x86/topology: Fix AMD core count
Subsystems (3)
Borislav Petkov
maintainerPATCH 1/4 x86/topology: Fix AMD core count
From: Peter Zijlstra <[email protected]>
It turns out AMD gets x86_max_cores wrong when there are compute units.
The issue is that Linux assumes:
nr_logical_cpus = nr_cores * nr_siblings
But AMD reports its CU unit as 2 cores, but then sets num_smp_siblings to 2 as well.
Boris: fixup ras/mce_amd_inj.c too, to compute the Node Base Core properly, according to the new nomenclature.
--- arch/x86/include/asm/smp.h | 1 + arch/x86/kernel/cpu/amd.c | 8 ++++---- arch/x86/ras/mce_amd_inj.c | 3 ++- 3 files changed, 7 insertions(+), 5 deletions(-)
arch/x86/include/asm/smp.h +1 -0
@@ -155,6 +155,7 @@ static inline int wbinvd_on_all_cpus(void)wbinvd (); return 0; } +#define smp_num_siblings 1#endif /* CONFIG_SMP */ extern unsigned disabled_cpus;
arch/x86/kernel/cpu/amd.c +4 -4
@@ -310,9 +310,9 @@ static void amd_get_topology(struct cpuinfo_x86 *c) node_id = ecx &7 ;/* get compute unit information */ - smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; + cores_per_cu = smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; + c->x86_max_cores /= smp_num_siblings; c->compute_unit_id = ebx &0xff ; - cores_per_cu += ((ebx >>8 ) &3 ); }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -328,8 +328,8 @@ static void amd_get_topology(struct cpuinfo_x86 *c)u32 cus_per_node;set_cpu_cap (c,X86_FEATURE_AMD_DCM ); - cores_per_node = c->x86_max_cores / nodes_per_socket; - cus_per_node = cores_per_node / cores_per_cu; + cus_per_node = c->x86_max_cores / nodes_per_socket; + cores_per_node = cus_per_node * cores_per_cu;/* store NodeID, use llc_shared_map to store sibling info */ per_cpu (cpu_llc_id, cpu) = node_id;
arch/x86/ras/mce_amd_inj.c +2 -1
@@ -20,6 +20,7 @@#include <linux/pci.h> #include <asm/mce.h> +#include <asm/smp.h> #include <asm/amd_nb.h> #include <asm/irq_vectors.h>
@@ -206,7 +207,7 @@ static u32 get_nbc_for_node(int node_id)struct cpuinfo_x86 *c = &boot_cpu_data;u32 cores_per_node; - cores_per_node = c->x86_max_cores /amd_get_nodes_per_socket (); + cores_per_node = (c->x86_max_cores * smp_num_siblings) /amd_get_nodes_per_socket ();return cores_per_node * node_id; }
Signature · 1 line
2.7.3
Borislav Petkov
maintainerPATCH 2/4 perf/amd: Cleanup Fam10h NB event constraints
From: Peter Zijlstra <[email protected]>
Avoid allocating the AMD NB event constraints data structure when not needed. This gets rid of x86_max_cores usage and avoids allocating this on AMD Core Perfctr supporting hardware (which has separate MSRs for NB events).
