A series patches about sched priority.

PATCH Applied Dongsheng Yang 5 participants 32 messages Mar 11, 2014 – Apr 18, 2014

Subsystems (22)

Memory management SCSI PowerPC Block layer +18 more subsystems

lore

Share & tools

lore

Dongsheng Yang

author

2014-03-11 10:19 UTC #0 lore raw

v3 cover A series patches about sched priority.

Hi all,
This patchset is all about priority.

1. Add two inline functions in prio.h named nice_to_rlimit and rlimit_to_nice. They are converting the value between nice value [-20, 19] and rlimit style value [1, 40].
2. Add a macro in ioprio.h named NICE_TO_IOPRIO.
It convert nice value [-20, 19] to io priority [0, 7].
3. Others are all about replace hardcoding value about nice to MIN_NICE or MAX_NICE.

Changelog:
-v2:
*Splite [2/16] into two patches, one is about the whitespace changes, and
the other is about nice_to_rlimit().
-v1:
*Implement nice_to_rlimit and rlimit_to_nice with inline functions
rather than macro.
*Add a patch[16/16] to replace opened code implement with nice_to_rlimit().

Dongsheng Yang (16):
sched/prio: Add a inline function named nice_to_rlimit() in prio.h. kernel/sys: Replace opened code implementation with nice_to_rlimit().
workqueue: Replace hardcoding of -20 with MIN_NICE.
locktorture: Replace hardcoding of 19 with MAX_NICE. tools/mq_perf_tests: Replace hardcoding of -20 with MIN_NICE.
mm: Replace hardcoding of 19 with MAX_NICE.
ioprio: Add a macro named NICE_TO_IOPRIO.
fs/hearbeat: Replace hardcoding of -20 with MIN_NICE. driver/block: Replace hardcoding of -20 with MIN_NICE. driver/char: Replace hardcoding of 19 with MAX_NICE. drivers/s390: Replace hardcoding of 19 with MAX_NICE. sched/prio: Add an inline function named rlimit_to_nice in prio.h. driver/staging/android: Use rlimit_to_nice to replace opened code
implementation.
driver/staging/lustre: Replace hardcoding of -20 with MIN_NICE. driver/scsi: Replace hardcoding of -20 with MIN_NICE.
sched: Get rid of opened code implementation of funtion

    nice_to_rlimit().

Joe Perches (1):
kernel/sys: Fix the indent issue in switch.

 drivers/block/loop.c                           |   2 +-
 drivers/block/nbd.c                            |   2 +-
 drivers/block/pktcdvd.c                        |   2 +-
 drivers/char/ipmi/ipmi_si_intf.c               |   2 +-
 drivers/s390/crypto/ap_bus.c                   |   2 +-
 drivers/scsi/bnx2fc/bnx2fc_fcoe.c              |   4 +-
 drivers/scsi/bnx2i/bnx2i_hwi.c                 |   2 +-
 drivers/scsi/fcoe/fcoe.c                       |   2 +-
 drivers/scsi/ibmvscsi/ibmvfc.c                 |   2 +-
 drivers/scsi/ibmvscsi/ibmvscsi.c               |   2 +-
 drivers/scsi/lpfc/lpfc_hbadisc.c               |   2 +-
 drivers/scsi/qla2xxx/qla_os.c                  |   2 +-
 drivers/staging/android/binder.c               |   4 +-
 drivers/staging/lustre/lustre/llite/lloop.c    |   2 +-
 fs/ocfs2/cluster/heartbeat.c                   |   2 +-
 include/linux/ioprio.h                         |   7 +-
 include/linux/sched/prio.h                     |  16 ++
 kernel/locking/locktorture.c                   |   2 +-
 kernel/sched/core.c                            |   2 +-
 kernel/sys.c                                   | 206 ++++++++++++-------------
 kernel/workqueue.c                             |   6 +-
 mm/huge_memory.c                               |   2 +-
 tools/testing/selftests/mqueue/mq_perf_tests.c |   4 +-
 23 files changed, 150 insertions(+), 129 deletions(-)

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #1 lore raw

PATCH v3 1/17 sched/prio: Add a inline function named nice_to_rlimit() in prio.h.

This patch add an inline function named nice_to_rlimit() in prio.h to convert nice value [19,-20] to rlimit style value [1,40].

Signed-off-by Dongsheng Yang <[email protected]>

---
 include/linux/sched/prio.h | 8 ++++++++
 1 file changed, 8 insertions(+)

include/linux/sched/prio.h +8 -0

index ac32258..29756f8 100644
@@ -41,4 +41,12 @@
 #define TASK_USER_PRIO(p)	USER_PRIO((p)->static_prio)
 #define MAX_USER_PRIO		(USER_PRIO(MAX_PRIO))
 
+/*
+ * Convert nice value [19,-20] to rlimit style value [1,40].
+ */
+static inline long nice_to_rlimit(long nice)
+{
+	return (MAX_NICE - nice + 1);
+}
+
 #endif /* _SCHED_PRIO_H */

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:16 UTC #2 lore raw

PATCH v3 2/17 kernel/sys: Fix the indent issue in switch.

In reply to Dongsheng Yang

From: Joe Perches <[email protected]>

Fix the indent issue and add a comment line for /* fall-through */.

Signed-off-by Joe Perches <[email protected]>

Signed-off-by Dongsheng Yang <[email protected]>

---
 kernel/sys.c | 204 +++++++++++++++++++++++++++++------------------------------
 1 file changed, 102 insertions(+), 102 deletions(-)

kernel/sys.c +102 ~~-102~~

index adaeab6..61a2ae9 100644
@@ -182,39 +182,39 @@ SYSCALL_DEFINE3(setpriority, int, which, int, who, int, niceval)
 	rcu_read_lock();
 	read_lock(&tasklist_lock);
 	switch (which) {
-		case PRIO_PROCESS:
-			if (who)
-				p = find_task_by_vpid(who);
-			else
-				p = current;
-			if (p)
-				error = set_one_prio(p, niceval, error);
-			break;
-		case PRIO_PGRP:
-			if (who)
-				pgrp = find_vpid(who);
-			else
-				pgrp = task_pgrp(current);
-			do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
+	case PRIO_PROCESS:
+		if (who)
+			p = find_task_by_vpid(who);
+		else
+			p = current;
+		if (p)
+			error = set_one_prio(p, niceval, error);
+		break;
+	case PRIO_PGRP:
+		if (who)
+			pgrp = find_vpid(who);
+		else
+			pgrp = task_pgrp(current);
+		do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
+			error = set_one_prio(p, niceval, error);
+		} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
+		break;
+	case PRIO_USER:
+		uid = make_kuid(cred->user_ns, who);
+		user = cred->user;
+		if (!who)
+			uid = cred->uid;
+		else if (!uid_eq(uid, cred->uid) &&
+			 !(user = find_user(uid)))
+			goto out_unlock;	/* No processes for this user */
+
+		do_each_thread(g, p) {
+			if (uid_eq(task_uid(p), uid))
 				error = set_one_prio(p, niceval, error);
-			} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
-			break;
-		case PRIO_USER:
-			uid = make_kuid(cred->user_ns, who);
-			user = cred->user;
-			if (!who)
-				uid = cred->uid;
-			else if (!uid_eq(uid, cred->uid) &&
-				 !(user = find_user(uid)))
-				goto out_unlock;	/* No processes for this user */
-
-			do_each_thread(g, p) {
-				if (uid_eq(task_uid(p), uid))
-					error = set_one_prio(p, niceval, error);
-			} while_each_thread(g, p);
-			if (!uid_eq(uid, cred->uid))
-				free_uid(user);		/* For find_user() */
-			break;
+		} while_each_thread(g, p);
+		if (!uid_eq(uid, cred->uid))
+			free_uid(user);		/* For find_user() */
+		break;
 	}
 out_unlock:
 	read_unlock(&tasklist_lock);

