sh: Populate initial secondary CPU info from boot_cpu_data.
[safe/jmp/linux-2.6] / arch / powerpc / kernel / prom.c
index 5a24415..d4405b9 100644 (file)
@@ -16,7 +16,6 @@
 #undef DEBUG
 
 #include <stdarg.h>
-#include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
 #include <linux/bitops.h>
 #include <linux/module.h>
 #include <linux/kexec.h>
+#include <linux/debugfs.h>
+#include <linux/irq.h>
+#include <linux/lmb.h>
 
 #include <asm/prom.h>
 #include <asm/rtas.h>
-#include <asm/lmb.h>
 #include <asm/page.h>
 #include <asm/processor.h>
 #include <asm/irq.h>
@@ -50,6 +51,9 @@
 #include <asm/machdep.h>
 #include <asm/pSeries_reconfig.h>
 #include <asm/pci-bridge.h>
+#include <asm/phyp_dump.h>
+#include <asm/kexec.h>
+#include <mm/mmu_decl.h>
 
 #ifdef DEBUG
 #define DBG(fmt...) printk(KERN_ERR fmt)
@@ -62,7 +66,7 @@ static int __initdata dt_root_addr_cells;
 static int __initdata dt_root_size_cells;
 
 #ifdef CONFIG_PPC64
-static int __initdata iommu_is_off;
+int __initdata iommu_is_off;
 int __initdata iommu_force_on;
 unsigned long tce_alloc_start, tce_alloc_end;
 #endif
@@ -75,434 +79,13 @@ static struct boot_param_header *initial_boot_params __initdata;
 struct boot_param_header *initial_boot_params;
 #endif
 
-static struct device_node *allnodes = NULL;
+extern struct device_node *allnodes;   /* temporary while merging */
 
-/* use when traversing tree through the allnext, child, sibling,
- * or parent members of struct device_node.
- */
-static DEFINE_RWLOCK(devtree_lock);
+extern rwlock_t devtree_lock;  /* temporary while merging */
 
 /* export that to outside world */
 struct device_node *of_chosen;
 
