From 607b45e9a216e89a63351556e488eea06be0ff48 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Fri, 11 Apr 2014 11:44:39 +0200 Subject: sched, powerpc: Create a dedicated topology table Create a dedicated topology table for handling asymetric feature of powerpc. Signed-off-by: Vincent Guittot Reviewed-by: Preeti U Murthy Signed-off-by: Peter Zijlstra Cc: Andy Fleming Cc: Anton Blanchard Cc: Benjamin Herrenschmidt Cc: Grant Likely Cc: Linus Torvalds Cc: Michael Ellerman Cc: Paul Gortmaker Cc: Paul Mackerras Cc: Preeti U. Murthy Cc: Rob Herring Cc: Srivatsa S. Bhat Cc: Toshi Kani Cc: Vasant Hegde Cc: tony.luck@intel.com Cc: fenghua.yu@intel.com Cc: schwidefsky@de.ibm.com Cc: cmetcalf@tilera.com Cc: dietmar.eggemann@arm.com Cc: devicetree@vger.kernel.org Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1397209481-28542-4-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar --- arch/powerpc/kernel/smp.c | 31 +++++++++++++++++++++++-------- 1 file changed, 23 insertions(+), 8 deletions(-) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index e2a4232c5871..10ffffef0414 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -766,6 +766,28 @@ int setup_profiling_timer(unsigned int multiplier) return 0; } +#ifdef CONFIG_SCHED_SMT +/* cpumask of CPUs with asymetric SMT dependancy */ +static const int powerpc_smt_flags(void) +{ + int flags = SD_SHARE_CPUPOWER | SD_SHARE_PKG_RESOURCES; + + if (cpu_has_feature(CPU_FTR_ASYM_SMT)) { + printk_once(KERN_INFO "Enabling Asymmetric SMT scheduling\n"); + flags |= SD_ASYM_PACKING; + } + return flags; +} +#endif + +static struct sched_domain_topology_level powerpc_topology[] = { +#ifdef CONFIG_SCHED_SMT + { cpu_smt_mask, powerpc_smt_flags, SD_INIT_NAME(SMT) }, +#endif + { cpu_cpu_mask, SD_INIT_NAME(DIE) }, + { NULL, }, +}; + void __init smp_cpus_done(unsigned int max_cpus) { cpumask_var_t old_mask; @@ -790,15 +812,8 @@ void __init smp_cpus_done(unsigned int max_cpus) dump_numa_cpu_topology(); -} + set_sched_topology(powerpc_topology); -int arch_sd_sibling_asym_packing(void) -{ - if (cpu_has_feature(CPU_FTR_ASYM_SMT)) { - printk_once(KERN_INFO "Enabling Asymmetric SMT scheduling\n"); - return SD_ASYM_PACKING; - } - return 0; } #ifdef CONFIG_HOTPLUG_CPU -- cgit v1.2.3 From 8c272261194dfda11cc046fbe808e052f6f284eb Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Mon, 19 May 2014 11:14:23 -0700 Subject: powerpc/numa: Enable USE_PERCPU_NUMA_NODE_ID Based off 3bccd996 for ia64, convert powerpc to use the generic per-CPU topology tracking, specifically: initialize per cpu numa_node entry in start_secondary remove the powerpc cpu_to_node() define CONFIG_USE_PERCPU_NUMA_NODE_ID if NUMA Signed-off-by: Nishanth Aravamudan Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index e2a4232c5871..d7252adea759 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -390,6 +390,7 @@ void smp_prepare_boot_cpu(void) #ifdef CONFIG_PPC64 paca[boot_cpuid].__current = current; #endif + set_numa_node(numa_cpu_lookup_table[boot_cpuid]); current_set[boot_cpuid] = task_thread_info(current); } @@ -750,6 +751,11 @@ void start_secondary(void *unused) } traverse_core_siblings(cpu, true); + /* + * numa_node_id() works after this. + */ + set_numa_node(numa_cpu_lookup_table[cpu]); + smp_wmb(); notify_cpu_starting(cpu); set_cpu_online(cpu, true); -- cgit v1.2.3 From 64bb80d87f01ec01c76863b61b457e0904387f2f Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Fri, 16 May 2014 16:41:20 -0700 Subject: powerpc/numa: Enable CONFIG_HAVE_MEMORYLESS_NODES Based off fd1197f1 for ia64, enable CONFIG_HAVE_MEMORYLESS_NODES if NUMA. Initialize the local memory node in start_secondary. With this commit and the preceding to enable CONFIG_USER_PERCPU_NUMA_NODE_ID, which is a prerequisite, in a PowerKVM guest with the following topology: numactl --hardware available: 3 nodes (0-2) node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 