[PATCH] (1-2) governors for 60-bk

Ed Tomlinson (tomlins@cam.org)
Thu, 13 Feb 2003 08:18:24 -0500


Here is the patch for governors

ptg-C4

Ed Tomlinson

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
# ChangeSet 1.994 -> 1.995
# include/linux/sched.h 1.129 -> 1.130
# kernel/fork.c 1.104 -> 1.105
# kernel/user.c 1.6 -> 1.7
# kernel/sched.c 1.158 -> 1.159
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/02/12 ed@oscar.et.ca 1.995
# governors for thread groups and users
# --------------------------------------------
#
diff -Nru a/include/linux/sched.h b/include/linux/sched.h
--- a/include/linux/sched.h Wed Feb 12 12:13:01 2003
+++ b/include/linux/sched.h Wed Feb 12 12:13:01 2003
@@ -177,6 +177,11 @@

#include <linux/aio.h>

+struct ptg_struct { /* pseudo thread groups */
+ atomic_t active; /* number of tasks in run queues */
+ atomic_t count; /* number of refs */
+};
+
struct mm_struct {
struct vm_area_struct * mmap; /* list of VMAs */
struct rb_root mm_rb;
@@ -277,6 +282,7 @@
struct user_struct {
atomic_t __count; /* reference count */
atomic_t processes; /* How many processes does this user have? */
+ atomic_t active; /* How many active processes does this user have? */
atomic_t files; /* How many open files does this user have? */

/* Hash table maintenance information */
@@ -322,6 +328,8 @@
struct list_head ptrace_list;

struct mm_struct *mm, *active_mm;
+ struct ptg_struct * ptgroup; /* pseudo thread group for this task */
+ atomic_t *governor; /* the atomic_t that governs this task */

/* task state */
struct linux_binfmt *binfmt;
diff -Nru a/kernel/fork.c b/kernel/fork.c
--- a/kernel/fork.c Wed Feb 12 12:13:01 2003
+++ b/kernel/fork.c Wed Feb 12 12:13:01 2003
@@ -74,6 +74,14 @@

void __put_task_struct(struct task_struct *tsk)
{
+ if (tsk->ptgroup && atomic_sub_and_test(1,&tsk->ptgroup->count)) {
+ kfree(tsk->ptgroup);
+ tsk->ptgroup = NULL;
+ tsk->governor = &tsk->user->active;
+ if (tsk == current)
+ atomic_inc(tsk->governor);
+ }
chedule_tunables-C4
if (tsk != current) {
free_thread_info(tsk->thread_info);
kmem_cache_free(task_struct_cachep,tsk);
@@ -450,6 +458,7 @@

tsk->mm = NULL;
tsk->active_mm = NULL;
+ tsk->ptgroup = NULL;

/*
* Are we cloning a kernel thread?
@@ -715,6 +724,29 @@
p->flags = new_flags;
}

+static inline int setup_governor(unsigned long clone_flags, struct task_struct *p)
+{
+ if ( ((clone_flags & CLONE_VM) && (clone_flags & CLONE_FILES)) ||
+ (clone_flags & CLONE_THREAD)) {
+ if (current->ptgroup)
+ atomic_inc(&current->ptgroup->count);
+ else {
+ current->ptgroup = kmalloc(sizeof(struct ptg_struct), GFP_ATOMIC);
+ if (!current->ptgroup)
+ return 1;
+ /* printk(KERN_INFO "ptgroup - pid %u\n",current->pid); */
+ atomic_set(&current->ptgroup->count,2);
+ atomic_set(&current->ptgroup->active,1);
+ atomic_dec(current->governor);
+ current->governor = &current->ptgroup->active;
+ }
+ p->ptgroup = current->ptgroup;
+ p->governor = &p->ptgroup->active;
+ } else
+ p->governor = &p->user->active;
+ return 0;
+}
+
asmlinkage int sys_set_tid_address(int *tidptr)
{
current->clear_child_tid = tidptr;
@@ -855,6 +887,12 @@
goto bad_fork_cleanup_mm;
retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
if (retval)
+ goto bad_fork_cleanup_namespace;
+ /*
+ * Setup the governor pointer for the new process, allocating a new ptg as
+ * required if the process is a member of a thread group.
+ */
+ if (setup_governor(clone_flags, p))
goto bad_fork_cleanup_namespace;

if (clone_flags & CLONE_CHILD_SETTID)
diff -Nru a/kernel/sched.c b/kernel/sched.c
--- a/kernel/sched.c Wed Feb 12 12:13:01 2003
+++ b/kernel/sched.c Wed Feb 12 12:13:01 2003
@@ -62,6 +62,8 @@
#define MAX_TIMESLICE (300 * HZ / 1000)
#define CHILD_PENALTY 95
#define PARENT_PENALTY 100
+#define THREAD_GOVERNOR 15 /* allow threads groups 1.5 full timeslices */
+#define USER_GOVERNOR 300 /* allow user 30 full timeslices */
#define EXIT_WEIGHT 3
#define PRIO_BONUS_RATIO 25
#define INTERACTIVE_DELTA 2
@@ -121,9 +123,23 @@
#define BASE_TIMESLICE(p) (MIN_TIMESLICE + \
((MAX_TIMESLICE - MIN_TIMESLICE) * (MAX_PRIO-1-(p)->static_prio)/(MAX_USER_PRIO - 1)))

