From 6b314d0e11924c803bf8cd944e87fd58cdb5088c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 2 Dec 2009 18:58:05 +0100 Subject: sched: Remove sysctl.sched_features Since we've had a much saner debugfs interface to this, remove the sysctl one. Signed-off-by: Peter Zijlstra LKML-Reference: [ v2: build fix ] Signed-off-by: Ingo Molnar --- kernel/sysctl.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4dbf93a52ee9..e5cc53514caa 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -314,14 +314,6 @@ static struct ctl_table kern_table[] = { .strategy = &sysctl_intvec, .extra1 = &zero, }, - { - .ctl_name = CTL_UNNUMBERED, - .procname = "sched_features", - .data = &sysctl_sched_features, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = &proc_dointvec, - }, { .ctl_name = CTL_UNNUMBERED, .procname = "sched_migration_cost", -- cgit v1.2.2 From 1983a922a1bc843806b9a36cf3a370b242783140 Mon Sep 17 00:00:00 2001 From: Christian Ehrhardt Date: Mon, 30 Nov 2009 12:16:47 +0100 Subject: sched: Make tunable scaling style configurable As scaling now takes place on all kind of cpu add/remove events a user that configures values via proc should be able to configure if his set values are still rescaled or kept whatever happens. As the comments state that log2 was just a second guess that worked the interface is not just designed for on/off, but to choose a scaling type. Currently this allows none, log and linear, but more important it allwos us to keep the interface even if someone has an even better idea how to scale the values. Signed-off-by: Christian Ehrhardt Signed-off-by: Peter Zijlstra LKML-Reference: <1259579808-11357-3-git-send-email-ehrhardt@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- kernel/sysctl.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index e5cc53514caa..d10406e5fdfe 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -251,6 +251,8 @@ static int min_sched_granularity_ns = 100000; /* 100 usecs */ static int max_sched_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_wakeup_granularity_ns; /* 0 usecs */ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ +static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE; +static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1; #endif static struct ctl_table kern_table[] = { @@ -304,6 +306,18 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "sched_tunable_scaling", + .data = &sysctl_sched_tunable_scaling, + .maxlen = sizeof(enum sched_tunable_scaling), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &min_sched_tunable_scaling, + .extra2 = &max_sched_tunable_scaling, + }, + { .ctl_name = CTL_UNNUMBERED, .procname = "sched_shares_thresh", -- cgit v1.2.2 From acb4a848da821a095ae9e4d8b22ae2d9633ba5cd Mon Sep 17 00:00:00 2001 From: Christian Ehrhardt Date: Mon, 30 Nov 2009 12:16:48 +0100 Subject: sched: Update normalized values on user updates via proc The normalized values are also recalculated in case the scaling factor changes. This patch updates the internally used scheduler tuning values that are normalized to one cpu in case a user sets new values via sysfs. Together with patch 2 of this series this allows to let user configured values scale (or not) to cpu add/remove events taking place later. Signed-off-by: Christian Ehrhardt Signed-off-by: Peter Zijlstra LKML-Reference: <1259579808-11357-4-git-send-email-ehrhardt@linux.vnet.ibm.com> [ v2: fix warning ] Signed-off-by: Ingo Molnar --- kernel/sysctl.