s390/mm: provide minimal setup_per_cpu_areas() implementation

s390 allows to enable CONFIG_NUMA, mainly to enable a couple of system
calls which are only present if NUMA is enabled. The NUMA specific system
calls are required by a couple of applications, which wouldn't work if the
system calls wouldn't be present.

The NUMA implementation itself maps all CPUs and memory to node 0. A
special case is the generic percpu setup code, which doesn't expect an s390
like implementation and therefore emits a message/warning:
"percpu: cpu 0 has no node -1 or node-local memory".

In order to get rid of this message, and also to provide sane CPU to node
and CPU distance mappings implement a minimal setup_per_cpu_areas()
function, which is very close to the generic variant.
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>

arch/s390/Kconfig

@@ -197,6 +197,7 @@ config S390
 	select HAVE_RSEQ
 	select HAVE_SAMPLE_FTRACE_DIRECT
 	select HAVE_SAMPLE_FTRACE_DIRECT_MULTI
+	select HAVE_SETUP_PER_CPU_AREA
 	select HAVE_SOFTIRQ_ON_OWN_STACK
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_VIRT_CPU_ACCOUNTING
@@ -208,6 +209,7 @@ config S390
 	select MMU_GATHER_MERGE_VMAS
 	select MODULES_USE_ELF_RELA
 	select NEED_DMA_MAP_STATE	if PCI
+	select NEED_PER_CPU_EMBED_FIRST_CHUNK
 	select NEED_SG_DMA_LENGTH	if PCI
 	select OLD_SIGACTION
 	select OLD_SIGSUSPEND3

arch/s390/mm/init.c

@@ -31,6 +31,7 @@
 #include <linux/cma.h>
 #include <linux/gfp.h>
 #include <linux/dma-direct.h>
+#include <linux/percpu.h>
 #include <asm/processor.h>
 #include <linux/uaccess.h>
 #include <asm/pgalloc.h>
@@ -222,6 +223,41 @@ unsigned long memory_block_size_bytes(void)
 	return max_t(unsigned long, MIN_MEMORY_BLOCK_SIZE, sclp.rzm);
 }
 
+unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(__per_cpu_offset);
+
+static int __init pcpu_cpu_distance(unsigned int from, unsigned int to)
+{
+	return LOCAL_DISTANCE;
+}
+
+static int __init pcpu_cpu_to_node(int cpu)
+{
+	return 0;
+}
+
+void __init setup_per_cpu_areas(void)
+{
+	unsigned long delta;
+	unsigned int cpu;
+	int rc;
+
+	/*
+	 * Always reserve area for module percpu variables.  That's
+	 * what the legacy allocator did.
+	 */
+	rc = pcpu_embed_first_chunk(PERCPU_MODULE_RESERVE,
+				    PERCPU_DYNAMIC_RESERVE, PAGE_SIZE,
+				    pcpu_cpu_distance,
+				    pcpu_cpu_to_node);
+	if (rc < 0)
+		panic("Failed to initialize percpu areas.");
+
+	delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
+	for_each_possible_cpu(cpu)
+		__per_cpu_offset[cpu] = delta + pcpu_unit_offsets[cpu];
+}
+
 #ifdef CONFIG_MEMORY_HOTPLUG
 
 #ifdef CONFIG_CMA