[PATCH v2] properly reserve in bootmem the lmb reserved regions that cross NUMA nodes

[PATCH v2] properly reserve in bootmem the lmb reserved regions that cross NUMA nodes

[Jon Tollefson]

If there are multiple reserved memory blocks via lmb_reserve() that are
contiguous addresses and on different NUMA nodes we are losing track of which 
address ranges to reserve in bootmem on which node.  I discovered this 
when I only recently got to try 16GB huge pages on a system with more 
then 2 nodes.

When scanning the device tree in early boot we call lmb_reserve() with 
the addresses of the 16G pages that we find so that the memory doesn't 
get used for something else.  For example the addresses for the pages 
could be 4000000000, 4400000000, 4800000000, 4C00000000, etc - 8 pages, 
one on each of eight nodes.  In the lmb after all the pages have been 
reserved it will look something like the following:

lmb_dump_all:
    memory.cnt            = 0x2
    memory.size           = 0x3e80000000
    memory.region[0x0].base       = 0x0
                      .size     = 0x1e80000000
    memory.region[0x1].base       = 0x4000000000
                      .size     = 0x2000000000
    reserved.cnt          = 0x5
    reserved.size         = 0x3e80000000
    reserved.region[0x0].base       = 0x0
                      .size     = 0x7b5000
    reserved.region[0x1].base       = 0x2a00000
                      .size     = 0x78c000
    reserved.region[0x2].base       = 0x328c000
                      .size     = 0x43000
    reserved.region[0x3].base       = 0xf4e8000
                      .size     = 0xb18000
    reserved.region[0x4].base       = 0x4000000000
                      .size     = 0x2000000000


The reserved.region[0x4] contains the 16G pages.  In 
arch/powerpc/mm/num.c: do_init_bootmem() we loop through each of the 
node numbers looking for the reserved regions that belong to the 
particular node.  It is not able to identify region 0x4 as being a part 
of each of the 8 nodes.  It is assuming that a reserved region is only
on a single node.

This patch takes out the reserved region loop from inside
the loop that goes over each node.  It looks up the active region containing
the start of the reserved region.  If it extends past that active region then
it adjusts the size and gets the next active region containing it.


Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
---

Changes:
    -style changes as suggested by Adam Litke


Please consider for 2.6.28.


 arch/powerpc/mm/numa.c |   63 ++++++++++++++++++++++++++++---------------------
 include/linux/mm.h     |    2 +
 mm/page_alloc.c        |   19 ++++++++++++++
 3 files changed, 57 insertions(+), 27 deletions(-)


diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index d9a1813..9a3b0c9 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -837,36 +837,45 @@ void __init do_init_bootmem(void)
 				  start_pfn, end_pfn);
 
 		free_bootmem_with_active_regions(nid, end_pfn);
+	}
 
-		/* Mark reserved regions on this node */
-		for (i = 0; i < lmb.reserved.cnt; i++) {
-			unsigned long physbase = lmb.reserved.region[i].base;
-			unsigned long size = lmb.reserved.region[i].size;
-			unsigned long start_paddr = start_pfn << PAGE_SHIFT;
-			unsigned long end_paddr = end_pfn << PAGE_SHIFT;
-
-			if (early_pfn_to_nid(physbase >> PAGE_SHIFT) != nid &&
-			    early_pfn_to_nid((physbase+size-1) >> PAGE_SHIFT) != nid)
-				continue;
-
-			if (physbase < end_paddr &&
-			    (physbase+size) > start_paddr) {
-				/* overlaps */
-				if (physbase < start_paddr) {
-					size -= start_paddr - physbase;
-					physbase = start_paddr;
-				}
-
-				if (size > end_paddr - physbase)
-					size = end_paddr - physbase;
-
-				dbg("reserve_bootmem %lx %lx\n", physbase,
-				    size);
-				reserve_bootmem_node(NODE_DATA(nid), physbase,
-						     size, BOOTMEM_DEFAULT);
-			}
+	/* Mark reserved regions */
+	for (i = 0; i < lmb.reserved.cnt; i++) {
+		unsigned long physbase = lmb.reserved.region[i].base;
+		unsigned long size = lmb.reserved.region[i].size;
+		unsigned long start_pfn = physbase >> PAGE_SHIFT;
+		unsigned long end_pfn = ((physbase + size - 1) >> PAGE_SHIFT);
+		struct node_active_region *node_ar;
+
+		node_ar = get_node_active_region(start_pfn);
+		while (start_pfn < end_pfn && node_ar != NULL) {
+			/*
+			 * if reserved region extends past active region
+			 * then trim size to active region
+			 */
+			if (end_pfn >= node_ar->end_pfn)
+				size = (node_ar->end_pfn << PAGE_SHIFT)
+					- (start_pfn << PAGE_SHIFT);
+			dbg("reserve_bootmem %lx %lx nid=%d\n", physbase, size,
+				node_ar->nid);
+			reserve_bootmem_node(NODE_DATA(node_ar->nid), physbase,
+						size, BOOTMEM_DEFAULT);
+			/*
+			 * if reserved region extends past the active region
+			 * then get next active region that contains
+			 *        this reserved region
+			 */
+			if (end_pfn >= node_ar->end_pfn) {
+				start_pfn = node_ar->end_pfn;
+				physbase = start_pfn << PAGE_SHIFT;
+				node_ar = get_node_active_region(start_pfn);
+			} else
+				break;
 		}
 
