From 01341fbd0d8d4e717fc1231cdffe00343088ce0b Mon Sep 17 00:00:00 2001 From: Yunfeng Ye Date: Thu, 19 Nov 2020 14:21:25 +0800 Subject: workqueue: Kick a worker based on the actual activation of delayed works In realtime scenario, We do not want to have interference on the isolated cpu cores. but when invoking alloc_workqueue() for percpu wq on the housekeeping cpu, it kick a kworker on the isolated cpu. alloc_workqueue pwq_adjust_max_active wake_up_worker The comment in pwq_adjust_max_active() said: "Need to kick a worker after thawed or an unbound wq's max_active is bumped" So it is unnecessary to kick a kworker for percpu's wq when invoking alloc_workqueue(). this patch only kick a worker based on the actual activation of delayed works. Signed-off-by: Yunfeng Ye Reviewed-by: Lai Jiangshan Signed-off-by: Tejun Heo --- kernel/workqueue.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 437935e7a199..0695c7895c89 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3728,17 +3728,24 @@ static void pwq_adjust_max_active(struct pool_workqueue *pwq) * is updated and visible. */ if (!freezable || !workqueue_freezing) { + bool kick = false; + pwq->max_active = wq->saved_max_active; while (!list_empty(&pwq->delayed_works) && - pwq->nr_active < pwq->max_active) + pwq->nr_active < pwq->max_active) { pwq_activate_first_delayed(pwq); + kick = true; + } /* * Need to kick a worker after thawed or an unbound wq's - * max_active is bumped. It's a slow path. Do it always. + * max_active is bumped. In realtime scenarios, always kicking a + * worker will cause interference on the isolated cpu cores, so + * let's kick iff work items were activated. */ - wake_up_worker(pwq->pool); + if (kick) + wake_up_worker(pwq->pool); } else { pwq->max_active = 0; } -- cgit v1.3-14-g43fede From 58315c96651152b9f438e5e56c910994234e2c7a Mon Sep 17 00:00:00 2001 From: Bhaskar Chowdhury Date: Mon, 9 Nov 2020 16:01:11 +0530 Subject: kernel: cgroup: Mundane spelling fixes throughout the file Few spelling fixes throughout the file. Signed-off-by: Bhaskar Chowdhury Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index e41c21819ba0..690f477d537c 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -244,7 +244,7 @@ bool cgroup_ssid_enabled(int ssid) * * The default hierarchy is the v2 interface of cgroup and this function * can be used to test whether a cgroup is on the default hierarchy for - * cases where a subsystem should behave differnetly depending on the + * cases where a subsystem should behave differently depending on the * interface version. * * List of changed behaviors: @@ -262,7 +262,7 @@ bool cgroup_ssid_enabled(int ssid) * "cgroup.procs" instead. * * - "cgroup.procs" is not sorted. pids will be unique unless they got - * recycled inbetween reads. + * recycled in-between reads. * * - "release_agent" and "notify_on_release" are removed. Replacement * notification mechanism will be implemented. @@ -345,7 +345,7 @@ static bool cgroup_is_mixable(struct cgroup *cgrp) return !cgroup_parent(cgrp); } -/* can @cgrp become a thread root? should always be true for a thread root */ +/* can @cgrp become a thread root? Should always be true for a thread root */ static bool cgroup_can_be_thread_root(struct cgroup *cgrp) { /* mixables don't care */ @@ -530,7 +530,7 @@ static struct cgroup_subsys_state *cgroup_e_css_by_mask(struct cgroup *cgrp, * the root css is returned, so this function always returns a valid css. * * The returned css is not guaranteed to be online, and therefore it is the - * callers responsiblity to tryget a reference for it. + * callers responsibility to try get a reference for it. */ struct cgroup_subsys_state *cgroup_e_css(struct cgroup *cgrp, struct cgroup_subsys *ss) @@ -702,7 +702,7 @@ EXPORT_SYMBOL_GPL(of_css); ; \ else -/* walk live descendants