From a62eaf151d9cb478d127cfbc2e93c498869785b0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 16 Feb 2006 23:41:58 +0100 Subject: [PATCH] x86_64: Add boot option to disable randomized mappings and cleanup AMD SimNow!'s JIT doesn't like them at all in the guest. For distribution installation it's easiest if it's a boot time option. Also I moved the variable to a more appropiate place and make it independent from sysctl And marked __read_mostly which it is. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- mm/memory.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'mm') diff --git a/mm/memory.c b/mm/memory.c index 2bee1f2..9abc600 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -82,6 +82,16 @@ EXPORT_SYMBOL(num_physpages); EXPORT_SYMBOL(high_memory); EXPORT_SYMBOL(vmalloc_earlyreserve); +int randomize_va_space __read_mostly = 1; + +static int __init disable_randmaps(char *s) +{ + randomize_va_space = 0; + return 0; +} +__setup("norandmaps", disable_randmaps); + + /* * If a p?d_bad entry is found while walking page tables, report * the error, before resetting entry to p?d_none. Usually (but -- cgit v1.1 From dd942ae331425812930cd01766178b7e28e65f2d Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 17 Feb 2006 01:39:16 +0100 Subject: [PATCH] Handle all and empty zones when setting up custom zonelists for mbind The memory allocator doesn't like empty zones (which have an uninitialized freelist), so a x86-64 system with a node fully in GFP_DMA32 only would crash on mbind. Fix that up by putting all possible zones as fallback into the zonelist and skipping the empty ones. In fact the code always enough allocated space for all zones, but only used it for the highest. This change just uses all the memory that was allocated before. This should work fine for now, but whoever implements node hot removal needs to fix this somewhere else too (or make sure zone datastructures by itself never go away, only their memory) Signed-off-by: Andi Kleen Acked-by: Christoph Lameter Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) (limited to 'mm') diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 3bd7fb7..323fdcf 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -132,19 +132,29 @@ static int mpol_check_policy(int mode, nodemask_t *nodes) } return nodes_subset(*nodes, node_online_map) ? 0 : -EINVAL; } + /* Generate a custom zonelist for the BIND policy. */ static struct zonelist *bind_zonelist(nodemask_t *nodes) { struct zonelist *zl; - int num, max, nd; + int num, max, nd, k; max = 1 + MAX_NR_ZONES * nodes_weight(*nodes); - zl = kmalloc(sizeof(void *) * max, GFP_KERNEL); + zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL); if (!zl) return NULL; num = 0; - for_each_node_mask(nd, *nodes) - zl->zones[num++] = &NODE_DATA(nd)->node_zones[policy_zone]; + /* First put in the highest zones from all nodes, then all the next + lower zones etc. Avoid empty zones because the memory allocator + doesn't like them. If you implement node hot removal you + have to fix that. */ + for (k = policy_zone; k >= 0; k--) { + for_each_node_mask(nd, *nodes) { + struct zone *z = &NODE_DATA(nd)->node_zones[k]; + if (z->present_pages > 0) + zl->zones[num++] = z; + } + } zl->zones[num] = NULL; return zl; } -- cgit v1.1 From 4cf808eb443ead42777a0230b73aec0cee7fb298 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 17 Feb 2006 20:38:21 +0100 Subject: [PATCH] Handle holes in node mask in node fallback list setup Change the find_next_best_node algorithm to correctly skip over holes in the node online mask. Previously it would not handle missing nodes correctly and cause crashes at boot. [Written by Linus, tested by AK] Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 62c1225..208812b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1541,29 +1541,29 @@ static int __initdata node_load[MAX_NUMNODES]; */ static int __init find_next_best_node(int node, nodemask_t *used_node_mask) { - int i, n, val; + int n, val; int min_val = INT_MAX; int best_node = -1; - for_each_online_node(i) { - cpumask_t tmp; + /* Use the local node if we haven't already */ + if (!node_isset(node, *used_node_mask)) { + node_set(node, *used_node_mask); + return node; + } - /* Start from local node */ - n = (node+i) % num_online_nodes(); + for_each_online_node(n) { + cpumask_t tmp; /* Don't want a node to appear more than once */ if (node_isset(n, *used_node_mask)) continue; - /* Use the local node if we haven't already */ - if (!node_isset(node, *used_node_mask)) { - best_node = node; - break; - } - /* Use the distance array to find the distance */ val = node_distance(node, n); + /* Penalize nodes under us ("prefer the next node") */ + val += (n < node); + /* Give preference to headless and unused nodes */ tmp = node_to_cpumask(n); if (!cpus_empty(tmp)) -- cgit v1.1 From 636f13c174dd7c84a437d3c3e8fa66f03f7fda63 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Fri, 17 Feb 2006 13:59:36 -0800 Subject: [PATCH] sys_mbind sanity checking Make sure maxnodes is safe size before calculating nlongs in get_nodes(). Signed-off-by: Chris Wright Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'mm') diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 323fdcf..bedfa4f 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -808,6 +808,8 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask, nodes_clear(*nodes); if (maxnode == 0 || !nmask) return 0; + if (maxnode > PAGE_SIZE) + return -EINVAL; nlongs = BITS_TO_LONGS(maxnode); if ((maxnode % BITS_PER_LONG) == 0) -- cgit v1.1 From 9827b781f20828e5ceb911b879f268f78fe90815 Mon Sep 17 00:00:00 2001 From: Kurt Garloff Date: Mon, 20 Feb 2006 18:27:51 -0800 Subject: [PATCH] OOM kill: children accounting In the badness() calculation, there's currently this piece of code: /* * Processes which fork a lot of child processes are likely * a good choice. We add the vmsize of the children if they * have an own mm. This prevents forking servers to flood the * machine with an endless amount of children */ list_for_each(tsk, &p->children) { struct task_struct *chld; chld = list_entry(tsk, struct task_struct, sibling); if (chld->mm = p->mm && chld->mm) points += chld->mm->total_vm; } The intention is clear: If some server (apache) keeps spawning new children and we run OOM, we want to kill the father rather than picking a child. This -- to some degree -- also helps a bit with getting fork bombs under control, though I'd consider this a desirable side-effect rather than a feature. There's one problem with this: No matter how many or few children there are, if just one of them misbehaves, and all others (including the father) do everything right, we still always kill the whole family. This hits in real life; whether it's javascript in konqueror resulting in kdeinit (and thus the whole KDE session) being hit or just a classical server that spawns children. Sidenote: The killer does kill all direct children as well, not only the selected father, see oom_kill_process(). The idea in attached patch is that we do want to account the memory consumption of the (direct) children to the father -- however not fully. This maintains the property that fathers with too many children will still very likely be picked, whereas a single misbehaving child has the chance to be picked by the OOM killer. In the patch I account only half (rounded up) of the children's vm_size to the parent. This means that if one child eats more mem than the rest of the family, it will be picked, otherwise it's still the father and thus the whole family that gets selected. This is heuristics -- we could debate whether accounting for a fourth would be better than for half of it. Or -- if people would consider it worth the trouble -- make it a sysctl. For now I sticked to accounting for half, which should IMHO be a significant improvement. The patch does one more thing: As users tend to be irritated by the choice of killed processes (mainly because the children are killed first, despite some of them having a very low OOM score), I added some more output: The selected (father) process will be reported first and it's oom_score printed to syslog. Description: Only account for half of children's vm size in oom score calculation This should still give the parent enough point in case of fork bombs. If any child however has more than 50% of the vm size of all children together, it'll get a higher score and be elected. This patch also makes the kernel display the oom_score. Signed-off-by: Kurt Garloff Cc: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/oom_kill.