-struct device_node *dflt_interrupt_controller;
-int num_interrupt_controllers;
-
-/*
- * Wrapper for allocating memory for various data that needs to be
- * attached to device nodes as they are processed at boot or when
- * added to the device tree later (e.g. DLPAR).  At boot there is
- * already a region reserved so we just increment *mem_start by size;
- * otherwise we call kmalloc.
- */
-static void * prom_alloc(unsigned long size, unsigned long *mem_start)
-{
-       unsigned long tmp;
-
-       if (!mem_start)
-               return kmalloc(size, GFP_KERNEL);
-
-       tmp = *mem_start;
-       *mem_start += size;
-       return (void *)tmp;
-}
-
-/*
- * Find the device_node with a given phandle.
- */
-static struct device_node * find_phandle(phandle ph)
-{
-       struct device_node *np;
-
-       for (np = allnodes; np != 0; np = np->allnext)
-               if (np->linux_phandle == ph)
-                       return np;
-       return NULL;
-}
-
-/*
- * Find the interrupt parent of a node.
- */
-static struct device_node * __devinit intr_parent(struct device_node *p)
-{
-       phandle *parp;
-
-       parp = (phandle *) get_property(p, "interrupt-parent", NULL);
-       if (parp == NULL)
-               return p->parent;
-       p = find_phandle(*parp);
-       if (p != NULL)
-               return p;
-       /*
-        * On a powermac booted with BootX, we don't get to know the
-        * phandles for any nodes, so find_phandle will return NULL.
-        * Fortunately these machines only have one interrupt controller
-        * so there isn't in fact any ambiguity.  -- paulus
-        */
-       if (num_interrupt_controllers == 1)
-               p = dflt_interrupt_controller;
-       return p;
-}
-
-/*
- * Find out the size of each entry of the interrupts property
- * for a node.
- */
-int __devinit prom_n_intr_cells(struct device_node *np)
-{
-       struct device_node *p;
-       unsigned int *icp;
-
-       for (p = np; (p = intr_parent(p)) != NULL; ) {
-               icp = (unsigned int *)
-                       get_property(p, "#interrupt-cells", NULL);
-               if (icp != NULL)
-                       return *icp;
-               if (get_property(p, "interrupt-controller", NULL) != NULL
-                   || get_property(p, "interrupt-map", NULL) != NULL) {
-                       printk("oops, node %s doesn't have #interrupt-cells\n",
-                              p->full_name);
-                       return 1;
-               }
-       }
-#ifdef DEBUG_IRQ
-       printk("prom_n_intr_cells failed for %s\n", np->full_name);
-#endif
-       return 1;
-}
-
-/*
- * Map an interrupt from a device up to the platform interrupt
- * descriptor.
- */
-static int __devinit map_interrupt(unsigned int **irq, struct device_node **ictrler,
-                                  struct device_node *np, unsigned int *ints,
-                                  int nintrc)
-{
-       struct device_node *p, *ipar;
-       unsigned int *imap, *imask, *ip;
-       int i, imaplen, match;
-       int newintrc = 0, newaddrc = 0;
-       unsigned int *reg;
-       int naddrc;
-
-       reg = (unsigned int *) get_property(np, "reg", NULL);
-       naddrc = prom_n_addr_cells(np);
-       p = intr_parent(np);
-       while (p != NULL) {
-               if (get_property(p, "interrupt-controller", NULL) != NULL)
-                       /* this node is an interrupt controller, stop here */
-                       break;
-               imap = (unsigned int *)
-                       get_property(p, "interrupt-map", &imaplen);
-               if (imap == NULL) {
-                       p = intr_parent(p);
-                       continue;
-               }
-               imask = (unsigned int *)
-                       get_property(p, "interrupt-map-mask", NULL);
-               if (imask == NULL) {
-                       printk("oops, %s has interrupt-map but no mask\n",
-                              p->full_name);
-                       return 0;
-               }
-               imaplen /= sizeof(unsigned int);
-               match = 0;
-               ipar = NULL;
-               while (imaplen > 0 && !match) {
-                       /* check the child-interrupt field */
-                       match = 1;
-                       for (i = 0; i < naddrc && match; ++i)
-                               match = ((reg[i] ^ imap[i]) & imask[i]) == 0;
-                       for (; i < naddrc + nintrc && match; ++i)
-                               match = ((ints[i-naddrc] ^ imap[i]) & imask[i]) == 0;
-                       imap += naddrc + nintrc;
-                       imaplen -= naddrc + nintrc;
-                       /* grab the interrupt parent */
-                       ipar = find_phandle((phandle) *imap++);
-                       --imaplen;
-                       if (ipar == NULL && num_interrupt_controllers == 1)
-                               /* cope with BootX not giving us phandles */
-                               ipar = dflt_interrupt_controller;
-                       if (ipar == NULL) {
-                               printk("oops, no int parent %x in map of %s\n",
-                                      imap[-1], p->full_name);
-                               return 0;
-                       }
-                       /* find the parent's # addr and intr cells */
-                       ip = (unsigned int *)
-                               get_property(ipar, "#interrupt-cells", NULL);
-                       if (ip == NULL) {
-                               printk("oops, no #interrupt-cells on %s\n",
-                                      ipar->full_name);
-                               return 0;
-                       }
-                       newintrc = *ip;
-                       ip = (unsigned int *)
-                               get_property(ipar, "#address-cells", NULL);
-                       newaddrc = (ip == NULL)? 0: *ip;
-                       imap += newaddrc + newintrc;
-                       imaplen -= newaddrc + newintrc;
-               }
-               if (imaplen < 0) {
-                       printk("oops, error decoding int-map on %s, len=%d\n",
-                              p->full_name, imaplen);
-                       return 0;
-               }
-               if (!match) {
-#ifdef DEBUG_IRQ
-                       printk("oops, no match in %s int-map for %s\n",
-                              p->full_name, np->full_name);
-#endif
-                       return 0;
-               }
-               p = ipar;
-               naddrc = newaddrc;
-               nintrc = newintrc;
-               ints = imap - nintrc;
-               reg = ints - naddrc;
-       }
-       if (p == NULL) {
-#ifdef DEBUG_IRQ
-               printk("hmmm, int tree for %s doesn't have ctrler\n",
-                      np->full_name);
-#endif
-               return 0;
-       }
-       *irq = ints;
-       *ictrler = p;
-       return nintrc;
-}
-
-static unsigned char map_isa_senses[4] = {
-       IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE,
-       IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE,
-       IRQ_SENSE_EDGE  | IRQ_POLARITY_NEGATIVE,
-       IRQ_SENSE_EDGE  | IRQ_POLARITY_POSITIVE
-};
-
-static unsigned char map_mpic_senses[4] = {
-       IRQ_SENSE_EDGE  | IRQ_POLARITY_POSITIVE,
-       IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE,
-       /* 2 seems to be used for the 8259 cascade... */
-       IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE,
-       IRQ_SENSE_EDGE  | IRQ_POLARITY_NEGATIVE,
-};
-
-static int __devinit finish_node_interrupts(struct device_node *np,
-                                           unsigned long *mem_start,
-                                           int measure_only)
-{
-       unsigned int *ints;
-       int intlen, intrcells, intrcount;
-       int i, j, n, sense;
-       unsigned int *irq, virq;
-       struct device_node *ic;
-       int trace = 0;
-
