use lmb with x86
Subsystems (9)
cover use lmb with x86
extend lmb to have find_lmb_area()
Thanks
Yinghai
PATCH 1/4 x86: do not free zero sized per cpu areas
From: Ian Campbell <[email protected]>
This avoids an infinite loop in free_early_partial().
Add a warning to free_early_partial to catch future problems.
-v5: put back start > end back into WARN_ONCE() -v6: use one line for if according to linus
--- kernel/early_res.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-)
kernel/early_res.c +6 -0
@@ -333,6 +333,12 @@ void __init free_early_partial(u64 start, u64 end)struct early_res *r;int i; +if (start == end) +return ; + +if (WARN_ONCE (start > end,"free_early_partial: wrong range [%#llx, %#llx]\n" , start, end)) +return ; + try_next: i =find_overlapped_early (start, end);if (i >= max_early_res)
Signature · 1 line
1.6.4.2
PATCH 2/4 x86: add find_e820_area_node
In reply to Yinghai LuDavid Miller pointed out that early_res have problem to find node data on correct node when we have
node0: [0, 2g), [4g, 6g), [10g, 14g)
node1: [6g, 10g), [14g, 18g)
the cross node case
the problem is there for x86 bits even before we are using early_res for bootmem replacement. after early_res for bootmem replacement, alloc_bootmem_node still can get range on correct node
this patch is fixing problem before bootmem or early_res replacement for bootmem.
now only user is for x86 64bit numa to find node data.
the point is use early_node_map with find_e820_area_node()
--- arch/x86/include/asm/e820.h | 1 + arch/x86/kernel/e820.c | 15 +++++++++++++++ arch/x86/mm/numa_64.c | 4 ++-- include/linux/mm.h | 2 ++ mm/page_alloc.c | 37 +++++++++++++++++++++++-------------- 5 files changed, 43 insertions(+), 16 deletions(-)
arch/x86/include/asm/e820.h +1 -0
@@ -116,6 +116,7 @@ extern unsigned long end_user_pfn;extern u64 find_e820_area (u64 start,u64 end,u64 size,u64 align);extern u64 find_e820_area_size (u64 start,u64 *sizep,u64 align); +u64 find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align);extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align);#include <linux/early_res.h>
arch/x86/kernel/e820.c +15 -0
@@ -787,6 +787,21 @@ u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align)return -1ULL ; } +u64 __init find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align) +{ +u64 addr; +/* +* need to call this function after e820_register_active_regions +* so early_node_map[] is set +*/ + addr =find_memory_core_early (nid, size, align, start, end); +if (addr !=-1ULL ) +return addr; + +/* fallback, should already have start end in the node range */ +return find_e820_area (start, end, size, align); +} +/* * pre allocated 4k and reserved it in e820 */
arch/x86/mm/numa_64.c +2 -2
@@ -174,7 +174,7 @@ static void * __init early_node_mem(int nodeid, unsigned long start,if (start < (MAX_DMA32_PFN <<PAGE_SHIFT ) && end > (MAX_DMA32_PFN <<PAGE_SHIFT )) start =MAX_DMA32_PFN <<PAGE_SHIFT ; - mem =find_e820_area (start, end, size, align); + mem =find_e820_area_node (nodeid, start, end, size, align);if (mem !=-1L )return __va (mem);
@@ -184,7 +184,7 @@ static void * __init early_node_mem(int nodeid, unsigned long start, start =MAX_DMA32_PFN <<PAGE_SHIFT ; else start =MAX_DMA_PFN <<PAGE_SHIFT ; - mem =find_e820_area (start, end, size, align); + mem =find_e820_area_node (nodeid, start, end, size, align);if (mem !=-1L )return __va (mem);
include/linux/mm.h +2 -0
@@ -1160,6 +1160,8 @@ extern void free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn);int add_from_early_node_map (struct range *range,int az,int nr_range,int nid); +u64 __init find_memory_core_early (int nid,u64 size,u64 align, +u64 goal,u64 limit);void *__alloc_memory_core_early (int nodeid,u64 size,u64 align,u64 goal,u64 limit);typedef int (*work_fn_t )(unsigned long ,unsigned long ,void *);
mm/page_alloc.c +23 -14
@@ -3408,12 +3408,11 @@ int __init add_from_early_node_map(struct range *range, int az,return nr_range; } -#ifdef CONFIG_NO_BOOTMEM -void *__init __alloc_memory_core_early (int nid,u64 size,u64 align, +#ifdef CONFIG_HAVE_EARLY_RES +u64 __init find_memory_core_early (int nid,u64 size,u64 align,u64 goal,u64 limit) {int i; -void *ptr;/* need to go over early_node_map to find out good range for node */ for_each_active_range_index_in_nid (i, nid) {
@@ -3430,20 +3429,30 @@ void * __init __alloc_memory_core_early(int nid, u64 size, u64 align,if (addr ==-1ULL )continue ; -#if 0 -printk (KERN_DEBUG "alloc (nid=%d %llx - %llx) (%llx - %llx) %llx %llx => %llx\n" , - nid, - ei_start, ei_last, goal, limit, size, - align, addr); + return addr; + } + +return -1ULL ; +}#endif - ptr =phys_to_virt (addr); -memset (ptr,0 , size); -reserve_early_without_check (addr, addr + size,"BOOTMEM" ); - return ptr; - } +#ifdef CONFIG_NO_BOOTMEM +void *__init __alloc_memory_core_early (int nid,u64 size,u64 align, +u64 goal,u64 limit) +{ +void *ptr; -return NULL ; +u64 addr; + + addr =find_memory_core_early (nid, size, align, goal, limit); + +if (addr ==-1ULL ) +return NULL ; + + ptr =phys_to_virt (addr); +memset (ptr,0 , size); +reserve_early_without_check (addr, addr + size,"BOOTMEM" ); +return ptr; }#endif
Signature · 1 line
1.6.4.2
PATCH 3/4 x86: add sanitize_e820_map
In reply to Yinghai Luso we could not take e820.map directly.