--- arch/x86/events/amd/core.c | 21 ++++++++++++++++++--- arch/x86/events/perf_event.h | 5 +++++ 2 files changed, 23 insertions(+), 3 deletions(-)
arch/x86/events/amd/core.c +18 -3
@@ -369,7 +369,7 @@ static int amd_pmu_cpu_prepare(int cpu)WARN_ON_ONCE (cpuc->amd_nb); -if (boot_cpu_data.x86_max_cores <2 ) +if (!x86_pmu.amd_nb_constraints)return NOTIFY_OK ; cpuc->amd_nb =amd_alloc_nb (cpu);
@@ -388,7 +388,7 @@ static void amd_pmu_cpu_starting(int cpu) cpuc->perf_ctr_virt_mask =AMD64_EVENTSEL_HOSTONLY ; -if (boot_cpu_data.x86_max_cores <2 ) +if (!x86_pmu.amd_nb_constraints)return ; nb_id =amd_get_nb_id (cpu);
@@ -414,7 +414,7 @@ static void amd_pmu_cpu_dead(int cpu) {struct cpu_hw_events *cpuhw; -if (boot_cpu_data.x86_max_cores <2 ) +if (!x86_pmu.amd_nb_constraints)return ; cpuhw = &per_cpu (cpu_hw_events, cpu);
@@ -648,6 +648,8 @@ static __initconst const struct x86_pmu amd_pmu = { .cpu_prepare = amd_pmu_cpu_prepare, .cpu_starting = amd_pmu_cpu_starting, .cpu_dead = amd_pmu_cpu_dead, + + .amd_nb_constraints = 1, };static int __init amd_core_pmu_init (void )
@@ -674,6 +676,11 @@ static int __init amd_core_pmu_init(void) x86_pmu.eventsel =MSR_F15H_PERF_CTL ; x86_pmu.perfctr =MSR_F15H_PERF_CTR ; x86_pmu.num_counters =AMD64_NUM_COUNTERS_CORE ; +/* +* AMD Core perfctr has separate MSRs for the NB events, see +* the amd/uncore.c driver. +*/ + x86_pmu.amd_nb_constraints =0 ;pr_cont ("core perfctr, " );return 0 ;
@@ -693,6 +700,14 @@ __init int amd_pmu_init(void)if (ret)return ret; +if (num_possible_cpus () ==1 ) { +/* +* No point in allocating data structures to serialize +* against other CPUs, when there is only the one CPU. +*/ + x86_pmu.amd_nb_constraints =0 ; + } +/* Events are common for all AMDs */ memcpy (hw_cache_event_ids, amd_hw_cache_event_ids,sizeof (hw_cache_event_ids));
arch/x86/events/perf_event.h +5 -0
@@ -608,6 +608,11 @@ struct x86_pmu {atomic_t lbr_exclusive[x86_lbr_exclusive_max];/* +* AMD bits +*/ + unsigned int amd_nb_constraints : 1; + +/* * Extra registers for events */ struct extra_reg *extra_regs;
Signature · 1 line
2.7.3
Borislav Petkov
maintainerPATCH 3/4 x86/cpu: Get rid of compute_unit_id
In reply to Borislav PetkovFrom: Borislav Petkov <[email protected]>
It is cpu_core_id anyway.
--- arch/x86/include/asm/processor.h | 2 -- arch/x86/kernel/amd_nb.c | 6 ++---- arch/x86/kernel/cpu/amd.c | 10 +++------- arch/x86/kernel/smpboot.c | 2 +- 4 files changed, 6 insertions(+), 14 deletions(-)
arch/x86/include/asm/processor.h +0 -2
@@ -132,8 +132,6 @@ struct cpuinfo_x86 {u16 logical_proc_id;/* Core id: */ u16 cpu_core_id; -/* Compute unit id */ -u8 compute_unit_id;/* Index into per_cpu list: */ u16 cpu_index;u32 microcode;
arch/x86/kernel/amd_nb.c +2 -4
@@ -170,15 +170,13 @@ int amd_get_subcaches(int cpu) {struct pci_dev *link =node_to_amd_nb (amd_get_nb_id (cpu))->link;unsigned int mask; -int cuid;if (!amd_nb_has_feature (AMD_NB_L3_PARTITIONING ))return 0 ;pci_read_config_dword (link,0x1d4 , &mask); - cuid =cpu_data (cpu).compute_unit_id; -return (mask >> (4 * cuid)) &0xf ; +return (mask >> (4 *cpu_data (cpu).cpu_core_id)) &0xf ; }int amd_set_subcaches (int cpu,unsigned long mask)
@@ -204,7 +202,7 @@ int amd_set_subcaches(int cpu, unsigned long mask)pci_write_config_dword (nb->misc,0x1b8 , reg & ~0x180000 ); } - cuid =cpu_data (cpu).compute_unit_id; + cuid =cpu_data (cpu).cpu_core_id; mask <<=4 * cuid; mask |= (0xf ^ (1 << cuid)) <<26 ;
arch/x86/kernel/cpu/amd.c +3 -7
@@ -297,7 +297,6 @@ static int nearby_node(int apicid)#ifdef CONFIG_SMP static void amd_get_topology (struct cpuinfo_x86 *c) { -u32 cores_per_cu =1 ;u8 node_id;int cpu =smp_processor_id ();
