* [PATCH] x86: e820 merge parse mem/memmap
2008-06-04 2:35 ` [PATCH] x86: 32 bit use e820_register_active_regions Yinghai Lu
@ 2008-06-04 7:39 ` Yinghai Lu
2008-06-04 10:27 ` [PATCH] x86: 32 bit use e820_register_active_regions Ingo Molnar
` (3 subsequent siblings)
4 siblings, 0 replies; 19+ messages in thread
From: Yinghai Lu @ 2008-06-04 7:39 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
after we get 32 bit support e820_register_active_regions
we could merge parse mem/memmap
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/arch/x86/kernel/e820.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820.c
+++ linux-2.6/arch/x86/kernel/e820.c
@@ -873,3 +873,89 @@ u64 __init e820_hole_size(u64 start, u64
}
return end - start - ((u64)ram << PAGE_SHIFT);
}
+
+static void early_panic(char *msg)
+{
+ early_printk(msg);
+ panic(msg);
+}
+
+/* "mem=nopentium" disables the 4MB page tables. */
+static int __init parse_memopt(char *p)
+{
+ u64 mem_size;
+
+ if (!p)
+ return -EINVAL;
+
+#ifdef CONFIG_X86_32
+ if (!strcmp(p, "nopentium")) {
+ setup_clear_cpu_cap(X86_FEATURE_PSE);
+ return 0;
+ }
+#endif
+
+ mem_size = memparse(p, &p);
+ end_user_pfn = mem_size>>PAGE_SHIFT;
+ return 0;
+}
+early_param("mem", parse_memopt);
+
+static int userdef __initdata;
+
+static int __init parse_memmap_opt(char *p)
+{
+ char *oldp;
+ u64 start_at, mem_size;
+
+ if (!strcmp(p, "exactmap")) {
+#ifdef CONFIG_CRASH_DUMP
+ /*
+ * If we are doing a crash dump, we still need to know
+ * the real mem size before original memory map is
+ * reset.
+ */
+ e820_register_active_regions(0, 0, -1UL);
+ saved_max_pfn = e820_end_of_ram();
+ remove_all_active_ranges();
+#endif
+ e820.nr_map = 0;
+ userdef = 1;
+ return 0;
+ }
+
+ oldp = p;
+ mem_size = memparse(p, &p);
+ if (p == oldp)
+ return -EINVAL;
+
+ userdef = 1;
+ if (*p == '@') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_RAM);
+ } else if (*p == '#') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_ACPI);
+ } else if (*p == '$') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_RESERVED);
+ } else {
+ end_user_pfn = (mem_size >> PAGE_SHIFT);
+ }
+ return *p == '\0' ? 0 : -EINVAL;
+}
+early_param("memmap", parse_memmap_opt);
+
+void __init finish_e820_parsing(void)
+{
+ if (userdef) {
+ int nr = e820.nr_map;
+
+ if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
+ early_panic("Invalid user supplied memory map");
+ e820.nr_map = nr;
+
+ printk(KERN_INFO "user-defined physical RAM map:\n");
+ e820_print_map("user");
+ }
+}
Index: linux-2.6/arch/x86/kernel/e820_32.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820_32.c
+++ linux-2.6/arch/x86/kernel/e820_32.c
@@ -207,36 +207,6 @@ void __init init_iomem_resources(struct
}
}
-void __init limit_regions(unsigned long long size)
-{
- unsigned long long current_addr;
- int i;
-
- e820_print_map("limit_regions start");
- for (i = 0; i < e820.nr_map; i++) {
- current_addr = e820.map[i].addr + e820.map[i].size;
- if (current_addr < size)
- continue;
-
- if (e820.map[i].type != E820_RAM)
- continue;
-
- if (e820.map[i].addr >= size) {
- /*
- * This region starts past the end of the
- * requested size, skip it completely.
- */
- e820.nr_map = i;
- } else {
- e820.nr_map = i + 1;
- e820.map[i].size -= current_addr - size;
- }
- e820_print_map("limit_regions endfor");
- return;
- }
- e820_print_map("limit_regions endfunc");
-}
-
/* Overridden in paravirt.c if CONFIG_PARAVIRT */
char * __init __attribute__((weak)) memory_setup(void)
{
@@ -249,93 +219,3 @@ void __init setup_memory_map(void)
e820_print_map(memory_setup());
}
-static int __initdata user_defined_memmap;
-
-/*
- * "mem=nopentium" disables the 4MB page tables.
- * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
- * to <mem>, overriding the bios size.
- * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
- * <start> to <start>+<mem>, overriding the bios size.
- *
- * HPA tells me bootloaders need to parse mem=, so no new
- * option should be mem= [also see Documentation/i386/boot.txt]
- */
-static int __init parse_mem(char *arg)
-{
- if (!arg)
- return -EINVAL;
-
- if (strcmp(arg, "nopentium") == 0) {
- setup_clear_cpu_cap(X86_FEATURE_PSE);
- } else {
- /* If the user specifies memory size, we
- * limit the BIOS-provided memory map to
- * that size. exactmap can be used to specify
- * the exact map. mem=number can be used to
- * trim the existing memory map.