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) (limited to 'mm') diff --git a/mm/oom_kill.c b/mm/oom_kill.c index b05ab8f..949eba1 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -58,15 +58,17 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) /* * Processes which fork a lot of child processes are likely - * a good choice. We add the vmsize of the children if they + * a good choice. We add half the vmsize of the children if they * have an own mm. This prevents forking servers to flood the - * machine with an endless amount of children + * machine with an endless amount of children. In case a single + * child is eating the vast majority of memory, adding only half + * to the parents will make the child our kill candidate of choice. */ list_for_each(tsk, &p->children) { struct task_struct *chld; chld = list_entry(tsk, struct task_struct, sibling); if (chld->mm != p->mm && chld->mm) - points += chld->mm->total_vm; + points += chld->mm->total_vm/2 + 1; } /* @@ -136,12 +138,12 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) * * (not docbooked, we don't want this one cluttering up the manual) */ -static struct task_struct * select_bad_process(void) +static struct task_struct *select_bad_process(unsigned long *ppoints) { - unsigned long maxpoints = 0; struct task_struct *g, *p; struct task_struct *chosen = NULL; struct timespec uptime; + *ppoints = 0; do_posix_clock_monotonic_gettime(&uptime); do_each_thread(g, p) { @@ -169,9 +171,9 @@ static struct task_struct * select_bad_process(void) return p; points = badness(p, uptime.tv_sec); - if (points > maxpoints || !chosen) { + if (points > *ppoints || !chosen) { chosen = p; - maxpoints = points; + *ppoints = points; } } while_each_thread(g, p); return chosen; @@ -237,12 +239,15 @@ static struct mm_struct *oom_kill_task(task_t *p) return mm; } -static struct mm_struct *oom_kill_process(struct task_struct *p) +static struct mm_struct *oom_kill_process(struct task_struct *p, + unsigned long points) { struct mm_struct *mm; struct task_struct *c; struct list_head *tsk; + printk(KERN_ERR "Out of Memory: Kill process %d (%s) score %li and " + "children.\n", 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); @@ -267,6 +272,7 @@ void out_of_memory(gfp_t gfp_mask, int order) { struct mm_struct *mm = NULL; task_t * p; + unsigned long points; if (printk_ratelimit()) { printk("oom-killer: gfp_mask=0x%x, order=%d\n", @@ -278,7 +284,7 @@ void out_of_memory(gfp_t gfp_mask, int order) cpuset_lock(); read_lock(&tasklist_lock); retry: - p = select_bad_process(); + p = select_bad_process(&points); if (PTR_ERR(p) == -1UL) goto out; @@ -290,7 +296,7 @@ retry: panic("Out of memory and no killable processes...\n"); } - mm = oom_kill_process(p); + mm = oom_kill_process(p, points); if (!mm) goto retry; -- cgit v1.1 From 9b0f8b040acd8dfd23860754c0d09ff4f44e2cbc Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 20 Feb 2006 18:27:52 -0800 Subject: [PATCH] Terminate process that fails on a constrained allocation Some allocations are restricted to a limited set of nodes (due to memory policies or cpuset constraints). If the page allocator is not able to find enough memory then that does not mean that overall system memory is low. In particular going postal and more or less randomly shooting at processes is not likely going to help the situation but may just lead to suicide (the whole system coming down). It is better to signal to the process that no memory exists given the constraints that the process (or the configuration