interval * 10);
> -		queue_delayed_work(system_long_wq, &sbi->mdb_work, delay);
> +		queue_delayed_work(system_dfl_long_wq, &sbi->mdb_work, delay);
>  		sbi->work_queued = 1;
>  	}
>  	spin_unlock(&sbi->work_lock);

Looks good.

Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>

Thanks,
Slava.

[RFC PATCH 5/6] fuse: dax: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/fuse/dax.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/fuse/dax.c b/fs/fuse/dax.c
index f6cf00a8938c..8b53625ac7ab 100644
--- a/fs/fuse/dax.c
+++ b/fs/fuse/dax.c
@@ -113,7 +113,7 @@ __kick_dmap_free_worker(struct fuse_conn_dax *fcd, unsigned long delay_ms)
 	free_threshold = max_t(unsigned long, fcd->nr_ranges * FUSE_DAX_RECLAIM_THRESHOLD / 100,
 			     1);
 	if (fcd->nr_free_ranges < free_threshold)
-		queue_delayed_work(system_long_wq, &fcd->free_work,
+		queue_delayed_work(system_dfl_long_wq, &fcd->free_work,
 				   msecs_to_jiffies(delay_ms));
 }
 
-- 
2.53.0

[RFC PATCH 6/6] affs: Move long delayed work on system_dfl_long_wq

[Marco Crivellari]

Currently the code enqueue work items using {queue|mod}_delayed_work(),
using system_long_wq. This workqueue should be used when long works are
expected and it is a per-cpu workqueue.

The function(s) end up calling __queue_delayed_work(), which set a global
timer that could fire anywhere, enqueuing the work where the timer fired.

Unbound works could benefit from scheduler task placement, to optimize
performance and power consumption. Long work shouldn't stick to a single
CPU.

Recently, a new unbound workqueue specific for long running work has
been added:

    c116737e972e ("workqueue: Add system_dfl_long_wq for long unbound works")

Since the workqueue work doesn't rely on per-cpu variables, there is no
obvious reason that justify the use of a per-cpu workqueue. So change
system_long_wq with system_dfl_long_wq so that the work may benefit from
scheduler task placement.

Cc: David Sterba <dsterba@suse.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 fs/affs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/affs/super.c b/fs/affs/super.c
index 079f36e1ddec..061da7b795bd 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -88,7 +88,7 @@ void affs_mark_sb_dirty(struct super_block *sb)
 	spin_lock(&sbi->work_lock);
 	if (!sbi->work_queued) {
 	       delay = msecs_to_jiffies(dirty_writeback_interval * 10);
-	       queue_delayed_work(system_long_wq, &sbi->sb_work, delay);
+	       queue_delayed_work(system_dfl_long_wq, &sbi->sb_work, delay);
 	       sbi->work_queued = 1;
 	}
 	spin_unlock(&sbi->work_lock);
-- 
2.53.0