The patch titled cpuset: rewrite update_tasks_nodemask() has been added to the -mm tree. Its filename is cpuset-rewrite-update_tasks_nodemask.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: cpuset: rewrite update_tasks_nodemask() From: Li Zefan <lizf@xxxxxxxxxxxxxx> This patch uses cgroup_scan_tasks() to rebind tasks' vmas to new cpuset's mems_allowed. Not only simplify the code largely, but also avoid allocating an array to hold mm pointers of all the tasks in the cpuset. This array can be big (size > PAGESIZE) if we have lots of tasks in that cpuset, thus has a chance to fail the allocation when under memory stress. Signed-off-by: Li Zefan <lizf@xxxxxxxxxxxxxx> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> Cc: Paul Menage <menage@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- kernel/cpuset.c | 109 ++++++++++++++++------------------------------ 1 file changed, 39 insertions(+), 70 deletions(-) diff -puN kernel/cpuset.c~cpuset-rewrite-update_tasks_nodemask kernel/cpuset.c --- a/kernel/cpuset.c~cpuset-rewrite-update_tasks_nodemask +++ a/kernel/cpuset.c @@ -1026,6 +1026,31 @@ static void cpuset_migrate_mm(struct mm_ mutex_unlock(&callback_mutex); } +/* + * Rebind task's vmas to cpuset's new mems_allowed, and migrate pages to new + * nodes if memory_migrate flag is set. Called with cgroup_mutex held. + */ +static void cpuset_change_nodemask(struct task_struct *p, + struct cgroup_scanner *scan) +{ + struct mm_struct *mm; + struct cpuset *cs; + int migrate; + const nodemask_t *oldmem = scan->data; + + mm = get_task_mm(p); + if (!mm) + return; + + cs = cgroup_cs(scan->cg); + migrate = is_memory_migrate(cs); + + mpol_rebind_mm(mm, &cs->mems_allowed); + if (migrate) + cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed); + mmput(mm); +} + static void *cpuset_being_rebound; /** @@ -1038,88 +1063,32 @@ static void *cpuset_being_rebound; */ static int update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem) { - struct task_struct *p; - struct mm_struct **mmarray; - int i, n, ntasks; - int migrate; - int fudge; - struct cgroup_iter it; int retval; + struct cgroup_scanner scan; cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ - fudge = 10; /* spare mmarray[] slots */ - fudge += cpumask_weight(cs->cpus_allowed);/* imagine 1 fork-bomb/cpu */ - retval = -ENOMEM; + scan.cg = cs->css.cgroup; + scan.test_task = NULL; + scan.process_task = cpuset_change_nodemask; + scan.heap = NULL; + scan.data = (nodemask_t *)oldmem; /* - * Allocate mmarray[] to hold mm reference for each task - * in cpuset cs. Can't kmalloc GFP_KERNEL while holding - * tasklist_lock. We could use GFP_ATOMIC, but with a - * few more lines of code, we can retry until we get a big - * enough mmarray[] w/o using GFP_ATOMIC. - */ - while (1) { - ntasks = cgroup_task_count(cs->css.cgroup); /* guess */ - ntasks += fudge; - mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL); - if (!mmarray) - goto done; - read_lock(&tasklist_lock); /* block fork */ - if (cgroup_task_count(cs->css.cgroup) <= ntasks) - break; /* got enough */ - read_unlock(&tasklist_lock); /* try again */ - kfree(mmarray); - } - - n = 0; - - /* Load up mmarray[] with mm reference for each task in cpuset. */ - cgroup_iter_start(cs->css.cgroup, &it); - while ((p = cgroup_iter_next(cs->css.cgroup, &it))) { - struct mm_struct *mm; - - if (n >= ntasks) { - printk(KERN_WARNING - "Cpuset mempolicy rebind incomplete.\n"); - break; - } - mm = get_task_mm(p); - if (!mm) - continue; - mmarray[n++] = mm; - } - cgroup_iter_end(cs->css.cgroup, &it); - read_unlock(&tasklist_lock); - - /* - * Now that we've dropped the tasklist spinlock, we can - * rebind the vma mempolicies of each mm in mmarray[] to their - * new cpuset, and release that mm. The mpol_rebind_mm() - * call takes mmap_sem, which we couldn't take while holding - * tasklist_lock. Forks can happen again now - the mpol_dup() - * cpuset_being_rebound check will catch such forks, and rebind - * their vma mempolicies too. Because we still hold the global - * cgroup_mutex, we know that no other rebind effort will - * be contending for the global variable cpuset_being_rebound. + * The mpol_rebind_mm() call takes mmap_sem, which we couldn't + * take while holding tasklist_lock. Forks can happen - the + * mpol_dup() cpuset_being_rebound check will catch such forks, + * and rebind their vma mempolicies too. Because we still hold + * the global cgroup_mutex, we know that no other rebind effort + * will be contending for the global variable cpuset_being_rebound. * It's ok if we rebind the same mm twice; mpol_rebind_mm() * is idempotent. Also migrate pages in each mm to new nodes. */ - migrate = is_memory_migrate(cs); - for (i = 0; i < n; i++) { - struct mm_struct *mm = mmarray[i]; - - mpol_rebind_mm(mm, &cs->mems_allowed); - if (migrate) - cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed); - mmput(mm); - } + retval = cgroup_scan_tasks(&scan); /* We're done rebinding vmas to this cpuset's new mems_allowed. */ - kfree(mmarray); cpuset_being_rebound = NULL; - retval = 0; -done: + return retval; } _ Patches currently in -mm which might be from lizf@xxxxxxxxxxxxxx are linux-next.patch vfs-add-missing-unlock-in-sget.patch relax-ns_can_attach-checks-to-allow-attaching-to-grandchild-cgroups.patch cgroup-css-id-support.patch cgroup-fix-frequent-ebusy-at-rmdir.patch cgroups-more-documentation-for-remount-and-release_agent.patch memcg-use-css-id.patch memcg-hierarchical-stat.patch memcg-fix-shrinking-memory-to-return-ebusy-by-fixing-retry-algorithm.patch memcg-fix-oom-killer-under-memcg.patch memcg-fix-oom-killer-under-memcg-fix2.patch memcg-fix-oom-killer-under-memcg-fix.patch memcg-show-memcg-information-during-oom.patch memcg-show-memcg-information-during-oom-fix2.patch memcg-show-memcg-information-during-oom-fix.patch memcg-show-memcg-information-during-oom-fix-fix.patch memcg-show-memcg-information-during-oom-fix-fix-checkpatch-fixes.patch use-css-id-in-swap_cgroup-for-saving-memory-v4.patch cpuset-fix-possible-races-in-cpu-memory-hotplug.patch cgroups-add-data-field-to-struct-cgroup_scanner.patch cpuset-rewrite-update_tasks_nodemask.patch cpuset-avoid-changing-cpusets-mems-when-errno-returned.patch cpuset-remove-struct-cpuset_hotplug_scanner.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html