[v5,12/17] xen/arm: enable device tree based NUMA in system init

Commit Message

Henry Wang May 29, 2023, 2:19 a.m. UTC

From: Wei Chen <[e-mail]>

In this patch, we can start to create NUMA system that is
based on device tree.

Signed-off-by: Wei Chen <[e-mail]>
Signed-off-by: Henry Wang <[e-mail]>
---
v4 -> v5:
1. Fix coding style: label indented by 1 space.
v3 -> v4:
1. No change.
v2 -> v3:
1. No change.
v1 -> v2:
1. replace ~0 by INVALID_PADDR.
2. only print error messages for invalid dtb data.
3. remove unnecessary return.
4. remove the parameter of numa_init.
---
 xen/arch/arm/include/asm/numa.h |  5 +++
 xen/arch/arm/numa.c             | 57 +++++++++++++++++++++++++++++++++
 xen/arch/arm/setup.c            |  7 ++++
 3 files changed, 69 insertions(+)

Patch

diff --git a/xen/arch/arm/include/asm/numa.h b/xen/arch/arm/include/asm/numa.h
index 15308f5a36..55ac4665db 100644
--- a/xen/arch/arm/include/asm/numa.h
+++ b/xen/arch/arm/include/asm/numa.h
@@ -45,6 +45,7 @@  extern void numa_set_distance(nodeid_t from, nodeid_t to,
                               unsigned int distance);
 extern void numa_detect_cpu_node(unsigned int cpu);
 extern int numa_device_tree_init(const void *fdt);
+extern void numa_init(void);
 
 #else
 
@@ -90,6 +91,10 @@  static inline void numa_detect_cpu_node(unsigned int cpu)
 {
 }
 
+static inline void numa_init(void)
+{
+}
+
 #endif
 
 #define arch_want_default_dmazone() (false)
diff --git a/xen/arch/arm/numa.c b/xen/arch/arm/numa.c
index 08e15ebbb0..13a167fc4f 100644
--- a/xen/arch/arm/numa.c
+++ b/xen/arch/arm/numa.c
@@ -18,7 +18,11 @@ 
  *
  */
 #include <xen/init.h>
+#include <xen/device_tree.h>
+#include <xen/nodemask.h>
 #include <xen/numa.h>
+#include <xen/pfn.h>
+#include <xen/acpi.h>
 
 enum dt_numa_status __ro_after_init device_tree_numa = DT_NUMA_DEFAULT;
 
@@ -104,6 +108,59 @@  unsigned char __node_distance(nodeid_t from, nodeid_t to)
     return node_distance_map[from][to];
 }
 
+void __init numa_init(void)
+{
+    unsigned int idx;
+    paddr_t ram_start = INVALID_PADDR;
+    paddr_t ram_size = 0;
+    paddr_t ram_end = 0;
+
+    /* NUMA has been turned off through Xen parameters */
+    if ( numa_off )
+        goto mem_init;
+
+    /* Initialize NUMA from device tree when system is not ACPI booted */
+    if ( acpi_disabled )
+    {
+        int ret = numa_device_tree_init(device_tree_flattened);
+        if ( ret )
+        {
+            numa_off = true;
+            if ( ret == -EINVAL )
+                printk(XENLOG_WARNING
+                       "Init NUMA from device tree failed, ret=%d\n", ret);
+        }
+    }
+    else
+    {
+        /* We don't support NUMA for ACPI boot currently */
+        printk(XENLOG_WARNING
+               "ACPI NUMA has not been supported yet, NUMA off!\n");
+        numa_off = true;
+    }
+
+ mem_init:
+    /*
+     * Find the minimal and maximum address of RAM, NUMA will
+     * build a memory to node mapping table for the whole range.
+     */
+    ram_start = bootinfo.mem.bank[0].start;
+    ram_size  = bootinfo.mem.bank[0].size;
+    ram_end   = ram_start + ram_size;
+    for ( idx = 1 ; idx < bootinfo.mem.nr_banks; idx++ )
+    {
+        paddr_t bank_start = bootinfo.mem.bank[idx].start;
+        paddr_t bank_size = bootinfo.mem.bank[idx].size;
+        paddr_t bank_end = bank_start + bank_size;
+
+        ram_size  = ram_size + bank_size;
+        ram_start = min(ram_start, bank_start);
+        ram_end   = max(ram_end, bank_end);
+    }
+
+    numa_initmem_init(PFN_UP(ram_start), PFN_DOWN(ram_end));
+}
+
 int __init arch_get_ram_range(unsigned int idx, paddr_t *start, paddr_t *end)
 {
     if ( idx >= bootinfo.mem.nr_banks )
diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c
index ab9eb6fb80..bfcb0c7b6b 100644
--- a/xen/arch/arm/setup.c
+++ b/xen/arch/arm/setup.c
@@ -1121,6 +1121,13 @@  void __init start_xen(unsigned long boot_phys_offset,
     /* Parse the ACPI tables for possible boot-time configuration */
     acpi_boot_table_init();
 
+    /*
+     * Try to initialize NUMA system, if failed, the system will
+     * fallback to uniform system which means system has only 1
+     * NUMA node.
+     */
+    numa_init();
+
     end_boot_allocator();
 
     /*