X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=mm%2Foom_kill.c;h=41b4e362221d943ba1759d9de3d740222f067c98;hb=954ffcb35f5aca428661d29b96c4eee82b3c19cd;hp=bada3d03119ff01381d6d47c1b828b59c3867106;hpb=89fa30242facca249aead2aac03c4c69764f911c;p=safe%2Fjmp%2Flinux-2.6 diff --git a/mm/oom_kill.c b/mm/oom_kill.c index bada3d0..41b4e36 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -15,7 +15,9 @@ * kernel subsystems and hints as to where to find out what things do. */ +#include #include +#include #include #include #include @@ -60,12 +62,6 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) } /* - * swapoff can easily use up all memory, so kill those first. - */ - if (p->flags & PF_SWAPOFF) - return ULONG_MAX; - - /* * The memory size of the process is the basis for the badness. */ points = mm->total_vm; @@ -76,6 +72,12 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) task_unlock(p); /* + * swapoff can easily use up all memory, so kill those first. + */ + if (p->flags & PF_SWAPOFF) + return ULONG_MAX; + + /* * Processes which fork a lot of child processes are likely * a good choice. We add half the vmsize of the children if they * have an own mm. This prevents forking servers to flood the @@ -146,14 +148,16 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) * Adjust the score by oomkilladj. */ if (p->oomkilladj) { - if (p->oomkilladj > 0) + if (p->oomkilladj > 0) { + if (!points) + points = 1; points <<= p->oomkilladj; - else + } else points >>= -(p->oomkilladj); } #ifdef DEBUG - printk(KERN_DEBUG "OOMkill: task %d (%s) got %d points\n", + printk(KERN_DEBUG "OOMkill: task %d (%s) got %lu points\n", p->pid, p->comm, points); #endif return points; @@ -173,10 +177,10 @@ static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask) { #ifdef CONFIG_NUMA struct zone **z; - nodemask_t nodes = node_online_map; + nodemask_t nodes = node_states[N_HIGH_MEMORY]; for (z = zonelist->zones; *z; z++) - if (cpuset_zone_allowed(*z, gfp_mask)) + if (cpuset_zone_allowed_softwall(*z, gfp_mask)) node_clear(zone_to_nid(*z), nodes); else return CONSTRAINT_CPUSET; @@ -204,16 +208,30 @@ static struct task_struct *select_bad_process(unsigned long *ppoints) do_posix_clock_monotonic_gettime(&uptime); do_each_thread(g, p) { unsigned long points; - int releasing; - /* skip kernel threads */ + /* + * skip kernel threads and tasks which have already released + * their mm. + */ if (!p->mm) continue; - /* skip the init task with pid == 1 */ - if (p->pid == 1) + /* skip the init task */ + if (is_init(p)) continue; /* + * This task already has access to memory reserves and is + * being killed. Don't allow any other task access to the + * memory reserve. + * + * Note: this may have a chance of deadlock if it gets + * blocked waiting for another task which itself is waiting + * for memory. Is there a better alternative? + */ + if (test_tsk_thread_flag(p, TIF_MEMDIE)) + return ERR_PTR(-1UL); + + /* * This is in the process of releasing memory so wait for it * to finish before killing some other task by mistake. * @@ -221,21 +239,16 @@ static struct task_struct *select_bad_process(unsigned long *ppoints) * go ahead if it is exiting: this will simply set TIF_MEMDIE, * which will allow it to gain access to memory reserves in * the process of exiting and releasing its resources. - * Otherwise we could get an OOM deadlock. + * Otherwise we could get an easy OOM deadlock. */ - releasing = test_tsk_thread_flag(p, TIF_MEMDIE) || - p->flags & PF_EXITING; - if (releasing) { - /* PF_DEAD tasks have already released their mm */ - if (p->flags & PF_DEAD) - continue; - if (p->flags & PF_EXITING && p == current) { - chosen = p; - *ppoints = ULONG_MAX; - break; - } - return ERR_PTR(-1UL); + if (p->flags & PF_EXITING) { + if (p != current) + return ERR_PTR(-1UL); + + chosen = p; + *ppoints = ULONG_MAX; } + if (p->oomkilladj == OOM_DISABLE) continue; @@ -245,6 +258,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints) *ppoints = points; } } while_each_thread(g, p); + return chosen; } @@ -253,27 +267,22 @@ static struct task_struct *select_bad_process(unsigned long *ppoints) * flag though it's unlikely that we select a process with CAP_SYS_RAW_IO * set. */ -static void __oom_kill_task(struct task_struct *p, const char *message) +static void __oom_kill_task(struct task_struct *p, int verbose) { - if (p->pid == 1) { + if (is_init(p)) { WARN_ON(1); printk(KERN_WARNING "tried to kill init!