@@ -244,47 +244,47 @@ SYSCALL_DEFINE2(getpriority, int, which, int, who)
 	rcu_read_lock();
 	read_lock(&tasklist_lock);
 	switch (which) {
-		case PRIO_PROCESS:
-			if (who)
-				p = find_task_by_vpid(who);
-			else
-				p = current;
-			if (p) {
+	case PRIO_PROCESS:
+		if (who)
+			p = find_task_by_vpid(who);
+		else
+			p = current;
+		if (p) {
+			niceval = 20 - task_nice(p);
+			if (niceval > retval)
+				retval = niceval;
+		}
+		break;
+	case PRIO_PGRP:
+		if (who)
+			pgrp = find_vpid(who);
+		else
+			pgrp = task_pgrp(current);
+		do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
+			niceval = 20 - task_nice(p);
+			if (niceval > retval)
+				retval = niceval;
+		} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
+		break;
+	case PRIO_USER:
+		uid = make_kuid(cred->user_ns, who);
+		user = cred->user;
+		if (!who)
+			uid = cred->uid;
+		else if (!uid_eq(uid, cred->uid) &&
+			 !(user = find_user(uid)))
+			goto out_unlock;	/* No processes for this user */
+
+		do_each_thread(g, p) {
+			if (uid_eq(task_uid(p), uid)) {
 				niceval = 20 - task_nice(p);
 				if (niceval > retval)
 					retval = niceval;
 			}
-			break;
-		case PRIO_PGRP:
-			if (who)
-				pgrp = find_vpid(who);
-			else
-				pgrp = task_pgrp(current);
-			do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
-				niceval = 20 - task_nice(p);
-				if (niceval > retval)
-					retval = niceval;
-			} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);
-			break;
-		case PRIO_USER:
-			uid = make_kuid(cred->user_ns, who);
-			user = cred->user;
-			if (!who)
-				uid = cred->uid;
-			else if (!uid_eq(uid, cred->uid) &&
-				 !(user = find_user(uid)))
-				goto out_unlock;	/* No processes for this user */
-
-			do_each_thread(g, p) {
-				if (uid_eq(task_uid(p), uid)) {
-					niceval = 20 - task_nice(p);
-					if (niceval > retval)
-						retval = niceval;
-				}
-			} while_each_thread(g, p);
-			if (!uid_eq(uid, cred->uid))
-				free_uid(user);		/* for find_user() */
-			break;
+		} while_each_thread(g, p);
+		if (!uid_eq(uid, cred->uid))
+			free_uid(user);		/* for find_user() */
+		break;
 	}
 out_unlock:
 	read_unlock(&tasklist_lock);

@@ -1541,41 +1541,41 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
 		return;
 
 	switch (who) {
-		case RUSAGE_BOTH:
-		case RUSAGE_CHILDREN:
-			utime = p->signal->cutime;
-			stime = p->signal->cstime;
-			r->ru_nvcsw = p->signal->cnvcsw;
-			r->ru_nivcsw = p->signal->cnivcsw;
-			r->ru_minflt = p->signal->cmin_flt;
-			r->ru_majflt = p->signal->cmaj_flt;
-			r->ru_inblock = p->signal->cinblock;
-			r->ru_oublock = p->signal->coublock;
-			maxrss = p->signal->cmaxrss;
-
-			if (who == RUSAGE_CHILDREN)
-				break;
-
-		case RUSAGE_SELF:
-			thread_group_cputime_adjusted(p, &tgutime, &tgstime);
-			utime += tgutime;
-			stime += tgstime;
-			r->ru_nvcsw += p->signal->nvcsw;
-			r->ru_nivcsw += p->signal->nivcsw;
-			r->ru_minflt += p->signal->min_flt;
-			r->ru_majflt += p->signal->maj_flt;
-			r->ru_inblock += p->signal->inblock;
-			r->ru_oublock += p->signal->oublock;
-			if (maxrss < p->signal->maxrss)
-				maxrss = p->signal->maxrss;
-			t = p;
-			do {
-				accumulate_thread_rusage(t, r);
-			} while_each_thread(p, t);
+	case RUSAGE_BOTH:
+	case RUSAGE_CHILDREN:
+		utime = p->signal->cutime;
+		stime = p->signal->cstime;
+		r->ru_nvcsw = p->signal->cnvcsw;
+		r->ru_nivcsw = p->signal->cnivcsw;
+		r->ru_minflt = p->signal->cmin_flt;
+		r->ru_majflt = p->signal->cmaj_flt;
+		r->ru_inblock = p->signal->cinblock;
+		r->ru_oublock = p->signal->coublock;
+		maxrss = p->signal->cmaxrss;
+
+		if (who == RUSAGE_CHILDREN)
 			break;
+		/* fall-through */
+	case RUSAGE_SELF:
+		thread_group_cputime_adjusted(p, &tgutime, &tgstime);
+		utime += tgutime;
+		stime += tgstime;
+		r->ru_nvcsw += p->signal->nvcsw;
+		r->ru_nivcsw += p->signal->nivcsw;
+		r->ru_minflt += p->signal->min_flt;
+		r->ru_majflt += p->signal->maj_flt;
+		r->ru_inblock += p->signal->inblock;
+		r->ru_oublock += p->signal->oublock;
+		if (maxrss < p->signal->maxrss)
+			maxrss = p->signal->maxrss;
+		t = p;
+		do {
+			accumulate_thread_rusage(t, r);
+		} while_each_thread(p, t);
+		break;
 
-		default:
-			BUG();
+	default:
+		BUG();
 	}
 	unlock_task_sighand(p, &flags);

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #3 lore raw

PATCH v3 3/17 kernel/sys: Replace opened code implementation with nice_to_rlimit().