@@ -310,9 +309,9 @@ static void amd_get_topology(struct cpuinfo_x86 *c) node_id = ecx &7 ;/* get compute unit information */ - cores_per_cu = smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; + smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; c->x86_max_cores /= smp_num_siblings; - c->compute_unit_id = ebx &0xff ; + c->cpu_core_id = ebx &0xff ; }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -324,19 +323,16 @@ static void amd_get_topology(struct cpuinfo_x86 *c)/* fixup multi-node processor information */ if (nodes_per_socket >1 ) { -u32 cores_per_node;u32 cus_per_node;set_cpu_cap (c,X86_FEATURE_AMD_DCM ); cus_per_node = c->x86_max_cores / nodes_per_socket; - cores_per_node = cus_per_node * cores_per_cu;/* store NodeID, use llc_shared_map to store sibling info */ per_cpu (cpu_llc_id, cpu) = node_id;/* core id has to be in the [0 .. cores_per_node - 1] range */ - c->cpu_core_id %= cores_per_node; - c->compute_unit_id %= cus_per_node; + c->cpu_core_id %= cus_per_node; } }#endif
arch/x86/kernel/smpboot.c +1 -1
@@ -422,7 +422,7 @@ static bool match_smt(struct cpuinfo_x86 *c, struct cpuinfo_x86 *o)if (c->phys_proc_id == o->phys_proc_id &&per_cpu (cpu_llc_id, cpu1) ==per_cpu (cpu_llc_id, cpu2) && - c->compute_unit_id == o->compute_unit_id) + c->cpu_core_id == o->cpu_core_id)return topology_sane (c, o,"smt" ); } else if (c->phys_proc_id == o->phys_proc_id &&
Signature · 1 line
2.7.3
tip-bot for Borislav Petkov
merge bot[tip:x86/urgent] x86/cpu: Get rid of compute_unit_id
x86/cpu: Get rid of compute_unit_id
Commit-ID: 8196dab4fc159943df6baaac04973bb1accb7100 Gitweb: http://git.kernel.org/tip/8196dab4fc159943df6baaac04973bb1accb7100 Author: Borislav Petkov <[email protected]> AuthorDate: Fri, 25 Mar 2016 15:52:36 +0100 Committer: Thomas Gleixner <[email protected]> CommitDate: Tue, 29 Mar 2016 10:45:04 +0200
x86/cpu: Get rid of compute_unit_id
It is cpu_core_id anyway.
--- arch/x86/include/asm/processor.h | 2 -- arch/x86/kernel/amd_nb.c | 6 ++---- arch/x86/kernel/cpu/amd.c | 10 +++------- arch/x86/kernel/smpboot.c | 2 +- 4 files changed, 6 insertions(+), 14 deletions(-)
arch/x86/include/asm/processor.h +0 -2
@@ -132,8 +132,6 @@ struct cpuinfo_x86 {u16 logical_proc_id;/* Core id: */ u16 cpu_core_id; -/* Compute unit id */ -u8 compute_unit_id;/* Index into per_cpu list: */ u16 cpu_index;u32 microcode;
arch/x86/kernel/amd_nb.c +2 -4
@@ -170,15 +170,13 @@ int amd_get_subcaches(int cpu) {struct pci_dev *link =node_to_amd_nb (amd_get_nb_id (cpu))->link;unsigned int mask; -int cuid;if (!amd_nb_has_feature (AMD_NB_L3_PARTITIONING ))return 0 ;pci_read_config_dword (link,0x1d4 , &mask); - cuid =cpu_data (cpu).compute_unit_id; -return (mask >> (4 * cuid)) &0xf ; +return (mask >> (4 *cpu_data (cpu).cpu_core_id)) &0xf ; }int amd_set_subcaches (int cpu,unsigned long mask)
@@ -204,7 +202,7 @@ int amd_set_subcaches(int cpu, unsigned long mask)pci_write_config_dword (nb->misc,0x1b8 , reg & ~0x180000 ); } - cuid =cpu_data (cpu).compute_unit_id; + cuid =cpu_data (cpu).cpu_core_id; mask <<=4 * cuid; mask |= (0xf ^ (1 << cuid)) <<26 ;
arch/x86/kernel/cpu/amd.c +3 -7
@@ -300,7 +300,6 @@ static int nearby_node(int apicid)#ifdef CONFIG_SMP static void amd_get_topology (struct cpuinfo_x86 *c) { -u32 cores_per_cu =1 ;u8 node_id;int cpu =smp_processor_id ();
@@ -312,9 +311,9 @@ static void amd_get_topology(struct cpuinfo_x86 *c) node_id = ecx &7 ;/* get compute unit information */ - cores_per_cu = smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; + smp_num_siblings = ((ebx >>8 ) &3 ) +1 ; c->x86_max_cores /= smp_num_siblings; - c->compute_unit_id = ebx &0xff ; + c->cpu_core_id = ebx &0xff ; }else if (cpu_has (c,X86_FEATURE_NODEID_MSR )) {u64 value;
@@ -325,19 +324,16 @@ static void amd_get_topology(struct cpuinfo_x86 *c)/* fixup multi-node processor information */ if (nodes_per_socket >1 ) { -u32 cores_per_node;u32 cus_per_node;set_cpu_cap (c,X86_FEATURE_AMD_DCM ); cus_per_node = c->x86_max_cores / nodes_per_socket; - cores_per_node = cus_per_node * cores_per_cu;/* store NodeID, use llc_shared_map to store sibling info */ per_cpu (cpu_llc_id, cpu) = node_id;/* core id has to be in the [0 .. cores_per_node - 1] range */ - c->cpu_core_id %= cores_per_node; - c->compute_unit_id %= cus_per_node; + c->cpu_core_id %= cus_per_node; } }#endif
arch/x86/kernel/smpboot.c +1 -1