and could change e820_saved to initdata
--- arch/x86/include/asm/e820.h | 5 ++--- arch/x86/kernel/e820.c | 26 ++++++++++++++++++-------- arch/x86/kernel/efi.c | 2 +- arch/x86/kernel/setup.c | 10 +++++----- arch/x86/xen/setup.c | 4 +--- 5 files changed, 27 insertions(+), 20 deletions(-)
arch/x86/include/asm/e820.h +2 -3
@@ -75,15 +75,14 @@ struct e820map {#ifdef __KERNEL__/* see comment in arch/x86/kernel/e820.c */ extern struct e820map e820; -extern struct e820map e820_saved;extern unsigned long pci_mem_start;extern int e820_any_mapped (u64 start,u64 end,unsigned type);extern int e820_all_mapped (u64 start,u64 end,unsigned type);extern void e820_add_region (u64 start,u64 size,int type);extern void e820_print_map (char *who); -extern int -sanitize_e820_map (struct e820entry *biosmap,int max_nr_map,u32 *pnr_map); +int sanitize_e820_map (void ); +void save_e820_map (void );extern u64 e820_update_range (u64 start,u64 size,unsigned old_type,unsigned new_type); extern u64 e820_remove_range(u64 start, u64 size, unsigned old_type,
arch/x86/kernel/e820.c +18 -8
@@ -35,7 +35,7 @@* next kernel with full memory. */ struct e820map e820; -struct e820map e820_saved; +static struct e820map __initdata e820_saved;/* For PCI or other memory-mapped resources */ unsigned long pci_mem_start =0xaeedbabe ;
@@ -224,7 +224,7 @@ void __init e820_print_map(char *who)* ______________________4_ */ -int __init sanitize_e820_map (struct e820entry *biosmap,int max_nr_map, +static int __init __sanitize_e820_map (struct e820entry *biosmap,int max_nr_map,u32 *pnr_map) { struct change_member {
@@ -383,6 +383,11 @@ int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map,return 0 ; } +int __init sanitize_e820_map (void ) +{ +return __sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +} +static int __init __append_e820_map (struct e820entry *biosmap,int nr_map) {while (nr_map) {
@@ -555,7 +560,7 @@ void __init update_e820(void)u32 nr_map; nr_map = e820.nr_map; -if (sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &nr_map)) +if (__sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &nr_map))return ; e820.nr_map = nr_map;printk (KERN_INFO "modified physical RAM map:\n" );
@@ -566,7 +571,7 @@ static void __init update_e820_saved(void)u32 nr_map; nr_map = e820_saved.nr_map; -if (sanitize_e820_map (e820_saved.map,ARRAY_SIZE (e820_saved.map), &nr_map)) +if (__sanitize_e820_map (e820_saved.map,ARRAY_SIZE (e820_saved.map), &nr_map))return ; e820_saved.nr_map = nr_map; }
@@ -661,7 +666,7 @@ void __init parse_e820_ext(struct setup_data *sdata, unsigned long pa_data) sdata =early_ioremap (pa_data, map_len); extmap = (struct e820entry *)(sdata->data);__append_e820_map (extmap, entries); -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map ();if (map_len >PAGE_SIZE )early_iounmap (sdata, map_len);printk (KERN_INFO "extended physical RAM map:\n" );
@@ -1043,7 +1048,7 @@ void __init finish_e820_parsing(void)if (userdef) {u32 nr = e820.nr_map; -if (sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &nr) <0 ) +if (__sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &nr) <0 )early_panic ("Invalid user supplied memory map" ); e820.nr_map = nr;
@@ -1173,7 +1178,7 @@ char *__init default_machine_specific_memory_setup(void)* the next section from 1mb->appropriate_mem_k */ new_nr = boot_params.e820_entries; -sanitize_e820_map (boot_params.e820_map, +__sanitize_e820_map (boot_params.e820_map,ARRAY_SIZE (boot_params.e820_map), &new_nr); boot_params.e820_entries = new_nr;
@@ -1200,12 +1205,17 @@ char *__init default_machine_specific_memory_setup(void)return who; } +void __init save_e820_map (void ) +{ +memcpy (&e820_saved, &e820,sizeof (struct e820map )); +} +void __init setup_memory_map (void ) {char *who; who = x86_init.resources.memory_setup (); -memcpy (&e820_saved, &e820,sizeof (struct e820map )); +save_e820_map ();printk (KERN_INFO "BIOS-provided physical RAM map:\n" );e820_print_map (who); }
arch/x86/kernel/efi.c +1 -1
@@ -272,7 +272,7 @@ static void __init do_add_efi_memmap(void) }e820_add_region (start, size, e820_type); } -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map (); }void __init efi_reserve_early (void )
arch/x86/kernel/setup.c +5 -5
@@ -461,8 +461,8 @@ static void __init e820_reserve_setup_data(void)if (!found)return ; -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); -memcpy (&e820_saved, &e820,sizeof (struct e820map )); +sanitize_e820_map (); +save_e820_map ();printk (KERN_INFO "extended physical RAM map:\n" );e820_print_map ("reserve setup_data" ); }
@@ -614,7 +614,7 @@ static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) d->ident);e820_update_range (0 ,0x10000 ,E820_RAM ,E820_RESERVED ); -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map ();return 0 ; }
@@ -683,7 +683,7 @@ static void __init trim_bios_range(void)* take them out. */ e820_remove_range (BIOS_BEGIN ,BIOS_END -BIOS_BEGIN ,E820_RAM ,1 ); -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map (); }/*
@@ -854,7 +854,7 @@ void __init setup_arch(char **cmdline_p)if (ppro_with_ram_bug ()) {e820_update_range (0x70000000ULL ,0x40000ULL ,E820_RAM ,E820_RESERVED ); -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map ();printk (KERN_INFO "fixed physical RAM map:\n" );e820_print_map ("bad_ppro" ); }
arch/x86/xen/setup.c +1 -3
@@ -43,8 +43,6 @@ char * __init xen_memory_setup(void) max_pfn =min (MAX_DOMAIN_PAGES , max_pfn); - e820.nr_map =0 ; -e820_add_region (0 ,PFN_PHYS ((u64 )max_pfn),E820_RAM );/*
@@ -65,7 +63,7 @@ char * __init xen_memory_setup(void)__pa (xen_start_info->pt_base), "XEN START INFO"); -sanitize_e820_map (e820.map,ARRAY_SIZE (e820.map), &e820.nr_map); +sanitize_e820_map ();return "Xen" ; }
Signature · 1 line
1.6.4.2
RFC PATCH 4/4 x86: use lmb to replace early_res
In reply to Yinghai Lustill keep kernel/early_res.c for the extension.
should move those file to lib/lmb.c later?
-v2: fix NO_BOOTMEM hang with printk
--- arch/x86/Kconfig | 1 + arch/x86/include/asm/e820.h | 38 ++- arch/x86/include/asm/lmb.h | 8 + arch/x86/kernel/e820.c | 163 ++---------- arch/x86/kernel/head.c | 2 +- arch/x86/kernel/head32.c | 4 +- arch/x86/kernel/head64.c | 2 + arch/x86/kernel/setup.c | 2 + arch/x86/kernel/setup_percpu.c | 6 - include/linux/early_res.h | 9 +- include/linux/lmb.h | 5 +- kernel/early_res.c | 593 +++++++++++++++------------------------ lib/lmb.c | 11 +- mm/page_alloc.c | 2 +- mm/sparse-vmemmap.c | 4 +- 15 files changed, 317 insertions(+), 533 deletions(-) create mode 100644 arch/x86/include/asm/lmb.h
arch/x86/Kconfig +1 -0
@@ -27,6 +27,7 @@ config X86 select HAVE_PERF_EVENTS if (!M386 && !M486) select HAVE_IOREMAP_PROT select HAVE_KPROBES + select HAVE_LMB select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS
arch/x86/include/asm/e820.h +26 -12
@@ -113,22 +113,36 @@ static inline void early_memtest(unsigned long start, unsigned long end)extern unsigned long end_user_pfn; -extern u64 find_e820_area (u64 start,u64 end,u64 size,u64 align); -extern u64 find_e820_area_size (u64 start,u64 *sizep,u64 align); -u64 find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align); -extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align);#include <linux/early_res.h> +static inline u64 find_e820_area (u64 start,u64 end,u64 size,u64 align) +{ +return find_lmb_area (start, end, size, align); +} +static inline u64 find_e820_area_size (u64 start,u64 *sizep,u64 align) +{ +return find_lmb_area_size (start, sizep, align); +} +static inline u64 +find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align) +{ +return find_lmb_area_node (nid, start, end, size, align); +} +extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align);extern unsigned long e820_end_of_ram_pfn (void );extern unsigned long e820_end_of_low_ram_pfn (void ); -extern int e820_find_active_region (const struct e820entry *ei, -unsigned long start_pfn, -unsigned long last_pfn, -unsigned long *ei_startpfn, -unsigned long *ei_endpfn); -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); +static inline void e820_register_active_regions (int nid, +unsigned long start_pfn, +unsigned long end_pfn) +{ +lmb_register_active_regions (nid, start_pfn, end_pfn); +} +static inline u64 e820_hole_size (u64 start,u64 end) +{ +return lmb_hole_size (start, end); +} +void init_lmb_memory (void ); +void fill_lmb_memory (void );extern void finish_e820_parsing (void );extern void e820_reserve_resources (void );extern void e820_reserve_resources_late (void );