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

---
 kernel/sys.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

kernel/sys.c +3 -3

index 61a2ae9..1bfe79a 100644
@@ -250,7 +250,7 @@ SYSCALL_DEFINE2(getpriority, int, which, int, who)
 		else
 			p = current;
 		if (p) {
-			niceval = 20 - task_nice(p);
+			niceval = nice_to_rlimit(task_nice(p));
 			if (niceval > retval)
 				retval = niceval;
 		}

@@ -261,7 +261,7 @@ SYSCALL_DEFINE2(getpriority, int, which, int, who)
 		else
 			pgrp = task_pgrp(current);
 		do_each_pid_thread(pgrp, PIDTYPE_PGID, p) {
-			niceval = 20 - task_nice(p);
+			niceval = nice_to_rlimit(task_nice(p));
 			if (niceval > retval)
 				retval = niceval;
 		} while_each_pid_thread(pgrp, PIDTYPE_PGID, p);

@@ -277,7 +277,7 @@ SYSCALL_DEFINE2(getpriority, int, which, int, who)
 
 		do_each_thread(g, p) {
 			if (uid_eq(task_uid(p), uid)) {
-				niceval = 20 - task_nice(p);
+				niceval = nice_to_rlimit(task_nice(p));
 				if (niceval > retval)
 					retval = niceval;
 			}

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:18 UTC #4 lore raw

PATCH v3 4/17 workqueue: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc Tejun Heo <[email protected]>

---
 kernel/workqueue.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

kernel/workqueue.c +3 -3

index 3fa5b8f..7528bec 100644
@@ -100,10 +100,10 @@ enum {
 
 	/*
 	 * Rescue workers are used only on emergencies and shared by
-	 * all cpus.  Give -20.
+	 * all cpus.  Give MIN_NICE.
 	 */
-	RESCUER_NICE_LEVEL	= -20,
-	HIGHPRI_NICE_LEVEL	= -20,
+	RESCUER_NICE_LEVEL	= MIN_NICE,
+	HIGHPRI_NICE_LEVEL	= MIN_NICE,
 
 	WQ_NAME_LEN		= 24,
 };

Signature · 1 line

1.8.2.1

Tejun Heo

maintainer Applied

2014-03-11 12:22 UTC #5 lore raw

Re: [PATCH 04/17] workqueue: Replace hardcoding of -20 with MIN_NICE.

Quoted 2 lines · Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>
cc Tejun Heo <[email protected]>

Applied to wq/for-3.15.

Thanks.

Signature · 1 line

tejun

Tejun Heo

maintainer

2014-03-11 12:55 UTC #6 lore raw

Re: [PATCH 04/17] workqueue: Replace hardcoding of -20 with MIN_NICE.

Quoted 5 lines · Tejun Heo

Quoted 2 lines · Dongsheng Yang
Signed-off-by Dongsheng Yang <[email protected]>
cc Tejun Heo <[email protected]>
Applied to wq/for-3.15.

And reverted as MIN_NICE is not defined yet. Please feel free to add my Acked-by and route it together with the prerequisite changes.

Thanks.

Signature · 1 line

tejun

Dongsheng Yang

author

2014-03-11 12:58 UTC #7 lore raw

Re: [PATCH 04/17] workqueue: Replace hardcoding of -20 with MIN_NICE.

Hi Tejun,

Quoted 7 lines · Tejun Heo

Quoted 4 lines · Tejun Heo
Quoted 2 lines · Dongsheng Yang
Signed-off-by Dongsheng Yang <[email protected]>
cc Tejun Heo <[email protected]>
Applied to wq/for-3.15.
And reverted as MIN_NICE is not defined yet. Please feel free to add my Acked-by and route it together with the prerequisite changes.

The MIN_NICE is just in tip tree for now. So I think this patch should go through tip.

Thank you for your ack.

Quoted 3 lines

Thanks.

tip-bot for Dongsheng Yang

merge bot

2014-04-18 13:24 UTC #8 lore raw

[tip:sched/core] sched, treewide: Replace hardcoded nice values with MIN_NICE/MAX_NICE

In reply to Dongsheng Yang

sched, treewide: Replace hardcoded nice values with MIN_NICE/MAX_NICE

Commit-ID:  8698a745d800c59cd5a576398bdeccd578ac66f1
Gitweb:     http://git.kernel.org/tip/8698a745d800c59cd5a576398bdeccd578ac66f1
Author:     Dongsheng Yang <[email protected]>
AuthorDate: Tue, 11 Mar 2014 18:09:12 +0800
Committer:  Ingo Molnar <[email protected]>
CommitDate: Fri, 18 Apr 2014 12:07:24 +0200

sched, treewide: Replace hardcoded nice values with MIN_NICE/MAX_NICE

Replace various -20/+19 hardcoded nice values with MIN_NICE/MAX_NICE.

Signed-off-by Dongsheng Yang <[email protected]>

Acked-by Tejun Heo <[email protected]>

Signed-off-by Peter Zijlstra <[email protected]>

Link http://lkml.kernel.org/r/ff13819fd09b7a5dba5ab5ae797f2e7019bdfa17.1394532288.git.yangds.fnst@cn.fujitsu.com open here

Cc [email protected]

[ Consolidated the patches, twiddled the changelog. ]

Signed-off-by Ingo Molnar <[email protected]>

---
 drivers/block/loop.c                        | 2 +-
 drivers/block/nbd.c                         | 2 +-
 drivers/block/pktcdvd.c                     | 2 +-
 drivers/char/ipmi/ipmi_si_intf.c            | 2 +-
 drivers/s390/crypto/ap_bus.c                | 2 +-
 drivers/scsi/bnx2fc/bnx2fc_fcoe.c           | 4 ++--
 drivers/scsi/bnx2i/bnx2i_hwi.c              | 2 +-
 drivers/scsi/fcoe/fcoe.c                    | 2 +-
 drivers/scsi/ibmvscsi/ibmvfc.c              | 2 +-
 drivers/scsi/ibmvscsi/ibmvscsi.c            | 2 +-
 drivers/scsi/lpfc/lpfc_hbadisc.c            | 2 +-
 drivers/scsi/qla2xxx/qla_os.c               | 2 +-
 drivers/staging/android/binder.c            | 2 +-
 drivers/staging/lustre/lustre/llite/lloop.c | 2 +-
 fs/ocfs2/cluster/heartbeat.c                | 2 +-
 kernel/locking/locktorture.c                | 2 +-
 kernel/workqueue.c                          | 6 +++---
 mm/huge_memory.c                            | 2 +-
 18 files changed, 21 insertions(+), 21 deletions(-)

drivers/block/loop.c +1 -1

index 66e8c3b..c8bf270 100644
@@ -548,7 +548,7 @@ static int loop_thread(void *data)
 	struct loop_device *lo = data;
 	struct bio *bio;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (!kthread_should_stop() || !bio_list_empty(&lo->lo_bio_list)) {

drivers/block/nbd.c +1 -1

index 55298db..2a1f26b 100644
@@ -533,7 +533,7 @@ static int nbd_thread(void *data)
 	struct nbd_device *nbd = data;
 	struct request *req;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	while (!kthread_should_stop() || !list_empty(&nbd->waiting_queue)) {
 		/* wait for something to do */
 		wait_event_interruptible(nbd->waiting_wq,

drivers/block/pktcdvd.c +1 -1

index a2af73d..ef166ad 100644
@@ -1463,7 +1463,7 @@ static int kcdrwd(void *foobar)
 	struct packet_data *pkt;
 	long min_sleep_time, residue;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_freezable();
 
 	for (;;) {

drivers/char/ipmi/ipmi_si_intf.c +1 -1

index b7efd3c..0f20d36 100644
@@ -998,7 +998,7 @@ static int ipmi_thread(void *data)
 	struct timespec busy_until;
 
 	ipmi_si_set_not_busy(&busy_until);
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 	while (!kthread_should_stop()) {
 		int busy_wait;

drivers/s390/crypto/ap_bus.c +1 -1

index ab3baa7..8eec165 100644
@@ -1803,7 +1803,7 @@ static int ap_poll_thread(void *data)
 	int requests;
 	struct ap_device *ap_dev;
 