of the process) has placed on the allocation behavior. The process may be killed but then the sysadmin or developer can investigate the situation. The solution is similar to what we do when running out of hugepages. This patch adds a check before we kill processes. At that point performance considerations do not matter much so we just scan the zonelist and reconstruct a list of nodes. If the list of nodes does not contain all online nodes then this is a constrained allocation and we should kill the current process. Signed-off-by: Christoph Lameter Cc: Nick Piggin Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/oom_kill.c | 103 ++++++++++++++++++++++++++++++++++++++++++-------------- mm/page_alloc.c | 2 +- 2 files changed, 79 insertions(+), 26 deletions(-) (limited to 'mm') diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 949eba1..8123fad 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -133,6 +133,36 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) } /* + * Types of limitations to the nodes from which allocations may occur + */ +#define CONSTRAINT_NONE 1 +#define CONSTRAINT_MEMORY_POLICY 2 +#define CONSTRAINT_CPUSET 3 + +/* + * Determine the type of allocation constraint. + */ +static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask) +{ +#ifdef CONFIG_NUMA + struct zone **z; + nodemask_t nodes = node_online_map; + + for (z = zonelist->zones; *z; z++) + if (cpuset_zone_allowed(*z, gfp_mask)) + node_clear((*z)->zone_pgdat->node_id, + nodes); + else + return CONSTRAINT_CPUSET; + + if (!nodes_empty(nodes)) + return CONSTRAINT_MEMORY_POLICY; +#endif + + return CONSTRAINT_NONE; +} + +/* * Simple selection loop. We chose the process with the highest * number of 'points'. We expect the caller will lock the tasklist. * @@ -184,7 +214,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints) * CAP_SYS_RAW_IO set, send SIGTERM instead (but it's unlikely that * we select a process with CAP_SYS_RAW_IO set). */ -static void __oom_kill_task(task_t *p) +static void __oom_kill_task(task_t *p, const char *message) { if (p->pid == 1) { WARN_ON(1); @@ -200,8 +230,8 @@ static void __oom_kill_task(task_t *p) return; } task_unlock(p); - printk(KERN_ERR "Out of Memory: Killed process %d (%s).\n", - p->pid, p->comm); + printk(KERN_ERR "%s: Killed process %d (%s).\n", + message, p->pid, p->comm); /* * We give our sacrificial lamb high priority and access to @@ -214,7 +244,7 @@ static void __oom_kill_task(task_t *p) force_sig(SIGKILL, p); } -static struct mm_struct *oom_kill_task(task_t *p) +static struct mm_struct *oom_kill_task(task_t *p, const char *message) { struct mm_struct *mm = get_task_mm(p); task_t * g, * q; @@ -226,21 +256,21 @@ static struct mm_struct *oom_kill_task(task_t *p) return NULL; } - __oom_kill_task(p); + __oom_kill_task(p, message); /* * kill all processes that share the ->mm (i.e. all threads), * but are in a different thread group */ do_each_thread(g, q) if (q->mm == mm && q->tgid != p->tgid) - __oom_kill_task(q); + __oom_kill_task(q, message); while_each_thread(g, q); return mm; } static struct mm_struct *oom_kill_process(struct task_struct *p, - unsigned long points) + unsigned long points, const char *message) { struct mm_struct *mm; struct task_struct *c; @@ -253,11 +283,11 @@ static struct mm_struct *oom_kill_process(struct task_struct *p, c = list_entry(tsk, struct task_struct, sibling); if (c->mm == p->mm) continue; - mm = oom_kill_task(c); + mm = oom_kill_task(c, message); if (mm) return mm; } - return oom_kill_task(p); + return oom_kill_task(p, message); } /** @@ -268,10 +298,10 @@ static struct mm_struct *oom_kill_process(struct task_struct *p, * OR try to be smart about which process to kill. Note that we * don't have to be perfect here, we just have to be good. */ -void out_of_memory(gfp_t gfp_mask, int order) +void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order) { struct mm_struct *mm = NULL; - task_t * p; + task_t *p; unsigned long points; if (printk_ratelimit()) { @@ -283,25 +313,48 @@ void