X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=arch%2Fpowerpc%2Fmm%2Fnuma.c;h=7393bd76d698406f85a70b423113c8e163afacae;hb=7021d86afa6f3a8bf76218ac97f5847a6d985730;hp=39328da32edf0ee0948160cb4393d09b3f491c46;hpb=8342681d3e1f5fbd404c21b5e10f87277411f3eb;p=safe%2Fjmp%2Flinux-2.6 diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index 39328da..7393bd7 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -39,7 +39,6 @@ EXPORT_SYMBOL(numa_cpu_lookup_table); EXPORT_SYMBOL(numa_cpumask_lookup_table); EXPORT_SYMBOL(node_data); -static bootmem_data_t __initdata plat_node_bdata[MAX_NUMNODES]; static int min_common_depth; static int n_mem_addr_cells, n_mem_size_cells; @@ -90,6 +89,48 @@ static int __cpuinit fake_numa_create_new_node(unsigned long end_pfn, return 0; } +/* + * get_active_region_work_fn - A helper function for get_node_active_region + * Returns datax set to the start_pfn and end_pfn if they contain + * the initial value of datax->start_pfn between them + * @start_pfn: start page(inclusive) of region to check + * @end_pfn: end page(exclusive) of region to check + * @datax: comes in with ->start_pfn set to value to search for and + * goes out with active range if it contains it + * Returns 1 if search value is in range else 0 + */ +static int __init get_active_region_work_fn(unsigned long start_pfn, + unsigned long end_pfn, void *datax) +{ + struct node_active_region *data; + data = (struct node_active_region *)datax; + + if (start_pfn <= data->start_pfn && end_pfn > data->start_pfn) { + data->start_pfn = start_pfn; + data->end_pfn = end_pfn; + return 1; + } + return 0; + +} + +/* + * get_node_active_region - Return active region containing start_pfn + * Active range returned is empty if none found. + * @start_pfn: The page to return the region for. + * @node_ar: Returned set to the active region containing start_pfn + */ +static void __init get_node_active_region(unsigned long start_pfn, + struct node_active_region *node_ar) +{ + int nid = early_pfn_to_nid(start_pfn); + + node_ar->nid = nid; + node_ar->start_pfn = start_pfn; + node_ar->end_pfn = start_pfn; + work_with_active_regions(nid, get_active_region_work_fn, node_ar); +} + static void __cpuinit map_cpu_to_node(int cpu, int node) { numa_cpu_lookup_table[cpu] = node; @@ -151,6 +192,21 @@ static const int *of_get_associativity(struct device_node *dev) return of_get_property(dev, "ibm,associativity", NULL); } +/* + * Returns the property linux,drconf-usable-memory if + * it exists (the property exists only in kexec/kdump kernels, + * added by kexec-tools) + */ +static const u32 *of_get_usable_memory(struct device_node *memory) +{ + const u32 *prop; + u32 len; + prop = of_get_property(memory, "linux,drconf-usable-memory", &len); + if (!prop || len < sizeof(unsigned int)) + return 0; + return prop; +} + /* Returns nid in the range [0..MAX_NUMNODES-1], or -1 if no useful numa * info is found. */ @@ -472,12 +528,10 @@ static unsigned long __init numa_enforce_memory_limit(unsigned long start, /* * We use lmb_end_of_DRAM() in here instead of memory_limit because * we've already adjusted it for the limit and it takes care of - * having memory holes below the limit. + * having memory holes below the limit. Also, in the case of + * iommu_is_off, memory_limit is not set but is implicitly enforced. */ - if (! memory_limit) - return size; - if (start + size <= lmb_end_of_DRAM()) return size; @@ -488,14 +542,29 @@ static unsigned long __init numa_enforce_memory_limit(unsigned long start, } /* + * Reads the counter for a given entry in + * linux,drconf-usable-memory property + */ +static inline int __init read_usm_ranges(const u32 **usm) +{ + /* + * For each lmb in ibm,dynamic-memory a corresponding + * entry in linux,drconf-usable-memory property contains + * a counter followed by that many (base, size) duple. + * read the counter from linux,drconf-usable-memory + */ + return read_n_cells(n_mem_size_cells, usm); +} + +/* * Extract NUMA information from the ibm,dynamic-reconfiguration-memory * node. This