- */
- unsigned long long mem_size;
-
- mem_size = memparse(arg, &arg);
- limit_regions(mem_size);
- user_defined_memmap = 1;
- }
- return 0;
-}
-early_param("mem", parse_mem);
-
-static int __init parse_memmap(char *arg)
-{
- if (!arg)
- return -EINVAL;
-
- if (strcmp(arg, "exactmap") == 0) {
-#ifdef CONFIG_CRASH_DUMP
- /* If we are doing a crash dump, we
- * still need to know the real mem
- * size before original memory map is
- * reset.
- */
- e820_register_active_regions(0, 0, -1UL);
- saved_max_pfn = e820_end_of_ram();
- remove_all_active_ranges();
-#endif
- e820.nr_map = 0;
- user_defined_memmap = 1;
- } else {
- /* If the user specifies memory size, we
- * limit the BIOS-provided memory map to
- * that size. exactmap can be used to specify
- * the exact map. mem=number can be used to
- * trim the existing memory map.
- */
- unsigned long long start_at, mem_size;
-
- mem_size = memparse(arg, &arg);
- if (*arg == '@') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_RAM);
- } else if (*arg == '#') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_ACPI);
- } else if (*arg == '$') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_RESERVED);
- } else {
- limit_regions(mem_size);
- user_defined_memmap = 1;
- }
- }
- return 0;
-}
-early_param("memmap", parse_memmap);
-
-void __init finish_e820_parsing(void)
-{
- if (user_defined_memmap) {
- printk(KERN_INFO "user-defined physical RAM map:\n");
- e820_print_map("user");
- }
-}
-
Index: linux-2.6/arch/x86/kernel/e820_64.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820_64.c
+++ linux-2.6/arch/x86/kernel/e820_64.c
@@ -96,75 +96,6 @@ char *__init machine_specific_memory_set
return who;
}
-static int __init parse_memopt(char *p)
-{
- if (!p)
- return -EINVAL;
- end_user_pfn = memparse(p, &p);
- end_user_pfn >>= PAGE_SHIFT;
- return 0;
-}
-early_param("mem", parse_memopt);
-
-static int userdef __initdata;
-
-static int __init parse_memmap_opt(char *p)
-{
- char *oldp;
- unsigned long long start_at, mem_size;
-
- if (!strcmp(p, "exactmap")) {
-#ifdef CONFIG_CRASH_DUMP
- /*
- * If we are doing a crash dump, we still need to know
- * the real mem size before original memory map is
- * reset.
- */
- e820_register_active_regions(0, 0, -1UL);
- saved_max_pfn = e820_end_of_ram();
- remove_all_active_ranges();
-#endif
- e820.nr_map = 0;
- userdef = 1;
- return 0;
- }
-
- oldp = p;
- mem_size = memparse(p, &p);
- if (p == oldp)
- return -EINVAL;
-
- userdef = 1;
- if (*p == '@') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_RAM);
- } else if (*p == '#') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_ACPI);
- } else if (*p == '$') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_RESERVED);
- } else {
- end_user_pfn = (mem_size >> PAGE_SHIFT);
- }
- return *p == '\0' ? 0 : -EINVAL;
-}
-early_param("memmap", parse_memmap_opt);
-
-void __init finish_e820_parsing(void)
-{
- if (userdef) {
- int nr = e820.nr_map;
-
- if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
- early_panic("Invalid user supplied memory map");
- e820.nr_map = nr;
-
- printk(KERN_INFO "user-defined physical RAM map:\n");
- e820_print_map("user");
- }
-}
-
int __init arch_get_ram_range(int slot, u64 *addr, u64 *size)
{
int i;
Index: linux-2.6/include/asm-x86/e820.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820.h
+++ linux-2.6/include/asm-x86/e820.h
@@ -97,6 +97,8 @@ extern int e820_find_active_region(const
extern void e820_register_active_regions(int nid, unsigned long start_pfn,
unsigned long end_pfn);
extern u64 e820_hole_size(u64 start, u64 end);
+extern void finish_e820_parsing(void);
+
#endif /* __ASSEMBLY__ */
#define ISA_START_ADDRESS 0xa0000
Index: linux-2.6/include/asm-x86/e820_32.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820_32.h
+++ linux-2.6/include/asm-x86/e820_32.h
@@ -19,9 +19,7 @@
#ifndef __ASSEMBLY__
extern void setup_memory_map(void);
-extern void finish_e820_parsing(void);
-extern void limit_regions(unsigned long long size);
extern void init_iomem_resources(struct resource *code_resource,
struct resource *data_resource,
struct resource *bss_resource);
Index: linux-2.6/include/asm-x86/e820_64.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820_64.h
+++ linux-2.6/include/asm-x86/e820_64.h
@@ -22,8 +22,6 @@ extern int is_memory_any_valid(unsigned
extern int e820_all_non_reserved(unsigned long start, unsigned long end);
extern int is_memory_all_valid(unsigned long start, unsigned long end);
-extern void finish_e820_parsing(void);
-
#endif/*!__ASSEMBLY__*/
#endif/*__E820_HEADER*/
^ permalink raw reply [flat|nested] 19+ messages in thread* Re: [PATCH] x86: 32 bit use e820_register_active_regions
2008-06-04 2:35 ` [PATCH] x86: 32 bit use e820_register_active_regions Yinghai Lu
2008-06-04 7:39 ` [PATCH] x86: e820 merge parse mem/memmap Yinghai Lu
@ 2008-06-04 10:27 ` Ingo Molnar
2008-06-04 20:21 ` [PATCH] x86: e820 max_arch_pfn typo fix for 64 bit Yinghai Lu
` (2 subsequent siblings)
4 siblings, 0 replies; 19+ messages in thread
From: Ingo Molnar @ 2008-06-04 10:27 UTC (permalink / raw)
To: Yinghai Lu
Cc: Thomas Gleixner, H. Peter Anvin, Andrew Morton,
linux-kernel@vger.kernel.org
* Yinghai Lu <yhlu.kernel@gmail.com> wrote:
> so 32bit is more like 64bit, and more smart with combining e820 and numa.