-       //#define TRACE(fmt...) do { if (trace) { printk(fmt); mdelay(1000); } } while(0)
-#define TRACE(fmt...)
-
-       if (!strcmp(np->name, "smu-doorbell"))
-               trace = 1;
-
-       TRACE("Finishing SMU doorbell ! num_interrupt_controllers = %d\n",
-             num_interrupt_controllers);
-
-       if (num_interrupt_controllers == 0) {
-               /*
-                * Old machines just have a list of interrupt numbers
-                * and no interrupt-controller nodes.
-                */
-               ints = (unsigned int *) get_property(np, "AAPL,interrupts",
-                                                    &intlen);
-               /* XXX old interpret_pci_props looked in parent too */
-               /* XXX old interpret_macio_props looked for interrupts
-                  before AAPL,interrupts */
-               if (ints == NULL)
-                       ints = (unsigned int *) get_property(np, "interrupts",
-                                                            &intlen);
-               if (ints == NULL)
-                       return 0;
-
-               np->n_intrs = intlen / sizeof(unsigned int);
-               np->intrs = prom_alloc(np->n_intrs * sizeof(np->intrs[0]),
-                                      mem_start);
-               if (!np->intrs)
-                       return -ENOMEM;
-               if (measure_only)
-                       return 0;
-
-               for (i = 0; i < np->n_intrs; ++i) {
-                       np->intrs[i].line = *ints++;
-                       np->intrs[i].sense = IRQ_SENSE_LEVEL
-                               | IRQ_POLARITY_NEGATIVE;
-               }
-               return 0;
-       }
-
-       ints = (unsigned int *) get_property(np, "interrupts", &intlen);
-       TRACE("ints=%p, intlen=%d\n", ints, intlen);
-       if (ints == NULL)
-               return 0;
-       intrcells = prom_n_intr_cells(np);
-       intlen /= intrcells * sizeof(unsigned int);
-       TRACE("intrcells=%d, new intlen=%d\n", intrcells, intlen);
-       np->intrs = prom_alloc(intlen * sizeof(*(np->intrs)), mem_start);
-       if (!np->intrs)
-               return -ENOMEM;
-
-       if (measure_only)
-               return 0;
-
-       intrcount = 0;
-       for (i = 0; i < intlen; ++i, ints += intrcells) {
-               n = map_interrupt(&irq, &ic, np, ints, intrcells);
-               TRACE("map, irq=%d, ic=%p, n=%d\n", irq, ic, n);
-               if (n <= 0)
-                       continue;
-
-               /* don't map IRQ numbers under a cascaded 8259 controller */
-               if (ic && device_is_compatible(ic, "chrp,iic")) {
-                       np->intrs[intrcount].line = irq[0];
-                       sense = (n > 1)? (irq[1] & 3): 3;
-                       np->intrs[intrcount].sense = map_isa_senses[sense];
-               } else {
-                       virq = virt_irq_create_mapping(irq[0]);
-                       TRACE("virq=%d\n", virq);
-#ifdef CONFIG_PPC64
-                       if (virq == NO_IRQ) {
-                               printk(KERN_CRIT "Could not allocate interrupt"
-                                      " number for %s\n", np->full_name);
-                               continue;
-                       }
-#endif
-                       np->intrs[intrcount].line = irq_offset_up(virq);
-                       sense = (n > 1)? (irq[1] & 3): 1;
-
-                       /* Apple uses bits in there in a different way, let's
-                        * only keep the real sense bit on macs
-                        */
-                       if (_machine == PLATFORM_POWERMAC)
-                               sense &= 0x1;
-                       np->intrs[intrcount].sense = map_mpic_senses[sense];
-               }
-
-#ifdef CONFIG_PPC64
-               /* We offset irq numbers for the u3 MPIC by 128 in PowerMac */
-               if (_machine == PLATFORM_POWERMAC && ic && ic->parent) {
-                       char *name = get_property(ic->parent, "name", NULL);
-                       if (name && !strcmp(name, "u3"))
-                               np->intrs[intrcount].line += 128;
-                       else if (!(name && (!strcmp(name, "mac-io") ||
-                                           !strcmp(name, "u4"))))
-                               /* ignore other cascaded controllers, such as
-                                  the k2-sata-root */
-                               break;
-               }
-#endif /* CONFIG_PPC64 */
-               if (n > 2) {
-                       printk("hmmm, got %d intr cells for %s:", n,
-                              np->full_name);
-                       for (j = 0; j < n; ++j)
-                               printk(" %d", irq[j]);
-                       printk("\n");
-               }
-               ++intrcount;
-       }
-       np->n_intrs = intrcount;
-
-       return 0;
-}
-
-static int __devinit finish_node(struct device_node *np,
-                                unsigned long *mem_start,
-                                int measure_only)
-{
-       struct device_node *child;
-       int rc = 0;
-
-       rc = finish_node_interrupts(np, mem_start, measure_only);
-       if (rc)
-               goto out;
-
-       for (child = np->child; child != NULL; child = child->sibling) {
-               rc = finish_node(child, mem_start, measure_only);
-               if (rc)
-                       goto out;
-       }
-out:
-       return rc;
-}
-
-static void __init scan_interrupt_controllers(void)
-{
-       struct device_node *np;
-       int n = 0;
-       char *name, *ic;
-       int iclen;
-
-       for (np = allnodes; np != NULL; np = np->allnext) {
-               ic = get_property(np, "interrupt-controller", &iclen);
-               name = get_property(np, "name", NULL);
-               /* checking iclen makes sure we don't get a false
-                  match on /chosen.interrupt_controller */
-               if ((name != NULL
-                    && strcmp(name, "interrupt-controller") == 0)
-                   || (ic != NULL && iclen == 0
-                       && strcmp(name, "AppleKiwi"))) {
-                       if (n == 0)
-                               dflt_interrupt_controller = np;
-                       ++n;
-               }
-       }
-       num_interrupt_controllers = n;
-}
-
-/**
- * finish_device_tree is called once things are running normally
- * (i.e. with text and data mapped to the address they were linked at).
- * It traverses the device tree and fills in some of the additional,
- * fields in each node like {n_}addrs and {n_}intrs, the virt interrupt
- * mapping is also initialized at this point.
- */
-void __init finish_device_tree(void)
-{
-       unsigned long start, end, size = 0;
-
-       DBG(" -> finish_device_tree\n");
-
-#ifdef CONFIG_PPC64
-       /* Initialize virtual IRQ map */
-       virt_irq_init();
-#endif
-       scan_interrupt_controllers();
-
-       /*
-        * Finish device-tree (pre-parsing some properties etc...)
-        * We do this in 2 passes. One with "measure_only" set, which
-        * will only measure the amount of memory needed, then we can
-        * allocate that memory, and call finish_node again. However,
-        * we must be careful as most routines will fail nowadays when
-        * prom_alloc() returns 0, so we must make sure our first pass
-        * doesn't start at 0. We pre-initialize size to 16 for that
-        * reason and then remove those additional 16 bytes
-        */
-       size = 16;
-       finish_node(allnodes, &size, 1);
-       size -= 16;
-
-       if (0 == size)
-               end = start = 0;
-       else
-               end = start = (unsigned long)__va(lmb_alloc(size, 128));
-
-       finish_node(allnodes, &end, 0);
-       BUG_ON(end != start + size);
-
-       DBG(" <- finish_device_tree\n");
-}
-
 static inline char *find_flat_dt_string(u32 offset)
 {
        return ((char *)initial_boot_params) +
@@ -570,6 +153,18 @@ int __init of_scan_flat_dt(int (*it)(unsigned long node,
        return rc;
 }
 