@@ -422,7 +422,7 @@ static bool match_smt(struct cpuinfo_x86 *c, struct cpuinfo_x86 *o)if (c->phys_proc_id == o->phys_proc_id &&per_cpu (cpu_llc_id, cpu1) ==per_cpu (cpu_llc_id, cpu2) && - c->compute_unit_id == o->compute_unit_id) + c->cpu_core_id == o->cpu_core_id)return topology_sane (c, o,"smt" ); } else if (c->phys_proc_id == o->phys_proc_id &&
Borislav Petkov
maintainerPATCH 4/4 x86/Documentation: Start documenting x86 topology
In reply to Borislav PetkovFrom: Borislav Petkov <[email protected]>
This should contain important aspects of how we represent the system topology on x86. If people have questions about it and this file doesn't answer it, then it must be updated.
--- Documentation/x86/topology.txt | 202 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 202 insertions(+) create mode 100644 Documentation/x86/topology.txt
Documentation/x86/topology.txt new file +202 -0
@@ -0,0 +1,202 @@ +x86 Topology +============ + +This documents and clarifies the main aspects of x86 topology modelling +and representation in the kernel. Update/change when doing changes to +the respective code. + +Started by Thomas Gleixner <[email protected]> and Borislav Petkov <[email protected]>. + +The main aim of the topology facilities is to present adequate +interfaces to code which needs to know/query/use the structure of the +running system wrt threads, cores, packages, etc. + +The kernel does not care about the concept of physical sockets because +a socket has no relevance to software. It's an electromechanical +component. In the past a socket always contained a single package +(see below), but with the advent of Multi Chip Modules (MCM) a socket +can hold more than one package. So there might be still references to +sockets in the code, but they are of historical nature and should be +cleaned up. + +The topology of a system is described in the units of: + + - packages + - cores + - threads + +* Package: + + Packages contain a number of cores plus shared resources, e.g. DRAM + controller, shared caches etc. + + AMD nomenclature for package is 'Node'. + + Package-related topology information in the kernel: + + - cpuinfo_x86.x86_max_cores: + + The number of cores in a package. This information is retrieved via CPUID. + + - cpuinfo_x86.phys_proc_id: + + The physical ID of the package. This information is retrieved via CPUID + and deduced from the APIC IDs of the cores in the package. + + - cpuinfo_x86.logical_id: + + The logical ID of the package. As we do not trust BIOSes to enumerate the + packages in a consistent way, we introduced the concept of logical package + ID so we can sanely calculate the number of maximum possible packages in + the system and have the packages enumerated linearly. + + - topology_max_packages(): + + The maximum possible number of packages in the system. Helpful for per + package facilities to preallocate per package information. + + +* Cores: + + A core consists of 1 or more threads. It does not matter whether the threads + are SMT- or CMT-type threads. + + AMDs nomenclature for a CMT core is "Compute Unit". The kernel always uses + "core". + + Core-related topology information in the kernel: + + - smp_num_siblings: + + The number of threads in a core. The number of threads in a package can be + calculated by: + + threads_per_package = cpuinfo_x86.x86_max_cores * smp_num_siblings + + +* Threads: + + A thread is a single scheduling unit. It's the equivalent to a logical Linux + CPU. + + AMDs nomenclature for CMT threads is "Compute Unit Core". The kernel always + uses "thread". + + Thread-related topology information in the kernel: + + - topology_core_cpumask(): + + The cpumask contains all online threads in the package to which a thread + belongs. + + The number of online threads is also printed in /proc/cpuinfo "siblings." + + - topology_sibling_mask(): + + The cpumask contains all online threads in the core to which a thread + belongs. + + - topology_logical_package_id(): + + The logical package ID to which a thread belongs. + + - topology_physical_package_id(): + + The physical package ID to