All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] aoe: Avoid flush_scheduled_work() usage
@ 2022-04-18 14:03 Tetsuo Handa
  2022-04-18 22:58 ` kernel test robot
  2022-04-18 23:31 ` [PATCH v2] " Tetsuo Handa
  0 siblings, 2 replies; 5+ messages in thread
From: Tetsuo Handa @ 2022-04-18 14:03 UTC (permalink / raw)
  To: Justin Sanders, Jens Axboe; +Cc: linux-block

Flushing system-wide workqueues is dangerous and will be forbidden.
Replace system_wq with local aoe_wq.

Link: https://lkml.kernel.org/r/49925af7-78a8-a3dd-bce6-cfc02e1a9236@I-love.SAKURA.ne.jp
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
---
Note: This patch is only compile tested.

 drivers/block/aoe/aoe.h     | 2 ++
 drivers/block/aoe/aoeblk.c  | 2 +-
 drivers/block/aoe/aoecmd.c  | 2 +-
 drivers/block/aoe/aoedev.c  | 4 ++--
 drivers/block/aoe/aoemain.c | 9 ++++++++-
 5 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h
index 84d0fcebd6af..749ae1246f4c 100644
--- a/drivers/block/aoe/aoe.h
+++ b/drivers/block/aoe/aoe.h
@@ -244,3 +244,5 @@ void aoenet_exit(void);
 void aoenet_xmit(struct sk_buff_head *);
 int is_aoe_netif(struct net_device *ifp);
 int set_aoe_iflist(const char __user *str, size_t size);
+
+extern struct workqueue_struct *aoe_wq;
diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
index 8a91fcac6f82..348adf335217 100644
--- a/drivers/block/aoe/aoeblk.c
+++ b/drivers/block/aoe/aoeblk.c
@@ -435,7 +435,7 @@ aoeblk_gdalloc(void *vp)
 err:
 	spin_lock_irqsave(&d->lock, flags);
 	d->flags &= ~DEVFL_GD_NOW;
-	schedule_work(&d->work);
+	queue_work(aoe_wq, &d->work);
 	spin_unlock_irqrestore(&d->lock, flags);
 }
 
diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c
index 384073ef2323..d7317425be51 100644
--- a/drivers/block/aoe/aoecmd.c
+++ b/drivers/block/aoe/aoecmd.c
@@ -968,7 +968,7 @@ ataid_complete(struct aoedev *d, struct aoetgt *t, unsigned char *id)
 		d->flags |= DEVFL_NEWSIZE;
 	else
 		d->flags |= DEVFL_GDALLOC;
-	schedule_work(&d->work);
+	queue_work(aoe_wq, &d->work);
 }
 
 static void
diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c
index c5753c6bfe80..b381d1c3ef32 100644
--- a/drivers/block/aoe/aoedev.c
+++ b/drivers/block/aoe/aoedev.c
@@ -321,7 +321,7 @@ flush(const char __user *str, size_t cnt, int exiting)
 			specified = 1;
 	}
 
-	flush_scheduled_work();
+	flush_workqueue(aoe_wq);
 	/* pass one: do aoedev_downdev, which might sleep */
 restart1:
 	spin_lock_irqsave(&devlist_lock, flags);
@@ -520,7 +520,7 @@ freetgt(struct aoedev *d, struct aoetgt *t)
 void
 aoedev_exit(void)
 {
-	flush_scheduled_work();
+	flush_workqueue(aoe_wq);
 	flush(NULL, 0, EXITING);
 }
 
diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c
index 1e4e2971171c..892d627a2379 100644
--- a/drivers/block/aoe/aoemain.c
+++ b/drivers/block/aoe/aoemain.c
@@ -35,6 +35,7 @@ aoe_exit(void)
 	aoechr_exit();
 	aoedev_exit();
 	aoeblk_exit();		/* free cache after de-allocating bufs */
+	destroy_workqueue(aoe_wq);
 }
 
 static int __init
@@ -42,9 +43,13 @@ aoe_init(void)
 {
 	int ret;
 
