X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=kernel%2Fcpuset.c;h=1f06e76901067ffa1c08d586d0128f64ad6d6c65;hb=a57ebc90f1350296edded12d33d7c278831bc3bf;hp=805fb9097318b761b60b98e8ad8f11a7d3756fcf;hpb=49302d0c42592b37f49ae96e0f06a3599cf5a8a0;p=linux-2.6

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 805fb90973..1f06e76901 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -398,21 +398,31 @@ static int cpuset_path(const struct cpuset *cs, char *buf, int buflen)
  * to continue to serve a useful existence.  Next time it's released,
  * we will get notified again, if it still has 'notify_on_release' set.
  *
- * Note final arg to call_usermodehelper() is 0 - that means
- * don't wait.  Since we are holding the global cpuset_sem here,
- * and we are asking another thread (started from keventd) to rmdir a
- * cpuset, we can't wait - or we'd deadlock with the removing thread
- * on cpuset_sem.
+ * The final arg to call_usermodehelper() is 0, which means don't
+ * wait.  The separate /sbin/cpuset_release_agent task is forked by
+ * call_usermodehelper(), then control in this thread returns here,
+ * without waiting for the release agent task.  We don't bother to
+ * wait because the caller of this routine has no use for the exit
+ * status of the /sbin/cpuset_release_agent task, so no sense holding
+ * our caller up for that.
+ *
+ * The simple act of forking that task might require more memory,
+ * which might need cpuset_sem.  So this routine must be called while
+ * cpuset_sem is not held, to avoid a possible deadlock.  See also
+ * comments for check_for_release(), below.
  */
 
-static int cpuset_release_agent(char *cpuset_str)
+static void cpuset_release_agent(const char *pathbuf)
 {
 	char *argv[3], *envp[3];
 	int i;
 
+	if (!pathbuf)
+		return;
+
 	i = 0;
 	argv[i++] = "/sbin/cpuset_release_agent";
-	argv[i++] = cpuset_str;
+	argv[i++] = (char *)pathbuf;
 	argv[i] = NULL;
 
 	i = 0;
@@ -421,17 +431,29 @@ static int cpuset_release_agent(char *cpuset_str)
 	envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin";
 	envp[i] = NULL;
 
-	return call_usermodehelper(argv[0], argv, envp, 0);
+	call_usermodehelper(argv[0], argv, envp, 0);
+	kfree(pathbuf);
 }
 
 /*
  * Either cs->count of using tasks transitioned to zero, or the
  * cs->children list of child cpusets just became empty.  If this
  * cs is notify_on_release() and now both the user count is zero and
- * the list of children is empty, send notice to user land.
+ * the list of children is empty, prepare cpuset path in a kmalloc'd
+ * buffer, to be returned via ppathbuf, so that the caller can invoke
+ * cpuset_release_agent() with it later on, once cpuset_sem is dropped.
+ * Call here with cpuset_sem held.
+ *
+ * This check_for_release() routine is responsible for kmalloc'ing
+ * pathbuf.  The above cpuset_release_agent() is responsible for
+ * kfree'ing pathbuf.  The caller of these routines is responsible
+ * for providing a pathbuf pointer, initialized to NULL, then
+ * calling check_for_release() with cpuset_sem held and the address
+ * of the pathbuf pointer, then dropping cpuset_sem, then calling
+ * cpuset_release_agent() with pathbuf, as set by check_for_release().
  */
 
-static void check_for_release(struct cpuset *cs)
+static void check_for_release(struct cpuset *cs, char **ppathbuf)
 {
 	if (notify_on_release(cs) && atomic_read(&cs->count) == 0 &&
 	    list_empty(&cs->children)) {
@@ -441,10 +463,9 @@ static void check_for_release(struct cpuset *cs)
 		if (!buf)
 			return;
 		if (cpuset_path(cs, buf, PAGE_SIZE) < 0)
-			goto out;
-		cpuset_release_agent(buf);
-out:
-		kfree(buf);
+			kfree(buf);
+		else
+			*ppathbuf = buf;
 	}
 }
 
@@ -606,6 +627,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial)
  * Call with cpuset_sem held.  May nest a call to the
  * lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
  */
+
 static void update_cpu_domains(struct cpuset *cur)
 {
 	struct cpuset *c, *par = cur->parent;
@@ -727,14 +749,14 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
 	return 0;
 }
 