+	}
+
+	for_each_online_node(nid) {
 		sparse_memory_present_with_active_regions(nid);
 	}
 }
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 72a15dc..d186a7e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1020,6 +1020,8 @@ extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
 extern void remove_all_active_ranges(void);
 extern unsigned long absent_pages_in_range(unsigned long start_pfn,
 						unsigned long end_pfn);
+struct node_active_region *get_node_active_region(
+						unsigned long start_pfn);
 extern void get_pfn_range_for_nid(unsigned int nid,
 			unsigned long *start_pfn, unsigned long *end_pfn);
 extern unsigned long find_min_pfn_with_active_regions(void);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 27b8681..fbbb759 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3074,6 +3074,25 @@ static void __meminit account_node_boundary(unsigned int nid,
 		unsigned long *start_pfn, unsigned long *end_pfn) {}
 #endif
 
+/**
+ * get_node_active_region - Return active region containing start_pfn
+ * @start_pfn The page to return the region for.
+ *
+ * It will return NULL if active region is not found.
+ */
+struct node_active_region *
+__meminit get_node_active_region(unsigned long start_pfn)
+{
+	int i;
+	for (i = 0; i < nr_nodemap_entries; i++) {
+		unsigned long node_start_pfn = early_node_map[i].start_pfn;
+		unsigned long node_end_pfn = early_node_map[i].end_pfn;
+
+		if (node_start_pfn <= start_pfn && node_end_pfn > start_pfn)
+			return &early_node_map[i];
+	}
+	return NULL;
+}
 
 /**
  * get_pfn_range_for_nid - Return the start and end page frames for a node

Re: [PATCH v2] properly reserve in bootmem the lmb reserved regions that cross NUMA nodes

[Benjamin Herrenschmidt]

Minor nits ...

One is, you add this helper to mm/page_alloc.c, which means that I'll
need some ack from Hugh or Andrew before I can merge that via the
powerpc tree... Unless there's another user, I'd rather keep the
helper function in powerpc code for now, it can be moved to common
code later if somebody needs something similar.

> +	/* Mark reserved regions */
> +	for (i = 0; i < lmb.reserved.cnt; i++) {
> +		unsigned long physbase = lmb.reserved.region[i].base;
> +		unsigned long size = lmb.reserved.region[i].size;
> +		unsigned long start_pfn = physbase >> PAGE_SHIFT;
> +		unsigned long end_pfn = ((physbase + size - 1) >> PAGE_SHIFT);
> +		struct node_active_region *node_ar;

I'm not too happy wit the fact that something called "end_pfn" is
sometimes inclusive and sometime exclusive.

IE. From your implementation of get_node_active_region() it looks like
early_node_map[i].end_pfn isn't part of the range (exclusive) while
in your loop, the way you define end_pfn to be base + size - 1 means
it's part of the range (inclusive). That subtle distinction makes it
harder to understand the logic and is bug prone.

> +		node_ar = get_node_active_region(start_pfn);
> +		while (start_pfn < end_pfn && node_ar != NULL) {
> +			/*
> +			 * if reserved region extends past active region
> +			 * then trim size to active region
> +			 */
> +			if (end_pfn >= node_ar->end_pfn)

So the above test is correct, but it took me two more brain cells to
figure it out than necessary :-)

> +				size = (node_ar->end_pfn << PAGE_SHIFT)
> +					- (start_pfn << PAGE_SHIFT);
> +			dbg("reserve_bootmem %lx %lx nid=%d\n", physbase, size,
> +				node_ar->nid);
> +			reserve_bootmem_node(NODE_DATA(node_ar->nid), physbase,
> +						size, BOOTMEM_DEFAULT);
> +			/*
> +			 * if reserved region extends past the active region
> +			 * then get next active region that contains
> +			 *        this reserved region
> +			 */
> +			if (end_pfn >= node_ar->end_pfn) {
> +				start_pfn = node_ar->end_pfn;
> +				physbase = start_pfn << PAGE_SHIFT;
> +				node_ar = get_node_active_region(start_pfn);
> +			} else
> +				break;
>  		}
Minor nit but the above would look nicer if you wrote instead

			if (end_pfn < node_ar->end_pfn)
				break;
			start_pfn = ...
 
> +	}
> +
> +	for_each_online_node(nid) {
>  		sparse_memory_present_with_active_regions(nid);
>  	}
>  }

And you can remove the { and } above.

Cheers,
Ben.