+	aoe_wq = alloc_workqueue("aoe_wq", 0, 0);
+	if (!aoe_wq)
+		return -ENOMEM;
+
 	ret = aoedev_init();
 	if (ret)
-		return ret;
+		goto dev_fail;
 	ret = aoechr_init();
 	if (ret)
 		goto chr_fail;
@@ -77,6 +82,8 @@ aoe_init(void)
 	aoechr_exit();
  chr_fail:
 	aoedev_exit();
+ dev_fail:
+	destroy_workqueue(aoe_wq);
 
 	printk(KERN_INFO "aoe: initialisation failure.\n");
 	return ret;
-- 
2.32.0



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH] aoe: Avoid flush_scheduled_work() usage
  2022-04-18 14:03 [PATCH] aoe: Avoid flush_scheduled_work() usage Tetsuo Handa
@ 2022-04-18 22:58 ` kernel test robot
  2022-04-18 23:31 ` [PATCH v2] " Tetsuo Handa
  1 sibling, 0 replies; 5+ messages in thread
From: kernel test robot @ 2022-04-18 22:58 UTC (permalink / raw)
  To: Tetsuo Handa, Justin Sanders, Jens Axboe; +Cc: kbuild-all, linux-block

Hi Tetsuo,

I love your patch! Yet something to improve:

[auto build test ERROR on axboe-block/for-next]
[also build test ERROR on v5.18-rc3 next-20220414]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/intel-lab-lkp/linux/commits/Tetsuo-Handa/aoe-Avoid-flush_scheduled_work-usage/20220418-231118
base:   https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git for-next
config: arc-randconfig-r023-20220419 (https://download.01.org/0day-ci/archive/20220419/202204190659.PmftNpOE-lkp@intel.com/config)
compiler: arceb-elf-gcc (GCC) 11.2.0
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # https://github.com/intel-lab-lkp/linux/commit/b1ae05b7bd5a2a2abad4aba03f55e7f4b4599789
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Tetsuo-Handa/aoe-Avoid-flush_scheduled_work-usage/20220418-231118
        git checkout b1ae05b7bd5a2a2abad4aba03f55e7f4b4599789
        # save the config file to linux build tree
        mkdir build_dir
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross O=build_dir ARCH=arc SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

   arceb-elf-ld: drivers/block/aoe/aoeblk.o: in function `queue_work':
>> include/linux/workqueue.h:502: undefined reference to `aoe_wq'
>> arceb-elf-ld: include/linux/workqueue.h:502: undefined reference to `aoe_wq'
   arceb-elf-ld: drivers/block/aoe/aoecmd.o: in function `queue_work':
>> include/linux/workqueue.h:502: undefined reference to `aoe_wq'
>> arceb-elf-ld: include/linux/workqueue.h:502: undefined reference to `aoe_wq'
   arceb-elf-ld: drivers/block/aoe/aoedev.o: in function `flush':
>> drivers/block/aoe/aoedev.c:324: undefined reference to `aoe_wq'
   arceb-elf-ld: drivers/block/aoe/aoedev.o:drivers/block/aoe/aoedev.c:324: more undefined references to `aoe_wq' follow


vim +502 include/linux/workqueue.h

dcd989cb73ab0f Tejun Heo    2010-06-29  475  
8425e3d5bdbe8e Tejun Heo    2013-03-13  476  /**
8425e3d5bdbe8e Tejun Heo    2013-03-13  477   * queue_work - queue work on a workqueue
8425e3d5bdbe8e Tejun Heo    2013-03-13  478   * @wq: workqueue to use
8425e3d5bdbe8e Tejun Heo    2013-03-13  479   * @work: work to queue
8425e3d5bdbe8e Tejun Heo    2013-03-13  480   *
8425e3d5bdbe8e Tejun Heo    2013-03-13  481   * Returns %false if @work was already on a queue, %true otherwise.
8425e3d5bdbe8e Tejun Heo    2013-03-13  482   *
8425e3d5bdbe8e Tejun Heo    2013-03-13  483   * We queue the work to the CPU on which it was submitted, but if the CPU dies
8425e3d5bdbe8e Tejun Heo    2013-03-13  484   * it can be processed by another CPU.
dbb92f88648d62 Andrea Parri 2020-01-22  485   *
dbb92f88648d62 Andrea Parri 2020-01-22  486   * Memory-ordering properties:  If it returns %true, guarantees that all stores
dbb92f88648d62 Andrea Parri 2020-01-22  487   * preceding the call to queue_work() in the program order will be visible from
dbb92f88648d62 Andrea Parri 2020-01-22  488   * the CPU which will execute @work by the time such work executes, e.g.,
dbb92f88648d62 Andrea Parri 2020-01-22  489   *
dbb92f88648d62 Andrea Parri 2020-01-22  490   * { x is initially 0 }
dbb92f88648d62 Andrea Parri 2020-01-22  491   *
dbb92f88648d62 Andrea Parri 2020-01-22  492   *   CPU0				CPU1
dbb92f88648d62 Andrea Parri 2020-01-22  493   *
dbb92f88648d62 Andrea Parri 2020-01-22  494   *   WRITE_ONCE(x, 1);			[ @work is being executed ]
dbb92f88648d62 Andrea Parri 2020-01-22  495   *   r0 = queue_work(wq, work);		  r1 = READ_ONCE(x);
dbb92f88648d62 Andrea Parri 2020-01-22  496   *
dbb92f88648d62 Andrea Parri 2020-01-22  497   * Forbids: r0 == true && r1 == 0
8425e3d5bdbe8e Tejun Heo    2013-03-13  498   */
8425e3d5bdbe8e Tejun Heo    2013-03-13  499  static inline bool queue_work(struct workqueue_struct *wq,
8425e3d5bdbe8e Tejun Heo    2013-03-13  500  			      struct work_struct *work)
8425e3d5bdbe8e Tejun Heo    2013-03-13  501  {
8425e3d5bdbe8e Tejun Heo    2013-03-13 @502  	return queue_work_on(WORK_CPU_UNBOUND, wq, work);
8425e3d5bdbe8e Tejun Heo    2013-03-13  503  }
8425e3d5bdbe8e Tejun Heo    2013-03-13  504  