+unsigned long __init of_get_flat_dt_root(void)
+{
+       unsigned long p = ((unsigned long)initial_boot_params) +
+               initial_boot_params->off_dt_struct;
+
+       while(*((u32 *)p) == OF_DT_NOP)
+               p += 4;
+       BUG_ON (*((u32 *)p) != OF_DT_BEGIN_NODE);
+       p += 4;
+       return _ALIGN(p + strlen((char *)p) + 1, 4);
+}
+
 /**
  * This  function can be used within scan_flattened_dt callback to get
  * access to properties
@@ -612,6 +207,25 @@ void* __init of_get_flat_dt_prop(unsigned long node, const char *name,
        } while(1);
 }
 
+int __init of_flat_dt_is_compatible(unsigned long node, const char *compat)
+{
+       const char* cp;
+       unsigned long cplen, l;
+
+       cp = of_get_flat_dt_prop(node, "compatible", &cplen);
+       if (cp == NULL)
+               return 0;
+       while (cplen > 0) {
+               if (strncasecmp(cp, compat, strlen(compat)) == 0)
+                       return 1;
+               l = strlen(cp) + 1;
+               cp += l;
+               cplen -= l;
+       }
+
+       return 0;
+}
+
 static void *__init unflatten_dt_alloc(unsigned long *mem, unsigned long size,
                                       unsigned long align)
 {
@@ -686,7 +300,7 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
 #ifdef DEBUG
                                if ((strlen(p) + l + 1) != allocl) {
                                        DBG("%s: p: %d, l: %d, a: %d\n",
-                                           pathp, strlen(p), l, allocl);
+                                           pathp, (int)strlen(p), l, allocl);
                                }
 #endif
                                p += strlen(p);
@@ -775,18 +389,19 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
                if (allnextpp) {
                        pp->name = "name";
                        pp->length = sz;
-                       pp->value = (unsigned char *)(pp + 1);
+                       pp->value = pp + 1;
                        *prev_pp = pp;
                        prev_pp = &pp->next;
                        memcpy(pp->value, ps, sz - 1);
                        ((char *)pp->value)[sz - 1] = 0;
-                       DBG("fixed up name for %s -> %s\n", pathp, pp->value);
+                       DBG("fixed up name for %s -> %s\n", pathp,
+                               (char *)pp->value);
                }
        }
        if (allnextpp) {
                *prev_pp = NULL;
-               np->name = get_property(np, "name", NULL);
-               np->type = get_property(np, "device_type", NULL);
+               np->name = of_get_property(np, "name", NULL);
+               np->type = of_get_property(np, "device_type", NULL);
 