arch/x86/include/asm/lmb.h new file +8 -0
@@ -0,0 +1,8 @@ +#ifndef _X86_LMB_H +#define _X86_LMB_H + +#define LMB_DBG(fmt...) printk(fmt) + +#define LMB_REAL_LIMIT 0 + +#endif
arch/x86/kernel/e820.c +23 -140
193 lines not on this page · raw message
arch/x86/kernel/head.c +1 -1
@@ -51,5 +51,5 @@ void __init reserve_ebda_region(void) lowmem =0x9f000 ;/* reserve all memory between lowmem and the 1MB mark */ -reserve_early_overlap_ok (lowmem,0x100000 ,"BIOS reserved" ); +reserve_early (lowmem,0x100000 ,"BIOS reserved" ); }
arch/x86/kernel/head32.c +3 -1
@@ -29,13 +29,15 @@ static void __init i386_default_early_setup(void)void __init i386_start_kernel (void ) { + +init_lmb_memory ();#ifdef CONFIG_X86_TRAMPOLINE /* * But first pinch a few for the stack/trampoline stuff * FIXME: Don't need the extra page at 4K, but need to fix * trampoline before removing it. (see the GDT stuff) */ -reserve_early_overlap_ok (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE , +reserve_early (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE ,"EX TRAMPOLINE" );#endif
arch/x86/kernel/head64.c +2 -0
@@ -96,6 +96,8 @@ void __init x86_64_start_kernel(char * real_mode_data)void __init x86_64_start_reservations (char *real_mode_data) { +init_lmb_memory (); +copy_bootdata (__va (real_mode_data));reserve_early (__pa_symbol (&_text),__pa_symbol (&__bss_stop),"TEXT DATA BSS" );
arch/x86/kernel/setup.c +2 -0
@@ -894,6 +894,8 @@ void __init setup_arch(char **cmdline_p) max_pfn_mapped =KERNEL_IMAGE_SIZE >>PAGE_SHIFT ;#endif +fill_lmb_memory (); +#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION setup_bios_corruption_check ();#endif
arch/x86/kernel/setup_percpu.c +0 -6
@@ -137,13 +137,7 @@ static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align)static void __init pcpu_fc_free (void *ptr,size_t size) { -#ifdef CONFIG_NO_BOOTMEM -u64 start =__pa (ptr); -u64 end = start + size; -free_early_partial (start, end); -#else free_bootmem (__pa (ptr), size); -#endif }static int __init pcpu_cpu_distance (unsigned int from,unsigned int to)
include/linux/early_res.h +6 -3
@@ -5,15 +5,18 @@extern void reserve_early (u64 start,u64 end,char *name);extern void reserve_early_overlap_ok (u64 start,u64 end,char *name);extern void free_early (u64 start,u64 end); -void free_early_partial (u64 start,u64 end);extern void early_res_to_bootmem (u64 start,u64 end); -void reserve_early_without_check (u64 start,u64 end,char *name);u64 find_early_area (u64 ei_start,u64 ei_last,u64 start,u64 end,u64 size,u64 align);u64 find_early_area_size (u64 ei_start,u64 ei_last,u64 start,u64 *sizep,u64 align); -u64 find_fw_memmap_area (u64 start,u64 end,u64 size,u64 align); +u64 find_lmb_area (u64 start,u64 end,u64 size,u64 align); +u64 find_lmb_area_size (u64 start,u64 *sizep,u64 align); +u64 find_lmb_area_node (int nid,u64 start,u64 end,u64 size,u64 align); +void lmb_register_active_regions (int nid,unsigned long start_pfn, +unsigned long last_pfn); +u64 lmb_hole_size (u64 start,u64 end);u64 get_max_mapped (void );#include <linux/range.h> int get_free_all_memory_range (struct range **rangep,int nodeid);
include/linux/lmb.h +4 -1
@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };struct lmb {
@@ -37,6 +38,8 @@ struct lmb { };extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];extern void __init lmb_init (void );extern void __init lmb_analyze (void );
kernel/early_res.c +228 -365
737 lines not on this page · raw message
lib/lmb.c +10 -1
@@ -18,6 +18,8 @@#define LMB_ALLOC_ANYWHERE 0struct lmb lmb; +struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];static int lmb_debug;
@@ -106,6 +108,11 @@ static void lmb_coalesce_regions(struct lmb_region *rgn,void __init lmb_init (void ) { + lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +/* Create a dummy zero size LMB which will get coalesced away later. * This simplifies the lmb_add() code below... */
@@ -169,7 +176,7 @@ static long lmb_add_region(struct lmb_region *rgn, u64 base, u64 size)if (coalesced)return coalesced; -if (rgn->cnt >=MAX_LMB_REGIONS ) +if (rgn->cnt >= (rgn->region_array_size -1 ))return -1 ;/* Couldn't coalesce the LMB, so add it to the sorted table. */
@@ -539,3 +546,5 @@ int lmb_find(struct lmb_property *res) }return -1 ; } + +
mm/page_alloc.c +1 -1
@@ -3451,7 +3451,7 @@ void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, ptr =phys_to_virt (addr);memset (ptr,0 , size); -reserve_early_without_check (addr, addr + size,"BOOTMEM" ); +reserve_early (addr, addr + size,"BOOTMEM" ); return ptr; }#endif
mm/sparse-vmemmap.c +2 -2
@@ -225,8 +225,8 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,char name[15 ];snprintf (name,sizeof (name),"MEMMAP %d" , nodeid); -reserve_early_without_check (__pa (vmemmap_buf_start), -__pa (vmemmap_buf), name); +reserve_early (__pa (vmemmap_buf_start), +__pa (vmemmap_buf), name); }#else free_bootmem (__pa (vmemmap_buf), vmemmap_buf_end - vmemmap_buf);
Signature · 1 line
1.6.4.2
Re: [RFC PATCH -v2 4/4] x86: use lmb to replace early_res
Quoted 5 lines · Yinghai Lu
still keep
kernel/early_res.cfor the extension.should move those file to
lib/lmb.clater?-v2: fix
NO_BOOTMEMhang with printk
Would be nice to track known TODO items - that way people dont feel that you are ignoring (or have missed) their feedback. For example are the list of 4 improvements i suggested in the previous mail planned, or do you disagree with some of that?
Thanks,
Ingo
RFC PATCH 1/2 lmb: seperate region array from lmb_region struct
In reply to Yinghai Lulmb_init() connect them back.
also add region_array_size in lmb_region to tack the region array size.
-v3: seperate lmb core change to seperated patch
--- include/linux/lmb.h | 5 ++++- lib/lmb.c | 11 ++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-)
include/linux/lmb.h +4 -1
@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };struct lmb {
@@ -37,6 +38,8 @@ struct lmb { };extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];extern void __init lmb_init (void );extern void __init lmb_analyze (void );
lib/lmb.c +8 -1
@@ -18,6 +18,8 @@#define LMB_ALLOC_ANYWHERE 0struct lmb lmb; +struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];static int lmb_debug;
@@ -106,6 +108,11 @@ static void lmb_coalesce_regions(structvoid __init lmb_init (void ) { + lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +/* Create a dummy zero size LMB which will get coalesced away later. * This simplifies the lmb_add() code below... */
@@ -169,7 +176,7 @@ static long lmb_add_region(struct lmb_reif (coalesced)return coalesced; -if (rgn->cnt >=MAX_LMB_REGIONS ) +if (rgn->cnt >= (rgn->region_array_size -1 ))return -1 ;/* Couldn't coalesce the LMB, so add it to the sorted table. */
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
lib/lmb.c static void lmb_coalesce_regions(struct · line 115
Quoted 7 lines · Yinghai Lu
void
__initlmb_init(void)
{lib/lmb.c +5
-0+ lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +
That's rather unreadable and has random whitespace noise.