assumes n_mem_{addr,size}_cells have been set. */ static void __init parse_drconf_memory(struct device_node *memory) { - const u32 *dm; - unsigned int n, rc; - unsigned long lmb_size, size; + const u32 *dm, *usm; + unsigned int n, rc, ranges, is_kexec_kdump = 0; + unsigned long lmb_size, base, size, sz; int nid; struct assoc_arrays aa; @@ -511,6 +580,11 @@ static void __init parse_drconf_memory(struct device_node *memory) if (rc) return; + /* check if this is a kexec/kdump kernel */ + usm = of_get_usable_memory(memory); + if (usm != NULL) + is_kexec_kdump = 1; + for (; n != 0; --n) { struct of_drconf_cell drmem; @@ -522,21 +596,31 @@ static void __init parse_drconf_memory(struct device_node *memory) || !(drmem.flags & DRCONF_MEM_ASSIGNED)) continue; - nid = of_drconf_to_nid_single(&drmem, &aa); + base = drmem.base_addr; + size = lmb_size; + ranges = 1; - fake_numa_create_new_node( - ((drmem.base_addr + lmb_size) >> PAGE_SHIFT), + if (is_kexec_kdump) { + ranges = read_usm_ranges(&usm); + if (!ranges) /* there are no (base, size) duple */ + continue; + } + do { + if (is_kexec_kdump) { + base = read_n_cells(n_mem_addr_cells, &usm); + size = read_n_cells(n_mem_size_cells, &usm); + } + nid = of_drconf_to_nid_single(&drmem, &aa); + fake_numa_create_new_node( + ((base + size) >> PAGE_SHIFT), &nid); - - node_set_online(nid); - - size = numa_enforce_memory_limit(drmem.base_addr, lmb_size); - if (!size) - continue; - - add_active_range(nid, drmem.base_addr >> PAGE_SHIFT, - (drmem.base_addr >> PAGE_SHIFT) - + (size >> PAGE_SHIFT)); + node_set_online(nid); + sz = numa_enforce_memory_limit(base, size); + if (sz) + add_active_range(nid, base >> PAGE_SHIFT, + (base >> PAGE_SHIFT) + + (sz >> PAGE_SHIFT)); + } while (--ranges); } } @@ -738,42 +822,50 @@ static void __init dump_numa_memory_topology(void) * required. nid is the preferred node and end is the physical address of * the highest address in the node. * - * Returns the physical address of the memory. + * Returns the virtual address of the memory. */ -static void __init *careful_allocation(int nid, unsigned long size, +static void __init *careful_zallocation(int nid, unsigned long size, unsigned long align, unsigned long end_pfn) { + void *ret; int new_nid; - unsigned long ret = __lmb_alloc_base(size, align, end_pfn << PAGE_SHIFT); + unsigned long ret_paddr; + + ret_paddr = __lmb_alloc_base(size, align, end_pfn << PAGE_SHIFT); /* retry over all memory */ - if (!ret) - ret = __lmb_alloc_base(size, align, lmb_end_of_DRAM()); + if (!ret_paddr) + ret_paddr = __lmb_alloc_base(size, align, lmb_end_of_DRAM()); - if (!ret) - panic("numa.c: cannot allocate %lu bytes on node %d", + if (!ret_paddr) + panic("numa.c: cannot allocate %lu bytes for node %d", size, nid); + ret = __va(ret_paddr); + /* - * If the memory came from a previously allocated node, we must - * retry with the bootmem allocator. + * We initialize the nodes in numeric order: 0, 1, 2... + * and hand over control from the LMB allocator to the + * bootmem allocator. If this function is called for + * node 5, then we know that all nodes <5 are using the + * bootmem allocator instead of the LMB allocator. + * + * So, check the nid from which this allocation came + * and double check to see if we need to use bootmem + * instead of the LMB. We don't free the LMB memory + * since it would be useless. */ - new_nid = early_pfn_to_nid(ret >> PAGE_SHIFT); + new_nid = early_pfn_to_nid(ret_paddr >> PAGE_SHIFT); if (new_nid < nid) { - ret = (unsigned long)__alloc_bootmem_node(NODE_DATA(new_nid), + ret = __alloc_bootmem_node(NODE_DATA(new_nid), size, align, 0); - if (!ret) - panic("numa.c: cannot allocate %lu bytes on node %d", - size, new_nid); - - ret = __pa(ret); - - dbg("alloc_bootmem %lx %lx\n", ret, size); + dbg("alloc_bootmem %p %lx\n", ret, size); } - return (void *)ret; + memset(ret, 0, size); + return ret; } static struct notifier_block __cpuinitdata ppc64_numa_nb = { @@ -781,10 +873,77 @@ static struct notifier_block __cpuinitdata ppc64_numa_nb = { .priority = 1 /* Must