-static int attach_task(struct cpuset *cs, char *buf)
+static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
 {
 	pid_t pid;
 	struct task_struct *tsk;
 	struct cpuset *oldcs;
 	cpumask_t cpus;
 
-	if (sscanf(buf, "%d", &pid) != 1)
+	if (sscanf(pidbuf, "%d", &pid) != 1)
 		return -EIO;
 	if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))
 		return -ENOSPC;
@@ -777,7 +799,7 @@ static int attach_task(struct cpuset *cs, char *buf)
 
 	put_task_struct(tsk);
 	if (atomic_dec_and_test(&oldcs->count))
-		check_for_release(oldcs);
+		check_for_release(oldcs, ppathbuf);
 	return 0;
 }
 
@@ -801,6 +823,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
 	struct cftype *cft = __d_cft(file->f_dentry);
 	cpuset_filetype_t type = cft->private;
 	char *buffer;
+	char *pathbuf = NULL;
 	int retval = 0;
 
 	/* Crude upper limit on largest legitimate cpulist user might write. */
@@ -841,7 +864,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
 		retval = update_flag(CS_NOTIFY_ON_RELEASE, cs, buffer);
 		break;
 	case FILE_TASKLIST:
-		retval = attach_task(cs, buffer);
+		retval = attach_task(cs, buffer, &pathbuf);
 		break;
 	default:
 		retval = -EINVAL;
@@ -852,6 +875,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
 		retval = nbytes;
 out2:
 	up(&cpuset_sem);
+	cpuset_release_agent(pathbuf);
 out1:
 	kfree(buffer);
 	return retval;
@@ -1357,6 +1381,7 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
 	struct cpuset *cs = dentry->d_fsdata;
 	struct dentry *d;
 	struct cpuset *parent;
+	char *pathbuf = NULL;
 
 	/* the vfs holds both inode->i_sem already */
 
@@ -1376,7 +1401,7 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
 		update_cpu_domains(cs);
 	list_del(&cs->sibling);	/* delete my sibling from parent->children */
 	if (list_empty(&parent->children))
-		check_for_release(parent);
+		check_for_release(parent, &pathbuf);
 	spin_lock(&cs->dentry->d_lock);
 	d = dget(cs->dentry);
 	cs->dentry = NULL;
@@ -1384,6 +1409,7 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
 	cpuset_d_remove_dir(d);
 	dput(d);
 	up(&cpuset_sem);
+	cpuset_release_agent(pathbuf);
 	return 0;
 }
 
@@ -1483,10 +1509,13 @@ void cpuset_exit(struct task_struct *tsk)
 	task_unlock(tsk);
 
 	if (notify_on_release(cs)) {
+		char *pathbuf = NULL;
+
 		down(&cpuset_sem);
 		if (atomic_dec_and_test(&cs->count))
-			check_for_release(cs);
+			check_for_release(cs, &pathbuf);
 		up(&cpuset_sem);
+		cpuset_release_agent(pathbuf);
 	} else {
 		atomic_dec(&cs->count);
 	}
@@ -1570,17 +1599,114 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl)
 	return 0;
 }
 