>
> Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
applied, thanks.
Ingo
^ permalink raw reply [flat|nested] 19+ messages in thread* [PATCH] x86: e820 max_arch_pfn typo fix for 64 bit
2008-06-04 2:35 ` [PATCH] x86: 32 bit use e820_register_active_regions Yinghai Lu
2008-06-04 7:39 ` [PATCH] x86: e820 merge parse mem/memmap Yinghai Lu
2008-06-04 10:27 ` [PATCH] x86: 32 bit use e820_register_active_regions Ingo Molnar
@ 2008-06-04 20:21 ` Yinghai Lu
2008-06-04 22:47 ` H. Peter Anvin
2008-06-06 21:43 ` [PATCH] x86: shrink pages should check all Yinghai Lu
2008-06-10 19:55 ` [PATCH] x86: e820 merge parse mem/memmap Yinghai Lu
4 siblings, 1 reply; 19+ messages in thread
From: Yinghai Lu @ 2008-06-04 20:21 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
should use right shift
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/arch/x86/kernel/e820.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820.c
+++ linux-2.6/arch/x86/kernel/e820.c
@@ -771,7 +771,7 @@ u64 __init early_reserve_e820(u64 startt
# define MAX_ARCH_PFN (1ULL<<(32-PAGE_SHIFT))
# endif
#else /* CONFIG_X86_32 */
-# define MAX_ARCH_PFN MAXMEM<<PAGE_SHIFT
+# define MAX_ARCH_PFN MAXMEM>>PAGE_SHIFT
#endif
/*
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] x86: e820 max_arch_pfn typo fix for 64 bit
2008-06-04 20:21 ` [PATCH] x86: e820 max_arch_pfn typo fix for 64 bit Yinghai Lu
@ 2008-06-04 22:47 ` H. Peter Anvin
0 siblings, 0 replies; 19+ messages in thread
From: H. Peter Anvin @ 2008-06-04 22:47 UTC (permalink / raw)
To: Yinghai Lu
Cc: Ingo Molnar, Thomas Gleixner, Andrew Morton,
linux-kernel@vger.kernel.org
Yinghai Lu wrote:
> should use right shift
>
> Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
>
> Index: linux-2.6/arch/x86/kernel/e820.c
> ===================================================================
> --- linux-2.6.orig/arch/x86/kernel/e820.c
> +++ linux-2.6/arch/x86/kernel/e820.c
> @@ -771,7 +771,7 @@ u64 __init early_reserve_e820(u64 startt
> # define MAX_ARCH_PFN (1ULL<<(32-PAGE_SHIFT))
> # endif
> #else /* CONFIG_X86_32 */
> -# define MAX_ARCH_PFN MAXMEM<<PAGE_SHIFT
> +# define MAX_ARCH_PFN MAXMEM>>PAGE_SHIFT
> #endif
>
Should also use parens (operator inside a macro).
-hpa
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH] x86: shrink pages should check all
2008-06-04 2:35 ` [PATCH] x86: 32 bit use e820_register_active_regions Yinghai Lu
` (2 preceding siblings ...)
2008-06-04 20:21 ` [PATCH] x86: e820 max_arch_pfn typo fix for 64 bit Yinghai Lu
@ 2008-06-06 21:43 ` Yinghai Lu
2008-06-07 1:53 ` [PATCH] x86: numa32 use find_e820_area to find KVA ram on node Yinghai Lu
` (2 more replies)
2008-06-10 19:55 ` [PATCH] x86: e820 merge parse mem/memmap Yinghai Lu
4 siblings, 3 replies; 19+ messages in thread
From: Yinghai Lu @ 2008-06-06 21:43 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
we are uing register_e820_active_regions instead of add_active_range directly.
so end_pfn could be different between the value in early_node_map to node_end_pfn.
need to make shrink_active_range more smart.
shrink_active_range is only used x86 32 bit, or need to move back in some file
in arch/x86?