Should be something like:
lmb.memory.region = lmb_memory_region; lmb.memory.region_array_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_array_size = ARRAY_SIZE(lmb_reserved_region);
also, i'd suggest to shorten region_array_size to region_size (we know it's an array), so it would become:
lmb.memory.region = lmb_memory_region; lmb.memory.region_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_size = ARRAY_SIZE(lmb_reserved_region);
lib/lmb.c static long lmb_add_region(struct lmb_re · line 180
Quoted 3 lines
lib/lmb.c +1
-1-if (rgn->cnt >=MAX_LMB_REGIONS ) +if (rgn->cnt >= (rgn->region_array_size -1 ))return -1 ;
'x >= y-1' is equivalent to 'x > y', so that should be:
if (rgn->cnt > rgn->region_size) Ingo
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 20 lines · Ingo Molnar
lib/lmb.c static void lmb_coalesce_regions(struct · line 115
Quoted 7 lines · Yinghai Lu
void
__initlmb_init(void)
{lib/lmb.c +5
-0+ lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +That's rather unreadable and has random whitespace noise.
Should be something like:
lmb.memory.region = lmb_memory_region; lmb.memory.region_array_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_array_size = ARRAY_SIZE(lmb_reserved_region);
Quoted 6 lines
also, i'd suggest to shorten
region_array_sizetoregion_size(we know it's an array), so it would become:lmb.memory.region = lmb_memory_region; lmb.memory.region_size = ARRAY_SIZE(lmb_memory_region);
I don't mean to be pedantic, but the LMB code already has a lot of region.size references so region_size looks a bit awkward. All of the accessors in linux/lmb.h use region_nr as the array index, so perhaps nr_regions would be less ambiguous.
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 24 lines · Paul Mundt
Quoted 23 lines · Ingo Molnar
lib/lmb.c static void lmb_coalesce_regions(struct · line 115
Quoted 7 lines · Yinghai Lu
void
__initlmb_init(void)
{lib/lmb.c +5
-0+ lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +That's rather unreadable and has random whitespace noise.
Should be something like:
lmb.memory.region = lmb_memory_region; lmb.memory.region_array_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_array_size = ARRAY_SIZE(lmb_reserved_region);also, i'd suggest to shorten
region_array_sizetoregion_size(we know it's an array), so it would become:
Quoted 7 lines
Quoted 3 lines
lmb.memory.region = lmb_memory_region; lmb.memory.region_size = ARRAY_SIZE(lmb_memory_region);I don't mean to be pedantic, but the LMB code already has a lot of region.size references so
region_sizelooks a bit awkward. All of the accessors inlinux/lmb.huseregion_nras the array index, so perhapsnr_regionswould be less ambiguous.
@@ -26,7 +26,8 @@ struct lmb_property {
struct lmb_region {
unsigned long cnt;
u64 size;
- struct lmb_property region[MAX_LMB_REGIONS+1];
+ struct lmb_property *region;
+ unsigned long region_array_size;
};
cnt is number of slots used.
size is memory size
can we use rgn_sz for region array size?
YH
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 22 lines · Yinghai Lu
Quoted 11 lines · Paul Mundt
Quoted 6 lines · Ingo Molnar
also, i'd suggest to shorten
region_array_sizetoregion_size(we know it's an array), so it would become:lmb.memory.region = lmb_memory_region; lmb.memory.region_size = ARRAY_SIZE(lmb_memory_region);I don't mean to be pedantic, but the LMB code already has a lot of region.size references so
region_sizelooks a bit awkward. All of the accessors inlinux/lmb.huseregion_nras the array index, so perhapsnr_regionswould be less ambiguous.include/linux/lmb.h +2
-1@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };
Quoted 3 lines
cnt is number of slots used.
size is memory size
I'm fully aware of what each of these things are, and I have no idea what any of that has to do with the point I raised above.
Quoted 2 lines
can we use
rgn_szfor region array size?
No.
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
In reply to Ingo MolnarQuoted 14 lines · Ingo Molnar
lib/lmb.c static void lmb_coalesce_regions(struct · line 115
Quoted 7 lines · Yinghai Lu
void
__initlmb_init(void)
{lib/lmb.c +5
-0+ lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +That's rather unreadable and has random whitespace noise.
Should be something like:
Quoted 7 lines
lmb.memory.region = lmb_memory_region; lmb.memory.region_array_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_array_size = ARRAY_SIZE(lmb_reserved_region);also, i'd suggest to shorten
region_array_sizetoregion_size(we know it's an array), so it would become:
I dislike those arrays anyways. See my other message about turning them into lists, which would get rid of capacity constraints completely. What do you think ?
Cheers,
Ben.
RFC PATCH v4 1/2 lmb: seperate region array from lmb_region struct
lmb_init() connect them back.
also add nr_regions in lmb_region to tack the region array size.
-v3: seperate lmb core change to seperated patch
-v4: according to Ingo, change >= x -1 to > x
change region_array_size to nr_regions
--- include/linux/lmb.h | 5 ++++- lib/lmb.c | 9 ++++++++- 2 files changed, 12 insertions(+), 2 deletions(-)
include/linux/lmb.h +4 -1
@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long nr_regions; };struct lmb {
@@ -37,6 +38,8 @@ struct lmb { };extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];extern void __init lmb_init (void );extern void __init lmb_analyze (void );
lib/lmb.c +8 -1
@@ -18,6 +18,8 @@#define LMB_ALLOC_ANYWHERE 0struct lmb lmb; +struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];static int lmb_debug;
@@ -106,6 +108,11 @@ static void lmb_coalesce_regions(structvoid __init lmb_init (void ) { + lmb.memory.region = lmb_memory_region; + lmb.reserved.region = lmb_reserved_region; + lmb.memory.nr_regions =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.nr_regions =ARRAY_SIZE (lmb_reserved_region); +/* Create a dummy zero size LMB which will get coalesced away later. * This simplifies the lmb_add() code below... */
@@ -169,7 +176,7 @@ static long lmb_add_region(struct lmb_reif (coalesced)return coalesced; -if (rgn->cnt >=MAX_LMB_REGIONS ) +if (rgn->cnt > rgn->nr_regions)return -1 ;/* Couldn't coalesce the LMB, so add it to the sorted table. */
RFC PATCH 2/2 x86: use lmb to replace early_res
In reply to Benjamin Herrenschmidtstill keep kernel/early_res.c for the extension.
should move those file to lib/lmb.c later?
in early_res.c
1. change find_e820_area_xxx, to find_lmb_area_xxx
2. e820_register_active_regions to lmb_register_active_regions.
3. reserve_early will call lmb_reserve directly.
4. free_early will call lmb_free directly.
5. remove functions that are used by old reserve_early and free_early
6. get_free_all_memory_range use lmb.reserved.