                if (!np->name)
                        np->name = "<NULL>";
@@ -805,6 +420,44 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
        return mem;
 }
 
+static int __init early_parse_mem(char *p)
+{
+       if (!p)
+               return 1;
+
+       memory_limit = PAGE_ALIGN(memparse(p, &p));
+       DBG("memory limit = 0x%llx\n", (unsigned long long)memory_limit);
+
+       return 0;
+}
+early_param("mem", early_parse_mem);
+
+/**
+ * move_device_tree - move tree to an unused area, if needed.
+ *
+ * The device tree may be allocated beyond our memory limit, or inside the
+ * crash kernel region for kdump. If so, move it out of the way.
+ */
+static void __init move_device_tree(void)
+{
+       unsigned long start, size;
+       void *p;
+
+       DBG("-> move_device_tree\n");
+
+       start = __pa(initial_boot_params);
+       size = initial_boot_params->totalsize;
+
+       if ((memory_limit && (start + size) > memory_limit) ||
+                       overlaps_crashkernel(start, size)) {
+               p = __va(lmb_alloc_base(size, PAGE_SIZE, lmb.rmo_size));
+               memcpy(p, initial_boot_params, size);
+               initial_boot_params = (struct boot_param_header *)p;
+               DBG("Moved device tree to 0x%p\n", p);
+       }
+
+       DBG("<- move_device_tree\n");
+}
 
 /**
  * unflattens the device-tree passed by the firmware, creating the
@@ -854,13 +507,170 @@ void __init unflatten_device_tree(void)
        DBG(" <- unflatten_device_tree()\n");
 }
 