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 	while (1) {
 		if (ap_suspend_flag)
 			return 0;

drivers/scsi/bnx2fc/bnx2fc_fcoe.c +2 -2

index 6287f6a..3455cc5 100644
@@ -464,7 +464,7 @@ static int bnx2fc_l2_rcv_thread(void *arg)
 	struct fcoe_percpu_s *bg = arg;
 	struct sk_buff *skb;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {
 		schedule();

@@ -602,7 +602,7 @@ int bnx2fc_percpu_io_thread(void *arg)
 	struct bnx2fc_work *work, *tmp;
 	LIST_HEAD(work_list);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {
 		schedule();

drivers/scsi/bnx2i/bnx2i_hwi.c +1 -1

index b5ffd28..d6d491c 100644
@@ -1870,7 +1870,7 @@ int bnx2i_percpu_io_thread(void *arg)
 	struct bnx2i_work *work, *tmp;
 	LIST_HEAD(work_list);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (!kthread_should_stop()) {
 		spin_lock_bh(&p->p_work_lock);

drivers/scsi/fcoe/fcoe.c +1 -1

index f317000..843a679 100644
@@ -1872,7 +1872,7 @@ static int fcoe_percpu_receive_thread(void *arg)
 
 	skb_queue_head_init(&tmp);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 retry:
 	while (!kthread_should_stop()) {

drivers/scsi/ibmvscsi/ibmvfc.c +1 -1

index 23f5ba5..8dd4768 100644
@@ -4515,7 +4515,7 @@ static int ibmvfc_work(void *data)
 	struct ibmvfc_host *vhost = data;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (1) {
 		rc = wait_event_interruptible(vhost->work_wait_q,

drivers/scsi/ibmvscsi/ibmvscsi.c +1 -1

index fa76440..2ebfb2b 100644
@@ -2213,7 +2213,7 @@ static int ibmvscsi_work(void *data)
 	struct ibmvscsi_host_data *hostdata = data;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (1) {
 		rc = wait_event_interruptible(hostdata->work_wait_q,

drivers/scsi/lpfc/lpfc_hbadisc.c +1 -1

index 59b51c5..294c072 100644
@@ -731,7 +731,7 @@ lpfc_do_work(void *p)
 	struct lpfc_hba *phba = p;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	current->flags |= PF_NOFREEZE;
 	phba->data_flags = 0;

drivers/scsi/qla2xxx/qla_os.c +1 -1

index 19e99cc..afc8481 100644
@@ -4828,7 +4828,7 @@ qla2x00_do_dpc(void *data)
 	ha = (struct qla_hw_data *)data;
 	base_vha = pci_get_drvdata(ha->pdev);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {

drivers/staging/android/binder.c +1 -1

index cfe4bc8..179b21b 100644
@@ -441,7 +441,7 @@ static void binder_set_nice(long nice)
 		     "%d: nice value %ld not allowed use %ld instead\n",
 		      current->pid, nice, min_nice);
 	set_user_nice(current, min_nice);
-	if (min_nice < 20)
+	if (min_nice <= MAX_NICE)
 		return;
 	binder_user_error("%d RLIMIT_NICE not set\n", current->pid);
 }

drivers/staging/lustre/lustre/llite/lloop.c +1 -1

index f78eda2..d4c2cd9 100644
@@ -407,7 +407,7 @@ static int loop_thread(void *data)
 	int refcheck;
 	int ret = 0;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	lo->lo_state = LLOOP_BOUND;

fs/ocfs2/cluster/heartbeat.c +1 -1

index bf482df..7303929 100644
@@ -1107,7 +1107,7 @@ static int o2hb_thread(void *data)
 
 	mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread running\n");
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	/* Pin node */
 	o2nm_depend_this_node();

kernel/locking/locktorture.c +1 -1

index f26b1a1..23343be 100644
@@ -216,7 +216,7 @@ static int lock_torture_writer(void *arg)
 	static DEFINE_TORTURE_RANDOM(rand);
 
 	VERBOSE_TOROUT_STRING("lock_torture_writer task started");
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 
 	do {
 		schedule_timeout_uninterruptible(1);

kernel/workqueue.c +3 -3

index 0ee63af..c30c01b 100644
@@ -100,10 +100,10 @@ enum {
 
 	/*
 	 * Rescue workers are used only on emergencies and shared by
-	 * all cpus.  Give -20.
+	 * all cpus.  Give MIN_NICE.
 	 */
-	RESCUER_NICE_LEVEL	= -20,
-	HIGHPRI_NICE_LEVEL	= -20,
+	RESCUER_NICE_LEVEL	= MIN_NICE,
+	HIGHPRI_NICE_LEVEL	= MIN_NICE,
 
 	WQ_NAME_LEN		= 24,
 };

mm/huge_memory.c +1 -1

index 1546655..dcdb6f9 100644
@@ -2803,7 +2803,7 @@ static int khugepaged(void *none)
 	struct mm_slot *mm_slot;
 
 	set_freezable();
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 
 	while (!kthread_should_stop()) {
 		khugepaged_do_scan();

Dongsheng Yang

author

2014-03-11 10:17 UTC #9 lore raw

PATCH v3 5/17 locktorture: Replace hardcoding of 19 with MAX_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

---
 kernel/locking/locktorture.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

kernel/locking/locktorture.c +1 -1

index f26b1a1..23343be 100644
@@ -216,7 +216,7 @@ static int lock_torture_writer(void *arg)
 	static DEFINE_TORTURE_RANDOM(rand);
 
 	VERBOSE_TOROUT_STRING("lock_torture_writer task started");
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 
 	do {
 		schedule_timeout_uninterruptible(1);

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:19 UTC #10 lore raw

PATCH v3 6/17 tools/mq_perf_tests: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

---
 tools/testing/selftests/mqueue/mq_perf_tests.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

tools/testing/selftests/mqueue/mq_perf_tests.c +2 -2

index 2fadd4b..09dce51 100644
@@ -527,9 +527,9 @@ void increase_limits(void)
 	while (try_set(max_msgsize, cur_max_msgsize += 1024))
 		;
 	cur_max_msgsize = get(max_msgsize);
-	if (setpriority(PRIO_PROCESS, 0, -20) != 0)
+	if (setpriority(PRIO_PROCESS, 0, MIN_NICE) != 0)
 		shutdown(2, "setpriority()", __LINE__);
-	cur_nice = -20;
+	cur_nice = MIN_NICE;
 }
 
 int main(int argc, char *argv[])

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #11 lore raw

PATCH v3 7/17 mm: Replace hardcoding of 19 with MAX_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Bob Liu <[email protected]>

cc "Aneesh Kumar K.V" <[email protected]>

cc "Kirill A. Shutemov" <[email protected]>

cc Mel Gorman <[email protected]>

cc Rik van Riel <[email protected]>

cc Andrew Morton <[email protected]>

---
 mm/huge_memory.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

mm/huge_memory.c +1 -1

index 1546655..dcdb6f9 100644
@@ -2803,7 +2803,7 @@ static int khugepaged(void *none)
 	struct mm_slot *mm_slot;
 
 	set_freezable();
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 
 	while (!kthread_should_stop()) {
 		khugepaged_do_scan();

Signature · 7 lines

1.8.2.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Dongsheng Yang

author

2014-03-11 10:16 UTC #12 lore raw

PATCH v3 8/17 ioprio: Add a macro named NICE_TO_IOPRIO.