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/arch/x86/mm/discontig_32.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/discontig_32.c
+++ linux-2.6/arch/x86/mm/discontig_32.c
@@ -279,7 +279,7 @@ static unsigned long calculate_numa_rema
node_end_pfn[nid] -= size;
node_remap_start_pfn[nid] = node_end_pfn[nid];
- shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]);
+ shrink_active_range(nid, node_end_pfn[nid]);
}
printk("Reserving total of %ld pages for numa KVA remap\n",
reserve_pages);
Index: linux-2.6/include/linux/mm.h
===================================================================
--- linux-2.6.orig/include/linux/mm.h
+++ linux-2.6/include/linux/mm.h
@@ -997,8 +997,7 @@ extern void free_area_init_node(int nid,
extern void free_area_init_nodes(unsigned long *max_zone_pfn);
extern void add_active_range(unsigned int nid, unsigned long start_pfn,
unsigned long end_pfn);
-extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
- unsigned long new_end_pfn);
+extern void shrink_active_range(unsigned int nid, unsigned long new_end_pfn);
extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
unsigned long end_pfn);
extern void remove_all_active_ranges(void);
Index: linux-2.6/mm/page_alloc.c
===================================================================
--- linux-2.6.orig/mm/page_alloc.c
+++ linux-2.6/mm/page_alloc.c
@@ -3592,25 +3592,51 @@ void __init add_active_range(unsigned in
/**
* shrink_active_range - Shrink an existing registered range of PFNs
* @nid: The node id the range is on that should be shrunk
- * @old_end_pfn: The old end PFN of the range
* @new_end_pfn: The new PFN of the range
*
* i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
- * The map is kept at the end physical page range that has already been
- * registered with add_active_range(). This function allows an arch to shrink
- * an existing registered range.
+ * The map is kept near the end physical page range that has already been
+ * registered. This function allows an arch to shrink an existing registered
+ * range.
*/
-void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
- unsigned long new_end_pfn)
+void __init shrink_active_range(unsigned int nid, unsigned long new_end_pfn)
{
- int i;
+ int i,j;
+ int removed = 0;
/* Find the old active region end and shrink */
- for_each_active_range_index_in_nid(i, nid)
- if (early_node_map[i].end_pfn == old_end_pfn) {
+ for_each_active_range_index_in_nid(i, nid) {
+ if (early_node_map[i].start_pfn >= new_end_pfn) {
+ /* clear it */
+ early_node_map[i].end_pfn = 0;
+ removed = 1;
+ continue;
+ }
+ if (early_node_map[i].end_pfn > new_end_pfn) {
early_node_map[i].end_pfn = new_end_pfn;
- break;
+ continue;
+ }
+ }
+
+ if (!removed)
+ return;
+
+ /* remove the blank ones */
+ for (i = nr_nodemap_entries - 1; i > 0; i--) {
+ if (early_node_map[i].nid != nid)
+ continue;
+ if (early_node_map[i].end_pfn)
+ continue;
+ /* we found it, get rid of it */
+ for (j = i; j < nr_nodemap_entries - 1; j++) {
+ early_node_map[j].nid = early_node_map[j+1].nid;
+ early_node_map[j].start_pfn = early_node_map[j+1].start_pfn;
+ early_node_map[j].end_pfn = early_node_map[j+1].end_pfn;
}
+ j = nr_nodemap_entries - 1;
+ memset(&early_node_map[j], 0, sizeof(early_node_map[j]));
+ nr_nodemap_entries--;
+ }
}
/**
^ permalink raw reply [flat|nested] 19+ messages in thread* [PATCH] x86: numa32 use find_e820_area to find KVA ram on node
2008-06-06 21:43 ` [PATCH] x86: shrink pages should check all Yinghai Lu
@ 2008-06-07 1:53 ` Yinghai Lu
2008-06-10 9:53 ` Ingo Molnar
2008-06-07 1:54 ` [PATCH] x86: fix fail with 64g above system with numa32 Yinghai Lu
2008-06-09 2:39 ` [PATCH] x86: shrink pages should check all v2 Yinghai Lu
2 siblings, 1 reply; 19+ messages in thread
From: Yinghai Lu @ 2008-06-07 1:53 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
don't assume we can use ram near end of every node.
esp some system has less memory. and they could have
kva address and kva ram all below max_low_pfn
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/arch/x86/mm/discontig_32.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/discontig_32.c
+++ linux-2.6/arch/x86/mm/discontig_32.c
@@ -225,17 +225,21 @@ static unsigned long calculate_numa_rema
{
int nid;
unsigned long size, reserve_pages = 0;
- unsigned long pfn;
for_each_online_node(nid) {
- unsigned old_end_pfn = node_end_pfn[nid];
+ u64 node_end_target;
+ u64 node_end_final;
/*
* The acpi/srat node info can show hot-add memroy zones
* where memory could be added but not currently present.
*/
+ printk("node %d pfn: [%lx - %lx]\n",
+ nid, node_start_pfn[nid], node_end_pfn[nid]);
if (node_start_pfn[nid] > max_pfn)
continue;
+ if (!node_end_pfn[nid])
+ continue;
if (node_end_pfn[nid] > max_pfn)
node_end_pfn[nid] = max_pfn;
@@ -247,37 +251,40 @@ static unsigned long calculate_numa_rema
/* now the roundup is correct, convert to PAGE_SIZE pages */
size = size * PTRS_PER_PTE;
- /*
- * Validate the region we are allocating only contains valid
- * pages.