7. early_res_to_bootmem use lmb.reserved
8. add fill_lmb_memory() to fill lmb.memory according e820 RAM entries
-v2: fix NO_BOOTMEM hang with printk
-v4: add add_lmb_memory that could increase lmb.memory.region size
change region_array_size to nr_regions
make sure some find_lmb_area<_size> are called after fill_lmb_memory
todo:
1. make early_memtest to depend on early_res. and move it to mm/
2. make all lmb user to use extend early_res/nobootmem
3. merge lmb.c and early_res.c, move it to mm/
4. make other platform to use lmb/early_res/nobootmem
5. remove BOOTMEM related code
--- arch/x86/Kconfig | 1 arch/x86/include/asm/e820.h | 38 +- arch/x86/include/asm/lmb.h | 8 arch/x86/kernel/check.c | 14 arch/x86/kernel/e820.c | 171 +---------- arch/x86/kernel/head.c | 2 arch/x86/kernel/head32.c | 5 arch/x86/kernel/head64.c | 2 arch/x86/kernel/setup.c | 9 arch/x86/kernel/setup_percpu.c | 6 arch/x86/mm/memtest.c | 5 arch/x86/mm/numa_64.c | 4 include/linux/early_res.h | 19 - kernel/early_res.c | 631 ++++++++++++++++------------------------- mm/page_alloc.c | 2 mm/sparse-vmemmap.c | 11 16 files changed, 344 insertions(+), 584 deletions(-)
arch/x86/Kconfig +1 -0
@@ -27,6 +27,7 @@ config X86 select HAVE_PERF_EVENTS if (!M386 && !M486) select HAVE_IOREMAP_PROT select HAVE_KPROBES + select HAVE_LMB select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS
arch/x86/include/asm/e820.h +22 -16
@@ -111,24 +111,30 @@ static inline void early_memtest(unsigne }#endif -extern unsigned long end_user_pfn; - -extern u64 find_e820_area (u64 start,u64 end,u64 size,u64 align); -extern u64 find_e820_area_size (u64 start,u64 *sizep,u64 align); -u64 find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align); -extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align); -#include <linux/early_res.h> -extern unsigned long e820_end_of_ram_pfn (void );extern unsigned long e820_end_of_low_ram_pfn (void ); -extern int e820_find_active_region (const struct e820entry *ei, -unsigned long start_pfn, -unsigned long last_pfn, -unsigned long *ei_startpfn, -unsigned long *ei_endpfn); -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); +#include <linux/early_res.h> +extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align); + +/* +* next three functions will be removed, esp find_e820_area() +* can not be used before fill_lmb_memory() +*/ +static inline u64 find_e820_area (u64 start,u64 end,u64 size,u64 align) +{ +return find_lmb_area (start, end, size, align); +} +static inline void e820_register_active_regions (int nid,unsigned long start_pfn,unsigned long end_pfn) +{ +lmb_register_active_regions (nid, start_pfn, end_pfn); +} +static inline u64 e820_hole_size (u64 start,u64 end) +{ +return lmb_hole_size (start, end); +} + +void init_lmb_memory (void ); +void fill_lmb_memory (void );extern void finish_e820_parsing (void );extern void e820_reserve_resources (void );extern void e820_reserve_resources_late (void );
arch/x86/kernel/e820.c +26 -145
213 lines not on this page · raw message
arch/x86/kernel/head.c +1 -1
@@ -51,5 +51,5 @@ void __init reserve_ebda_region(void) lowmem =0x9f000 ;/* reserve all memory between lowmem and the 1MB mark */ -reserve_early_overlap_ok (lowmem,0x100000 ,"BIOS reserved" ); +reserve_early (lowmem,0x100000 ,"BIOS reserved" ); }
arch/x86/kernel/head32.c +3 -2
@@ -29,14 +29,15 @@ static void __init i386_default_early_sevoid __init i386_start_kernel (void ) { +init_lmb_memory (); +#ifdef CONFIG_X86_TRAMPOLINE /* * But first pinch a few for the stack/trampoline stuff * FIXME: Don't need the extra page at 4K, but need to fix * trampoline before removing it. (see the GDT stuff) */ -reserve_early_overlap_ok (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE , -"EX TRAMPOLINE" ); +reserve_early (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE ,"EX TRAMPOLINE" );#endif reserve_early (__pa_symbol (&_text),__pa_symbol (&__bss_stop),"TEXT DATA BSS" );
arch/x86/kernel/head64.c +2 -0
@@ -96,6 +96,8 @@ void __init x86_64_start_kernel(char * rvoid __init x86_64_start_reservations (char *real_mode_data) { +init_lmb_memory (); +copy_bootdata (__va (real_mode_data));reserve_early (__pa_symbol (&_text),__pa_symbol (&__bss_stop),"TEXT DATA BSS" );
arch/x86/kernel/setup.c +6 -3
@@ -868,8 +868,6 @@ void __init setup_arch(char **cmdline_p)*/ max_pfn =e820_end_of_ram_pfn (); -/* preallocate 4k for mptable mpc */ -early_reserve_e820_mpc_new ();/* update e820 for memory not covered by WB MTRRs */ mtrr_bp_init ();if (mtrr_trim_uncached_memory (max_pfn))
@@ -894,6 +892,11 @@ void __init setup_arch(char **cmdline_p) max_pfn_mapped =KERNEL_IMAGE_SIZE >>PAGE_SHIFT ;#endif +fill_lmb_memory (); + +/* preallocate 4k for mptable mpc */ +early_reserve_e820_mpc_new (); +#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION setup_bios_corruption_check ();#endif
@@ -970,7 +973,7 @@ void __init setup_arch(char **cmdline_p)initmem_init (0 , max_pfn, acpi, k8);#ifndef CONFIG_NO_BOOTMEM -early_res_to_bootmem (0 , max_low_pfn<<PAGE_SHIFT ); +lmb_reserved_to_bootmem (0 , max_low_pfn<<PAGE_SHIFT );#endif dma32_reserve_bootmem ();
arch/x86/kernel/setup_percpu.c +0 -6
@@ -137,13 +137,7 @@ static void * __init pcpu_fc_alloc(unsigstatic void __init pcpu_fc_free (void *ptr,size_t size) { -#ifdef CONFIG_NO_BOOTMEM -u64 start =__pa (ptr); -u64 end = start + size; -free_early_partial (start, end); -#else free_bootmem (__pa (ptr), size); -#endif }static int __init pcpu_cpu_distance (unsigned int from,unsigned int to)
include/linux/early_res.h +11 -8
@@ -3,19 +3,22 @@#ifdef __KERNEL__extern void reserve_early (u64 start,u64 end,char *name); -extern void reserve_early_overlap_ok (u64 start,u64 end,char *name);extern void free_early (u64 start,u64 end); -void free_early_partial (u64 start,u64 end); -extern void early_res_to_bootmem (u64 start,u64 end); +void lmb_reserved_to_bootmem (u64 start,u64 end); -void reserve_early_without_check (u64 start,u64 end,char *name);u64 find_early_area (u64 ei_start,u64 ei_last,u64 start,u64 end,u64 size,u64 align); -u64 find_early_area_size (u64 ei_start,u64 ei_last,u64 start, -u64 *sizep,u64 align); -u64 find_fw_memmap_area (u64 start,u64 end,u64 size,u64 align); + +void add_lmb_memory (u64 start,u64 end); + +u64 find_lmb_area (u64 start,u64 end,u64 size,u64 align); +u64 find_lmb_area_size (u64 start,u64 *sizep,u64 align); +u64 find_lmb_area_node (int nid,u64 start,u64 end,u64 size,u64 align); +void lmb_register_active_regions (int nid,unsigned long start_pfn, +unsigned long last_pfn); +u64 lmb_hole_size (u64 start,u64 end);u64 get_max_mapped (void ); -#include <linux/range.h> +struct range ;int get_free_all_memory_range (struct range **rangep,int nodeid);#endif /* __KERNEL__ */
kernel/early_res.c +251 -380
790 lines not on this page · raw message
mm/page_alloc.c +1 -1
@@ -3451,7 +3451,7 @@ void * __init __alloc_memory_core_early( ptr =phys_to_virt (addr);memset (ptr,0 , size); -reserve_early_without_check (addr, addr + size,"BOOTMEM" ); +reserve_early (addr, addr + size,"BOOTMEM" ); return ptr; }#endif