In reply to Dongsheng Yang

As the task nice value is in [-20, 19] and the io priority is in [0, 7], and the convert method from niceval to ioprio is implemented with an opened code in task_nice_ioprio().

This patch move the implementation to a macro NICE_TO_IOPRIO, making it more readable and modular.

Signed-off-by Dongsheng Yang <[email protected]>

---
 include/linux/ioprio.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

include/linux/ioprio.h +6 -1

index beb9ce1..c0faa0b 100644
@@ -18,6 +18,11 @@
 #define ioprio_valid(mask)	(IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE)
 
 /*
+ * Convert the nice value [19,-20] to io priority value [0,7].
+ */
+#define NICE_TO_IOPRIO(nice)	(nice_to_rlimit(nice) / 5)
+
+/*
  * These are the io priority groups as implemented by CFQ. RT is the realtime
  * class, it always gets premium service. BE is the best-effort scheduling
  * class, the default for any process. IDLE is the idle scheduling class, it

@@ -52,7 +57,7 @@ enum {
  */
 static inline int task_nice_ioprio(struct task_struct *task)
 {
-	return (task_nice(task) + 20) / 5;
+	return NICE_TO_IOPRIO(task_nice(task));
 }
 
 /*

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-12 10:32 UTC #13 lore raw

Re: [PATCH 08/17] ioprio: Add a macro named NICE_TO_IOPRIO.

Quoted 17 lines · Dongsheng Yang

As the task nice value is in [-20, 19] and the io priority is in [0, 7], and the convert method from niceval to ioprio is implemented with an opened code in task_nice_ioprio().
This patch move the implementation to a macro NICE_TO_IOPRIO, making it more readable and modular.
Signed-off-by Dongsheng Yang <[email protected]>
---
  include/linux/ioprio.h | 7 ++++++-
  1 file changed, 6 insertions(+), 1 deletion(-)
include/linux/ioprio.h +0 -0
index beb9ce1..c0faa0b 100644

include/linux/ioprio.h line 24

Quoted 7 lines

include/linux/ioprio.h +4 -0

@@ -18,6 +18,11 @@
  #define ioprio_valid(mask)	(IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE)
 
  /*
+ * Convert the nice value [19,-20] to io priority value [0,7].
+ */
+#define NICE_TO_IOPRIO(nice)	(nice_to_rlimit(nice) / 5)
+

Hi Peter,
As you asked, why I added this macro in ioprio.h? And why it is better?

IMO, there is a policy to convert a nice value to io priority, that
divide the nice
range into 8 parts by 5. I think this is a common requirement for lots of io subsystem. So I implement this policy with NICE_TO_IOPRIO.

About function task_nice_iopiro(), it is to get the iopriority of a
task. It should get the
nice value at first and then use the NICE_TO_IOPRIO to convert it to ioprio in the policy.

When our policy changed, we need only update the center controller, NICE_TO_IOPRIO. Keep the others work well.

This is my point about this patch. I wish it explained better. Thanx.

Quoted 13 lines

include/linux/ioprio.h +2 -1

+/*
   * These are the io priority groups as implemented by CFQ. RT is the realtime
   * class, it always gets premium service. BE is the best-effort scheduling
   * class, the default for any process. IDLE is the idle scheduling class, it

@@ -52,7 +57,7 @@ enum {
   */
  static inline int task_nice_ioprio(struct task_struct *task)
  {
-	return (task_nice(task) + 20) / 5;
+	return NICE_TO_IOPRIO(task_nice(task));
  }
 
  /*

Peter Zijlstra

maintainer

2014-03-19 08:27 UTC #14 lore raw

Re: [PATCH 08/17] ioprio: Add a macro named NICE_TO_IOPRIO.

In reply to Dongsheng Yang

Quoted 29 lines · Dongsheng Yang

As the task nice value is in [-20, 19] and the io priority is in [0, 7], and the convert method from niceval to ioprio is implemented with an opened code in task_nice_ioprio().
This patch move the implementation to a macro NICE_TO_IOPRIO, making it more readable and modular.
Signed-off-by Dongsheng Yang <[email protected]>
---
 include/linux/ioprio.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
include/linux/ioprio.h +5 -0
index beb9ce1..c0faa0b 100644
@@ -18,6 +18,11 @@
 #define ioprio_valid(mask)	(IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE)
 
 /*
+ * Convert the nice value [19,-20] to io priority value [0,7].
+ */
+#define NICE_TO_IOPRIO(nice)	(nice_to_rlimit(nice) / 5)
+
+/*
  * These are the io priority groups as implemented by CFQ. RT is the realtime
  * class, it always gets premium service. BE is the best-effort scheduling
  * class, the default for any process. IDLE is the idle scheduling class, it

include/linux/ioprio.h enum { · line 61

Quoted 6 lines

include/linux/ioprio.h +1 -1

@@ -52,7 +57,7 @@ enum {
  */
 static inline int task_nice_ioprio(struct task_struct *task)
 {
-	return (task_nice(task) + 20) / 5;
+	return NICE_TO_IOPRIO(task_nice(task));
 }

I'm still not seeing the point of adding that macro; if you want more documentation; just add the comment.