- */
- for (pfn = node_end_pfn[nid] - size;
- pfn < node_end_pfn[nid]; pfn++)
- if (!page_is_ram(pfn))
- break;
+ node_end_target = round_down(node_end_pfn[nid] - size,
+ PTRS_PER_PTE);
+ node_end_target <<= PAGE_SHIFT;
+ do {
+ node_end_final = find_e820_area(node_end_target,
+ ((u64)node_end_pfn[nid])<<PAGE_SHIFT,
+ ((u64)size)<<PAGE_SHIFT,
+ LARGE_PAGE_BYTES);
+ node_end_target -= LARGE_PAGE_BYTES;
+ } while (node_end_final == -1ULL &&
+ (node_end_target>>PAGE_SHIFT) > (node_start_pfn[nid]));
- if (pfn != node_end_pfn[nid])
- size = 0;
+ if (node_end_final == -1ULL)
+ panic("Can not get kva ram\n");
printk("Reserving %ld pages of KVA for lmem_map of node %d\n",
size, nid);
node_remap_size[nid] = size;
node_remap_offset[nid] = reserve_pages;
reserve_pages += size;
- printk("Shrinking node %d from %ld pages to %ld pages\n",
- nid, node_end_pfn[nid], node_end_pfn[nid] - size);
+ printk("Shrinking node %d from %ld pages to %lld pages\n",
+ nid, node_end_pfn[nid], node_end_final>>PAGE_SHIFT);
- if (node_end_pfn[nid] & (PTRS_PER_PTE-1)) {
- /*
- * Align node_end_pfn[] and node_remap_start_pfn[] to
- * pmd boundary. remap_numa_kva will barf otherwise.
- */
- printk("Shrinking node %d further by %ld pages for proper alignment\n",
- nid, node_end_pfn[nid] & (PTRS_PER_PTE-1));
- size += node_end_pfn[nid] & (PTRS_PER_PTE-1);
- }
+ /*
+ * prevent kva address below max_low_pfn want it on system
+ * with less memory later.
+ * layout will be: KVA address , KVA RAM
+ */
+ if ((node_end_final>>PAGE_SHIFT) < max_low_pfn)
+ reserve_early(node_end_final,
+ node_end_final+(((u64)size)<<PAGE_SHIFT),
+ "KVA RAM");
- node_end_pfn[nid] -= size;
+ node_end_pfn[nid] = node_end_final>>PAGE_SHIFT;
node_remap_start_pfn[nid] = node_end_pfn[nid];
shrink_active_range(nid, node_end_pfn[nid]);
}
^ permalink raw reply [flat|nested] 19+ messages in thread* [PATCH] x86: fix fail with 64g above system with numa32
2008-06-06 21:43 ` [PATCH] x86: shrink pages should check all Yinghai Lu
2008-06-07 1:53 ` [PATCH] x86: numa32 use find_e820_area to find KVA ram on node Yinghai Lu
@ 2008-06-07 1:54 ` Yinghai Lu
2008-06-10 9:53 ` Ingo Molnar
2008-06-09 2:39 ` [PATCH] x86: shrink pages should check all v2 Yinghai Lu
2 siblings, 1 reply; 19+ messages in thread
From: Yinghai Lu @ 2008-06-07 1:54 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
Signed-off-by : Yinghai Lu <yhlu.kernel@gmail.com>
--- a/arch/x86/kernel/srat_32.c 2008-06-06 17:37:51.000000000 -0700
+++ b/arch/x86/kernel/srat_32.c 2008-06-06 17:38:38.000000000 -0700
@@ -251,7 +251,7 @@ static int __init acpi20_parse_srat(stru
for_each_online_node(nid) {
unsigned long start = node_start_pfn[nid];
- unsigned long end = node_end_pfn[nid];
+ unsigned long end = min(node_end_pfn[nid],max_pfn);
memory_present(nid, start, end);
node_remap_size[nid] = node_memmap_size_bytes(nid, start, end);
^ permalink raw reply [flat|nested] 19+ messages in thread* Re: [PATCH] x86: fix fail with 64g above system with numa32
2008-06-07 1:54 ` [PATCH] x86: fix fail with 64g above system with numa32 Yinghai Lu
@ 2008-06-10 9:53 ` Ingo Molnar
0 siblings, 0 replies; 19+ messages in thread
From: Ingo Molnar @ 2008-06-10 9:53 UTC (permalink / raw)
To: Yinghai Lu
Cc: Thomas Gleixner, H. Peter Anvin, Andrew Morton,
linux-kernel@vger.kernel.org
* Yinghai Lu <yhlu.kernel@gmail.com> wrote:
> +++ b/arch/x86/kernel/srat_32.c 2008-06-06 17:38:38.000000000 -0700
> @@ -251,7 +251,7 @@ static int __init acpi20_parse_srat(stru
>
> for_each_online_node(nid) {
> unsigned long start = node_start_pfn[nid];
> - unsigned long end = node_end_pfn[nid];
> + unsigned long end = min(node_end_pfn[nid],max_pfn);
>
> memory_present(nid, start, end);
applied to tip/x86/mpparse, thanks.