mm/sparse-vmemmap.c +0 -11
@@ -219,18 +219,7 @@ void __init sparse_mem_maps_populate_nodif (vmemmap_buf_start) {/* need to free left buf */ -#ifdef CONFIG_NO_BOOTMEM -free_early (__pa (vmemmap_buf_start),__pa (vmemmap_buf_end)); -if (vmemmap_buf_start < vmemmap_buf) { -char name[15 ]; - -snprintf (name,sizeof (name),"MEMMAP %d" , nodeid); -reserve_early_without_check (__pa (vmemmap_buf_start), -__pa (vmemmap_buf), name); - } -#else free_bootmem (__pa (vmemmap_buf), vmemmap_buf_end - vmemmap_buf); -#endif vmemmap_buf =NULL ; vmemmap_buf_end =NULL ; }
arch/x86/include/asm/lmb.h new file +8 -0
@@ -0,0 +1,8 @@ +#ifndef _X86_LMB_H +#define _X86_LMB_H + +#define LMB_DBG(fmt...) printk(fmt) + +#define LMB_REAL_LIMIT 0 + +#endif
arch/x86/kernel/check.c +8 -6
@@ -2,7 +2,8 @@#include <linux/sched.h> #include <linux/kthread.h> #include <linux/workqueue.h> -#include <asm/e820.h> +#include <linux/early_res.h> +#include <asm/proto.h> /*
@@ -18,10 +19,12 @@ static int __read_mostly memory_corruptistatic unsigned __read_mostly corruption_check_size =64 *1024 ;static unsigned __read_mostly corruption_check_period =60 ;/* seconds */ -static struct e820entry scan_areas[MAX_SCAN_AREAS ]; +static struct scan_area { +u64 addr; +u64 size; +} scan_areas[MAX_SCAN_AREAS ];static int num_scan_areas; -static __init int set_corruption_check (char *arg) {char *end;
@@ -81,7 +84,7 @@ void __init setup_bios_corruption_check(while (addr < corruption_check_size && num_scan_areas <MAX_SCAN_AREAS ) {u64 size; - addr =find_e820_area_size (addr, &size,PAGE_SIZE ); + addr =find_lmb_area_size (addr, &size,PAGE_SIZE );if (!(addr +1 ))break ;
@@ -92,7 +95,7 @@ void __init setup_bios_corruption_check(if ((addr + size) > corruption_check_size) size = corruption_check_size - addr; -e820_update_range (addr, size,E820_RAM ,E820_RESERVED ); +reserve_early (addr, addr + size,"SCAN RAM" ); scan_areas[num_scan_areas].addr = addr; scan_areas[num_scan_areas].size = size; num_scan_areas++;
@@ -105,7 +108,6 @@ void __init setup_bios_corruption_check(printk (KERN_INFO "Scanning %d areas for low memory corruption\n" , num_scan_areas); -update_e820 (); }
arch/x86/mm/memtest.c +2 -3
@@ -6,8 +6,7 @@#include <linux/smp.h> #include <linux/init.h> #include <linux/pfn.h> - -#include <asm/e820.h> +#include <linux/early_res.h> static u64 patterns[]__initdata = {0 ,
@@ -74,7 +73,7 @@ static void __init do_one_pass(u64 patteu64 size =0 ;while (start < end) { - start =find_e820_area_size (start, &size,1 ); + start =find_lmb_area_size (start, &size,1 );/* done ? */ if (start >= end)
arch/x86/mm/numa_64.c +2 -2
@@ -174,7 +174,7 @@ static void * __init early_node_mem(intif (start < (MAX_DMA32_PFN <<PAGE_SHIFT ) && end > (MAX_DMA32_PFN <<PAGE_SHIFT )) start =MAX_DMA32_PFN <<PAGE_SHIFT ; - mem =find_e820_area_node (nodeid, start, end, size, align); + mem =find_lmb_area_node (nodeid, start, end, size, align);if (mem !=-1L )return __va (mem);
@@ -184,7 +184,7 @@ static void * __init early_node_mem(int start =MAX_DMA32_PFN <<PAGE_SHIFT ; else start =MAX_DMA_PFN <<PAGE_SHIFT ; - mem =find_e820_area_node (nodeid, start, end, size, align); + mem =find_lmb_area_node (nodeid, start, end, size, align);if (mem !=-1L )return __va (mem);
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
In reply to Benjamin HerrenschmidtQuoted 20 lines · Benjamin Herrenschmidt
Quoted 19 lines · Ingo Molnar
lib/lmb.c static void lmb_coalesce_regions(struct · line 115
Quoted 7 lines · Yinghai Lu
void
__initlmb_init(void)
{lib/lmb.c +5
-0+ lmb.memory.region = lmb_memory_region; + lmb.memory.region_array_size =ARRAY_SIZE (lmb_memory_region); + lmb.reserved.region = lmb_reserved_region; + lmb.reserved.region_array_size =ARRAY_SIZE (lmb_reserved_region); +That's rather unreadable and has random whitespace noise.
Should be something like:
lmb.memory.region = lmb_memory_region; lmb.memory.region_array_size = ARRAY_SIZE(lmb_memory_region); lmb.reserved.region = lmb_reserved_region; lmb.reserved.region_array_size = ARRAY_SIZE(lmb_reserved_region);
Quoted 7 lines
Quoted 2 lines
also, i'd suggest to shorten
region_array_sizetoregion_size(we know it's an array), so it would become:I dislike those arrays anyways. See my other message about turning them into lists, which would get rid of capacity constraints completely. What do you think ?
2/2 introduce one new function that could double the array size
please check the v4.
the function rely on find_lmb_area().
it will check if there is enough space left, otherwise try to get new big array, and copy old array to new array.
final function like:
static void __init __check_and_double_region_array(struct lmb_region *type,
struct lmb_property *static_region,
u64 ex_start, u64 ex_end)
{
u64 start, end, size, mem;
struct lmb_property *new, *old;
unsigned long rgnsz = type->nr_regions;
/* do we have enough slots left ? */
if ((rgnsz - type->cnt) > max_t(unsigned long, rgnsz/8, 2))
return;
old = type->region;
/* double it */
mem = -1ULL;
size = sizeof(struct lmb_property) * rgnsz * 2;
if (old == static_region)
start = 0;
else
start = __pa(old) + sizeof(struct lmb_property) * rgnsz;
end = ex_start;
if (start + size < end)
mem = find_lmb_area(start, end, size,
sizeof(struct lmb_property));
if (mem == -1ULL) {
start = ex_end;
end = get_max_mapped();
if (start + size < end)
mem = find_lmb_area(start, end, size, sizeof(struct lmb_property));
}
if (mem == -1ULL)
panic("can not find more space for lmb.reserved.region array");
new = __va(mem);
/* copy old to new */
memcpy(&new[0], &old[0], sizeof(struct lmb_property) * rgnsz);
memset(&new[rgnsz], 0, sizeof(struct lmb_property) * rgnsz);
memset(&old[0], 0, sizeof(struct lmb_property) * rgnsz);
type->region = new;
type->nr_regions = rgnsz * 2;
printk(KERN_DEBUG "lmb.reserved.region array is doubled to %ld at [%llx - %llx]\n",
type->nr_regions, mem, mem + size - 1);
/* reserve new array and free old one */
lmb_reserve(mem, sizeof(struct lmb_property) * rgnsz * 2);
if (old != static_region)
lmb_free(__pa(old), sizeof(struct lmb_property) * rgnsz);
}void __init add_lmb_memory(u64 start, u64 end)
{
__check_and_double_region_array(&lmb.memory, &lmb_memory_region[0], start, end);
lmb_add(start, end - start);
}void __init reserve_early(u64 start, u64 end, char *name)
{
if (start == end)
return;
if (WARN_ONCE(start > end, "reserve_early: wrong range [%#llx, %#llx]\n", start, end))
return;
__check_and_double_region_array(&lmb.reserved, &lmb_reserved_region[0], start, end);
lmb_reserve(start, end - start);
}void __init free_early(u64 start, u64 end)
{
if (start == end)
return;
if (WARN_ONCE(start > end, "free_early: wrong range [%#llx, %#llx]\n", start, end))
return;
/* keep punching hole, could run out of slots too */
__check_and_double_region_array(&lmb.reserved, &lmb_reserved_region[0], start, end);
lmb_free(start, end - start);
}with those function, we can replace the bootmem in x86.
