psi: dont alloc memory for psi by default

Memory about struct psi_group is allocated by default for
each cgroup even if psi_disabled is true, in this case, these
allocated memory is waste, so alloc memory for struct psi_group
only when psi_disabled is false.

Signed-off-by: Chen Wandun <chenwandun@huawei.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Tejun Heo <tj@kernel.org>

diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index 1bfcfb1af3524f46388d240a4246218a6b042f6f..672de25e3ec83cb7b08b128d2f00ab3f8d322acb 100644 (file)
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -475,7 +475,7 @@ struct cgroup {
        struct work_struct release_agent_work;
 
        /* used to track pressure stalls */
-       struct psi_group psi;
+       struct psi_group *psi;
 
        /* used to store eBPF programs */
        struct cgroup_bpf bpf;

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 0d1ada8968d75ce833a2abb953d59ee912a845dc..ed53bfe7c46c4738df18744cc45ff6a204721e11 100644 (file)
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -674,7 +674,7 @@ static inline void pr_cont_cgroup_path(struct cgroup *cgrp)
 
 static inline struct psi_group *cgroup_psi(struct cgroup *cgrp)
 {
-       return &cgrp->psi;
+       return cgrp->psi;
 }
 
 bool cgroup_psi_enabled(void);

diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 1779ccddb734d00c53f6968c3445aaf5314aebf6..90a654cb8a1edfb8a632045317873d4b50fc85a0 100644 (file)
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -3609,21 +3609,21 @@ static int cpu_stat_show(struct seq_file *seq, void *v)
 static int cgroup_io_pressure_show(struct seq_file *seq, void *v)
 {
        struct cgroup *cgrp = seq_css(seq)->cgroup;
-       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
+       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : cgrp->psi;
 
        return psi_show(seq, psi, PSI_IO);
 }
 static int cgroup_memory_pressure_show(struct seq_file *seq, void *v)
 {
        struct cgroup *cgrp = seq_css(seq)->cgroup;
-       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
+       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : cgrp->psi;
 
        return psi_show(seq, psi, PSI_MEM);
 }
 static int cgroup_cpu_pressure_show(struct seq_file *seq, void *v)
 {
        struct cgroup *cgrp = seq_css(seq)->cgroup;
-       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
+       struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : cgrp->psi;
 
        return psi_show(seq, psi, PSI_CPU);
 }
@@ -3649,7 +3649,7 @@ static ssize_t cgroup_pressure_write(struct kernfs_open_file *of, char *buf,
                return -EBUSY;
        }
 
-       psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
+       psi = cgroup_ino(cgrp) == 1 ? &psi_system : cgrp->psi;
        new = psi_trigger_create(psi, buf, nbytes, res);
        if (IS_ERR(new)) {
                cgroup_put(cgrp);

diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index a337f3e3599733ce7ebe6d8576942f633edb1d41..ec66b40bdd403ba2b4e9d5bb8c4459fc2a0f5ef2 100644 (file)
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -957,10 +957,16 @@ int psi_cgroup_alloc(struct cgroup *cgroup)
        if (static_branch_likely(&psi_disabled))
                return 0;
 
-       cgroup->psi.pcpu = alloc_percpu(struct psi_group_cpu);
-       if (!cgroup->psi.pcpu)
+       cgroup->psi = kmalloc(sizeof(struct psi_group), GFP_KERNEL);
+       if (!cgroup->psi)
                return -ENOMEM;
-       group_init(&cgroup->psi);
+
+       cgroup->psi->pcpu = alloc_percpu(struct psi_group_cpu);
+       if (!cgroup->psi->pcpu) {
+               kfree(cgroup->psi);
+               return -ENOMEM;
+       }
+       group_init(cgroup->psi);
        return 0;
 }
 
@@ -969,10 +975,11 @@ void psi_cgroup_free(struct cgroup *cgroup)
        if (static_branch_likely(&psi_disabled))
                return;
 
-       cancel_delayed_work_sync(&cgroup->psi.avgs_work);
-       free_percpu(cgroup->psi.pcpu);
+       cancel_delayed_work_sync(&cgroup->psi->avgs_work);
+       free_percpu(cgroup->psi->pcpu);
        /* All triggers must be removed by now */
-       WARN_ONCE(cgroup->psi.poll_states, "psi: trigger leak\n");
+       WARN_ONCE(cgroup->psi->poll_states, "psi: trigger leak\n");
+       kfree(cgroup->psi);
 }
 
 /**