summaryrefslogtreecommitdiff
path: root/include/linux
diff options
context:
space:
mode:
authorTony Luck <tony.luck@intel.com>2015-06-24 16:58:09 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2015-06-24 17:49:44 -0700
commitfc6daaf93151877748f8096af6b3fddb147f22d6 (patch)
tree1892f34cca08d40af6598bccae87c42037c5ea80 /include/linux
parent6afdb859b71019143b8eecda02b8b29b03185055 (diff)
mm/memblock: add extra "flags" to memblock to allow selection of memory based on attribute
Some high end Intel Xeon systems report uncorrectable memory errors as a recoverable machine check. Linux has included code for some time to process these and just signal the affected processes (or even recover completely if the error was in a read only page that can be replaced by reading from disk). But we have no recovery path for errors encountered during kernel code execution. Except for some very specific cases were are unlikely to ever be able to recover. Enter memory mirroring. Actually 3rd generation of memory mirroing. Gen1: All memory is mirrored Pro: No s/w enabling - h/w just gets good data from other side of the mirror Con: Halves effective memory capacity available to OS/applications Gen2: Partial memory mirror - just mirror memory begind some memory controllers Pro: Keep more of the capacity Con: Nightmare to enable. Have to choose between allocating from mirrored memory for safety vs. NUMA local memory for performance Gen3: Address range partial memory mirror - some mirror on each memory controller Pro: Can tune the amount of mirror and keep NUMA performance Con: I have to write memory management code to implement The current plan is just to use mirrored memory for kernel allocations. This has been broken into two phases: 1) This patch series - find the mirrored memory, use it for boot time allocations 2) Wade into mm/page_alloc.c and define a ZONE_MIRROR to pick up the unused mirrored memory from mm/memblock.c and only give it out to select kernel allocations (this is still being scoped because page_alloc.c is scary). This patch (of 3): Add extra "flags" to memblock to allow selection of memory based on attribute. No functional changes Signed-off-by: Tony Luck <tony.luck@intel.com> Cc: Xishi Qiu <qiuxishi@huawei.com> Cc: Hanjun Guo <guohanjun@huawei.com> Cc: Xiexiuqi <xiexiuqi@huawei.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Naoya Horiguchi <nao.horiguchi@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include/linux')
-rw-r--r--include/linux/memblock.h41
1 files changed, 26 insertions, 15 deletions
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 9497ec7c77ea..7aeec0cb4c27 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -21,7 +21,10 @@
#define INIT_PHYSMEM_REGIONS 4
/* Definition of memblock flags. */
-#define MEMBLOCK_HOTPLUG 0x1 /* hotpluggable region */
+enum {
+ MEMBLOCK_NONE = 0x0, /* No special request */
+ MEMBLOCK_HOTPLUG = 0x1, /* hotpluggable region */
+};
struct memblock_region {
phys_addr_t base;
@@ -61,7 +64,7 @@ extern bool movable_node_enabled;
phys_addr_t memblock_find_in_range_node(phys_addr_t size, phys_addr_t align,
phys_addr_t start, phys_addr_t end,
- int nid);
+ int nid, ulong flags);
phys_addr_t memblock_find_in_range(phys_addr_t start, phys_addr_t end,
phys_addr_t size, phys_addr_t align);
phys_addr_t get_allocated_memblock_reserved_regions_info(phys_addr_t *addr);
@@ -85,11 +88,13 @@ int memblock_remove_range(struct memblock_type *type,
phys_addr_t base,
phys_addr_t size);
-void __next_mem_range(u64 *idx, int nid, struct memblock_type *type_a,
+void __next_mem_range(u64 *idx, int nid, ulong flags,
+ struct memblock_type *type_a,
struct memblock_type *type_b, phys_addr_t *out_start,
phys_addr_t *out_end, int *out_nid);
-void __next_mem_range_rev(u64 *idx, int nid, struct memblock_type *type_a,
+void __next_mem_range_rev(u64 *idx, int nid, ulong flags,
+ struct memblock_type *type_a,
struct memblock_type *type_b, phys_addr_t *out_start,
phys_addr_t *out_end, int *out_nid);
@@ -100,16 +105,17 @@ void __next_mem_range_rev(u64 *idx, int nid, struct memblock_type *type_a,