Thanks
Yinghai
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 5 lines · Yinghai Lu
I dislike those arrays anyways. See my other message about turning them into lists, which would get rid of capacity constraints completely. What do you think ?
Quoted 1 line
2/2 introduce one new function that could double the array size
It's still bloody arrays with fixed sizes, arbitrary limits and arbitrary waste of BSS space ;-) To be honest, I much prefer my idea of linked lists... But I'll let others speak.
I think your double array size looks more like a band-aid than a proper fix. If we are going to use LMB in the long run for bootmem, we need to properly address its capacity constraints, not just paper over the problem.
Cheers,
Ben.
Quoted 110 lines
please check the v4.
the function rely on
find_lmb_area().it will check if there is enough space left, otherwise try to get new big array, and
copy old array to new array.final function like:
static void
__init__check_and_double_region_array(structlmb_region*type,
structlmb_property*static_region, u64ex_start, u64ex_end){ u64 start, end, size, mem; struct lmb_property *new, *old; unsigned long rgnsz = type->nr_regions; /* do we have enough slots left ? */ if ((rgnsz - type->cnt) > max_t(unsigned long, rgnsz/8, 2)) return; old = type->region; /* double it */ mem = -1ULL; size = sizeof(struct lmb_property) * rgnsz * 2; if (old == static_region) start = 0; else start = __pa(old) + sizeof(struct lmb_property) * rgnsz; end = ex_start; if (start + size < end) mem = find_lmb_area(start, end, size, sizeof(struct lmb_property)); if (mem == -1ULL) { start = ex_end; end = get_max_mapped(); if (start + size < end) mem = find_lmb_area(start, end, size, sizeof(struct lmb_property)); } if (mem == -1ULL) panic("can not find more space for lmb.reserved.regionarray");
new = __va(mem); /* copy old to new */ memcpy(&new[0], &old[0], sizeof(struct lmb_property) * rgnsz); memset(&new[rgnsz], 0, sizeof(struct lmb_property) * rgnsz); memset(&old[0], 0, sizeof(struct lmb_property) * rgnsz); type->region = new; type->nr_regions = rgnsz * 2; printk(KERN_DEBUG "lmb.reserved.region array is doubled to %ld at [%llx - %llx]\n", type->nr_regions, mem, mem + size - 1); /* reserve new array and free old one */ lmb_reserve(mem, sizeof(struct lmb_property) * rgnsz * 2); if (old != static_region) lmb_free(__pa(old), sizeof(struct lmb_property) *rgnsz);
}void
__initadd_lmb_memory(u64 start, u64 end){
__check_and_double_region_array(&lmb.memory,&lmb_memory_region[0], start, end); lmb_add(start, end - start); }void
__initreserve_early(u64 start, u64 end, char *name){ if (start == end) return; if (WARN_ONCE(start > end, "reserve_early: wrong range [%#llx, %#llx]\n", start, end)) return; __check_and_double_region_array(&lmb.reserved, &lmb_reserved_region[0], start, end); lmb_reserve(start, end - start); }void
__initfree_early(u64 start, u64 end){ if (start == end) return; if (WARN_ONCE(start > end, "free_early: wrong range [%#llx, % #llx]\n", start, end)) return; /* keep punching hole, could run out of slots too */ __check_and_double_region_array(&lmb.reserved, &lmb_reserved_region[0], start, end); lmb_free(start, end - start); }with those function, we can replace the bootmem in x86.
Thanks
Yinghai
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
In reply to Yinghai LuQuoted 29 lines · Yinghai Lu
lmb_init()connect them back.also add
region_array_sizeinlmb_regionto tack the region array size.-v3: seperate lmb core change to seperated patch
Signed-off-by Yinghai Lu <[email protected]>--- include/linux/lmb.h | 5 ++++- lib/lmb.c | 11 ++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-)Index:
linux-2.6/include/linux/lmb.h
===================================================================include/linux/lmb.h +2
-1@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };struct lmb {
include/linux/lmb.h struct lmb { · line 42
Quoted 5 lines
include/linux/lmb.h +2
-0@@ -37,6 +38,8 @@ struct lmb { };extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];
Is there a reason why this needs to be exported ?
Thanks,
tglx
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 33 lines · Thomas Gleixner
Quoted 31 lines · Yinghai Lu
lmb_init()connect them back.also add
region_array_sizeinlmb_regionto tack the region array size.-v3: seperate lmb core change to seperated patch
Signed-off-by Yinghai Lu <[email protected]>--- include/linux/lmb.h | 5 ++++- lib/lmb.c | 11 ++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-)Index:
linux-2.6/include/linux/lmb.h
===================================================================include/linux/lmb.h +2
-1@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };struct lmb {@@ -37,6 +38,8 @@ struct lmb { };
Quoted 6 lines
include/linux/lmb.h struct lmb { · line 42
Quoted 3 lines
include/linux/lmb.h +2
-0extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];Is there a reason why this needs to be exported ?
lmb_reserved_region is used in kernel/early_res.c::__check_and_double_early_res/subtract_early_res/early_res_to_bootmem
later if those functions are moved lmb.c, we can make lmb_reserved_region to be static
same thing to lmb_memory_region, that will be used by add_lmb_memory() later.
YH
Re: [RFC PATCH -v3 1/2] lmb: seperate region array from lmb_region struct
Quoted 41 lines · Yinghai Lu
Quoted 33 lines · Thomas Gleixner
Quoted 31 lines · Yinghai Lu
lmb_init()connect them back.also add
region_array_sizeinlmb_regionto tack the region array size.-v3: seperate lmb core change to seperated patch
Signed-off-by Yinghai Lu <[email protected]>--- include/linux/lmb.h | 5 ++++- lib/lmb.c | 11 ++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-)Index:
linux-2.6/include/linux/lmb.h
===================================================================include/linux/lmb.h +2
-1@@ -26,7 +26,8 @@ struct lmb_property {struct lmb_region {unsigned long cnt;u64 size; -struct lmb_property region[MAX_LMB_REGIONS +1 ]; +struct lmb_property *region; +unsigned long region_array_size; };struct lmb {@@ -37,6 +38,8 @@ struct lmb { };Quoted 6 lines
include/linux/lmb.h struct lmb { · line 42
Quoted 3 lines
include/linux/lmb.h +2
-0extern struct lmb lmb; +extern struct lmb_property lmb_memory_region[MAX_LMB_REGIONS +1 ]; +extern struct lmb_property lmb_reserved_region[MAX_LMB_REGIONS +1 ];Is there a reason why this needs to be exported ?
Quoted 8 lines
lmb_reserved_regionis used in kernel/early_res.c::__check_and_double_early_res/subtract_early_res/early_res_to_bootmemlater if those functions are moved
lmb.c, we can makelmb_reserved_regionto be staticsame thing to
lmb_memory_region, that will be used byadd_lmb_memory()later.
I would strongly suggest for you to send the next (-v4) version to lkml only if it's 100% clean and complete in every known regard (with a good splitup and explanations in every commit), so that people dont repeat review or review an unclean aspect aspect that will go away in a next iteration.
Thanks,
Ingo
RFC PATCH 2/2 x86: use lmb to replace early_res
In reply to Yinghai Lustill keep kernel/early_res.c for the extension.
should move those file to lib/lmb.c later?
in early_res.c
1. change find_e820_area_xxx, to find_lmb_area_xxx
2. e820_register_active_regions to lmb_register_active_regions.
3. reserve_early will call lmb_reserve directly.
4. free_early will call lmb_free directly.
5. remove functions that are used by old reserve_early and free_early
6. get_free_all_memory_range use lmb.reserved.