run before sched domains notifier. */ }; +static void mark_reserved_regions_for_nid(int nid) +{ + struct pglist_data *node = NODE_DATA(nid); + int i; + + for (i = 0; i < lmb.reserved.cnt; i++) { + unsigned long physbase = lmb.reserved.region[i].base; + unsigned long size = lmb.reserved.region[i].size; + unsigned long start_pfn = physbase >> PAGE_SHIFT; + unsigned long end_pfn = ((physbase + size) >> PAGE_SHIFT); + struct node_active_region node_ar; + unsigned long node_end_pfn = node->node_start_pfn + + node->node_spanned_pages; + + /* + * Check to make sure that this lmb.reserved area is + * within the bounds of the node that we care about. + * Checking the nid of the start and end points is not + * sufficient because the reserved area could span the + * entire node. + */ + if (end_pfn <= node->node_start_pfn || + start_pfn >= node_end_pfn) + continue; + + get_node_active_region(start_pfn, &node_ar); + while (start_pfn < end_pfn && + node_ar.start_pfn < node_ar.end_pfn) { + unsigned long reserve_size = size; + /* + * if reserved region extends past active region + * then trim size to active region + */ + if (end_pfn > node_ar.end_pfn) + reserve_size = (node_ar.end_pfn << PAGE_SHIFT) + - (start_pfn << PAGE_SHIFT); + /* + * Only worry about *this* node, others may not + * yet have valid NODE_DATA(). + */ + if (node_ar.nid == nid) { + dbg("reserve_bootmem %lx %lx nid=%d\n", + physbase, reserve_size, node_ar.nid); + reserve_bootmem_node(NODE_DATA(node_ar.nid), + physbase, reserve_size, + BOOTMEM_DEFAULT); + } + /* + * if reserved region is contained in the active region + * then done. + */ + if (end_pfn <= node_ar.end_pfn) + break; + + /* + * reserved region extends past the active region + * get next active region that contains this + * reserved region + */ + start_pfn = node_ar.end_pfn; + physbase = start_pfn << PAGE_SHIFT; + size = size - reserve_size; + get_node_active_region(start_pfn, &node_ar); + } + } +} + + void __init do_init_bootmem(void) { int nid; - unsigned int i; min_low_pfn = 0; max_low_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT; @@ -801,22 +960,26 @@ void __init do_init_bootmem(void) for_each_online_node(nid) { unsigned long start_pfn, end_pfn; - unsigned long bootmem_paddr; + void *bootmem_vaddr; unsigned long bootmap_pages; get_pfn_range_for_nid(nid, &start_pfn, &end_pfn); - /* Allocate the node structure node local if possible */ - NODE_DATA(nid) = careful_allocation(nid, + /* + * Allocate the node structure node local if possible + * + * Be careful moving this around, as it relies on all + * previous nodes' bootmem to be initialized and have + * all reserved areas marked. + */ + NODE_DATA(nid) = careful_zallocation(nid, sizeof(struct pglist_data), SMP_CACHE_BYTES, end_pfn); - NODE_DATA(nid) = __va(NODE_DATA(nid)); - memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); dbg("node %d\n", nid); dbg("NODE_DATA() = %p\n", NODE_DATA(nid)); - NODE_DATA(nid)->bdata = &plat_node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; @@ -827,47 +990,23 @@ void __init do_init_bootmem(void) dbg("end_paddr = %lx\n", end_pfn << PAGE_SHIFT); bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn); - bootmem_paddr = (unsigned long)careful_allocation(nid, + bootmem_vaddr = careful_zallocation(nid, bootmap_pages << PAGE_SHIFT, PAGE_SIZE, end_pfn); - memset(__va(bootmem_paddr), 0, bootmap_pages << PAGE_SHIFT); - dbg("bootmap_paddr = %lx\n", bootmem_paddr); + dbg("bootmap_vaddr = %p\n", bootmem_vaddr); - init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT, + init_bootmem_node(NODE_DATA(nid), + __pa(bootmem_vaddr) >> PAGE_SHIFT, start_pfn, end_pfn); free_bootmem_with_active_regions(nid, end_pfn); - - /* Mark reserved regions on this node */ - for (i = 0; i < lmb.reserved.cnt; i++) { - unsigned long physbase = lmb.reserved.region[i].base; - unsigned long size = lmb.reserved.region[i].size; - unsigned long start_paddr = start_pfn << PAGE_SHIFT; - unsigned long end_paddr = end_pfn << PAGE_SHIFT; - - if (early_pfn_to_nid(physbase >> PAGE_SHIFT) != nid && - early_pfn_to_nid((physbase+size-1) >> PAGE_SHIFT) != nid) - continue; - - if (physbase < end_paddr && - (physbase+size) > start_paddr) { - /* overlaps */ - if (physbase < start_paddr) { - size -= start_paddr - physbase; - physbase = start_paddr; - } - - if (size > end_paddr - physbase) - size = end_paddr - physbase; - - dbg("reserve_bootmem %lx %lx\n", physbase, - size); - reserve_bootmem_node(NODE_DATA(nid), physbase, - size, BOOTMEM_DEFAULT); - } - } - + /* + * Be very careful about moving this around. Future + * calls to careful_zallocation() depend on this getting + * done correctly. + */ + mark_reserved_regions_for_nid(nid); sparse_memory_present_with_active_regions(nid); } } @@ -901,6 +1040,79 @@ early_param("numa", early_numa); #ifdef CONFIG_MEMORY_HOTPLUG /* + * Validate the node associated with the memory section we are + * trying to add. + */ +int valid_hot_add_scn(int *nid, unsigned long start, u32 lmb_size, + unsigned long scn_addr) +{ + nodemask_t nodes; + + if (*nid < 0 || !node_online(*nid)) + *nid = any_online_node(NODE_MASK_ALL); + + if ((scn_addr >= start) && (scn_addr < (start + lmb_size))) { + nodes_setall(nodes); + while (NODE_DATA(*nid)->node_spanned_pages == 0) { + node_clear(*nid, nodes); + *nid = any_online_node(nodes); + } + + return 1; + } + + return 0; +} + +/* + * Find the node associated with a hot added memory section represented + * by the ibm,dynamic-reconfiguration-memory node. + */ +static int hot_add_drconf_scn_to_nid(struct device_node *memory, + unsigned long scn_addr) +{ + const u32 *dm; + unsigned int n, rc; + unsigned long lmb_size; + int default_nid = any_online_node(NODE_MASK_ALL); + int nid; + struct assoc_arrays aa; + + n = of_get_drconf_memory(memory, &dm); + if (!n) + return default_nid;; + + lmb_size = of_get_lmb_size(memory); + if (!lmb_size) + return default_nid; + + rc = of_get_assoc_arrays(memory, &aa); + if (rc) + return default_nid; + + for (; n != 0; --n) { + struct of_drconf_cell drmem; + + read_drconf_cell(&drmem, &dm); + + /* skip this block if it is reserved or not assigned to + * this partition */ + if ((drmem.flags & DRCONF_MEM_RESERVED) + || !(drmem.flags & DRCONF_MEM_ASSIGNED)) + continue; + + nid = of_drconf_to_nid_single(&drmem, &aa); + + if (valid_hot_add_scn(&nid, drmem.base_addr, lmb_size, + scn_addr)) + return nid; + } + + BUG(); /* section address should be found above */ + return 0; +} + +/* * Find the node associated with a hot added memory section. Section * corresponds to a SPARSEMEM section, not an LMB. It is assumed that * sections are fully contained within a single LMB. @@ -908,12 +1120,17 @@ early_param("numa", early_numa); int hot_add_scn_to_nid(unsigned long scn_addr) { struct device_node *memory = NULL; - nodemask_t nodes; - int default_nid = any_online_node(NODE_MASK_ALL); int nid; if (!numa_enabled || (min_common_depth < 0)) - return default_nid; + return any_online_node(NODE_MASK_ALL); + + memory = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory"); + if (memory) { + nid = hot_add_drconf_scn_to_nid(memory, scn_addr); + of_node_put(memory); + return nid; + } while ((memory = of_find_node_by_type(memory, "memory")) != NULL) { unsigned long start, size; @@ -932,13 +1149,9 @@ ha_new_range: size = read_n_cells(n_mem_size_cells, &memcell_buf); nid = of_node_to_nid_single(memory); - /* Domains not present at boot default to 0 */ - if (nid < 0 || !node_online(nid)) - nid = default_nid; - - if ((scn_addr >= start) && (scn_addr < (start + size))) { + if (valid_hot_add_scn(&nid, start, size, scn_addr)) { of_node_put(memory); - goto got_nid; + return nid; } if (--ranges) /* process all ranges in cell */ @@ -946,14 +1159,5 @@ ha_new_range: } BUG(); /* section address should be found above */ return 0; - - /* Temporary code to ensure that returned node is not empty */ -got_nid: - nodes_setall(nodes); - while (NODE_DATA(nid)->node_spanned_pages == 0) { - node_clear(nid, nodes); - nid = any_online_node(nodes); - } - return nid; } #endif /* CONFIG_MEMORY_HOTPLUG */