which a thread belongs. + + - topology_core_id(); + + The ID of the core to which a thread belongs. It is also printed in /proc/cpuinfo + "core_id." + + + +System topology examples + +Note: + +The alternative Linux CPU enumeration depends on how the BIOS enumerates the +threads. Many BIOSes enumerate all threads 0 first and then all threads 1. +That has the "advantage" that the logical Linux CPU numbers of threads 0 stay +the same whether threads are enabled or not. That's merely an implementation +detail and has no practical impact. + +1) Single Package, Single Core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + +2) Single Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 3 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + +4) Dual Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 3 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 4 + -> [thread 1] -> Linux CPU 5 + -> [core 1] -> [thread 0] -> Linux CPU 6 + -> [thread 1] -> Linux CPU 7 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 4 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 5 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 6 + -> [core 1] -> [thread 0] -> Linux CPU 3 + -> [thread 1] -> Linux CPU 7 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + + [node 1] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 4 + -> [Compute Unit Core 1] -> Linux CPU 5 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 6 + -> [Compute Unit Core 1] -> Linux CPU 7 + +
Signature · 1 line
2.7.3
Borislav Petkov
maintainerPATCH 4/4 x86/Documentation: Start documenting x86 topology
From: Borislav Petkov <[email protected]>
This should contain important aspects of how we represent the system topology on x86. If people have questions about it and this file doesn't answer it, then it must be updated.
---
Let me add a reference to the generic topology document Documentation/cputopology.txt too.
Documentation/x86/topology.txt | 212 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 212 insertions(+) create mode 100644 Documentation/x86/topology.txt
Documentation/x86/topology.txt new file +212 -0
@@ -0,0 +1,212 @@ +x86 Topology +============ + +This documents and clarifies the main aspects of x86 topology modelling +and representation in the kernel. Update/change when doing changes to +the respective code. + +The architecture-agnostic topology definitions are in +Documentation/cputopology.txt. This file holds x86-specific +differences/specialities which must not necessarily apply to the generic +definitions. Thus, the way to read up on Linux topology on x86 is to +start with the generic one and look at this one in parallel for the x86 +specifics. + +Needless to say, code should use the generic functions - this file is +*only* here to *document* the inner workings of x86 topology. + +Started by Thomas Gleixner <[email protected]> and Borislav Petkov <[email protected]>. + +The main aim of the topology facilities is to present adequate +interfaces to code which needs to know/query/use the structure of the +running system wrt threads, cores, packages, etc. + +The kernel does not care about the concept of physical sockets because +a socket has no relevance to software. It's an electromechanical +component. In the past a socket always contained a single package +(see below), but with the advent of Multi Chip Modules (MCM) a socket +can hold more than one package. So there might be still references to +sockets in the code, but they are of historical nature and should be +cleaned up. + +The topology of a system is described in the units of: + + - packages + - cores + - threads + +* Package: + + Packages contain a number of cores plus shared resources, e.g. DRAM + controller, shared caches etc. + + AMD nomenclature for package is 'Node'. + + Package-related topology information in the kernel: + + - cpuinfo_x86.x86_max_cores: + + The number of cores in a package. This information is retrieved via CPUID. + + - cpuinfo_x86.phys_proc_id: + + The physical ID of the package. This