out_of_memory(gfp_t gfp_mask, int order) cpuset_lock(); read_lock(&tasklist_lock); + + /* + * Check if there were limitations on the allocation (only relevant for + * NUMA) that may require different handling. + */ + switch (constrained_alloc(zonelist, gfp_mask)) { + case CONSTRAINT_MEMORY_POLICY: + mm = oom_kill_process(current, points, + "No available memory (MPOL_BIND)"); + break; + + case CONSTRAINT_CPUSET: + mm = oom_kill_process(current, points, + "No available memory in cpuset"); + break; + + case CONSTRAINT_NONE: retry: - p = select_bad_process(&points); + /* + * Rambo mode: Shoot down a process and hope it solves whatever + * issues we may have. + */ + p = select_bad_process(&points); - if (PTR_ERR(p) == -1UL) - goto out; + if (PTR_ERR(p) == -1UL) + goto out; - /* Found nothing?!?! Either we hang forever, or we panic. */ - if (!p) { - read_unlock(&tasklist_lock); - cpuset_unlock(); - panic("Out of memory and no killable processes...\n"); - } + /* Found nothing?!?! Either we hang forever, or we panic. */ + if (!p) { + read_unlock(&tasklist_lock); + cpuset_unlock(); + panic("Out of memory and no killable processes...\n"); + } - mm = oom_kill_process(p, points); - if (!mm) - goto retry; + mm = oom_kill_process(p, points, "Out of memory"); + if (!mm) + goto retry; + + break; + } - out: - read_unlock(&tasklist_lock); +out: cpuset_unlock(); if (mm) mmput(mm); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 208812b..791690d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1015,7 +1015,7 @@ rebalance: if (page) goto got_pg; - out_of_memory(gfp_mask, order); + out_of_memory(zonelist, gfp_mask, order); goto restart; } -- cgit v1.1 From a9c930bac163c5e616ca0ba9378e7dc746c93227 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 20 Feb 2006 18:27:59 -0800 Subject: [PATCH] Fix units in mbind check maxnode is a bit index and can't be directly compared against a byte length like PAGE_SIZE Signed-off-by: Andi Kleen Cc: Chris Wright Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/mempolicy.c b/mm/mempolicy.c index bedfa4f..6422fe4 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -808,7 +808,7 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask, nodes_clear(*nodes); if (maxnode == 0 || !nmask) return 0; - if (maxnode > PAGE_SIZE) + if (maxnode > PAGE_SIZE*BITS_PER_BYTE) return -EINVAL; nlongs = BITS_TO_LONGS(maxnode); -- cgit v1.1 From 7a9166e3b037296366cea6f3c97f705d33e209e6 Mon Sep 17 00:00:00 2001 From: Luke Yang Date: Mon, 20 Feb 2006 18:28:07 -0800 Subject: [PATCH] Fix undefined symbols for nommu architecture Signed-off-by: Luke Yang Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/nommu.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index c10262d..99d2102 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -57,6 +57,8 @@ EXPORT_SYMBOL(vmalloc); EXPORT_SYMBOL(vfree); EXPORT_SYMBOL(vmalloc_to_page); EXPORT_SYMBOL(vmalloc_32); +EXPORT_SYMBOL(vmap); +EXPORT_SYMBOL(vunmap); /* * Handle all mappings that got truncated by a "truncate()" -- cgit v1.1 From fcab6f351305029fc5e3c632209d45cae57e4835 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 20 Feb 2006 18:28:10 -0800 Subject: [PATCH] mm/mempolicy.c: fix 'if ();' typo [akpm; it happens that the code was still correct, only inefficient ] Signed-off-by: Alexey Dobriyan Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 6422fe4..880831b 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -587,7 +587,7 @@ redo: } list_add(&page->lru, &newlist); nr_pages++; - if (nr_pages > MIGRATE_CHUNK_SIZE); + if (nr_pages > MIGRATE_CHUNK_SIZE) break; } err = migrate_pages(pagelist, &newlist, &moved, &failed); -- cgit v1.1 From b00dc3ad74fdb676552d46ee573b88e927240d0c Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 21 Feb 2006 23:49:47 +0000 Subject: [PATCH] tmpfs: fix mount mpol nodelist parsing I've been dissatisfied with the mpol_nodelist mount option which was added to tmpfs earlier in -rc. Replace it by mpol=policy:nodelist. And it was broken: a nodelist is a comma-separated list of