x86: make NUMA work on 32-bit again

On 32-bit NUMA, the memmap representing struct pages on each node is
allocated from node-local memory if possible. As only node-0 has memory from
ZONE_NORMAL, the memmap must be mapped into low memory. This is done by
reserving space in the Kernel Virtual Area (KVA) for the memmap belonging
to other nodes by taking pages from the end of ZONE_NORMAL and remapping
the other nodes memmap into those virtual addresses. The node boundaries
are then adjusted so that the region of pages is not used and it is marked
as reserved in the bootmem allocator.

This reserved portion of the KVA is PMD aligned althought
strictly speaking that requirement could be lifted (see thread at
http://lkml.org/lkml/2007/8/24/220). The problem is that when aligned, there
may be a portion of ZONE_NORMAL at the end that is not used for memmap and
does not have an initialised memmap nor is it marked reserved in the bootmem
allocator. Later in the boot process, these pages are freed and a storm of
Bad page state messages result.

This patch marks these pages reserved that are wasted due to alignment
in the bootmem allocator so they are not accidently freed. It is worth
noting that memory from node-0 is wasted where it could have been put into
ZONE_HIGHMEM on NUMA machines. Worse, the KVA is always reserved from the
location of real memory even when there is plenty of spare virtual address
space.

This patch also makes sure that reserve_bootmem() is not called with a
0-length size in numa_kva_reserve().  When this happens, it usually means
that a kernel built for Summit is being booted on a normal machine. The
resulting BUG_ON() is misleading so it is caught here.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index 88a7499..9f1d02c 100644 (file)
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -268,6 +268,7 @@ unsigned long __init setup_memory(void)
 {
        int nid;
        unsigned long system_start_pfn, system_max_low_pfn;
+       unsigned long wasted_pages;
 
        /*
         * When mapping a NUMA machine we allocate the node_mem_map arrays
@@ -292,7 +293,14 @@ unsigned long __init setup_memory(void)
                kva_start_pfn = PFN_DOWN(initrd_start - PAGE_OFFSET)
                        - kva_pages;
 #endif
-       kva_start_pfn -= kva_start_pfn & (PTRS_PER_PTE-1);
+
+       /*
+        * We waste pages past at the end of the KVA for no good reason other
+        * than how it is located. This is bad.
+        */
+       wasted_pages = kva_start_pfn & (PTRS_PER_PTE-1);
+       kva_start_pfn -= wasted_pages;
+       kva_pages += wasted_pages;
 
        system_max_low_pfn = max_low_pfn = find_max_low_pfn();
        printk("kva_start_pfn ~ %ld find_max_low_pfn() ~ %ld\n",
@@ -345,7 +353,8 @@ unsigned long __init setup_memory(void)
 
 void __init numa_kva_reserve(void)
 {
-       reserve_bootmem(PFN_PHYS(kva_start_pfn),PFN_PHYS(kva_pages));
+       if (kva_pages)
+               reserve_bootmem(PFN_PHYS(kva_start_pfn), PFN_PHYS(kva_pages));
 }
 
 void __init zone_sizes_init(void)