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index d10406e5fdfe..b9e5a45f1e28 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -253,6 +253,8 @@ static int min_wakeup_granularity_ns; /* 0 usecs */ static int max_wakeup_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE; static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1; +static int min_sched_shares_ratelimit = 100000; /* 100 usec */ +static int max_sched_shares_ratelimit = NSEC_PER_SEC; /* 1 second */ #endif static struct ctl_table kern_table[] = { @@ -271,7 +273,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_min_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, @@ -282,7 +284,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_latency, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &sched_nr_latency_handler, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_granularity_ns, .extra2 = &max_sched_granularity_ns, @@ -293,7 +295,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_wakeup_granularity, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_wakeup_granularity_ns, .extra2 = &max_wakeup_granularity_ns, @@ -304,7 +306,9 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_shares_ratelimit, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = &sched_proc_update_handler, + .extra1 = &min_sched_shares_ratelimit, + .extra2 = &max_sched_shares_ratelimit, }, { .ctl_name = CTL_UNNUMBERED, @@ -312,7 +316,7 @@ static struct ctl_table kern_table[] = { .data = &sysctl_sched_tunable_scaling, .maxlen = sizeof(enum sched_tunable_scaling), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = &sched_proc_update_handler, .strategy = &sysctl_intvec, .extra1 = &min_sched_tunable_scaling, .extra2 = &max_sched_tunable_scaling, -- cgit v1.2.2 From 06808b0827e1cd14eedc96bac2655d5b37ac246c Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 14 Dec 2009 17:58:21 -0800 Subject: hugetlb: derive huge pages nodes allowed from task mempolicy This patch derives a "nodes_allowed" node mask from the numa mempolicy of the task modifying the number of persistent huge pages to control the allocation, freeing and adjusting of surplus huge pages when the pool page count is modified via the new sysctl or sysfs attribute "nr_hugepages_mempolicy". The nodes_allowed mask is derived as follows: * For "default" [NULL] task mempolicy, a NULL nodemask_t pointer is produced. This will cause the hugetlb subsystem to use node_online_map as the "nodes_allowed". This preserves the behavior before this patch. * For "preferred" mempolicy, including explicit local allocation, a nodemask with the single preferred node will be produced. "local" policy will NOT track any internode migrations of the task adjusting nr_hugepages. * For "bind" and "interleave" policy, the mempolicy's nodemask will be used. * Other than to inform the construction of the nodes_allowed node mask, the actual mempolicy mode is ignored. That is, all modes behave like interleave over the resulting nodes_allowed mask with no "fallback". See the updated documentation [next patch] for more information about the implications of this patch. Examples: Starting with: Node 0 HugePages_Total: 0 Node 1 HugePages_Total: 0 Node 2 HugePages_Total: 0 Node 3 HugePages_Total: 0 Default behavior [with or without this patch] balances persistent hugepage allocation across nodes [with sufficient contiguous memory]: sysctl vm.nr_hugepages[_mempolicy]=32 yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 8 Node 3 HugePages_Total: 8 Of course, we only have nr_hugepages_mempolicy with the patch, but with default mempolicy, nr_hugepages_mempolicy behaves the same as nr_hugepages. Applying mempolicy--e.g., with numactl [using '-m' a.k.a. '--membind' because it allows multiple nodes to be specified and it's easy to type]--we can allocate huge pages on individual nodes or sets of nodes. So, starting from the condition above, with 8 huge pages per node, add 8 more to node 2 using: numactl -m 2 sysctl vm.nr_hugepages_mempolicy=40 This yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The incremental 8 huge pages were restricted to node 2 by the specified mempolicy. Similarly, we can use mempolicy to free persistent huge pages from specified nodes: numactl -m 0,1 sysctl vm.nr_hugepages_mempolicy=32 yields: Node 0 HugePages_Total: 4 Node 1 HugePages_Total: 4 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The 8 huge pages freed were balanced over nodes 0 and 1. [rientjes@google.com: accomodate reworked NODEMASK_ALLOC] Signed-off-by: David Rientjes Signed-off-by: Lee Schermerhorn