-- 
0-DAY CI Kernel Test Service
https://01.org/lkp

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH v2] aoe: Avoid flush_scheduled_work() usage
  2022-04-18 14:03 [PATCH] aoe: Avoid flush_scheduled_work() usage Tetsuo Handa
  2022-04-18 22:58 ` kernel test robot
@ 2022-04-18 23:31 ` Tetsuo Handa
  2022-05-01  4:09   ` Tetsuo Handa
  2022-05-01 12:42   ` Jens Axboe
  1 sibling, 2 replies; 5+ messages in thread
From: Tetsuo Handa @ 2022-04-18 23:31 UTC (permalink / raw)
  To: Justin Sanders, Jens Axboe; +Cc: linux-block

Flushing system-wide workqueues is dangerous and will be forbidden.
Replace system_wq with local aoe_wq.

Link: https://lkml.kernel.org/r/49925af7-78a8-a3dd-bce6-cfc02e1a9236@I-love.SAKURA.ne.jp
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
---
Changes in v2:
  Fix link error, reported by kernel test robot <lkp@intel.com>

 drivers/block/aoe/aoe.h     |  2 ++
 drivers/block/aoe/aoeblk.c  |  2 +-
 drivers/block/aoe/aoecmd.c  |  2 +-
 drivers/block/aoe/aoedev.c  |  4 ++--
 drivers/block/aoe/aoemain.c | 10 +++++++++-
 5 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h
index 84d0fcebd6af..749ae1246f4c 100644
--- a/drivers/block/aoe/aoe.h
+++ b/drivers/block/aoe/aoe.h
@@ -244,3 +244,5 @@ void aoenet_exit(void);
 void aoenet_xmit(struct sk_buff_head *);
 int is_aoe_netif(struct net_device *ifp);
 int set_aoe_iflist(const char __user *str, size_t size);
+
+extern struct workqueue_struct *aoe_wq;
diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
index 8a91fcac6f82..348adf335217 100644
--- a/drivers/block/aoe/aoeblk.c
+++ b/drivers/block/aoe/aoeblk.c
@@ -435,7 +435,7 @@ aoeblk_gdalloc(void *vp)
 err:
 	spin_lock_irqsave(&d->lock, flags);
 	d->flags &= ~DEVFL_GD_NOW;
-	schedule_work(&d->work);
+	queue_work(aoe_wq, &d->work);
 	spin_unlock_irqrestore(&d->lock, flags);
 }
 
diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c
index 384073ef2323..d7317425be51 100644
--- a/drivers/block/aoe/aoecmd.c
+++ b/drivers/block/aoe/aoecmd.c
@@ -968,7 +968,7 @@ ataid_complete(struct aoedev *d, struct aoetgt *t, unsigned char *id)
 		d->flags |= DEVFL_NEWSIZE;
 	else
 		d->flags |= DEVFL_GDALLOC;
-	schedule_work(&d->work);
+	queue_work(aoe_wq, &d->work);
 }
 
 static void
diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c
index c5753c6bfe80..b381d1c3ef32 100644
--- a/drivers/block/aoe/aoedev.c
+++ b/drivers/block/aoe/aoedev.c
@@ -321,7 +321,7 @@ flush(const char __user *str, size_t cnt, int exiting)
 			specified = 1;
 	}
 