Dongsheng Yang

author

2014-03-11 10:13 UTC #15 lore raw

PATCH v3 9/17 fs/hearbeat: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc ocfs2-devel at oss.oracle.com

cc Dong Fang <[email protected]>

---
 fs/ocfs2/cluster/heartbeat.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

fs/ocfs2/cluster/heartbeat.c +1 -1

index bf482df..7303929 100644
@@ -1107,7 +1107,7 @@ static int o2hb_thread(void *data)
 
 	mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread running\n");
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	/* Pin node */
 	o2nm_depend_this_node();

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:12 UTC #16 lore raw

PATCH v3 10/17 driver/block: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Wei Yongjun <[email protected]>

cc Tejun Heo <[email protected]>

---
 drivers/block/loop.c    | 2 +-
 drivers/block/nbd.c     | 2 +-
 drivers/block/pktcdvd.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

drivers/block/loop.c +1 -1

index 66e8c3b..c8bf270 100644
@@ -548,7 +548,7 @@ static int loop_thread(void *data)
 	struct loop_device *lo = data;
 	struct bio *bio;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (!kthread_should_stop() || !bio_list_empty(&lo->lo_bio_list)) {

drivers/block/nbd.c +1 -1

index 55298db..2a1f26b 100644
@@ -533,7 +533,7 @@ static int nbd_thread(void *data)
 	struct nbd_device *nbd = data;
 	struct request *req;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	while (!kthread_should_stop() || !list_empty(&nbd->waiting_queue)) {
 		/* wait for something to do */
 		wait_event_interruptible(nbd->waiting_wq,

drivers/block/pktcdvd.c +1 -1

index a2af73d..ef166ad 100644
@@ -1463,7 +1463,7 @@ static int kcdrwd(void *foobar)
 	struct packet_data *pkt;
 	long min_sleep_time, residue;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_freezable();
 
 	for (;;) {

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #17 lore raw

PATCH v3 11/17 driver/char: Replace hardcoding of 19 with MAX_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Corey Minyard <[email protected]>

---
 drivers/char/ipmi/ipmi_si_intf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/char/ipmi/ipmi_si_intf.c +1 -1

index 03f4189..03c0eed 100644
@@ -999,7 +999,7 @@ static int ipmi_thread(void *data)
 	struct timespec busy_until;
 
 	ipmi_si_set_not_busy(&busy_until);
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 	while (!kthread_should_stop()) {
 		int busy_wait;

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:09 UTC #18 lore raw

PATCH v3 12/17 drivers/s390: Replace hardcoding of 19 with MAX_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Heiko Carstens <[email protected]>

cc Martin Schwidefsky <[email protected]>

cc Ingo Tuchscherer <[email protected]>

---
 drivers/s390/crypto/ap_bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/s390/crypto/ap_bus.c +1 -1

index ab3baa7..8eec165 100644
@@ -1803,7 +1803,7 @@ static int ap_poll_thread(void *data)
 	int requests;
 	struct ap_device *ap_dev;
 
-	set_user_nice(current, 19);
+	set_user_nice(current, MAX_NICE);
 	while (1) {
 		if (ap_suspend_flag)
 			return 0;

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:15 UTC #19 lore raw

PATCH v3 13/17 sched/prio: Add an inline function named rlimit_to_nice in prio.h.

In reply to Dongsheng Yang

This patch add an inline function named rlimit_to_nice to convert rlimit style value in [1, 40] to nice value in [-20, 19].

Signed-off-by Dongsheng Yang <[email protected]>

---
 include/linux/sched/prio.h | 8 ++++++++
 1 file changed, 8 insertions(+)

include/linux/sched/prio.h +8 -0

index 29756f8..d9cf5a5 100644
@@ -49,4 +49,12 @@ static inline long nice_to_rlimit(long nice)
 	return (MAX_NICE - nice + 1);
 }
 
+/*
+ * Convert rlimit style value [1,40] to nice value [-20, 19].
+ */
+static inline long rlimit_to_nice(long prio)
+{
+	return (MAX_NICE - prio + 1);
+}
+
 #endif /* _SCHED_PRIO_H */

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:12 UTC #20 lore raw

PATCH v3 14/17 driver/staging/android: Use rlimit_to_nice to replace opened code implementation.

In reply to Dongsheng Yang

There is a macro rlimit_to_nice in linux/sched/prio.h to convert priority in rlimit to nice value.

This patch replace the opened implementation with rlimit_to_nice.

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Masanari Iida <[email protected]>

cc Bojan Prtvar <[email protected]>

cc Serban Constantinescu <[email protected]>

Signed-off-by Dongsheng Yang <[email protected]>

---
 drivers/staging/android/binder.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/staging/android/binder.c +2 -2

index 1432d95..19e844b 100644
@@ -436,12 +436,12 @@ static void binder_set_nice(long nice)
 		set_user_nice(current, nice);
 		return;
 	}
-	min_nice = 20 - current->signal->rlim[RLIMIT_NICE].rlim_cur;
+	min_nice = rlimit_to_nice(current->signal->rlim[RLIMIT_NICE].rlim_cur);
 	binder_debug(BINDER_DEBUG_PRIORITY_CAP,
 		     "%d: nice value %ld not allowed use %ld instead\n",
 		      current->pid, nice, min_nice);
 	set_user_nice(current, min_nice);
-	if (min_nice < 20)
+	if (min_nice <= MAX_NICE)
 		return;
 	binder_user_error("%d RLIMIT_NICE not set\n", current->pid);
 }

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #21 lore raw

PATCH v3 15/17 driver/staging/lustre: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc Xiong Zhou <[email protected]>

cc Andreas Dilger <[email protected]>

cc Cyril Roelandt <[email protected]>

cc Kent Overstreet <[email protected]>

cc Peng Tao <[email protected]>

---
 drivers/staging/lustre/lustre/llite/lloop.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/staging/lustre/lustre/llite/lloop.c +1 -1

index 0718905..e40fc84 100644
@@ -407,7 +407,7 @@ static int loop_thread(void *data)
 	int refcheck;
 	int ret = 0;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	lo->lo_state = LLOOP_BOUND;

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:11 UTC #22 lore raw

PATCH v3 16/17 driver/scsi: Replace hardcoding of -20 with MIN_NICE.

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

cc [email protected]

cc James Smart <[email protected]>

cc Robert Jennings <[email protected]>

cc Robert Love <[email protected]>

---
 drivers/scsi/bnx2fc/bnx2fc_fcoe.c | 4 ++--
 drivers/scsi/bnx2i/bnx2i_hwi.c    | 2 +-
 drivers/scsi/fcoe/fcoe.c          | 2 +-
 drivers/scsi/ibmvscsi/ibmvfc.c    | 2 +-
 drivers/scsi/ibmvscsi/ibmvscsi.c  | 2 +-
 drivers/scsi/lpfc/lpfc_hbadisc.c  | 2 +-
 drivers/scsi/qla2xxx/qla_os.c     | 2 +-
 7 files changed, 8 insertions(+), 8 deletions(-)

drivers/scsi/bnx2fc/bnx2fc_fcoe.c +2 -2

index 9b94850..d24067a 100644
@@ -464,7 +464,7 @@ static int bnx2fc_l2_rcv_thread(void *arg)
 	struct fcoe_percpu_s *bg = arg;
 	struct sk_buff *skb;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {
 		schedule();