in preorder */ +/* walk live descendants in pre order */ #define cgroup_for_each_live_descendant_pre(dsct, d_css, cgrp) \ css_for_each_descendant_pre((d_css), cgroup_css((cgrp), NULL)) \ if (({ lockdep_assert_held(&cgroup_mutex); \ @@ -936,7 +936,7 @@ void put_css_set_locked(struct css_set *cset) WARN_ON_ONCE(!list_empty(&cset->threaded_csets)); - /* This css_set is dead. unlink it and release cgroup and css refs */ + /* This css_set is dead. Unlink it and release cgroup and css refs */ for_each_subsys(ss, ssid) { list_del(&cset->e_cset_node[ssid]); css_put(cset->subsys[ssid]); @@ -1061,7 +1061,7 @@ static struct css_set *find_existing_css_set(struct css_set *old_cset, /* * Build the set of subsystem state objects that we want to see in the - * new css_set. while subsystems can change globally, the entries here + * new css_set. While subsystems can change globally, the entries here * won't change, so no need for locking. */ for_each_subsys(ss, i) { @@ -1151,7 +1151,7 @@ static void link_css_set(struct list_head *tmp_links, struct css_set *cset, /* * Always add links to the tail of the lists so that the lists are - * in choronological order. + * in chronological order. */ list_move_tail(&link->cset_link, &cgrp->cset_links); list_add_tail(&link->cgrp_link, &cset->cgrp_links); @@ -4137,7 +4137,7 @@ struct cgroup_subsys_state *css_next_child(struct cgroup_subsys_state *pos, * implies that if we observe !CSS_RELEASED on @pos in this RCU * critical section, the one pointed to by its next pointer is * guaranteed to not have finished its RCU grace period even if we - * have dropped rcu_read_lock() inbetween iterations. + * have dropped rcu_read_lock() in-between iterations. * * If @pos has CSS_RELEASED set, its next pointer can't be * dereferenced; however, as each css is given a monotonically @@ -4385,7 +4385,7 @@ static struct css_set *css_task_iter_next_css_set(struct css_task_iter *it) } /** - * css_task_iter_advance_css_set - advance a task itererator to the next css_set + * css_task_iter_advance_css_set - advance a task iterator to the next css_set * @it: the iterator to advance * * Advance @it to the next css_set to walk. @@ -6320,7 +6320,7 @@ struct cgroup_subsys_state *css_from_id(int id, struct cgroup_subsys *ss) * * Find the cgroup at @path on the default hierarchy, increment its * reference count and return it. Returns pointer to the found cgroup on - * success, ERR_PTR(-ENOENT) if @path doens't exist and ERR_PTR(-ENOTDIR) + * success, ERR_PTR(-ENOENT) if @path doesn't exist and ERR_PTR(-ENOTDIR) * if @path points to a non-directory. */ struct cgroup *cgroup_get_from_path(const char *path) -- cgit v1.3-14-g43fede From 5a7b5f32c5aa628841502d19a813c633ff6ecbe4 Mon Sep 17 00:00:00 2001 From: Hui Su Date: Fri, 6 Nov 2020 22:47:40 +0800 Subject: cgroup/cgroup.c: replace 'of->kn->priv' with of_cft() we have supplied the inline function: of_cft() in cgroup.h. So replace the direct use 'of->kn->priv' with inline func of_cft(), which is more readable. Signed-off-by: Hui Su Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 690f477d537c..385f80cf7d0c 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -3657,7 +3657,7 @@ static ssize_t cgroup_freeze_write(struct kernfs_open_file *of, static int cgroup_file_open(struct kernfs_open_file *of) { - struct cftype *cft = of->kn->priv; + struct cftype *cft = of_cft(of); if (cft->open) return cft->open(of); @@ -3666,7 +3666,7 @@ static int cgroup_file_open(struct kernfs_open_file *of) static void cgroup_file_release(struct kernfs_open_file *of) { - struct cftype *cft = of->kn->priv; + struct cftype *cft = of_cft(of); if (cft->release) cft->release(of); @@ -3677,7 +3677,7 @@ static ssize_t cgroup_file_write(struct kernfs_open_file *of, char *buf, { struct cgroup_namespace *ns = current->nsproxy->cgroup_ns; struct cgroup *cgrp = of->kn->parent->priv; - struct cftype *cft = of->kn->priv; + struct cftype *cft = of_cft(of); struct cgroup_subsys_state *css; int ret; @@ -3727,7 +3727,7 @@ static ssize_t cgroup_file_write(struct kernfs_open_file *of, char *buf, static __poll_t cgroup_file_poll(struct kernfs_open_file *of, poll_table *pt) { - struct cftype *cft = of->kn->priv; + struct cftype *cft = of_cft(of); if (cft->poll) return cft->poll(of, pt); -- cgit v1.3-14-g43fede From 2d18e54dd8662442ef5898c6bdadeaf90b3cebbc Mon Sep 17 00:00:00 2001 From: Qinglang Miao Date: Thu, 10 Dec 2020 09:29:43 +0800 Subject: cgroup: Fix memory leak when parsing multiple source parameters A memory leak is found in cgroup1_parse_param() when multiple source parameters overwrite fc->source in the fs_context struct without free. unreferenced object 0xffff888100d930e0 (size 16): comm "mount", pid 520, jiffies 4303326831 (age 152.783s) hex dump (first 16 bytes): 74 65 73 74 6c 65 61 6b 00 00 00 00 00 00 00 00 testleak........ backtrace: [<000000003e5023ec>] kmemdup_nul+0x2d/0xa0 [<00000000377dbdaa>] vfs_parse_fs_string+0xc0/0x150 [<00000000cb2b4882>] generic_parse_monolithic+0x15a/0x1d0 [<000000000f750198>] path_mount+0xee1/0x1820 [<0000000004756de2>] do_mount+0xea/0x100 [<0000000094cafb0a>] __x64_sys_mount+0x14b/0x1f0 Fix this bug by permitting a single source parameter and rejecting with an error all subsequent ones. Fixes: 8d2451f4994f ("cgroup1: switch to option-by-option parsing") Reported-by: Hulk Robot Signed-off-by: Qinglang Miao Reviewed-by: Zefan Li Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup-v1.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c index 191c329e482a..32596fdbcd5b 100644 --- a/kernel/cgroup/cgroup-v1.c +++ b/kernel/cgroup/cgroup-v1.c @@ -908,6 +908,8 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param) opt = fs_parse(fc, cgroup1_fs_parameters, param, &result); if (opt == -ENOPARAM) { if (strcmp(param->key, "source") == 0) { + if (fc->source) + return invalf(fc, "Multiple sources not supported"); fc->source = param->string; param->string = NULL; return 0; -- cgit v1.3-14-g43fede From b1b6b5a30dce872f500dc43f067cba8e7f86fc7d Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 30 Dec 2020 21:34:16 +0000 Subject: kernel/io_uring: cancel io_uring before task works For cancelling io_uring requests it needs either to be able to run currently enqueued task_works or having it shut down by that moment. Otherwise io_uring_cancel_files() may be waiting for requests that won't ever complete. Go with the first way and do cancellations before setting PF_EXITING and so before putting the task_work infrastructure into a transition state where task_work_run() would better not be called. Cc: stable@vger.kernel.org # 5.5+ Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/file.c | 2 -- kernel/exit.c | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/fs/file.c b/fs/file.c index c0b60961c672..dab120b71e44 100644 --- a/fs/file.c +++ b/fs/file.c @@ -21,7 +21,6 @@ #include #include #include -#include unsigned int sysctl_nr_open __read_mostly = 1024*1024; unsigned int sysctl_nr_open_min = BITS_PER_LONG; @@ -428,7 +427,6 @@ void exit_files(struct task_struct *tsk) struct files_struct * files = tsk->files; if (files) { - io_uring_files_cancel(files); task_lock(tsk); tsk->files = NULL; task_unlock(tsk); diff --git a/kernel/exit.c b/kernel/exit.c index 3594291a8542..04029e35e69a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -63,6 +63,7 @@ #include #include #include +#include #include #include @@ -776,6 +777,7 @@ void __noreturn do_exit(long code) schedule(); } + io_uring_files_cancel(tsk->files); exit_signals(tsk); /* sets PF_EXITING */ /* sync mm's RSS info before statistics gathering */ -- cgit v1.3-14-g43fede