+static int next;
static inline unsigned int task_timeslice(task_t *p)
{
- return BASE_TIMESLICE(p);
+ int slice = BASE_TIMESLICE(p);
+ int threads = atomic_read(p->governor) * 10;
+ int govern = threads;
+ if (p->user->uid)
+ govern = (p->ptgroup) ? THREAD_GOVERNOR : USER_GOVERNOR;
+ if (threads > govern) {
+ slice = (slice * govern) / threads;
+ slice = (slice > MIN_TIMESLICE) ? slice : MIN_TIMESLICE;
+ }
+/* if (jiffies > next) {
+ printk(KERN_INFO "uid %d pid %d ptg %x gov %x threads %d lim %d slice %d\n",p->uid, p->pid, p->ptgroup, p->governor, threads/10, govern, slice);
+ next = jiffies + HZ*300;
+ } */
+ return slice;
}

/*
@@ -196,16 +212,18 @@
rq->node_nr_running = &node_nr_running[0];
}

-static inline void nr_running_inc(runqueue_t *rq)
+static inline void nr_running_inc(task_t *p, runqueue_t *rq)
{
atomic_inc(rq->node_nr_running);
rq->nr_running++;
+ atomic_inc(p->governor);
}

-static inline void nr_running_dec(runqueue_t *rq)
+static inline void nr_running_dec(task_t *p, runqueue_t *rq)
{
atomic_dec(rq->node_nr_running);
rq->nr_running--;
+ atomic_dec(p->governor);
}

__init void node_nr_running_init(void)
@@ -219,8 +237,8 @@
#else /* !CONFIG_NUMA */

# define nr_running_init(rq) do { } while (0)
-# define nr_running_inc(rq) do { (rq)->nr_running++; } while (0)
-# define nr_running_dec(rq) do { (rq)->nr_running--; } while (0)
+# define nr_running_inc(p, rq) do { (rq)->nr_running++; atomic_inc((p)->governor); } while (0)
+# define nr_running_dec(p, rq) do { (rq)->nr_running--; atomic_dec((p)->governor); } while (0)

#endif /* CONFIG_NUMA */

@@ -341,7 +359,7 @@
p->prio = effective_prio(p);
}
enqueue_task(p, array);
- nr_running_inc(rq);
+ nr_running_inc(p, rq);
}

/*
@@ -349,7 +367,7 @@
*/
static inline void deactivate_task(struct task_struct *p, runqueue_t *rq)
{
- nr_running_dec(rq);
+ nr_running_dec(p, rq);
if (p->state == TASK_UNINTERRUPTIBLE)
rq->nr_uninterruptible++;
dequeue_task(p, p->array);
@@ -910,9 +928,9 @@
static inline void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p, runqueue_t *this_rq, int this_cpu)
{
dequeue_task(p, src_array);
- nr_running_dec(src_rq);
+ nr_running_dec(p, src_rq);
set_task_cpu(p, this_cpu);
- nr_running_inc(this_rq);
+ nr_running_inc(p, this_rq);
enqueue_task(p, this_rq->active);
/*
* Note that idle threads have a prio of MAX_PRIO, for this test
@@ -2450,6 +2468,7 @@
rq->curr = current;
rq->idle = current;
set_task_cpu(current, smp_processor_id());
+ current->governor = &current->user->active;
wake_up_forked_process(current);

init_timers();
diff -Nru a/kernel/user.c b/kernel/user.c
--- a/kernel/user.c Wed Feb 12 12:13:01 2003
+++ b/kernel/user.c Wed Feb 12 12:13:01 2003
@@ -30,6 +30,7 @@
struct user_struct root_user = {
.__count = ATOMIC_INIT(1),
.processes = ATOMIC_INIT(1),
+ .active = ATOMIC_INIT(1),
.files = ATOMIC_INIT(0)
};

@@ -96,6 +97,7 @@
new->uid = uid;
atomic_set(&new->__count, 1);
atomic_set(&new->processes, 0);
+ atomic_set(&new->active, 0);
atomic_set(&new->files, 0);

/*
@@ -130,6 +132,11 @@
atomic_inc(&new_user->processes);
atomic_dec(&old_user->processes);
current->user = new_user;
+ if (!current->ptgroup) {
+ atomic_dec(current->governor);
+ current->governor = &current->user->active;
+ atomic_inc(current->governor);
+ }
free_uid(old_user);
}

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