Ingo
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH] x86: shrink pages should check all v2
2008-06-06 21:43 ` [PATCH] x86: shrink pages should check all Yinghai Lu
2008-06-07 1:53 ` [PATCH] x86: numa32 use find_e820_area to find KVA ram on node Yinghai Lu
2008-06-07 1:54 ` [PATCH] x86: fix fail with 64g above system with numa32 Yinghai Lu
@ 2008-06-09 2:39 ` Yinghai Lu
2008-06-09 10:15 ` Ingo Molnar
2 siblings, 1 reply; 19+ messages in thread
From: Yinghai Lu @ 2008-06-09 2:39 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
we are uing register_e820_active_regions instead of add_active_range directly.
so end_pfn could be different between the value in early_node_map to node_end_pfn.
need to make shrink_active_range more smart.
shrink_active_range is only used x86 32 bit, or need to move back in some file
in arch/x86?
v2: use memcpy instead
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/include/linux/mm.h
===================================================================
--- linux-2.6.orig/include/linux/mm.h
+++ linux-2.6/include/linux/mm.h
@@ -997,8 +997,7 @@ extern void free_area_init_node(int nid,
extern void free_area_init_nodes(unsigned long *max_zone_pfn);
extern void add_active_range(unsigned int nid, unsigned long start_pfn,
unsigned long end_pfn);
-extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
- unsigned long new_end_pfn);
+extern void shrink_active_range(unsigned int nid, unsigned long new_end_pfn);
extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
unsigned long end_pfn);
extern void remove_all_active_ranges(void);
Index: linux-2.6/mm/page_alloc.c
===================================================================
--- linux-2.6.orig/mm/page_alloc.c
+++ linux-2.6/mm/page_alloc.c
@@ -3592,25 +3592,49 @@ void __init add_active_range(unsigned in
/**
* shrink_active_range - Shrink an existing registered range of PFNs
* @nid: The node id the range is on that should be shrunk
- * @old_end_pfn: The old end PFN of the range
* @new_end_pfn: The new PFN of the range
*
* i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
- * The map is kept at the end physical page range that has already been
- * registered with add_active_range(). This function allows an arch to shrink
- * an existing registered range.
+ * The map is kept near the end physical page range that has already been
+ * registered. This function allows an arch to shrink an existing registered
+ * range.
*/
-void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
- unsigned long new_end_pfn)
+void __init shrink_active_range(unsigned int nid, unsigned long new_end_pfn)
{
- int i;
+ int i, j;
+ int removed = 0;
/* Find the old active region end and shrink */
- for_each_active_range_index_in_nid(i, nid)
- if (early_node_map[i].end_pfn == old_end_pfn) {
+ for_each_active_range_index_in_nid(i, nid) {
+ if (early_node_map[i].start_pfn >= new_end_pfn) {
+ /* clear it */
+ early_node_map[i].end_pfn = 0;
+ removed = 1;
+ continue;
+ }
+ if (early_node_map[i].end_pfn > new_end_pfn) {
early_node_map[i].end_pfn = new_end_pfn;
- break;
+ continue;
}
+ }
+
+ if (!removed)
+ return;
+
+ /* remove the blank ones */
+ for (i = nr_nodemap_entries - 1; i > 0; i--) {
+ if (early_node_map[i].nid != nid)
+ continue;
+ if (early_node_map[i].end_pfn)
+ continue;
+ /* we found it, get rid of it */
+ for (j = i; j < nr_nodemap_entries - 1; j++)
+ memcpy(&early_node_map[j], &early_node_map[j+1],
+ sizeof(early_node_map[j]));
+ j = nr_nodemap_entries - 1;
+ memset(&early_node_map[j], 0, sizeof(early_node_map[j]));
+ nr_nodemap_entries--;
+ }
}
/**
Index: linux-2.6/arch/x86/mm/discontig_32.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/discontig_32.c
+++ linux-2.6/arch/x86/mm/discontig_32.c
@@ -279,7 +279,7 @@ static unsigned long calculate_numa_rema
node_end_pfn[nid] -= size;
node_remap_start_pfn[nid] = node_end_pfn[nid];
- shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]);
+ shrink_active_range(nid, node_end_pfn[nid]);
}
printk("Reserving total of %ld pages for numa KVA remap\n",
reserve_pages);
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH] x86: e820 merge parse mem/memmap
2008-06-04 2:35 ` [PATCH] x86: 32 bit use e820_register_active_regions Yinghai Lu
` (3 preceding siblings ...)