+/*
+ * ibm,pa-features is a per-cpu property that contains a string of
+ * attribute descriptors, each of which has a 2 byte header plus up
+ * to 254 bytes worth of processor attribute bits.  First header
+ * byte specifies the number of bytes following the header.
+ * Second header byte is an "attribute-specifier" type, of which
+ * zero is the only currently-defined value.
+ * Implementation:  Pass in the byte and bit offset for the feature
+ * that we are interested in.  The function will return -1 if the
+ * pa-features property is missing, or a 1/0 to indicate if the feature
+ * is supported/not supported.  Note that the bit numbers are
+ * big-endian to match the definition in PAPR.
+ */
+static struct ibm_pa_feature {
+       unsigned long   cpu_features;   /* CPU_FTR_xxx bit */
+       unsigned int    cpu_user_ftrs;  /* PPC_FEATURE_xxx bit */
+       unsigned char   pabyte;         /* byte number in ibm,pa-features */
+       unsigned char   pabit;          /* bit number (big-endian) */
+       unsigned char   invert;         /* if 1, pa bit set => clear feature */
+} ibm_pa_features[] __initdata = {
+       {0, PPC_FEATURE_HAS_MMU,        0, 0, 0},
+       {0, PPC_FEATURE_HAS_FPU,        0, 1, 0},
+       {CPU_FTR_SLB, 0,                0, 2, 0},
+       {CPU_FTR_CTRL, 0,               0, 3, 0},
+       {CPU_FTR_NOEXECUTE, 0,          0, 6, 0},
+       {CPU_FTR_NODSISRALIGN, 0,       1, 1, 1},
+       {CPU_FTR_CI_LARGE_PAGE, 0,      1, 2, 0},
+       {CPU_FTR_REAL_LE, PPC_FEATURE_TRUE_LE, 5, 0, 0},
+};
+
+static void __init scan_features(unsigned long node, unsigned char *ftrs,
+                                unsigned long tablelen,
+                                struct ibm_pa_feature *fp,
+                                unsigned long ft_size)
+{
+       unsigned long i, len, bit;
+
+       /* find descriptor with type == 0 */
+       for (;;) {
+               if (tablelen < 3)
+                       return;
+               len = 2 + ftrs[0];
+               if (tablelen < len)
+                       return;         /* descriptor 0 not found */
+               if (ftrs[1] == 0)
+                       break;
+               tablelen -= len;
+               ftrs += len;
+       }
+
+       /* loop over bits we know about */
+       for (i = 0; i < ft_size; ++i, ++fp) {
+               if (fp->pabyte >= ftrs[0])
+                       continue;
+               bit = (ftrs[2 + fp->pabyte] >> (7 - fp->pabit)) & 1;
+               if (bit ^ fp->invert) {
+                       cur_cpu_spec->cpu_features |= fp->cpu_features;
+                       cur_cpu_spec->cpu_user_features |= fp->cpu_user_ftrs;
+               } else {
+                       cur_cpu_spec->cpu_features &= ~fp->cpu_features;
+                       cur_cpu_spec->cpu_user_features &= ~fp->cpu_user_ftrs;
+               }
+       }
+}
+
+static void __init check_cpu_pa_features(unsigned long node)
+{
+       unsigned char *pa_ftrs;
+       unsigned long tablelen;
+
+       pa_ftrs = of_get_flat_dt_prop(node, "ibm,pa-features", &tablelen);
+       if (pa_ftrs == NULL)
+               return;
+
+       scan_features(node, pa_ftrs, tablelen,
+                     ibm_pa_features, ARRAY_SIZE(ibm_pa_features));
+}
+
+#ifdef CONFIG_PPC_STD_MMU_64
+static void __init check_cpu_slb_size(unsigned long node)
+{
+       u32 *slb_size_ptr;
+
+       slb_size_ptr = of_get_flat_dt_prop(node, "slb-size", NULL);
+       if (slb_size_ptr != NULL) {
+               mmu_slb_size = *slb_size_ptr;
+               return;
+       }
+       slb_size_ptr = of_get_flat_dt_prop(node, "ibm,slb-size", NULL);
+       if (slb_size_ptr != NULL) {
+               mmu_slb_size = *slb_size_ptr;
+       }
+}
+#else
+#define check_cpu_slb_size(node) do { } while(0)
+#endif
+
+static struct feature_property {
+       const char *name;
+       u32 min_value;
+       unsigned long cpu_feature;
+       unsigned long cpu_user_ftr;
+} feature_properties[] __initdata = {
+#ifdef CONFIG_ALTIVEC
+       {"altivec", 0, CPU_FTR_ALTIVEC, PPC_FEATURE_HAS_ALTIVEC},
+       {"ibm,vmx", 1, CPU_FTR_ALTIVEC, PPC_FEATURE_HAS_ALTIVEC},
+#endif /* CONFIG_ALTIVEC */
+#ifdef CONFIG_VSX
+       /* Yes, this _really_ is ibm,vmx == 2 to enable VSX */
+       {"ibm,vmx", 2, CPU_FTR_VSX, PPC_FEATURE_HAS_VSX},
+#endif /* CONFIG_VSX */
+#ifdef CONFIG_PPC64
+       {"ibm,dfp", 1, 0, PPC_FEATURE_HAS_DFP},
+       {"ibm,purr", 1, CPU_FTR_PURR, 0},
+       {"ibm,spurr", 1, CPU_FTR_SPURR, 0},
+#endif /* CONFIG_PPC64 */
+};
+
+#if defined(CONFIG_44x) && defined(CONFIG_PPC_FPU)
+static inline void identical_pvr_fixup(unsigned long node)
+{
+       unsigned int pvr;
+       char *model = of_get_flat_dt_prop(node, "model", NULL);
+
+       /*
+        * Since 440GR(x)/440EP(x) processors have the same pvr,
+        * we check the node path and set bit 28 in the cur_cpu_spec
+        * pvr for EP(x) processor version. This bit is always 0 in
+        * the "real" pvr. Then we call identify_cpu again with
+        * the new logical pvr to enable FPU support.
+        */
+       if (model && strstr(model, "440EP")) {
+               pvr = cur_cpu_spec->pvr_value | 0x8;
+               identify_cpu(0, pvr);
+               DBG("Using logical pvr %x for %s\n", pvr, model);
+       }
+}
+#else
+#define identical_pvr_fixup(node) do { } while(0)
+#endif
+
+static void __init check_cpu_feature_properties(unsigned long node)
+{
+       unsigned long i;
+       struct feature_property *fp = feature_properties;
+       const u32 *prop;
+
+       for (i = 0; i < ARRAY_SIZE(feature_properties); ++i, ++fp) {
+               prop = of_get_flat_dt_prop(node, fp->name, NULL);
+               if (prop && *prop >= fp->min_value) {
+                       cur_cpu_spec->cpu_features |= fp->cpu_feature;
+                       cur_cpu_spec->cpu_user_features |= fp->cpu_user_ftr;
+               }
+       }
+}
+
 static int __init early_init_dt_scan_cpus(unsigned long node,
                                          const char *uname, int depth,
                                          void *data)
 {
        static int logical_cpuid = 0;
        char *type = of_get_flat_dt_prop(node, "device_type", NULL);
-       u32 *prop, *intserv;
+       const u32 *prop;
+       const u32 *intserv;
        int i, nthreads;
        unsigned long len;
        int found = 0;
@@ -917,23 +727,31 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
                        intserv[i]);
                boot_cpuid = logical_cpuid;
                set_hard_smp_processor_id(boot_cpuid, intserv[i]);
-       }
 
