From mboxrd@z Thu Jan 1 00:00:00 1970 From: jbarnes@sgi.com (Jesse Barnes) Date: Tue, 23 Sep 2003 17:07:52 +0000 Subject: Re: [PATCH] split contig and discontig functions Message-Id: List-Id: References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org Oops, I forgot to make meminit.h protect against multiple inclusion and there are some other minor formatting stuff that I missed. I'll post another patch shortly. Jesse On Mon, Sep 22, 2003 at 06:28:23PM -0700, Jesse Barnes wrote: > This patch mostly shuffles around code and creates two new files: > contig.c and meminit.h. The former contains functions contiguous and > virtually contiguous machines while the latter has prototypes and > declarations for stuff used in contig.c, discontig.c, setup.c, etc. I > put meminit.h in include/asm-ia64 since it's used by files in both > arch/ia64/kernel and arch/ia64/mm. > > Oddly, CONFIG_DISCONTIGMEM wasn't compiling at all :). That's now fixed > under this scheme, though the incompatibility between > CONFIG_DISCONTIGMEM and CONFIG_VIRTUAL_MEMMAP remains. I'll take care > of that with another patch later. > > Also, a few people have mailed me with suggestions on how to deal with > the ifndef CONFIG_DISCONTIGMEM in arch/ia64/mm/Makefile differently, but > none of them have worked so far and what's there seems pretty clear to > me. > > arch/ia64/kernel/setup.c | 174 ++++----------------------------------------- > arch/ia64/mm/Makefile | 3 > arch/ia64/mm/contig.c | 157 ++++++++++++++++++++++++++++++++++++++++ > arch/ia64/mm/discontig.c | 83 +++++++++++++++++++-- > arch/ia64/mm/init.c | 52 ------------- > include/asm-ia64/meminit.h | 38 +++++++++ > include/asm-ia64/numa.h | 3 > > Thanks, > Jesse > > > diff -Nru a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c > --- a/arch/ia64/kernel/setup.c Mon Sep 22 18:23:36 2003 > +++ b/arch/ia64/kernel/setup.c Mon Sep 22 18:23:36 2003 > @@ -45,6 +45,7 @@ > #include > #include > #include > +#include > > #if defined(CONFIG_SMP) && (IA64_CPU_SIZE > PAGE_SIZE) > # error "struct cpuinfo_ia64 too big!" > @@ -84,90 +85,11 @@ > char saved_command_line[COMMAND_LINE_SIZE]; /* used in proc filesystem */ > > /* > - * Entries defined so far: > - * - boot param structure itself > - * - memory map > - * - initrd (optional) > - * - command line string > - * - kernel code & data > - * > - * More could be added if necessary > - */ > -#define IA64_MAX_RSVD_REGIONS 5 > - > -struct rsvd_region { > - unsigned long start; /* virtual address of beginning of element */ > - unsigned long end; /* virtual address of end of element + 1 */ > -}; > - > -/* > * We use a special marker for the end of memory and it uses the extra (+1) slot > */ > -static struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1]; > -static int num_rsvd_regions; > +struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1]; > +int num_rsvd_regions; > > -#define IGNORE_PFN0 1 /* XXX fix me: ignore pfn 0 until TLB miss handler is updated... */ > - > -#ifndef CONFIG_DISCONTIGMEM > - > -static unsigned long bootmap_start; /* physical address where the bootmem map is located */ > - > -static int > -find_max_pfn (unsigned long start, unsigned long end, void *arg) > -{ > - unsigned long *max_pfnp = arg, pfn; > - > - pfn = (PAGE_ALIGN(end - 1) - PAGE_OFFSET) >> PAGE_SHIFT; > - if (pfn > *max_pfnp) > - *max_pfnp = pfn; > - return 0; > -} > - > -#else /* CONFIG_DISCONTIGMEM */ > - > -/* > - * efi_memmap_walk() knows nothing about layout of memory across nodes. Find > - * out to which node a block of memory belongs. Ignore memory that we cannot > - * identify, and split blocks that run across multiple nodes. > - * > - * Take this opportunity to round the start address up and the end address > - * down to page boundaries. > - */ > -void > -call_pernode_memory (unsigned long start, unsigned long end, void *arg) > -{ > - unsigned long rs, re; > - void (*func)(unsigned long, unsigned long, int, int); > - int i; > - > - start = PAGE_ALIGN(start); > - end &= PAGE_MASK; > - if (start >= end) > - return; > - > - func = arg; > - > - if (!num_memblks) { > - /* > - * This machine doesn't have SRAT, so call func with > - * nid=0, bank=0. > - */ > - if (start < end) > - (*func)(start, end - start, 0, 0); > - return; > - } > - > - for (i = 0; i < num_memblks; i++) { > - rs = max(start, node_memblk[i].start_paddr); > - re = min(end, node_memblk[i].start_paddr+node_memblk[i].size); > - > - if (rs < re) > - (*func)(rs, re-rs, node_memblk[i].nid, > - node_memblk[i].bank); > - } > -} > - > -#endif /* CONFIG_DISCONTIGMEM */ > > /* > * Filter incoming memory segments based on the primitive map created from the boot > @@ -215,48 +137,6 @@ > return 0; > } > > - > -#ifndef CONFIG_DISCONTIGMEM > -/* > - * Find a place to put the bootmap and return its starting address in bootmap_start. > - * This address must be page-aligned. > - */ > -static int > -find_bootmap_location (unsigned long start, unsigned long end, void *arg) > -{ > - unsigned long needed = *(unsigned long *)arg; > - unsigned long range_start, range_end, free_start; > - int i; > - > -#if IGNORE_PFN0 > - if (start = PAGE_OFFSET) { > - start += PAGE_SIZE; > - if (start >= end) return 0; > - } > -#endif > - > - free_start = PAGE_OFFSET; > - > - for (i = 0; i < num_rsvd_regions; i++) { > - range_start = max(start, free_start); > - range_end = min(end, rsvd_region[i].start & PAGE_MASK); > - > - if (range_end <= range_start) continue; /* skip over empty range */ > - > - if (range_end - range_start >= needed) { > - bootmap_start = __pa(range_start); > - return 1; /* done */ > - } > - > - /* nothing more available in this segment */ > - if (range_end = end) return 0; > - > - free_start = PAGE_ALIGN(rsvd_region[i].end); > - } > - return 0; > -} > -#endif /* !CONFIG_DISCONTIGMEM */ > - > static void > sort_regions (struct rsvd_region *rsvd_region, int max) > { > @@ -275,10 +155,14 @@ > } > } > > -static void > -find_memory (void) > +/** > + * reserve_memory - setup reserved memory areas > + * > + * Setup the reserved memory areas set aside for the boot parameters, initrd, > + * etc. > + */ > +void reserve_memory (void) > { > - unsigned long bootmap_size; > int n = 0; > > /* > @@ -317,36 +201,16 @@ > num_rsvd_regions = n; > > sort_regions(rsvd_region, num_rsvd_regions); > +} > > -#ifdef CONFIG_DISCONTIGMEM > - { > - extern void discontig_mem_init (void); > - > - bootmap_size = max_pfn = 0; /* stop gcc warnings */ > - discontig_mem_init(); > - } > -#else /* !CONFIG_DISCONTIGMEM */ > - > - /* first find highest page frame number */ > - max_pfn = 0; > - efi_memmap_walk(find_max_pfn, &max_pfn); > - > - /* how many bytes to cover all the pages */ > - bootmap_size = bootmem_bootmap_pages(max_pfn) << PAGE_SHIFT; > - > - /* look for a location to hold the bootmap */ > - bootmap_start = ~0UL; > - efi_memmap_walk(find_bootmap_location, &bootmap_size); > - if (bootmap_start = ~0UL) > - panic("Cannot find %ld bytes for bootmap\n", bootmap_size); > - > - bootmap_size = init_bootmem(bootmap_start >> PAGE_SHIFT, max_pfn); > - > - /* Free all available memory, then mark bootmem-map as being in use. */ > - efi_memmap_walk(filter_rsvd_memory, free_bootmem); > - reserve_bootmem(bootmap_start, bootmap_size); > -#endif /* !CONFIG_DISCONTIGMEM */ > - > +/** > + * find_initrd - get initrd parameters from the boot parameter structure > + * > + * Grab the initrd start and end from the boot parameter struct given us by > + * the boot loader. > + */ > +void find_initrd(void) > +{ > #ifdef CONFIG_BLK_DEV_INITRD > if (ia64_boot_param->initrd_start) { > initrd_start = (unsigned long)__va(ia64_boot_param->initrd_start); > diff -Nru a/arch/ia64/mm/Makefile b/arch/ia64/mm/Makefile > --- a/arch/ia64/mm/Makefile Mon Sep 22 18:23:36 2003 > +++ b/arch/ia64/mm/Makefile Mon Sep 22 18:23:36 2003 > @@ -7,3 +7,6 @@ > obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o > obj-$(CONFIG_NUMA) += numa.o > obj-$(CONFIG_DISCONTIGMEM) += discontig.o > +ifndef CONFIG_DISCONTIGMEM > +obj-y += contig.o > +endif > diff -Nru