information is retrieved via CPUID + and deduced from the APIC IDs of the cores in the package. + + - cpuinfo_x86.logical_id: + + The logical ID of the package. As we do not trust BIOSes to enumerate the + packages in a consistent way, we introduced the concept of logical package + ID so we can sanely calculate the number of maximum possible packages in + the system and have the packages enumerated linearly. + + - topology_max_packages(): + + The maximum possible number of packages in the system. Helpful for per + package facilities to preallocate per package information. + + +* Cores: + + A core consists of 1 or more threads. It does not matter whether the threads + are SMT- or CMT-type threads. + + AMDs nomenclature for a CMT core is "Compute Unit". The kernel always uses + "core". + + Core-related topology information in the kernel: + + - smp_num_siblings: + + The number of threads in a core. The number of threads in a package can be + calculated by: + + threads_per_package = cpuinfo_x86.x86_max_cores * smp_num_siblings + + +* Threads: + + A thread is a single scheduling unit. It's the equivalent to a logical Linux + CPU. + + AMDs nomenclature for CMT threads is "Compute Unit Core". The kernel always + uses "thread". + + Thread-related topology information in the kernel: + + - topology_core_cpumask(): + + The cpumask contains all online threads in the package to which a thread + belongs. + + The number of online threads is also printed in /proc/cpuinfo "siblings." + + - topology_sibling_mask(): + + The cpumask contains all online threads in the core to which a thread + belongs. + + - topology_logical_package_id(): + + The logical package ID to which a thread belongs. + + - topology_physical_package_id(): + + The physical package ID to which a thread belongs. + + - topology_core_id(); + + The ID of the core to which a thread belongs. It is also printed in /proc/cpuinfo + "core_id." + + + +System topology examples + +Note: + +The alternative Linux CPU enumeration depends on how the BIOS enumerates the +threads. Many BIOSes enumerate all threads 0 first and then all threads 1. +That has the "advantage" that the logical Linux CPU numbers of threads 0 stay +the same whether threads are enabled or not. That's merely an implementation +detail and has no practical impact. + +1) Single Package, Single Core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + +2) Single Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 3 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + +4) Dual Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 3 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 4 + -> [thread 1] -> Linux CPU 5 + -> [core 1] -> [thread 0] -> Linux CPU 6 + -> [thread 1] -> Linux CPU 7 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 4 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 5 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 6 + -> [core 1] -> [thread 0] -> Linux CPU 3 + -> [thread 1] -> Linux CPU 7 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + + [node 1] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 4 + -> [Compute Unit Core 1] -> Linux CPU 5 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 6 + -> [Compute Unit Core 1] -> Linux CPU 7 + +
Signature · 8 lines
2.7.3
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.tip-bot for Borislav Petkov
merge bot[tip:x86/urgent] x86/Documentation: Start documenting x86 topology
x86/Documentation: Start documenting x86 topology
Commit-ID: f7be8610bca88e59dd2fd5d98fcbc5031ef0e079 Gitweb: http://git.kernel.org/tip/f7be8610bca88e59dd2fd5d98fcbc5031ef0e079 Author: Borislav Petkov <[email protected]> AuthorDate: Mon, 28 Mar 2016 11:56:09 +0200 Committer: Thomas Gleixner <[email protected]> CommitDate: Tue, 29 Mar 2016 10:45:04 +0200
x86/Documentation: Start documenting x86 topology
This should contain important aspects of how we represent the system topology on x86. If people have questions about it and this file doesn't answer it, then it must be updated.