-	flush_scheduled_work();
+	flush_workqueue(aoe_wq);
 	/* pass one: do aoedev_downdev, which might sleep */
 restart1:
 	spin_lock_irqsave(&devlist_lock, flags);
@@ -520,7 +520,7 @@ freetgt(struct aoedev *d, struct aoetgt *t)
 void
 aoedev_exit(void)
 {
-	flush_scheduled_work();
+	flush_workqueue(aoe_wq);
 	flush(NULL, 0, EXITING);
 }
 
diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c
index 1e4e2971171c..6238c4c87cfc 100644
--- a/drivers/block/aoe/aoemain.c
+++ b/drivers/block/aoe/aoemain.c
@@ -16,6 +16,7 @@ MODULE_DESCRIPTION("AoE block/char driver for 2.6.2 and newer 2.6 kernels");
 MODULE_VERSION(VERSION);
 
 static struct timer_list timer;
+struct workqueue_struct *aoe_wq;
 
 static void discover_timer(struct timer_list *t)
 {
@@ -35,6 +36,7 @@ aoe_exit(void)
 	aoechr_exit();
 	aoedev_exit();
 	aoeblk_exit();		/* free cache after de-allocating bufs */
+	destroy_workqueue(aoe_wq);
 }
 
 static int __init
@@ -42,9 +44,13 @@ aoe_init(void)
 {
 	int ret;
 
+	aoe_wq = alloc_workqueue("aoe_wq", 0, 0);
+	if (!aoe_wq)
+		return -ENOMEM;
+
 	ret = aoedev_init();
 	if (ret)
-		return ret;
+		goto dev_fail;
 	ret = aoechr_init();
 	if (ret)
 		goto chr_fail;
@@ -77,6 +83,8 @@ aoe_init(void)
 	aoechr_exit();
  chr_fail:
 	aoedev_exit();
+ dev_fail:
+	destroy_workqueue(aoe_wq);
 
 	printk(KERN_INFO "aoe: initialisation failure.\n");
 	return ret;
-- 
2.32.0



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] aoe: Avoid flush_scheduled_work() usage
  2022-04-18 23:31 ` [PATCH v2] " Tetsuo Handa
@ 2022-05-01  4:09   ` Tetsuo Handa
  2022-05-01 12:42   ` Jens Axboe
  1 sibling, 0 replies; 5+ messages in thread
From: Tetsuo Handa @ 2022-05-01  4:09 UTC (permalink / raw)
  To: Justin Sanders, Jens Axboe; +Cc: linux-block

Any questions?

On 2022/04/19 8:31, Tetsuo Handa wrote:
> Flushing system-wide workqueues is dangerous and will be forbidden.
> Replace system_wq with local aoe_wq.
> 
> Link: https://lkml.kernel.org/r/49925af7-78a8-a3dd-bce6-cfc02e1a9236@I-love.SAKURA.ne.jp
> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> ---
> Changes in v2:
>   Fix link error, reported by kernel test robot <lkp@intel.com>
> 
>  drivers/block/aoe/aoe.h     |  2 ++
>  drivers/block/aoe/aoeblk.c  |  2 +-
>  drivers/block/aoe/aoecmd.c  |  2 +-
>  drivers/block/aoe/aoedev.c  |  4 ++--
>  drivers/block/aoe/aoemain.c | 10 +++++++++-
>  5 files changed, 15 insertions(+), 5 deletions(-)

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] aoe: Avoid flush_scheduled_work() usage
  2022-04-18 23:31 ` [PATCH v2] " Tetsuo Handa
  2022-05-01  4:09   ` Tetsuo Handa
@ 2022-05-01 12:42   ` Jens Axboe
  1 sibling, 0 replies; 5+ messages in thread
From: Jens Axboe @ 2022-05-01 12:42 UTC (permalink / raw)
  To: justin, penguin-kernel; +Cc: linux-block

On Tue, 19 Apr 2022 08:31:55 +0900, Tetsuo Handa wrote:
> Flushing system-wide workqueues is dangerous and will be forbidden.
> Replace system_wq with local aoe_wq.
> 
> 

Applied, thanks!

[1/1] aoe: Avoid flush_scheduled_work() usage
      commit: 0b8d7622ab1859bec082bd01c5e11137195f3d52

Best regards,
-- 
Jens Axboe



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2022-05-01 12:42 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-18 14:03 [PATCH] aoe: Avoid flush_scheduled_work() usage Tetsuo Handa
2022-04-18 22:58 ` kernel test robot
2022-04-18 23:31 ` [PATCH v2] " Tetsuo Handa
2022-05-01  4:09   ` Tetsuo Handa
2022-05-01 12:42   ` Jens Axboe

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.