a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c > --- /dev/null Wed Dec 31 16:00:00 1969 > +++ b/arch/ia64/mm/contig.c Mon Sep 22 18:23:36 2003 > @@ -0,0 +1,157 @@ > +/* > + * This file is subject to the terms and conditions of the GNU General Public > + * License. See the file "COPYING" in the main directory of this archive > + * for more details. > + * > + * Copyright (C) 1998-2003 Hewlett-Packard Co > + * David Mosberger-Tang > + * Stephane Eranian > + * Copyright (C) 2000, Rohit Seth > + * Copyright (C) 1999 VA Linux Systems > + * Copyright (C) 1999 Walt Drummond > + * Copyright (C) 2003 Silicon Graphics, Inc. All rights reserved. > + * > + * Routines used by ia64 machines with contiguous (or virtually contiguous) > + * memory. > + */ > +#include > +#include > +#include > +#include > +#include > +#include > +#include > +#include > +#include > + > +/** > + * show_mem - display a memory statistics summary > + * > + * Just walks the pages in the system and describes where they're allocated. > + */ > +void show_mem(void) > +{ > + int i, total = 0, reserved = 0; > + int shared = 0, cached = 0; > + > + printk("Mem-info:\n"); > + show_free_areas(); > + > + printk("Free swap: %6dkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); > + i = max_mapnr; > + while (i-- > 0) { > + total++; > + if (PageReserved(mem_map+i)) > + reserved++; > + else if (PageSwapCache(mem_map+i)) > + cached++; > + else if (page_count(mem_map + i)) > + shared += page_count(mem_map + i) - 1; > + } > + printk("%d pages of RAM\n", total); > + printk("%d reserved pages\n", reserved); > + printk("%d pages shared\n", shared); > + printk("%d pages swap cached\n", cached); > + printk("%ld pages in page table cache\n", pgtable_cache_size); > +} > + > +/* physical address where the bootmem map is located */ > +unsigned long bootmap_start; > + > +/** > + * find_max_pfn - adjust the maximum page number callback > + * @start: start of range > + * @end: end of range > + * @arg: address of pointer to global max_pfn variable > + * > + * Passed as a callback function to efi_memmap_walk() to determine the highest > + * available page frame number in the system. > + */ > +int find_max_pfn(unsigned long start, unsigned long end, void *arg) > +{ > + unsigned long *max_pfnp = arg, pfn; > + > + pfn = (PAGE_ALIGN(end - 1) - PAGE_OFFSET) >> PAGE_SHIFT; > + if (pfn > *max_pfnp) > + *max_pfnp = pfn; > + return 0; > +} > + > +/** > + * find_bootmap_location - callback to find a memory area for the bootmap > + * @start: start of region > + * @end: end of region > + * @arg: unused callback data > + * > + * Find a place to put the bootmap and return its starting address in > + * bootmap_start. This address must be page-aligned. > + */ > +int find_bootmap_location(unsigned long start, unsigned long end, void *arg) > +{ > + unsigned long needed = *(unsigned long *)arg; > + unsigned long range_start, range_end, free_start; > + int i; > + > +#if IGNORE_PFN0 > + if (start = PAGE_OFFSET) { > + start += PAGE_SIZE; > + if (start >= end) return 0; > + } > +#endif > + > + free_start = PAGE_OFFSET; > + > + for (i = 0; i < num_rsvd_regions; i++) { > + range_start = max(start, free_start); > + range_end = min(end, rsvd_region[i].start & PAGE_MASK); > + > + if (range_end <= range_start) > + continue; /* skip over empty range */ > + > + if (range_end - range_start >= needed) { > + bootmap_start = __pa(range_start); > + return 1; /* done */ > + } > + > + /* nothing more available in this segment */ > + if (range_end = end) > + return 0; > + > + free_start = PAGE_ALIGN(rsvd_region[i].end); > + } > + return 0; > +} > + > +/** > + * find_memory - setup memory map > + * > + * Walk the EFI memory map and find usable memory for the system, taking > + * into account reserved areas. > + */ > +void find_memory(void) > +{ > + unsigned long bootmap_size; > + > + reserve_memory(); > + > + /* first find highest page frame number */ > + max_pfn = 0; > + efi_memmap_walk(find_max_pfn, &max_pfn); > + > + /* how many bytes to cover all