The patch titled cpuset: avoid changing cpuset's mems when errno returned has been added to the -mm tree. Its filename is cpuset-avoid-changing-cpusets-mems-when-errno-returned.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: avoid changing cpuset's mems when errno returned From: Li Zefan <lizf@xxxxxxxxxxxxxx> When writing to cpuset.mems, cpuset has to update its mems_allowed before calling update_tasks_nodemask(), but this function might return -ENOMEM. To avoid this rare case, we allocate the memory before changing mems_allowed, and then pass to update_tasks_nodemask(). Similar to what update_cpumask() does. 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 | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff -puN kernel/cpuset.c~cpuset-avoid-changing-cpusets-mems-when-errno-returned kernel/cpuset.c --- a/kernel/cpuset.c~cpuset-avoid-changing-cpusets-mems-when-errno-returned +++ a/kernel/cpuset.c @@ -1057,13 +1057,15 @@ static void *cpuset_being_rebound; * update_tasks_nodemask - Update the nodemasks of tasks in the cpuset. * @cs: the cpuset in which each task's mems_allowed mask needs to be changed * @oldmem: old mems_allowed of cpuset cs + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() * * Called with cgroup_mutex held - * Return 0 if successful, -errno if not. + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. */ -static int update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem) +static void update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem, + struct ptr_heap *heap) { - int retval; struct cgroup_scanner scan; cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ @@ -1071,7 +1073,7 @@ static int update_tasks_nodemask(struct scan.cg = cs->css.cgroup; scan.test_task = NULL; scan.process_task = cpuset_change_nodemask; - scan.heap = NULL; + scan.heap = heap; scan.data = (nodemask_t *)oldmem; /* @@ -1084,12 +1086,10 @@ static int update_tasks_nodemask(struct * It's ok if we rebind the same mm twice; mpol_rebind_mm() * is idempotent. Also migrate pages in each mm to new nodes. */ - retval = cgroup_scan_tasks(&scan); + cgroup_scan_tasks(&scan); /* We're done rebinding vmas to this cpuset's new mems_allowed. */ cpuset_being_rebound = NULL; - - return retval; } /* @@ -1110,6 +1110,7 @@ static int update_nodemask(struct cpuset { nodemask_t oldmem; int retval; + struct ptr_heap heap; /* * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY]; @@ -1144,12 +1145,18 @@ static int update_nodemask(struct cpuset if (retval < 0) goto done; + retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); + if (retval < 0) + goto done; + mutex_lock(&callback_mutex); cs->mems_allowed = trialcs->mems_allowed; cs->mems_generation = cpuset_mems_generation++; mutex_unlock(&callback_mutex); - retval = update_tasks_nodemask(cs, &oldmem); + update_tasks_nodemask(cs, &oldmem, &heap); + + heap_free(&heap); done: return retval; } @@ -2002,7 +2009,7 @@ static void scan_for_empty_cpusets(struc remove_tasks_in_empty_cpuset(cp); else { update_tasks_cpumask(cp, NULL); - update_tasks_nodemask(cp, &oldmems); + update_tasks_nodemask(cp, &oldmems, NULL); } } } _ 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