@@ -602,7 +602,7 @@ int bnx2fc_percpu_io_thread(void *arg)
 	struct bnx2fc_work *work, *tmp;
 	LIST_HEAD(work_list);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {
 		schedule();

drivers/scsi/bnx2i/bnx2i_hwi.c +1 -1

index e4cf23d..9aaa325 100644
@@ -1870,7 +1870,7 @@ int bnx2i_percpu_io_thread(void *arg)
 	struct bnx2i_work *work, *tmp;
 	LIST_HEAD(work_list);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (!kthread_should_stop()) {
 		spin_lock_bh(&p->p_work_lock);

drivers/scsi/fcoe/fcoe.c +1 -1

index f317000..843a679 100644
@@ -1872,7 +1872,7 @@ static int fcoe_percpu_receive_thread(void *arg)
 
 	skb_queue_head_init(&tmp);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 retry:
 	while (!kthread_should_stop()) {

drivers/scsi/ibmvscsi/ibmvfc.c +1 -1

index 23f5ba5..8dd4768 100644
@@ -4515,7 +4515,7 @@ static int ibmvfc_work(void *data)
 	struct ibmvfc_host *vhost = data;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (1) {
 		rc = wait_event_interruptible(vhost->work_wait_q,

drivers/scsi/ibmvscsi/ibmvscsi.c +1 -1

index fa76440..2ebfb2b 100644
@@ -2213,7 +2213,7 @@ static int ibmvscsi_work(void *data)
 	struct ibmvscsi_host_data *hostdata = data;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	while (1) {
 		rc = wait_event_interruptible(hostdata->work_wait_q,

drivers/scsi/lpfc/lpfc_hbadisc.c +1 -1

index 883ea2d..1f8f4a7 100644
@@ -733,7 +733,7 @@ lpfc_do_work(void *p)
 	struct lpfc_hba *phba = p;
 	int rc;
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 	current->flags |= PF_NOFREEZE;
 	phba->data_flags = 0;

drivers/scsi/qla2xxx/qla_os.c +1 -1

index 89a5300..0e45568 100644
@@ -4756,7 +4756,7 @@ qla2x00_do_dpc(void *data)
 	ha = (struct qla_hw_data *)data;
 	base_vha = pci_get_drvdata(ha->pdev);
 
-	set_user_nice(current, -20);
+	set_user_nice(current, MIN_NICE);
 
 	set_current_state(TASK_INTERRUPTIBLE);
 	while (!kthread_should_stop()) {

Signature · 1 line

1.8.2.1

Dongsheng Yang

author

2014-03-11 10:14 UTC #23 lore raw

PATCH v3 17/17 sched: Get rid of opened code implementation of funtion nice_to_rlimit().

In reply to Dongsheng Yang

Signed-off-by Dongsheng Yang <[email protected]>

---
 kernel/sched/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

kernel/sched/core.c +1 -1

index ee8004c..d2735eb 100644
@@ -3046,7 +3046,7 @@ EXPORT_SYMBOL(set_user_nice);
 int can_nice(const struct task_struct *p, const int nice)
 {
 	/* convert nice value [19,-20] to rlimit style value [1,40] */
-	int nice_rlim = 20 - nice;
+	int nice_rlim = nice_to_rlimit(nice);
 
 	return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) ||
 		capable(CAP_SYS_NICE));

Signature · 1 line

1.8.2.1

Ingo Molnar

2014-04-18 10:01 UTC #24 lore raw

Re: [PATCH 17/17] sched: Get rid of opened code implementation of funtion nice_to_rlimit().

So the patch title has two typos and a superfluous period.

I really don't want to apply patches that only get written but don't get read...

Thanks,

Ingo

Dongsheng Yang

author

2014-04-18 10:15 UTC #25 lore raw

Re: [PATCH 17/17] sched: Get rid of opened code implementation of funtion nice_to_rlimit().

Quoted 1 line · Ingo Molnar

So the patch title has two typos and a superfluous period.

Really sorry for that. :( I will update it right now.

Quoted 2 lines

I really don't want to apply patches that only get written but don't get read...

Spelling double-check was added in my checklist, wish it can make my work better. :)

Dongsheng

Quoted 5 lines

Thanks,
Ingo

Ingo Molnar

2014-04-18 10:19 UTC #26 lore raw

Re: [PATCH 17/17] sched: Get rid of opened code implementation of funtion nice_to_rlimit().

Quoted 4 lines · Dongsheng Yang

Quoted 1 line · Ingo Molnar
So the patch title has two typos and a superfluous period.
Really sorry for that. :( I will update it right now.

Please wait for the next -tip update before you resend the series - I've applied a handful of other patches from you, which will reduce the size of your queue.

Thanks,

Ingo

Dongsheng Yang

author

2014-04-18 10:24 UTC #27 lore raw

Re: [PATCH 17/17] sched: Get rid of opened code implementation of funtion nice_to_rlimit().

Quoted 8 lines · Ingo Molnar

Quoted 3 lines · Dongsheng Yang
Quoted 1 line · Ingo Molnar
So the patch title has two typos and a superfluous period.
Really sorry for that. :( I will update it right now.
Please wait for the next -tip update before you resend the series - I've applied a handful of other patches from you, which will reduce the size of your queue.

Okey, thanx !

Quoted 5 lines

Thanks,
Ingo

Dongsheng Yang

author

2014-03-12 10:32 UTC #28 lore raw

In reply to Dongsheng Yang

Hi Peter,
Let me post more explanation about this patchset.
This series of patches here is to git rid of hardcoding of nice value and collect the logic control in prio.h.

Currently, in tip tree, we can get 36 results with the below command:

# git grep -n -w -E "19|20"|grep -P "_nice" -i
drivers/block/loop.c:551: set_user_nice(current, -20);
drivers/block/nbd.c:536: set_user_nice(current, -20);
drivers/block/pktcdvd.c:1466: set_user_nice(current, -20);
drivers/char/ipmi/ipmi_si_intf.c:1002: set_user_nice(current, 19);
drivers/s390/crypto/ap_bus.c:1806: set_user_nice(current, 19);
drivers/scsi/bnx2fc/bnx2fc_fcoe.c:467: set_user_nice(current, -20);
drivers/scsi/bnx2fc/bnx2fc_fcoe.c:605: set_user_nice(current, -20);
drivers/scsi/bnx2i/bnx2i_hwi.c:1873: set_user_nice(current, -20);
drivers/scsi/fcoe/fcoe.c:1875: set_user_nice(current, -20);
drivers/scsi/ibmvscsi/ibmvfc.c:4518: set_user_nice(current, -20);
drivers/scsi/ibmvscsi/ibmvscsi.c:2216: set_user_nice(current, -20);
drivers/scsi/lpfc/lpfc_hbadisc.c:736: set_user_nice(current, -20);
drivers/scsi/qla2xxx/qla_os.c:4759: set_user_nice(current, -20);
drivers/staging/android/binder.c:439: min_nice = 20 - current->signal->rlim[RLIMIT_NICE].rlim_cur;
drivers/staging/android/binder.c:444: if (min_nice < 20)
drivers/staging/lustre/lustre/llite/lloop.c:410: set_user_nice(current, -20);
fs/ocfs2/cluster/heartbeat.c:1110: set_user_nice(current, -20);
include/linux/ioprio.h:55: return (task_nice(task) + 20) / 5;
kernel/rcu/torture.c:808: set_user_nice(current, 19);
kernel/rcu/torture.c:874: set_user_nice(current, 19);
kernel/rcu/torture.c:990: set_user_nice(current, 19);
kernel/rcu/torture.c:1587: set_user_nice(current, 19);
kernel/sched/core.c:3001: if (TASK_NICE(p) == nice || nice < -20 || nice > 19)
kernel/sched/core.c:3618: attr->sched_nice = clamp(attr->sched_nice, -20, 19);
kernel/sched/sched.h:32:#define PRIO_TO_NICE(prio) ((prio) - MAX_RT_PRIO - 20)
kernel/sys.c:253: niceval = 20 - task_nice(p);
kernel/sys.c:264: niceval = 20 - task_nice(p);
kernel/sys.c:280: niceval = 20 - task_nice(p);
kernel/trace/ring_buffer_benchmark.c:43:static int producer_nice = 19;
kernel/trace/ring_buffer_benchmark.c:44:static int consumer_nice = 19;
kernel/trace/ring_buffer_benchmark.c:311: producer_nice == 19 && consumer_nice == 19)
kernel/workqueue.c:105: RESCUER_NICE_LEVEL = -20,
kernel/workqueue.c:106: HIGHPRI_NICE_LEVEL = -20,
mm/huge_memory.c:2806: set_user_nice(current, 19);
tools/testing/selftests/mqueue/mq_perf_tests.c:532: cur_nice = -20;