the pages */ > + bootmap_size = bootmem_bootmap_pages(max_pfn) << PAGE_SHIFT; > + > + /* look for a location to hold the bootmap */ > + bootmap_start = ~0UL; > + efi_memmap_walk(find_bootmap_location, &bootmap_size); > + if (bootmap_start = ~0UL) > + panic("Cannot find %ld bytes for bootmap\n", bootmap_size); > + > + bootmap_size = init_bootmem(bootmap_start >> PAGE_SHIFT, max_pfn); > + > + /* Free all available memory, then mark bootmem-map as being in use. */ > + efi_memmap_walk(filter_rsvd_memory, free_bootmem); > + reserve_bootmem(bootmap_start, bootmap_size); > + > + find_initrd(); > +} > diff -Nru a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c > --- a/arch/ia64/mm/discontig.c Mon Sep 22 18:23:36 2003 > +++ b/arch/ia64/mm/discontig.c Mon Sep 22 18:23:36 2003 > @@ -1,5 +1,5 @@ > /* > - * Copyright (c) 2000 Silicon Graphics, Inc. All rights reserved. > + * Copyright (c) 2000, 2003 Silicon Graphics, Inc. All rights reserved. > * Copyright (c) 2001 Intel Corp. > * Copyright (c) 2001 Tony Luck > * Copyright (c) 2002 NEC Corp. > @@ -12,10 +12,13 @@ > > #include > #include > +#include > #include > #include > #include > #include > +#include > +#include > > > /* > @@ -27,9 +30,6 @@ > static long boot_pg_data[8*NR_NODES+sizeof(pg_data_t)] __initdata; > static pg_data_t *pg_data_ptr[NR_NODES] __initdata; > static bootmem_data_t bdata[NR_NODES][NR_BANKS_PER_NODE+1] __initdata; > - > -extern int filter_rsvd_memory (unsigned long start, unsigned long end, void *arg); > - > /* > * Return the compact node number of this cpu. Used prior to > * setting up the cpu_data area. > @@ -209,11 +209,12 @@ > * Called early in boot to setup the boot memory allocator, and to > * allocate the node-local pg_data & node-directory data structures.. > */ > -void __init > -discontig_mem_init(void) > +void __init find_memory(void) > { > int node; > > + reserve_memory(); > + > if (numnodes = 0) { > printk(KERN_ERR "node info missing!\n"); > numnodes = 1; > @@ -232,6 +233,8 @@ > efi_memmap_walk(filter_rsvd_memory, discontig_free_bootmem_node); > discontig_reserve_bootmem(); > allocate_pernode_structures(); > + > + find_initrd(); > } > > /* > @@ -305,3 +308,71 @@ > } > } > > +void show_mem(void) > +{ > + int i, reserved = 0; > + int shared = 0, cached = 0; > + pg_data_t *pgdat; > + > + printk("Mem-info:\n"); > + show_free_areas(); > + printk("Free swap: %6dkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); > + for_each_pgdat(pgdat) { > + printk("Node ID: %d\n", pgdat->node_id); > + for(i = 0; i < pgdat->node_spanned_pages; i++) { > + if (PageReserved(pgdat->node_mem_map+i)) > + reserved++; > + else if (PageSwapCache(pgdat->node_mem_map+i)) > + cached++; > + else if (page_count(pgdat->node_mem_map+i)) > + shared += page_count(pgdat->node_mem_map+i)-1; > + } > + printk("\t%ld pages of RAM\n", pgdat->node_present_pages); > + printk("\t%d reserved pages\n", reserved); > + printk("\t%d pages shared\n", shared); > + printk("\t%d pages swap cached\n", cached); > + } > + printk("Total of %ld pages in page table cache\n", pgtable_cache_size); > + printk("%d free buffer pages\n", nr_free_buffer_pages()); > +} > + > +/* > + * efi_memmap_walk() knows nothing about layout of memory across nodes. Find > + * out to which node a block of memory belongs. Ignore memory that we cannot > + * identify, and split blocks that run across multiple nodes. > + * > + * Take this opportunity to round the start address up and the end address > + * down to page boundaries. > + */ > +void call_pernode_memory(unsigned long start, unsigned long end, void *arg) > +{ > + unsigned long rs, re; > + void (*func)(unsigned long, unsigned long, int, int); > + int i; > + > + start = PAGE_ALIGN(start); > + end &= PAGE_MASK; > + if (start >= end) > + return; > + > + func = arg; > + > + if (!num_memblks) { > + /* > + * This machine doesn't have SRAT, so call func with > + * nid=0, bank=0. > + */ > + if (start < end) > + (*func)(start, end - start, 0, 0); > + return; > + } > + > + for (i = 0; i < num_memblks; i++) { > + rs = max(start, node_memblk[i].start_paddr); > + re = min(end, node_memblk[i].start_paddr+node_memblk[i].size); > + > + if (rs < re) > + (*func)(rs, re-rs, node_memblk[i].nid, > + node_memblk[i].bank); > + } > +} > diff -Nru a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c > --- a/arch/ia64/mm/init.c Mon Sep 22 18:23:36 2003 > +++ b/arch/ia64/mm/init.c Mon Sep 22 18:23:36 2003 > @@ -214,58 +214,6 @@ > } > } > > -void > -show_mem(void) > -{ > - int i, total = 0, reserved = 0; > - int shared = 0, cached = 0; > - > - printk("Mem-info:\n"); > - show_free_areas(); > - > -#ifdef CONFIG_DISCONTIGMEM > - { > - pg_data_t *pgdat; > - > - printk("Free swap: %6dkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); > - for_each_pgdat(pgdat) { > - printk("Node ID: %d\n", pgdat->node_id); > - for(i = 0; i < pgdat->node_spanned_pages; i++) { > - if (PageReserved(pgdat->node_mem_map+i)) > - reserved++; > - else if (PageSwapCache(pgdat->node_mem_map+i)) > - cached++; > - else if (page_count(pgdat->node_mem_map + i)) > - shared += page_count(pgdat->node_mem_map + i) - 1; > - } > - printk("\t%d pages of RAM\n", pgdat->node_spanned_pages); > - printk("\t%d reserved pages\n", reserved); > - printk("\t%d pages shared\n", shared); > - printk("\t%d pages swap cached\n", cached); > - } > - printk("Total of %ld pages in page table cache\n", pgtable_cache_size); > - printk("%d free buffer pages\n", nr_free_buffer_pages()); > - } > -#else /* !CONFIG_DISCONTIGMEM */ > - printk("Free swap: %6dkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); > - i = max_mapnr; > - while (i-- > 0) { > - total++; > - if (PageReserved(mem_map+i)) > - reserved++; > - else if (PageSwapCache(mem_map+i)) > - cached++; > - else if (page_count(mem_map + i)) > - shared += page_count(mem_map + i) - 1; > - } > - printk("%d pages of RAM\n", total); > - printk("%d reserved pages\n", reserved); > - printk("%d pages shared\n", shared); > - printk("%d pages swap cached\n", cached); > - printk("%ld pages in page table cache\n", pgtable_cache_size); > -#endif /* !CONFIG_DISCONTIGMEM */ > -} > - > /* > * This is like put_dirty_page() but installs a clean page in the kernel's page table. > */ > diff -Nru a/include/asm-ia64/meminit.h b/include/asm-ia64/meminit.h > --- /dev/null Wed Dec 31 16:00:00 1969 > +++ b/include/asm-ia64/meminit.h Mon Sep 22 18:23:36 2003 > @@ -0,0 +1,38 @@ > +/* > + * This file is subject to the terms and conditions of the GNU General Public > + * License. See the file "COPYING" in the main directory of this archive > + * for more details. > + */ > + > +/* > + * Entries defined so far: > + * - boot param structure itself > + * - memory map > + * - initrd (optional) > + * - command line string > + * - kernel code & data > + * > + * More could be added if necessary > + */ > +#define IA64_MAX_RSVD_REGIONS 5 > + > +struct rsvd_region { > + unsigned long start; /* virtual address of beginning of element */ > + unsigned long end; /* virtual address of end of element + 1 */ > +}; > + > +extern struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1]; > +extern int num_rsvd_regions; > + > +extern void find_memory(void); > +extern void reserve_memory(void); > +extern void find_initrd(void); > +extern int filter_rsvd_memory(unsigned long start, unsigned long end, > + void *arg); > + > +#ifdef CONFIG_DISCONTIGMEM > +extern void call_pernode_memory(unsigned long start, unsigned long end, void *arg); > +#endif > + > +#define IGNORE_PFN0 1 /* XXX fix me: ignore pfn 0 until TLB miss > + handler is updated... */ > diff -Nru a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h > --- a/include/asm-ia64/numa.h Mon Sep 22 18:23:36 2003 > +++ b/include/asm-ia64/numa.h Mon Sep 22 18:23:36 2003 > @@ -11,6 +11,9 @@ > #ifndef _ASM_IA64_NUMA_H > #define _ASM_IA64_NUMA_H > > +#include > +#include > + > #ifdef CONFIG_NUMA > > #ifdef CONFIG_DISCONTIGMEM > - > To unsubscribe from this list: send the line "unsubscribe linux-ia64" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html