\n"); return; } - task_lock(p); - if (!p->mm || p->mm == &init_mm) { + if (!p->mm) { WARN_ON(1); printk(KERN_WARNING "tried to kill an mm-less task!\n"); - task_unlock(p); return; } - task_unlock(p); - if (message) { - printk(KERN_ERR "%s: Killed process %d (%s).\n", - message, p->pid, p->comm); - } + if (verbose) + printk(KERN_ERR "Killed process %d (%s)\n", p->pid, p->comm); /* * We give our sacrificial lamb high priority and access to @@ -286,7 +295,7 @@ static void __oom_kill_task(struct task_struct *p, const char *message) force_sig(SIGKILL, p); } -static int oom_kill_task(struct task_struct *p, const char *message) +static int oom_kill_task(struct task_struct *p) { struct mm_struct *mm; struct task_struct *g, *q; @@ -302,18 +311,28 @@ static int oom_kill_task(struct task_struct *p, const char *message) * However, this is of no concern to us. */ - if (mm == NULL || mm == &init_mm) + if (mm == NULL) return 1; - __oom_kill_task(p, message); + /* + * Don't kill the process if any threads are set to OOM_DISABLE + */ + do_each_thread(g, q) { + if (q->mm == mm && q->oomkilladj == OOM_DISABLE) + return 1; + } while_each_thread(g, q); + + __oom_kill_task(p, 1); + /* * kill all processes that share the ->mm (i.e. all threads), - * but are in a different thread group + * but are in a different thread group. Don't let them have access + * to memory reserves though, otherwise we might deplete all memory. */ - do_each_thread(g, q) + do_each_thread(g, q) { if (q->mm == mm && q->tgid != p->tgid) - __oom_kill_task(q, message); - while_each_thread(g, q); + force_sig(SIGKILL, q); + } while_each_thread(g, q); return 0; } @@ -329,21 +348,22 @@ static int oom_kill_process(struct task_struct *p, unsigned long points, * its children or threads, just set TIF_MEMDIE so it can die quickly */ if (p->flags & PF_EXITING) { - __oom_kill_task(p, NULL); + __oom_kill_task(p, 0); return 0; } - printk(KERN_ERR "Out of Memory: Kill process %d (%s) score %li" - " and children.\n", p->pid, p->comm, points); + printk(KERN_ERR "%s: kill process %d (%s) score %li or a child\n", + message, p->pid, p->comm, points); + /* Try to kill a child first */ list_for_each(tsk, &p->children) { c = list_entry(tsk, struct task_struct, sibling); if (c->mm == p->mm) continue; - if (!oom_kill_task(c, message)) + if (!oom_kill_task(c)) return 0; } - return oom_kill_task(p, message); + return oom_kill_task(p); } static BLOCKING_NOTIFIER_HEAD(oom_notify_list); @@ -373,6 +393,7 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order) struct task_struct *p; unsigned long points = 0; unsigned long freed = 0; + int constraint; blocking_notifier_call_chain(&oom_notify_list, 0, &freed); if (freed > 0) @@ -387,14 +408,18 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order) show_mem(); } - cpuset_lock(); - read_lock(&tasklist_lock); + if (sysctl_panic_on_oom == 2) + panic("out of memory. Compulsory panic_on_oom is selected.\n"); /* * Check if there were limitations on the allocation (only relevant for * NUMA) that may require different handling. */ - switch (constrained_alloc(zonelist, gfp_mask)) { + constraint = constrained_alloc(zonelist, gfp_mask); + cpuset_lock(); + read_lock(&tasklist_lock); + + switch (constraint) { case CONSTRAINT_MEMORY_POLICY: oom_kill_process(current, points, "No available memory (MPOL_BIND)");