-#ifdef CONFIG_ALTIVEC
-       /* Check if we have a VMX and eventually update CPU features */
-       prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", NULL);
-       if (prop && (*prop) > 0) {
-               cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC;
-               cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
-       }
+               /*
+                * PAPR defines "logical" PVR values for cpus that
+                * meet various levels of the architecture:
+                * 0x0f000001   Architecture version 2.04
+                * 0x0f000002   Architecture version 2.05
+                * If the cpu-version property in the cpu node contains
+                * such a value, we call identify_cpu again with the
+                * logical PVR value in order to use the cpu feature
+                * bits appropriate for the architecture level.
+                *
+                * A POWER6 partition in "POWER6 architected" mode
+                * uses the 0x0f000002 PVR value; in POWER5+ mode
+                * it uses 0x0f000001.
+                */
+               prop = of_get_flat_dt_prop(node, "cpu-version", NULL);
+               if (prop && (*prop & 0xff000000) == 0x0f000000)
+                       identify_cpu(0, *prop);
 
-       /* Same goes for Apple's "altivec" property */
-       prop = (u32 *)of_get_flat_dt_prop(node, "altivec", NULL);
-       if (prop) {
-               cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC;
-               cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
+               identical_pvr_fixup(node);
        }
-#endif /* CONFIG_ALTIVEC */
+
+       check_cpu_feature_properties(node);
+       check_cpu_pa_features(node);
+       check_cpu_slb_size(node);
 
 #ifdef CONFIG_PPC_PSERIES
        if (nthreads > 1)
@@ -945,10 +763,39 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
        return 0;
 }
 
+#ifdef CONFIG_BLK_DEV_INITRD
+static void __init early_init_dt_check_for_initrd(unsigned long node)
+{
+       unsigned long l;
+       u32 *prop;
+
+       DBG("Looking for initrd properties... ");
+
+       prop = of_get_flat_dt_prop(node, "linux,initrd-start", &l);
+       if (prop) {
+               initrd_start = (unsigned long)__va(of_read_ulong(prop, l/4));
+
+               prop = of_get_flat_dt_prop(node, "linux,initrd-end", &l);
+               if (prop) {
+                       initrd_end = (unsigned long)
+                                       __va(of_read_ulong(prop, l/4));
+                       initrd_below_start_ok = 1;
+               } else {
+                       initrd_start = 0;
+               }
+       }
+
+       DBG("initrd_start=0x%lx  initrd_end=0x%lx\n", initrd_start, initrd_end);
+}
+#else
+static inline void early_init_dt_check_for_initrd(unsigned long node)
+{
+}
+#endif /* CONFIG_BLK_DEV_INITRD */
+
 static int __init early_init_dt_scan_chosen(unsigned long node,
                                            const char *uname, int depth, void *data)
 {
-       u32 *prop;
        unsigned long *lprop;
        unsigned long l;
        char *p;
@@ -959,14 +806,6 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
            (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
                return 0;
 