--- Documentation/x86/topology.txt | 208 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 208 insertions(+)
Documentation/x86/topology.txt new file +208 -0
@@ -0,0 +1,208 @@ +x86 Topology +============ + +This documents and clarifies the main aspects of x86 topology modelling and +representation in the kernel. Update/change when doing changes to the +respective code. + +The architecture-agnostic topology definitions are in +Documentation/cputopology.txt. This file holds x86-specific +differences/specialities which must not necessarily apply to the generic +definitions. Thus, the way to read up on Linux topology on x86 is to start +with the generic one and look at this one in parallel for the x86 specifics. + +Needless to say, code should use the generic functions - this file is *only* +here to *document* the inner workings of x86 topology. + +Started by Thomas Gleixner <[email protected]> and Borislav Petkov <[email protected]>. + +The main aim of the topology facilities is to present adequate interfaces to +code which needs to know/query/use the structure of the running system wrt +threads, cores, packages, etc. + +The kernel does not care about the concept of physical sockets because a +socket has no relevance to software. It's an electromechanical component. In +the past a socket always contained a single package (see below), but with the +advent of Multi Chip Modules (MCM) a socket can hold more than one package. So +there might be still references to sockets in the code, but they are of +historical nature and should be cleaned up. + +The topology of a system is described in the units of: + + - packages + - cores + - threads + +* Package: + + Packages contain a number of cores plus shared resources, e.g. DRAM + controller, shared caches etc. + + AMD nomenclature for package is 'Node'. + + Package-related topology information in the kernel: + + - cpuinfo_x86.x86_max_cores: + + The number of cores in a package. This information is retrieved via CPUID. + + - cpuinfo_x86.phys_proc_id: + + The physical ID of the package. This information is retrieved via CPUID + and deduced from the APIC IDs of the cores in the package. + + - cpuinfo_x86.logical_id: + + The logical ID of the package. As we do not trust BIOSes to enumerate the + packages in a consistent way, we introduced the concept of logical package + ID so we can sanely calculate the number of maximum possible packages in + the system and have the packages enumerated linearly. + + - topology_max_packages(): + + The maximum possible number of packages in the system. Helpful for per + package facilities to preallocate per package information. + + +* Cores: + + A core consists of 1 or more threads. It does not matter whether the threads + are SMT- or CMT-type threads. + + AMDs nomenclature for a CMT core is "Compute Unit". The kernel always uses + "core". + + Core-related topology information in the kernel: + + - smp_num_siblings: + + The number of threads in a core. The number of threads in a package can be + calculated by: + + threads_per_package = cpuinfo_x86.x86_max_cores * smp_num_siblings + + +* Threads: + + A thread is a single scheduling unit. It's the equivalent to a logical Linux + CPU. + + AMDs nomenclature for CMT threads is "Compute Unit Core". The kernel always + uses "thread". + + Thread-related topology information in the kernel: + + - topology_core_cpumask(): + + The cpumask contains all online threads in the package to which a thread + belongs. + + The number of online threads is also printed in /proc/cpuinfo "siblings." + + - topology_sibling_mask(): + + The cpumask contains all online threads in the core to which a thread + belongs. + + - topology_logical_package_id(): + + The logical package ID to which a thread belongs. + + - topology_physical_package_id(): + + The physical package ID to which a thread belongs. + + - topology_core_id(); + + The ID of the core to which a thread belongs. It is also printed in /proc/cpuinfo + "core_id." + + + +System topology examples + +Note: + +The alternative Linux CPU enumeration depends on how the BIOS enumerates the +threads. Many BIOSes enumerate all threads 0 first and then all threads 1. +That has the "advantage" that the logical Linux CPU numbers of threads 0 stay +the same whether threads are enabled or not. That's merely an implementation +detail and has no practical impact. + +1) Single Package, Single Core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + +2) Single Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 3 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + +4) Dual Package, Dual Core + + a) One thread per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [core 1] -> [thread 0] -> Linux CPU 1 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [core 1] -> [thread 0] -> Linux CPU 3 + + b) Two threads per core + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 1 + -> [core 1] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 3 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 4 + -> [thread 1] -> Linux CPU 5 + -> [core 1] -> [thread 0] -> Linux CPU 6 + -> [thread 1] -> Linux CPU 7 + + Alternative enumeration: + + [package 0] -> [core 0] -> [thread 0] -> Linux CPU 0 + -> [thread 1] -> Linux CPU 4 + -> [core 1] -> [thread 0] -> Linux CPU 1 + -> [thread 1] -> Linux CPU 5 + + [package 1] -> [core 0] -> [thread 0] -> Linux CPU 2 + -> [thread 1] -> Linux CPU 6 + -> [core 1] -> [thread 0] -> Linux CPU 3 + -> [thread 1] -> Linux CPU 7 + + AMD nomenclature for CMT systems: + + [node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0 + -> [Compute Unit Core 1] -> Linux CPU 1 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2 + -> [Compute Unit Core 1] -> Linux CPU 3 + + [node 1] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 4 + -> [Compute Unit Core 1] -> Linux CPU 5 + -> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 6 + -> [Compute Unit Core 1] -> Linux CPU 7