node 0 size: 1998 MB node 0 free: 521 MB node 1 cpus: 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 node 1 size: 0 MB node 1 free: 0 MB node 2 cpus: node 2 size: 2039 MB node 2 free: 1739 MB node distances: node 0 1 2 0: 10 40 40 1: 40 10 40 2: 40 40 10 the unreclaimable slab is reduced by close to 130M: Before: Slab: 418176 kB SReclaimable: 26624 kB SUnreclaim: 391552 kB After: Slab: 298944 kB SReclaimable: 31744 kB SUnreclaim: 267200 kB Signed-off-by: Nishanth Aravamudan Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index d7252adea759..4863ea14f270 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -755,6 +755,7 @@ void start_secondary(void *unused) * numa_node_id() works after this. */ set_numa_node(numa_cpu_lookup_table[cpu]); + set_numa_mem(local_memory_node(numa_cpu_lookup_table[cpu])); smp_wmb(); notify_cpu_starting(cpu); -- cgit v1.2.3 From 441c19c8a290f5f1e1b263691641124c84232b6e Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 May 2014 18:15:25 +1000 Subject: powerpc/kvm/book3s_hv: Rework the secondary inhibit code As part of the support for split core on POWER8, we want to be able to block splitting of the core while KVM VMs are active. The logic to do that would be exactly the same as the code we currently have for inhibiting onlining of secondaries. Instead of adding an identical mechanism to block split core, rework the secondary inhibit code to be a "HV KVM is active" check. We can then use that in both the cpu hotplug code and the upcoming split core code. Signed-off-by: Michael Ellerman Signed-off-by: Michael Neuling Acked-by: Alexander Graf Acked-by: Paul Mackerras Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 34 +++------------------------------- 1 file changed, 3 insertions(+), 31 deletions(-) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 4863ea14f270..5cdd9eb3b24c 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -458,38 +459,9 @@ int generic_check_cpu_restart(unsigned int cpu) return per_cpu(cpu_state, cpu) == CPU_UP_PREPARE; } -static atomic_t secondary_inhibit_count; - -/* - * Don't allow secondary CPU threads to come online - */ -void inhibit_secondary_onlining(void) -{ - /* - * This makes secondary_inhibit_count stable during cpu - * online/offline operations. - */ - get_online_cpus(); - - atomic_inc(&secondary_inhibit_count); - put_online_cpus(); -} -EXPORT_SYMBOL_GPL(inhibit_secondary_onlining); - -/* - * Allow secondary CPU threads to come online again - */ -void uninhibit_secondary_onlining(void) -{ - get_online_cpus(); - atomic_dec(&secondary_inhibit_count); - put_online_cpus(); -} -EXPORT_SYMBOL_GPL(uninhibit_secondary_onlining); - -static int secondaries_inhibited(void) +static bool secondaries_inhibited(void) { - return atomic_read(&secondary_inhibit_count); + return kvm_hv_mode_active(); } #else /* HOTPLUG_CPU */ -- cgit v1.2.3 From 6f5e40a3001d2497a134386a173e3ec3fdf2ad0b Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 May 2014 18:15:28 +1000 Subject: powerpc: Check cpu_thread_in_subcore() in __cpu_up() To support split core we need to change the check in __cpu_up() that determines if a cpu is allowed to come online. Currently we refuse to online cpus which are not the primary thread within their core. On POWER8 with split core support this check needs to instead refuse to online cpus which are not the primary thread within their *sub* core. On POWER7 and other systems that do not support split core, threads_per_subcore == threads_per_core and so the check is equivalent. Signed-off-by: Michael Ellerman Signed-off-by: Michael Neuling Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 5cdd9eb3b24c..6af946e9a984 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -490,7 +490,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle) * Don't allow secondary threads to come online if inhibited */ if (threads_per_core > 1 && secondaries_inhibited() && - cpu % threads_per_core != 0) + cpu_thread_in_subcore(cpu)) return -EBUSY; if (smp_ops == NULL || -- cgit v1.2.3 From 5d4dfddd4f02b028d6ddaaa04d75d3b0cad1c9ae Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 27 May 2014 13:50:41 -0400 Subject: sched: Rename capacity related flags It is better not to think about compute capacity as being equivalent to "CPU power". The upcoming "power aware" scheduler work may create confusion with the notion of energy consumption if "power" is used too liberally. Let's rename the following feature flags since they do relate to capacity: SD_SHARE_CPUPOWER -> SD_SHARE_CPUCAPACITY ARCH_POWER -> ARCH_CAPACITY NONTASK_POWER -> NONTASK_CAPACITY Signed-off-by: Nicolas Pitre Signed-off-by: Peter Zijlstra Cc: Vincent Guittot Cc: Daniel Lezcano Cc: Morten Rasmussen Cc: "Rafael J. Wysocki" Cc: linaro-kernel@lists.linaro.org Cc: Andy Fleming Cc: Anton Blanchard Cc: Benjamin Herrenschmidt Cc: Grant Likely Cc: Linus Torvalds Cc: Michael Ellerman Cc: Paul Gortmaker Cc: Paul Mackerras Cc: Preeti U Murthy Cc: Rob Herring Cc: Srivatsa S. Bhat Cc: Toshi Kani Cc: Vasant Hegde Cc: Vincent Guittot Cc: devicetree@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/n/tip-e93lpnxb87owfievqatey6b5@git.kernel.org Signed-off-by: Ingo Molnar --- arch/powerpc/kernel/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 10ffffef0414..c51d16379cba 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -770,7 +770,7 @@ int setup_profiling_timer(unsigned int multiplier) /* cpumask of CPUs with asymetric SMT dependancy */ static const int powerpc_smt_flags(void) { - int flags = SD_SHARE_CPUPOWER | SD_SHARE_PKG_RESOURCES; + int flags = SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES; if (cpu_has_feature(CPU_FTR_ASYM_SMT)) { printk_once(KERN_INFO "Enabling Asymmetric SMT scheduling\n"); -- cgit v1.2.3 From b6220ad66bcd4a50737eb3c08e9466aa44f3bc98 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Tue, 24 Jun 2014 18:05:29 -0700 Subject: sched: Fix compiler warnings Commit 143e1e28cb (sched: Rework sched_domain topology definition) introduced a number of functions with a return value of 'const int'. gcc doesn't know what to do with that and, if the kernel is compiled with W=1, complains with the following warnings whenever sched.h is included. include/linux/sched.h:875:25: warning: type qualifiers ignored on function return type include/linux/sched.h:882:25: warning: type qualifiers ignored on function return type include/linux/sched.h:889:25: warning: type qualifiers ignored on function return type include/linux/sched.h:1002:21: warning: type qualifiers ignored on function return type Commits fb2aa855 (sched, ARM: Create a dedicated scheduler topology table) and 607b45e9a (sched, powerpc: Create a dedicated topology table) introduce the same warning in the arm and powerpc code. Drop 'const' from the function declarations to fix the problem. The fix for all three patches has to be applied together to avoid compilation failures for the affected architectures. Acked-by: Vincent Guittot Acked-by: Benjamin Herrenschmidt Signed-off-by: Guenter Roeck Cc: Russell King Cc: Paul Mackerras Cc: Dietmar Eggemann Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1403658329-13196-1-git-send-email-linux@roeck-us.net Signed-off-by: Ingo Molnar --- arch/powerpc/kernel/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel/smp.c') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 51a3ff78838a..1007fb802e6b 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -747,7 +747,7 @@ int setup_profiling_timer(unsigned int multiplier) #ifdef CONFIG_SCHED_SMT /* cpumask of CPUs with asymetric SMT dependancy */ -static const int powerpc_smt_flags(void) +static int powerpc_smt_flags(void) { int flags = SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES; -- cgit v1.2.3