x86, numa: Fix cpu to node mapping for sparse node ids

NUMA boot code assumes that physical node ids start at 0, but the DIMMs
that the apic id represents may not be reachable.  If this is the case,
node 0 is never online and cpus never end up getting appropriately
assigned to a node.  This causes the cpumask of all online nodes to be
empty and machines crash with kernel code assuming online nodes have
valid cpus.

The fix is to appropriately map all the address ranges for physical nodes
and ensure the cpu to node mapping function checks all possible nodes (up
to MAX_NUMNODES) instead of simply checking nodes 0-N, where N is the
number of physical nodes, for valid address ranges.

This requires no longer "compressing" the address ranges of nodes in the
physical node map from 0-N, but rather leave indices in physnodes[] to
represent the actual node id of the physical node.  Accordingly, the
topology exported by both amd_get_nodes() and acpi_get_nodes() no longer
must return the number of nodes to iterate through; all such iterations
will now be to MAX_NUMNODES.

This change also passes the end address of system RAM (which may be
different from normal operation if mem= is specified on the command line)
before the physnodes[] array is populated.  ACPI parsed nodes are
truncated to fit within the address range that respect the mem=
boundaries and even some physical nodes may become unreachable in such
cases.

When NUMA emulation does succeed, any apicid to node mapping that exists
for unreachable nodes are given default values so that proximity domains
can still be assigned.  This is important for node_distance() to
function as desired.

Signed-off-by: David Rientjes <rientjes@google.com>
LKML-Reference: <alpine.DEB.2.00.1012221702090.3701@chino.kir.corp.google.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>

1 files changed, 3 insertions, 15 deletions

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index dd300c491f1..3d73201ba34 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -266,25 +266,24 @@ static char *cmdline __initdata;
 static int __init setup_physnodes(unsigned long start, unsigned long end,
 					int acpi, int amd)
 {
-	int nr_nodes = 0;
 	int ret = 0;
 	int i;
 
 	memset(physnodes, 0, sizeof(physnodes));
 #ifdef CONFIG_ACPI_NUMA
 	if (acpi)
-		nr_nodes = acpi_get_nodes(physnodes);
+		acpi_get_nodes(physnodes, start, end);
 #endif
 #ifdef CONFIG_AMD_NUMA
 	if (amd)
-		nr_nodes = amd_get_nodes(physnodes);
+		amd_get_nodes(physnodes);
 #endif
 	/*
 	 * Basic sanity checking on the physical node map: there may be errors
 	 * if the SRAT or AMD code incorrectly reported the topology or the mem=
 	 * kernel parameter is used.
 	 */
-	for (i = 0; i < nr_nodes; i++) {
+	for (i = 0; i < MAX_NUMNODES; i++) {
 		if (physnodes[i].start == physnodes[i].end)
 			continue;
 		if (physnodes[i].start > end) {
@@ -299,17 +298,6 @@ static int __init setup_physnodes(unsigned long start, unsigned long end,
 			physnodes[i].start = start;
 		if (physnodes[i].end > end)
 			physnodes[i].end = end;
-	}
-
-	/*
-	 * Remove all nodes that have no memory or were truncated because of the
-	 * limited address range.
-	 */
-	for (i = 0; i < nr_nodes; i++) {
-		if (physnodes[i].start == physnodes[i].end)
-			continue;
-		physnodes[ret].start = physnodes[i].start;
-		physnodes[ret].end = physnodes[i].end;
 		ret++;
 	}