+/*
+ * nearest_exclusive_ancestor() - Returns the nearest mem_exclusive
+ * ancestor to the specified cpuset.  Call while holding cpuset_sem.
+ * If no ancestor is mem_exclusive (an unusual configuration), then
+ * returns the root cpuset.
+ */
+static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs)
+{
+	while (!is_mem_exclusive(cs) && cs->parent)
+		cs = cs->parent;
+	return cs;
+}
+
 /**
- * cpuset_zone_allowed - is zone z allowed in current->mems_allowed
- * @z: zone in question
+ * cpuset_zone_allowed - Can we allocate memory on zone z's memory node?
+ * @z: is this zone on an allowed node?
+ * @gfp_mask: memory allocation flags (we use __GFP_HARDWALL)
  *
- * Is zone z allowed in current->mems_allowed, or is
- * the CPU in interrupt context? (zone is always allowed in this case)
- */
-int cpuset_zone_allowed(struct zone *z)
+ * If we're in interrupt, yes, we can always allocate.  If zone
+ * z's node is in our tasks mems_allowed, yes.  If it's not a
+ * __GFP_HARDWALL request and this zone's nodes is in the nearest
+ * mem_exclusive cpuset ancestor to this tasks cpuset, yes.
+ * Otherwise, no.
+ *
+ * GFP_USER allocations are marked with the __GFP_HARDWALL bit,
+ * and do not allow allocations outside the current tasks cpuset.
+ * GFP_KERNEL allocations are not so marked, so can escape to the
+ * nearest mem_exclusive ancestor cpuset.
+ *
+ * Scanning up parent cpusets requires cpuset_sem.  The __alloc_pages()
+ * routine only calls here with __GFP_HARDWALL bit _not_ set if
+ * it's a GFP_KERNEL allocation, and all nodes in the current tasks
+ * mems_allowed came up empty on the first pass over the zonelist.
+ * So only GFP_KERNEL allocations, if all nodes in the cpuset are
+ * short of memory, might require taking the cpuset_sem semaphore.
+ *
+ * The first loop over the zonelist in mm/page_alloc.c:__alloc_pages()
+ * calls here with __GFP_HARDWALL always set in gfp_mask, enforcing
+ * hardwall cpusets - no allocation on a node outside the cpuset is
+ * allowed (unless in interrupt, of course).
+ *
+ * The second loop doesn't even call here for GFP_ATOMIC requests
+ * (if the __alloc_pages() local variable 'wait' is set).  That check
+ * and the checks below have the combined affect in the second loop of
+ * the __alloc_pages() routine that:
+ *	in_interrupt - any node ok (current task context irrelevant)
+ *	GFP_ATOMIC   - any node ok
+ *	GFP_KERNEL   - any node in enclosing mem_exclusive cpuset ok
+ *	GFP_USER     - only nodes in current tasks mems allowed ok.
+ **/
+
+int cpuset_zone_allowed(struct zone *z, unsigned int __nocast gfp_mask)
+{
+	int node;			/* node that zone z is on */
+	const struct cpuset *cs;	/* current cpuset ancestors */
+	int allowed = 1;		/* is allocation in zone z allowed? */
+
+	if (in_interrupt())
+		return 1;
+	node = z->zone_pgdat->node_id;
+	if (node_isset(node, current->mems_allowed))
+		return 1;
+	if (gfp_mask & __GFP_HARDWALL)	/* If hardwall request, stop here */
+		return 0;
+
+	/* Not hardwall and node outside mems_allowed: scan up cpusets */
+	down(&cpuset_sem);
+	cs = current->cpuset;
+	if (!cs)
+		goto done;		/* current task exiting */
+	cs = nearest_exclusive_ancestor(cs);
+	allowed = node_isset(node, cs->mems_allowed);
+done:
+	up(&cpuset_sem);
+	return allowed;
+}
+
+/**
+ * cpuset_excl_nodes_overlap - Do we overlap @p's mem_exclusive ancestors?
+ * @p: pointer to task_struct of some other task.
+ *
+ * Description: Return true if the nearest mem_exclusive ancestor
+ * cpusets of tasks @p and current overlap.  Used by oom killer to
+ * determine if task @p's memory usage might impact the memory
+ * available to the current task.
+ *
+ * Acquires cpuset_sem - not suitable for calling from a fast path.
+ **/
+
+int cpuset_excl_nodes_overlap(const struct task_struct *p)
 {
-	return in_interrupt() ||
-		node_isset(z->zone_pgdat->node_id, current->mems_allowed);
+	const struct cpuset *cs1, *cs2;	/* my and p's cpuset ancestors */
+	int overlap = 0;		/* do cpusets overlap? */
+
+	down(&cpuset_sem);
+	cs1 = current->cpuset;
+	if (!cs1)
+		goto done;		/* current task exiting */
+	cs2 = p->cpuset;
+	if (!cs2)
+		goto done;		/* task p is exiting */
+	cs1 = nearest_exclusive_ancestor(cs1);
+	cs2 = nearest_exclusive_ancestor(cs2);
+	overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed);
+done:
+	up(&cpuset_sem);
+
+	return overlap;
 }
 
 /*