Acked-by: Mel Gorman Reviewed-by: Andi Kleen Cc: KAMEZAWA Hiroyuki Cc: Randy Dunlap Cc: Nishanth Aravamudan Cc: Adam Litke Cc: Andy Whitcroft Cc: Eric Whitney Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 554ac4894f0f..60fc93131095 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1051,7 +1051,7 @@ static struct ctl_table vm_table[] = { .extra2 = &one_hundred, }, #ifdef CONFIG_HUGETLB_PAGE - { + { .procname = "nr_hugepages", .data = NULL, .maxlen = sizeof(unsigned long), @@ -1059,7 +1059,18 @@ static struct ctl_table vm_table[] = { .proc_handler = hugetlb_sysctl_handler, .extra1 = (void *)&hugetlb_zero, .extra2 = (void *)&hugetlb_infinity, - }, + }, +#ifdef CONFIG_NUMA + { + .procname = "nr_hugepages_mempolicy", + .data = NULL, + .maxlen = sizeof(unsigned long), + .mode = 0644, + .proc_handler = &hugetlb_mempolicy_sysctl_handler, + .extra1 = (void *)&hugetlb_zero, + .extra2 = (void *)&hugetlb_infinity, + }, +#endif { .procname = "hugetlb_shm_group", .data = &sysctl_hugetlb_shm_group, -- cgit v1.2.2 From 70da2340fbc68e91e701762f785479ab495a0869 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Mon, 14 Dec 2009 17:59:52 -0800 Subject: 'sysctl_max_map_count' should be non-negative Jan Engelhardt reported we have this problem: setting max_map_count to a value large enough results in programs dying at first try. This is on 2.6.31.6: 15:59 borg:/proc/sys/vm # echo $[1<<31-1] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count 1073741824 15:59 borg:/proc/sys/vm # echo $[1<<31] >max_map_count 15:59 borg:/proc/sys/vm # cat max_map_count Killed This is because we have a chance to make 'max_map_count' negative. but it's meaningless. Make it only accept non-negative values. Reported-by: Jan Engelhardt Signed-off-by: WANG Cong Cc: Ingo Molnar Cc: Peter Zijlstra Cc: James Morris Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 60fc93131095..45e4bef0012a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1131,7 +1131,8 @@ static struct ctl_table vm_table[] = { .data = &sysctl_max_map_count, .maxlen = sizeof(sysctl_max_map_count), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dointvec, + .extra1 = &zero, }, #else { -- cgit v1.2.2 From 6e1415467614e854fee660ff6648bd10fa976e95 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 15 Dec 2009 19:27:45 +0000 Subject: NOMMU: Optimise away the {dac_,}mmap_min_addr tests In NOMMU mode clamp dac_mmap_min_addr to zero to cause the tests on it to be skipped by the compiler. We do this as the minimum mmap address doesn't make any sense in NOMMU mode. mmap_min_addr and round_hint_to_min() can be discarded entirely in NOMMU mode. Signed-off-by: David Howells Acked-by: Eric Paris Signed-off-by: James Morris --- kernel/sysctl.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 45e4bef0012a..856a24eadf7e 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1214,6 +1214,7 @@ static struct ctl_table vm_table[] = { .proc_handler = proc_dointvec_jiffies, }, #endif +#ifdef CONFIG_MMU { .procname = "mmap_min_addr", .data = &dac_mmap_min_addr, @@ -1221,6 +1222,7 @@ static struct ctl_table vm_table[] = { .mode = 0644, .proc_handler = mmap_min_addr_handler, }, +#endif #ifdef CONFIG_NUMA { .procname = "numa_zonelist_order", -- cgit v1.2.2 From 3e26120cc7c819c97bc07281ca1fb9017cfe9a39 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Thu, 17 Dec 2009 15:27:05 -0800 Subject: kernel/sysctl.c: fix the incomplete part of sysctl_max_map_count-should-be-non-negative.patch It is a mistake that we used 'proc_dointvec', it should be 'proc_dointvec_minmax', as in the original patch. Signed-off-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sysctl.c') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 45e4bef0012a..6665761c006d 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1131,7 +1131,7 @@ static struct ctl_table vm_table[] = { .data = &sysctl_max_map_count, .maxlen = sizeof(sysctl_max_map_count), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = &zero, }, #else -- cgit v1.2.2