2008-06-06 21:43 ` [PATCH] x86: shrink pages should check all Yinghai Lu
@ 2008-06-10 19:55 ` Yinghai Lu
4 siblings, 0 replies; 19+ messages in thread
From: Yinghai Lu @ 2008-06-10 19:55 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, Andrew Morton
Cc: linux-kernel@vger.kernel.org
resend. it seems Ingo missed it.
after we get 32 bit support e820_register_active_regions
we could merge parse mem/memmap
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Index: linux-2.6/arch/x86/kernel/e820.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820.c
+++ linux-2.6/arch/x86/kernel/e820.c
@@ -873,3 +873,89 @@ u64 __init e820_hole_size(u64 start, u64
}
return end - start - ((u64)ram << PAGE_SHIFT);
}
+
+static void early_panic(char *msg)
+{
+ early_printk(msg);
+ panic(msg);
+}
+
+/* "mem=nopentium" disables the 4MB page tables. */
+static int __init parse_memopt(char *p)
+{
+ u64 mem_size;
+
+ if (!p)
+ return -EINVAL;
+
+#ifdef CONFIG_X86_32
+ if (!strcmp(p, "nopentium")) {
+ setup_clear_cpu_cap(X86_FEATURE_PSE);
+ return 0;
+ }
+#endif
+
+ mem_size = memparse(p, &p);
+ end_user_pfn = mem_size>>PAGE_SHIFT;
+ return 0;
+}
+early_param("mem", parse_memopt);
+
+static int userdef __initdata;
+
+static int __init parse_memmap_opt(char *p)
+{
+ char *oldp;
+ u64 start_at, mem_size;
+
+ if (!strcmp(p, "exactmap")) {
+#ifdef CONFIG_CRASH_DUMP
+ /*
+ * If we are doing a crash dump, we still need to know
+ * the real mem size before original memory map is
+ * reset.
+ */
+ e820_register_active_regions(0, 0, -1UL);
+ saved_max_pfn = e820_end_of_ram();
+ remove_all_active_ranges();
+#endif
+ e820.nr_map = 0;
+ userdef = 1;
+ return 0;
+ }
+
+ oldp = p;
+ mem_size = memparse(p, &p);
+ if (p == oldp)
+ return -EINVAL;
+
+ userdef = 1;
+ if (*p == '@') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_RAM);
+ } else if (*p == '#') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_ACPI);
+ } else if (*p == '$') {
+ start_at = memparse(p+1, &p);
+ add_memory_region(start_at, mem_size, E820_RESERVED);
+ } else {
+ end_user_pfn = (mem_size >> PAGE_SHIFT);
+ }
+ return *p == '\0' ? 0 : -EINVAL;
+}
+early_param("memmap", parse_memmap_opt);
+
+void __init finish_e820_parsing(void)
+{
+ if (userdef) {
+ int nr = e820.nr_map;
+
+ if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
+ early_panic("Invalid user supplied memory map");
+ e820.nr_map = nr;
+
+ printk(KERN_INFO "user-defined physical RAM map:\n");
+ e820_print_map("user");
+ }
+}
Index: linux-2.6/arch/x86/kernel/e820_32.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820_32.c
+++ linux-2.6/arch/x86/kernel/e820_32.c
@@ -207,36 +207,6 @@ void __init init_iomem_resources(struct
}
}
-void __init limit_regions(unsigned long long size)
-{
- unsigned long long current_addr;
- int i;
-
- e820_print_map("limit_regions start");
- for (i = 0; i < e820.nr_map; i++) {
- current_addr = e820.map[i].addr + e820.map[i].size;
- if (current_addr < size)
- continue;
-
- if (e820.map[i].type != E820_RAM)
- continue;
-
- if (e820.map[i].addr >= size) {
- /*
- * This region starts past the end of the
- * requested size, skip it completely.
- */
- e820.nr_map = i;
- } else {
- e820.nr_map = i + 1;
- e820.map[i].size -= current_addr - size;
- }
- e820_print_map("limit_regions endfor");
- return;
- }
- e820_print_map("limit_regions endfunc");
-}
-
/* Overridden in paravirt.c if CONFIG_PARAVIRT */
char * __init __attribute__((weak)) memory_setup(void)
{
@@ -249,93 +219,3 @@ void __init setup_memory_map(void)
e820_print_map(memory_setup());
}
-static int __initdata user_defined_memmap;
-
-/*
- * "mem=nopentium" disables the 4MB page tables.
- * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
- * to <mem>, overriding the bios size.
- * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
- * <start> to <start>+<mem>, overriding the bios size.
- *
- * HPA tells me bootloaders need to parse mem=, so no new
- * option should be mem= [also see Documentation/i386/boot.txt]
- */
-static int __init parse_mem(char *arg)
-{
- if (!arg)
- return -EINVAL;
-
- if (strcmp(arg, "nopentium") == 0) {
- setup_clear_cpu_cap(X86_FEATURE_PSE);
- } else {
- /* If the user specifies memory size, we
- * limit the BIOS-provided memory map to
- * that size. exactmap can be used to specify
- * the exact map. mem=number can be used to
- * trim the existing memory map.
- */
- unsigned long long mem_size;
-
- mem_size = memparse(arg, &arg);
- limit_regions(mem_size);
- user_defined_memmap = 1;
- }
- return 0;
-}
-early_param("mem", parse_mem);
-
-static int __init parse_memmap(char *arg)
-{
- if (!arg)
- return -EINVAL;
-
- if (strcmp(arg, "exactmap") == 0) {
-#ifdef CONFIG_CRASH_DUMP
- /* If we are doing a crash dump, we
- * still need to know the real mem
- * size before original memory map is
- * reset.