7. early_res_to_bootmem use lmb.reserved
8. add fill_lmb_memory() to fill lmb.memory according e820 RAM entries
-v2: fix NO_BOOTMEM hang with printk
--- arch/x86/Kconfig | 1 arch/x86/include/asm/e820.h | 38 +- arch/x86/include/asm/lmb.h | 8 arch/x86/kernel/e820.c | 163 +---------- arch/x86/kernel/head.c | 2 arch/x86/kernel/head32.c | 5 arch/x86/kernel/head64.c | 2 arch/x86/kernel/setup.c | 2 arch/x86/kernel/setup_percpu.c | 6 include/linux/early_res.h | 9 kernel/early_res.c | 592 +++++++++++++++-------------------------- mm/page_alloc.c | 2 mm/sparse-vmemmap.c | 4 13 files changed, 301 insertions(+), 533 deletions(-)
arch/x86/Kconfig +1 -0
@@ -27,6 +27,7 @@ config X86 select HAVE_PERF_EVENTS if (!M386 && !M486) select HAVE_IOREMAP_PROT select HAVE_KPROBES + select HAVE_LMB select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS
arch/x86/include/asm/e820.h +26 -12
@@ -113,22 +113,36 @@ static inline void early_memtest(unsigneextern unsigned long end_user_pfn; -extern u64 find_e820_area (u64 start,u64 end,u64 size,u64 align); -extern u64 find_e820_area_size (u64 start,u64 *sizep,u64 align); -u64 find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align); -extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align);#include <linux/early_res.h> +static inline u64 find_e820_area (u64 start,u64 end,u64 size,u64 align) +{ +return find_lmb_area (start, end, size, align); +} +static inline u64 find_e820_area_size (u64 start,u64 *sizep,u64 align) +{ +return find_lmb_area_size (start, sizep, align); +} +static inline u64 +find_e820_area_node (int nid,u64 start,u64 end,u64 size,u64 align) +{ +return find_lmb_area_node (nid, start, end, size, align); +} +extern u64 early_reserve_e820 (u64 startt,u64 sizet,u64 align);extern unsigned long e820_end_of_ram_pfn (void );extern unsigned long e820_end_of_low_ram_pfn (void ); -extern int e820_find_active_region (const struct e820entry *ei, -unsigned long start_pfn, -unsigned long last_pfn, -unsigned long *ei_startpfn, -unsigned long *ei_endpfn); -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); +static inline void e820_register_active_regions (int nid, +unsigned long start_pfn, +unsigned long end_pfn) +{ +lmb_register_active_regions (nid, start_pfn, end_pfn); +} +static inline u64 e820_hole_size (u64 start,u64 end) +{ +return lmb_hole_size (start, end); +} +void init_lmb_memory (void ); +void fill_lmb_memory (void );extern void finish_e820_parsing (void );extern void e820_reserve_resources (void );extern void e820_reserve_resources_late (void );
arch/x86/kernel/e820.c +23 -140
193 lines not on this page · raw message
arch/x86/kernel/head.c +1 -1
@@ -51,5 +51,5 @@ void __init reserve_ebda_region(void) lowmem =0x9f000 ;/* reserve all memory between lowmem and the 1MB mark */ -reserve_early_overlap_ok (lowmem,0x100000 ,"BIOS reserved" ); +reserve_early (lowmem,0x100000 ,"BIOS reserved" ); }
arch/x86/kernel/head32.c +3 -2
@@ -29,14 +29,15 @@ static void __init i386_default_early_sevoid __init i386_start_kernel (void ) { +init_lmb_memory (); +#ifdef CONFIG_X86_TRAMPOLINE /* * But first pinch a few for the stack/trampoline stuff * FIXME: Don't need the extra page at 4K, but need to fix * trampoline before removing it. (see the GDT stuff) */ -reserve_early_overlap_ok (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE , -"EX TRAMPOLINE" ); +reserve_early (PAGE_SIZE ,PAGE_SIZE +PAGE_SIZE ,"EX TRAMPOLINE" );#endif reserve_early (__pa_symbol (&_text),__pa_symbol (&__bss_stop),"TEXT DATA BSS" );
arch/x86/kernel/head64.c +2 -0
@@ -96,6 +96,8 @@ void __init x86_64_start_kernel(char * rvoid __init x86_64_start_reservations (char *real_mode_data) { +init_lmb_memory (); +copy_bootdata (__va (real_mode_data));reserve_early (__pa_symbol (&_text),__pa_symbol (&__bss_stop),"TEXT DATA BSS" );
arch/x86/kernel/setup.c +2 -0
@@ -894,6 +894,8 @@ void __init setup_arch(char **cmdline_p) max_pfn_mapped =KERNEL_IMAGE_SIZE >>PAGE_SHIFT ;#endif +fill_lmb_memory (); +#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION setup_bios_corruption_check ();#endif
arch/x86/kernel/setup_percpu.c +0 -6
@@ -137,13 +137,7 @@ static void * __init pcpu_fc_alloc(unsigstatic void __init pcpu_fc_free (void *ptr,size_t size) { -#ifdef CONFIG_NO_BOOTMEM -u64 start =__pa (ptr); -u64 end = start + size; -free_early_partial (start, end); -#else free_bootmem (__pa (ptr), size); -#endif }static int __init pcpu_cpu_distance (unsigned int from,unsigned int to)
include/linux/early_res.h +6 -3
@@ -5,15 +5,18 @@extern void reserve_early (u64 start,u64 end,char *name);extern void reserve_early_overlap_ok (u64 start,u64 end,char *name);extern void free_early (u64 start,u64 end); -void free_early_partial (u64 start,u64 end);extern void early_res_to_bootmem (u64 start,u64 end); -void reserve_early_without_check (u64 start,u64 end,char *name);u64 find_early_area (u64 ei_start,u64 ei_last,u64 start,u64 end,u64 size,u64 align);u64 find_early_area_size (u64 ei_start,u64 ei_last,u64 start,u64 *sizep,u64 align); -u64 find_fw_memmap_area (u64 start,u64 end,u64 size,u64 align); +u64 find_lmb_area (u64 start,u64 end,u64 size,u64 align); +u64 find_lmb_area_size (u64 start,u64 *sizep,u64 align); +u64 find_lmb_area_node (int nid,u64 start,u64 end,u64 size,u64 align); +void lmb_register_active_regions (int nid,unsigned long start_pfn, +unsigned long last_pfn); +u64 lmb_hole_size (u64 start,u64 end);u64 get_max_mapped (void );#include <linux/range.h> int get_free_all_memory_range (struct range **rangep,int nodeid);
kernel/early_res.c +226 -366
737 lines not on this page · raw message
mm/page_alloc.c +1 -1
@@ -3451,7 +3451,7 @@ void * __init __alloc_memory_core_early( ptr =phys_to_virt (addr);memset (ptr,0 , size); -reserve_early_without_check (addr, addr + size,"BOOTMEM" ); +reserve_early (addr, addr + size,"BOOTMEM" ); return ptr; }#endif
mm/sparse-vmemmap.c +2 -2
@@ -225,8 +225,8 @@ void __init sparse_mem_maps_populate_nodchar name[15 ];snprintf (name,sizeof (name),"MEMMAP %d" , nodeid); -reserve_early_without_check (__pa (vmemmap_buf_start), -__pa (vmemmap_buf), name); +reserve_early (__pa (vmemmap_buf_start), +__pa (vmemmap_buf), name); }#else free_bootmem (__pa (vmemmap_buf), vmemmap_buf_end - vmemmap_buf);
arch/x86/include/asm/lmb.h new file +8 -0
@@ -0,0 +1,8 @@ +#ifndef _X86_LMB_H +#define _X86_LMB_H + +#define LMB_DBG(fmt...) printk(fmt) + +#define LMB_REAL_LIMIT 0 + +#endif