numbers and ranges; the mount options are a comma-separated list of token=values. Whoops, blindly strsep'ing on commas doesn't work so well: since we've no numeric tokens, and unlikely to add them, use that to distinguish. Move the mpol= parsing to shmem_parse_mpol under CONFIG_NUMA, reject all its options as invalid if not NUMA. /proc shows MPOL_PREFERRED as "prefer", so use that name for the policy instead of "preferred". Enforce that mpol=default has no nodelist; that mpol=prefer has one node only; that mpol=bind has a nodelist; but let mpol=interleave use node_online_map if no nodelist given. Describe this in tmpfs.txt. Signed-off-by: Hugh Dickins Acked-by: Robin Holt Acked-by: Andi Kleen Signed-off-by: Linus Torvalds --- mm/shmem.c | 81 ++++++++++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 69 insertions(+), 12 deletions(-) (limited to 'mm') diff --git a/mm/shmem.c b/mm/shmem.c index f7ac7b8..7c455fb 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -45,6 +45,7 @@ #include #include #include +#include #include #include #include @@ -874,6 +875,51 @@ redirty: } #ifdef CONFIG_NUMA +static int shmem_parse_mpol(char *value, int *policy, nodemask_t *policy_nodes) +{ + char *nodelist = strchr(value, ':'); + int err = 1; + + if (nodelist) { + /* NUL-terminate policy string */ + *nodelist++ = '\0'; + if (nodelist_parse(nodelist, *policy_nodes)) + goto out; + } + if (!strcmp(value, "default")) { + *policy = MPOL_DEFAULT; + /* Don't allow a nodelist */ + if (!nodelist) + err = 0; + } else if (!strcmp(value, "prefer")) { + *policy = MPOL_PREFERRED; + /* Insist on a nodelist of one node only */ + if (nodelist) { + char *rest = nodelist; + while (isdigit(*rest)) + rest++; + if (!*rest) + err = 0; + } + } else if (!strcmp(value, "bind")) { + *policy = MPOL_BIND; + /* Insist on a nodelist */ + if (nodelist) + err = 0; + } else if (!strcmp(value, "interleave")) { + *policy = MPOL_INTERLEAVE; + /* Default to nodes online if no nodelist */ + if (!nodelist) + *policy_nodes = node_online_map; + err = 0; + } +out: + /* Restore string for error message */ + if (nodelist) + *--nodelist = ':'; + return err; +} + static struct page *shmem_swapin_async(struct shared_policy *p, swp_entry_t entry, unsigned long idx) { @@ -926,6 +972,11 @@ shmem_alloc_page(gfp_t gfp, struct shmem_inode_info *info, return page; } #else +static inline int shmem_parse_mpol(char *value, int *policy, nodemask_t *policy_nodes) +{ + return 1; +} + static inline struct page * shmem_swapin(struct shmem_inode_info *info,swp_entry_t entry,unsigned long idx) { @@ -1859,7 +1910,23 @@ static int shmem_parse_options(char *options, int *mode, uid_t *uid, { char *this_char, *value, *rest; - while ((this_char = strsep(&options, ",")) != NULL) { + while (options != NULL) { + this_char = options; + for (;;) { + /* + * NUL-terminate this option: unfortunately, + * mount options form a comma-separated list, + * but mpol's nodelist may also contain commas. + */ + options = strchr(options, ','); + if (options == NULL) + break; + options++; + if (!isdigit(*options)) { + options[-1] = '\0'; + break; + } + } if (!*this_char) continue; if ((value = strchr(this_char,'=')) != NULL) { @@ -1910,18 +1977,8 @@ static int shmem_parse_options(char *options, int *mode, uid_t *uid, if (*rest) goto bad_val; } else if (!strcmp(this_char,"mpol")) { - if (!strcmp(value,"default")) - *policy = MPOL_DEFAULT; - else if (!strcmp(value,"preferred")) - *policy = MPOL_PREFERRED; - else if (!strcmp(value,"bind")) - *policy = MPOL_BIND; - else if (!strcmp(value,"interleave")) - *policy = MPOL_INTERLEAVE; - else + if (shmem_parse_mpol(value,policy,policy_nodes)) goto bad_val; - } else if (!strcmp(this_char,"mpol_nodelist")) { - nodelist_parse(value, *policy_nodes); } else { printk(KERN_ERR "tmpfs: Bad mount option %s\n", this_char); -- cgit v1.1 From 1e275d406bf6b88e4de6925cf594b64bb2ec49bc Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Fri, 24 Feb 2006 13:04:12 -0800 Subject: [PATCH] page migration: Fix MPOL_INTERLEAVE behavior for