It means there are 36 *men* who can decide the max_nice or min_nice.

This patchset collect this power back to prio.h, then the other subsystems should use the macros in prio.h, rather than decide it by themselves.

In addition, we use macro such as NICE_TO_IOPRIO to replace a expression “(task_nice(task) + 20) / 5”, I believe the name of macro means better.

After this patchset applied:

# git grep -n -w -E "19|20"|grep -P "_nice" -i
include/linux/sched/prio.h:4:#define MAX_NICE 19
include/linux/sched/prio.h:5:#define MIN_NICE -20

It looks more safe and more extensibal.

PS: Peter, I see you are busy and this series of patches seems really trivial. I wish you can give me your opinion about it when you have a time. No rush at all.

Thanx. :)

Quoted 67 lines · Dongsheng Yang

Hi all,
This patchset is all about priority.
1. Add two inline functions in prio.h named nice_to_rlimit and rlimit_to_nice. They are converting the value between nice value [-20, 19] and rlimit style value [1, 40].
2. Add a macro in ioprio.h named NICE_TO_IOPRIO.
    It convert nice value [-20, 19] to io priority [0, 7].
3. Others are all about replace hardcoding value about nice to MIN_NICE or MAX_NICE.
Changelog:
-v2:
*Splite [2/16] into two patches, one is about the whitespace changes, and
the other is about nice_to_rlimit().
-v1:
*Implement nice_to_rlimit and rlimit_to_nice with inline functions
rather than macro.
*Add a patch[16/16] to replace opened code implement with nice_to_rlimit().
Dongsheng Yang (16):
sched/prio: Add a inline function named nice_to_rlimit() in prio.h. kernel/sys: Replace opened code implementation with nice_to_rlimit().
workqueue: Replace hardcoding of -20 with MIN_NICE.
locktorture: Replace hardcoding of 19 with MAX_NICE. tools/mq_perf_tests: Replace hardcoding of -20 with MIN_NICE.
mm: Replace hardcoding of 19 with MAX_NICE.
ioprio: Add a macro named NICE_TO_IOPRIO.
fs/hearbeat: Replace hardcoding of -20 with MIN_NICE. driver/block: Replace hardcoding of -20 with MIN_NICE. driver/char: Replace hardcoding of 19 with MAX_NICE. drivers/s390: Replace hardcoding of 19 with MAX_NICE. sched/prio: Add an inline function named rlimit_to_nice in prio.h. driver/staging/android: Use rlimit_to_nice to replace opened code
implementation.
driver/staging/lustre: Replace hardcoding of -20 with MIN_NICE. driver/scsi: Replace hardcoding of -20 with MIN_NICE.
sched: Get rid of opened code implementation of funtion
     nice_to_rlimit().
Joe Perches (1):
kernel/sys: Fix the indent issue in switch.
  drivers/block/loop.c                           |   2 +-
  drivers/block/nbd.c                            |   2 +-
  drivers/block/pktcdvd.c                        |   2 +-
  drivers/char/ipmi/ipmi_si_intf.c               |   2 +-
  drivers/s390/crypto/ap_bus.c                   |   2 +-
  drivers/scsi/bnx2fc/bnx2fc_fcoe.c              |   4 +-
  drivers/scsi/bnx2i/bnx2i_hwi.c                 |   2 +-
  drivers/scsi/fcoe/fcoe.c                       |   2 +-
  drivers/scsi/ibmvscsi/ibmvfc.c                 |   2 +-
  drivers/scsi/ibmvscsi/ibmvscsi.c               |   2 +-
  drivers/scsi/lpfc/lpfc_hbadisc.c               |   2 +-
  drivers/scsi/qla2xxx/qla_os.c                  |   2 +-
  drivers/staging/android/binder.c               |   4 +-
  drivers/staging/lustre/lustre/llite/lloop.c    |   2 +-
  fs/ocfs2/cluster/heartbeat.c                   |   2 +-
  include/linux/ioprio.h                         |   7 +-
  include/linux/sched/prio.h                     |  16 ++
  kernel/locking/locktorture.c                   |   2 +-
  kernel/sched/core.c                            |   2 +-
  kernel/sys.c                                   | 206 ++++++++++++-------------
  kernel/workqueue.c                             |   6 +-
  mm/huge_memory.c                               |   2 +-
  tools/testing/selftests/mqueue/mq_perf_tests.c |   4 +-
  23 files changed, 150 insertions(+), 129 deletions(-)

Peter Zijlstra

maintainer

2014-03-19 08:36 UTC #29 lore raw

In reply to Dongsheng Yang

I picked up all except:

2 - whcih I don't think should go through a scheduler tree
6 - its tools/ bits, the changelog wasn't clear if the MIN/MAX_NICE
macros were uapi visible and I couldn't be bothered to figure it out.
8 - I don't like the pointless macro

Dongsheng Yang

author

2014-03-19 09:23 UTC #30 lore raw

Hi Peter,

Quoted 3 lines · Peter Zijlstra

I picked up all except:
2 - whcih I don't think should go through a scheduler tree

I will send it to sys maintainers.

Quoted 2 lines

6 - its tools/ bits, the changelog wasn't clear if the MIN/MAX_NICE
macros were uapi visible and I couldn't be bothered to figure it out.

Wow, yes it is not visible.

Quoted 1 line

8 - I don't like the pointless macro

Okey, Thanx Peter.

Quoted 7 lines

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Dongsheng Yang

author

2014-03-20 02:04 UTC #31 lore raw

In reply to Peter Zijlstra

Quoted 3 lines · Peter Zijlstra

I picked up all except:
2 - whcih I don't think should go through a scheduler tree

Peter, as you did not pick the patch 2, there will be a conflict when you apply [3/17]. I sent a v2 for [3/17], which does not depend to [2/17].

Please replace [3/17] with the V2 one.

Thanx.

Quoted 10 lines

6 - its tools/ bits, the changelog wasn't clear if the MIN/MAX_NICE
macros were uapi visible and I couldn't be bothered to figure it out.
8 - I don't like the pointless macro
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/