* @type_a: ptr to memblock_type to iterate
* @type_b: ptr to memblock_type which excludes from the iteration
* @nid: node selector, %NUMA_NO_NODE for all nodes
+ * @flags: pick from blocks based on memory attributes
* @p_start: ptr to phys_addr_t for start address of the range, can be %NULL
* @p_end: ptr to phys_addr_t for end address of the range, can be %NULL
* @p_nid: ptr to int for nid of the range, can be %NULL
*/
-#define for_each_mem_range(i, type_a, type_b, nid, \
+#define for_each_mem_range(i, type_a, type_b, nid, flags, \
p_start, p_end, p_nid) \
- for (i = 0, __next_mem_range(&i, nid, type_a, type_b, \
+ for (i = 0, __next_mem_range(&i, nid, flags, type_a, type_b, \
p_start, p_end, p_nid); \
i != (u64)ULLONG_MAX; \
- __next_mem_range(&i, nid, type_a, type_b, \
+ __next_mem_range(&i, nid, flags, type_a, type_b, \
p_start, p_end, p_nid))
/**
@@ -119,17 +125,18 @@ void __next_mem_range_rev(u64 *idx, int nid, struct memblock_type *type_a,
* @type_a: ptr to memblock_type to iterate
* @type_b: ptr to memblock_type which excludes from the iteration
* @nid: node selector, %NUMA_NO_NODE for all nodes
+ * @flags: pick from blocks based on memory attributes
* @p_start: ptr to phys_addr_t for start address of the range, can be %NULL
* @p_end: ptr to phys_addr_t for end address of the range, can be %NULL
* @p_nid: ptr to int for nid of the range, can be %NULL
*/
-#define for_each_mem_range_rev(i, type_a, type_b, nid, \
+#define for_each_mem_range_rev(i, type_a, type_b, nid, flags, \
p_start, p_end, p_nid) \
for (i = (u64)ULLONG_MAX, \
- __next_mem_range_rev(&i, nid, type_a, type_b, \
+ __next_mem_range_rev(&i, nid, flags, type_a, type_b,\
p_start, p_end, p_nid); \
i != (u64)ULLONG_MAX; \
- __next_mem_range_rev(&i, nid, type_a, type_b, \
+ __next_mem_range_rev(&i, nid, flags, type_a, type_b, \
p_start, p_end, p_nid))
#ifdef CONFIG_MOVABLE_NODE
@@ -181,13 +188,14 @@ void __next_mem_pfn_range(int *idx, int nid, unsigned long *out_start_pfn,
* @p_start: ptr to phys_addr_t for start address of the range, can be %NULL
* @p_end: ptr to phys_addr_t for end address of the range, can be %NULL
* @p_nid: ptr to int for nid of the range, can be %NULL
+ * @flags: pick from blocks based on memory attributes
*
* Walks over free (memory && !reserved) areas of memblock. Available as
* soon as memblock is initialized.
*/
-#define for_each_free_mem_range(i, nid, p_start, p_end, p_nid) \
+#define for_each_free_mem_range(i, nid, flags, p_start, p_end, p_nid) \
for_each_mem_range(i, &memblock.memory, &memblock.reserved, \
- nid, p_start, p_end, p_nid)
+ nid, flags, p_start, p_end, p_nid)
/**
* for_each_free_mem_range_reverse - rev-iterate through free memblock areas
@@ -196,13 +204,15 @@ void __next_mem_pfn_range(int *idx, int nid, unsigned long *out_start_pfn,
* @p_start: ptr to phys_addr_t for start address of the range, can be %NULL
* @p_end: ptr to phys_addr_t for end address of the range, can be %NULL
* @p_nid: ptr to int for nid of the range, can be %NULL
+ * @flags: pick from blocks based on memory attributes
*
* Walks over free (memory && !reserved) areas of memblock in reverse
* order. Available as soon as memblock is initialized.
*/
-#define for_each_free_mem_range_reverse(i, nid, p_start, p_end, p_nid) \
+#define for_each_free_mem_range_reverse(i, nid, flags, p_start, p_end, \
+ p_nid) \
for_each_mem_range_rev(i, &memblock.memory, &memblock.reserved, \
- nid, p_start, p_end, p_nid)
+ nid, flags, p_start, p_end, p_nid)
static inline void memblock_set_region_flags(struct memblock_region *r,
unsigned long flags)
@@ -273,7 +283,8 @@ static inline bool memblock_bottom_up(void) { return false; }
#define MEMBLOCK_ALLOC_ACCESSIBLE 0
phys_addr_t __init memblock_alloc_range(phys_addr_t size, phys_addr_t align,
- phys_addr_t start, phys_addr_t end);
+ phys_addr_t start, phys_addr_t end,
+ ulong flags);
phys_addr_t memblock_alloc_base(phys_addr_t size, phys_addr_t align,
phys_addr_t max_addr);
phys_addr_t __memblock_alloc_base(phys_addr_t size, phys_addr_t align,