migration via mbind() migrate_pages_to() allocates a list of new pages on the intended target node or with the intended policy and then uses the list of new pages as targets for the migration of a list of pages out of place. When the pages are allocated it is not clear which of the out of place pages will be moved to the new pages. So we cannot specify an address as needed by alloc_page_vma(). This causes problem for MPOL_INTERLEAVE which will currently allocate the pages on the first node of the set. If mbind is used with vma that has the policy of MPOL_INTERLEAVE then the interleaving of pages may be destroyed. This patch fixes that by generating a fake address for each alloc_page_vma which will result is a distribution of pages as prescribed by MPOL_INTERLEAVE. Lee also noted that the sequence of nodes for the new pages seems to be inverted. So we also invert the way the lists of pages for migration are build. Signed-off-by: Christoph Lameter Signed-off-by: Lee Schermerhorn Looks-ok-to: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'mm') diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 880831b..67af4ce 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -552,7 +552,7 @@ static void migrate_page_add(struct page *page, struct list_head *pagelist, */ if ((flags & MPOL_MF_MOVE_ALL) || page_mapcount(page) == 1) { if (isolate_lru_page(page)) - list_add(&page->lru, pagelist); + list_add_tail(&page->lru, pagelist); } } @@ -569,6 +569,7 @@ static int migrate_pages_to(struct list_head *pagelist, LIST_HEAD(moved); LIST_HEAD(failed); int err = 0; + unsigned long offset = 0; int nr_pages; struct page *page; struct list_head *p; @@ -576,8 +577,21 @@ static int migrate_pages_to(struct list_head *pagelist, redo: nr_pages = 0; list_for_each(p, pagelist) { - if (vma) - page = alloc_page_vma(GFP_HIGHUSER, vma, vma->vm_start); + if (vma) { + /* + * The address passed to alloc_page_vma is used to + * generate the proper interleave behavior. We fake + * the address here by an increasing offset in order + * to get the proper distribution of pages. + * + * No decision has been made as to which page + * a certain old page is moved to so we cannot + * specify the correct address. + */ + page = alloc_page_vma(GFP_HIGHUSER, vma, + offset + vma->vm_start); + offset += PAGE_SIZE; + } else page = alloc_pages_node(dest, GFP_HIGHUSER, 0); @@ -585,7 +599,7 @@ redo: err = -ENOMEM; goto out; } - list_add(&page->lru, &newlist); + list_add_tail(&page->lru, &newlist); nr_pages++; if (nr_pages > MIGRATE_CHUNK_SIZE) break; -- cgit v1.1 From d4f7796e9b387e471ab0e8ed4e0c2bd616b3c193 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Fri, 24 Feb 2006 13:04:22 -0800 Subject: [PATCH] vmscan: fix zone_reclaim - PF_SWAPWRITE needs to be set for RECLAIM_SWAP to be able to write out pages to swap. Currently RECLAIM_SWAP may not do that. - remove setting nr_reclaimed pages after slab reclaim since the slab shrinking code does not use that and the nr_reclaimed pages is just right for the intended follow up action. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/vmscan.c b/mm/vmscan.c index 1838c15..b0af759 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1908,7 +1908,12 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) sc.swap_cluster_max = SWAP_CLUSTER_MAX; cond_resched(); - p->flags |= PF_MEMALLOC; + /* + * We need to be able to allocate from the reserves for RECLAIM_SWAP + * and we also need to be able to write out pages for RECLAIM_WRITE + * and RECLAIM_SWAP. + */ + p->flags |= PF_MEMALLOC | PF_SWAPWRITE; reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; @@ -1932,11 +1937,10 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) * a long time. */ shrink_slab(sc.nr_scanned, gfp_mask, order); - sc.nr_reclaimed = 1; /* Avoid getting the off node timeout */ } p->reclaim_state = NULL; - current->flags &= ~PF_MEMALLOC; + current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE); if (sc.nr_reclaimed == 0) zone->last_unsuccessful_zone_reclaim = jiffies; -- cgit v1.1