- */
- e820_register_active_regions(0, 0, -1UL);
- saved_max_pfn = e820_end_of_ram();
- remove_all_active_ranges();
-#endif
- e820.nr_map = 0;
- user_defined_memmap = 1;
- } else {
- /* If the user specifies memory size, we
- * limit the BIOS-provided memory map to
- * that size. exactmap can be used to specify
- * the exact map. mem=number can be used to
- * trim the existing memory map.
- */
- unsigned long long start_at, mem_size;
-
- mem_size = memparse(arg, &arg);
- if (*arg == '@') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_RAM);
- } else if (*arg == '#') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_ACPI);
- } else if (*arg == '$') {
- start_at = memparse(arg+1, &arg);
- add_memory_region(start_at, mem_size, E820_RESERVED);
- } else {
- limit_regions(mem_size);
- user_defined_memmap = 1;
- }
- }
- return 0;
-}
-early_param("memmap", parse_memmap);
-
-void __init finish_e820_parsing(void)
-{
- if (user_defined_memmap) {
- printk(KERN_INFO "user-defined physical RAM map:\n");
- e820_print_map("user");
- }
-}
-
Index: linux-2.6/arch/x86/kernel/e820_64.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/e820_64.c
+++ linux-2.6/arch/x86/kernel/e820_64.c
@@ -96,75 +96,6 @@ char *__init machine_specific_memory_set
return who;
}
-static int __init parse_memopt(char *p)
-{
- if (!p)
- return -EINVAL;
- end_user_pfn = memparse(p, &p);
- end_user_pfn >>= PAGE_SHIFT;
- return 0;
-}
-early_param("mem", parse_memopt);
-
-static int userdef __initdata;
-
-static int __init parse_memmap_opt(char *p)
-{
- char *oldp;
- unsigned long long start_at, mem_size;
-
- if (!strcmp(p, "exactmap")) {
-#ifdef CONFIG_CRASH_DUMP
- /*
- * If we are doing a crash dump, we still need to know
- * the real mem size before original memory map is
- * reset.
- */
- e820_register_active_regions(0, 0, -1UL);
- saved_max_pfn = e820_end_of_ram();
- remove_all_active_ranges();
-#endif
- e820.nr_map = 0;
- userdef = 1;
- return 0;
- }
-
- oldp = p;
- mem_size = memparse(p, &p);
- if (p == oldp)
- return -EINVAL;
-
- userdef = 1;
- if (*p == '@') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_RAM);
- } else if (*p == '#') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_ACPI);
- } else if (*p == '$') {
- start_at = memparse(p+1, &p);
- add_memory_region(start_at, mem_size, E820_RESERVED);
- } else {
- end_user_pfn = (mem_size >> PAGE_SHIFT);
- }
- return *p == '\0' ? 0 : -EINVAL;
-}
-early_param("memmap", parse_memmap_opt);
-
-void __init finish_e820_parsing(void)
-{
- if (userdef) {
- int nr = e820.nr_map;
-
- if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
- early_panic("Invalid user supplied memory map");
- e820.nr_map = nr;
-
- printk(KERN_INFO "user-defined physical RAM map:\n");
- e820_print_map("user");
- }
-}
-
int __init arch_get_ram_range(int slot, u64 *addr, u64 *size)
{
int i;
Index: linux-2.6/include/asm-x86/e820.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820.h
+++ linux-2.6/include/asm-x86/e820.h
@@ -97,6 +97,8 @@ extern int e820_find_active_region(const
extern void e820_register_active_regions(int nid, unsigned long start_pfn,
unsigned long end_pfn);
extern u64 e820_hole_size(u64 start, u64 end);
+extern void finish_e820_parsing(void);
+
#endif /* __ASSEMBLY__ */
#define ISA_START_ADDRESS 0xa0000
Index: linux-2.6/include/asm-x86/e820_32.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820_32.h
+++ linux-2.6/include/asm-x86/e820_32.h
@@ -19,9 +19,7 @@
#ifndef __ASSEMBLY__
extern void setup_memory_map(void);
-extern void finish_e820_parsing(void);
-extern void limit_regions(unsigned long long size);
extern void init_iomem_resources(struct resource *code_resource,
struct resource *data_resource,
struct resource *bss_resource);
Index: linux-2.6/include/asm-x86/e820_64.h
===================================================================
--- linux-2.6.orig/include/asm-x86/e820_64.h
+++ linux-2.6/include/asm-x86/e820_64.h
@@ -22,8 +22,6 @@ extern int is_memory_any_valid(unsigned
extern int e820_all_non_reserved(unsigned long start, unsigned long end);
extern int is_memory_all_valid(unsigned long start, unsigned long end);
-extern void finish_e820_parsing(void);
-
#endif/*!__ASSEMBLY__*/
#endif/*__E820_HEADER*/
^ permalink raw reply [flat|nested] 19+ messages in thread