x86/topo: Add topology_num_nodes_per_package()

Use the MADT and SRAT table data to compute __num_nodes_per_package.

Specifically, SRAT has already been parsed in x86_numa_init(), which is called
before acpi_boot_init() which parses MADT. So both are available in
topology_init_possible_cpus().

This number is useful to divinate the various Intel CoD/SNC and AMD NPS modes,
since the platforms are failing to provide this otherwise.

Doing it this way is independent of the number of online CPUs and
other such shenanigans.

Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>
Reviewed-by: Ingo Molnar <[e-mail]>
Tested-by: Tony Luck <[e-mail]>
Tested-by: K Prateek Nayak <[e-mail]>
Tested-by: Zhang Rui <[e-mail]>
Tested-by: Chen Yu <[e-mail]>
Tested-by: Kyle Meyer <[e-mail]>
Link: https://patch.msgid.link/[e-mail]

arch/x86/include/asm/topology.h

···
 extern unsigned int __max_threads_per_core;
 extern unsigned int __num_threads_per_package;
 extern unsigned int __num_cores_per_package;
+extern unsigned int __num_nodes_per_package;
 
 const char *get_topology_cpu_type_name(struct cpuinfo_x86 *c);
 enum x86_topology_cpu_type get_topology_cpu_type(struct cpuinfo_x86 *c);
···
 static inline unsigned int topology_num_threads_per_package(void)
 {
 	return __num_threads_per_package;
+}
+
+static inline unsigned int topology_num_nodes_per_package(void)
+{
+	return __num_nodes_per_package;
 }
 
 #ifdef CONFIG_X86_LOCAL_APIC

arch/x86/kernel/cpu/common.c

···
 unsigned int __max_logical_packages __ro_after_init = 1;
 EXPORT_SYMBOL(__max_logical_packages);
 
+unsigned int __num_nodes_per_package __ro_after_init = 1;
+EXPORT_SYMBOL(__num_nodes_per_package);
+
 unsigned int __num_cores_per_package __ro_after_init = 1;
 EXPORT_SYMBOL(__num_cores_per_package);
 

arch/x86/kernel/cpu/topology.c

···
 #include <asm/mpspec.h>
 #include <asm/msr.h>
 #include <asm/smp.h>
+#include <asm/numa.h>
 
 #include "cpu.h"
 
···
 	set_nr_cpu_ids(allowed);
 
 	cnta = domain_weight(TOPO_PKG_DOMAIN);
-	cntb = domain_weight(TOPO_DIE_DOMAIN);
 	__max_logical_packages = cnta;
+
+	pr_info("Max. logical packages: %3u\n", __max_logical_packages);
+
+	cntb = num_phys_nodes();
+	__num_nodes_per_package = DIV_ROUND_UP(cntb, cnta);
+
+	pr_info("Max. logical nodes:    %3u\n", cntb);
+	pr_info("Num. nodes per package:%3u\n", __num_nodes_per_package);
+
+	cntb = domain_weight(TOPO_DIE_DOMAIN);
 	__max_dies_per_package = 1U << (get_count_order(cntb) - get_count_order(cnta));
 
-	pr_info("Max. logical packages: %3u\n", cnta);
 	pr_info("Max. logical dies:     %3u\n", cntb);
 	pr_info("Max. dies per package: %3u\n", __max_dies_per_package);