-       /* get platform type */
-       prop = (u32 *)of_get_flat_dt_prop(node, "linux,platform", NULL);
-       if (prop == NULL)
-               return 0;
-#ifdef CONFIG_PPC_MULTIPLATFORM
-       _machine = *prop;
-#endif
-
 #ifdef CONFIG_PPC64
        /* check if iommu is forced on or off */
        if (of_get_flat_dt_prop(node, "linux,iommu-off", NULL) != NULL)
@@ -975,6 +814,7 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
                iommu_force_on = 1;
 #endif
 
+       /* mem=x on the command line is the preferred mechanism */
        lprop = of_get_flat_dt_prop(node, "linux,memory-limit", NULL);
        if (lprop)
                memory_limit = *lprop;
@@ -988,69 +828,30 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
                tce_alloc_end = *lprop;
 #endif
 
-#ifdef CONFIG_PPC_RTAS
-       /* To help early debugging via the front panel, we retrieve a minimal
-        * set of RTAS infos now if available
-        */
-       {
-               u64 *basep, *entryp;
-
-               basep = of_get_flat_dt_prop(node, "linux,rtas-base", NULL);
-               entryp = of_get_flat_dt_prop(node, "linux,rtas-entry", NULL);
-               prop = of_get_flat_dt_prop(node, "linux,rtas-size", NULL);
-               if (basep && entryp && prop) {
-                       rtas.base = *basep;
-                       rtas.entry = *entryp;
-                       rtas.size = *prop;
-               }
-       }
-#endif /* CONFIG_PPC_RTAS */
-
 #ifdef CONFIG_KEXEC
-       lprop = (u64*)of_get_flat_dt_prop(node, "linux,crashkernel-base", NULL);
-       if (lprop)
-               crashk_res.start = *lprop;
+       lprop = of_get_flat_dt_prop(node, "linux,crashkernel-base", NULL);
+       if (lprop)
+               crashk_res.start = *lprop;
 
-       lprop = (u64*)of_get_flat_dt_prop(node, "linux,crashkernel-size", NULL);
-       if (lprop)
-               crashk_res.end = crashk_res.start + *lprop - 1;
+       lprop = of_get_flat_dt_prop(node, "linux,crashkernel-size", NULL);
+       if (lprop)
+               crashk_res.end = crashk_res.start + *lprop - 1;
 #endif
 
+       early_init_dt_check_for_initrd(node);
+
        /* Retreive command line */
        p = of_get_flat_dt_prop(node, "bootargs", &l);
        if (p != NULL && l > 0)
                strlcpy(cmd_line, p, min((int)l, COMMAND_LINE_SIZE));
 
 #ifdef CONFIG_CMDLINE
-       if (l == 0 || (l == 1 && (*p) == 0))
+       if (p == NULL || l == 0 || (l == 1 && (*p) == 0))
                strlcpy(cmd_line, CONFIG_CMDLINE, COMMAND_LINE_SIZE);
 #endif /* CONFIG_CMDLINE */
 
        DBG("Command line is: %s\n", cmd_line);
 
-       if (strstr(cmd_line, "mem=")) {
-               char *p, *q;
-               unsigned long maxmem = 0;
-
-               for (q = cmd_line; (p = strstr(q, "mem=")) != 0; ) {
-                       q = p + 4;
-                       if (p > cmd_line && p[-1] != ' ')
-                               continue;
-                       maxmem = simple_strtoul(q, &q, 0);
-                       if (*q == 'k' || *q == 'K') {
-                               maxmem <<= 10;
-                               ++q;
-                       } else if (*q == 'm' || *q == 'M') {
-                               maxmem <<= 20;
-                               ++q;
-                       } else if (*q == 'g' || *q == 'G') {
-                               maxmem <<= 30;
-                               ++q;
-                       }
-               }
-               memory_limit = maxmem;
-       }
-
        /* break now */
        return 1;
 }
@@ -1075,29 +876,92 @@ static int __init early_init_dt_scan_root(unsigned long node,
        return 1;
 }
 
-static unsigned long __init dt_mem_next_cell(int s, cell_t **cellp)
+static u64 __init dt_mem_next_cell(int s, cell_t **cellp)
 {
        cell_t *p = *cellp;
-       unsigned long r;
 
-       /* Ignore more than 2 cells */
-       while (s > sizeof(unsigned long) / 4) {
-               p++;
-               s--;
-       }
-       r = *p++;
-#ifdef CONFIG_PPC64
-       if (s > 1) {
-               r <<= 32;
-               r |= *(p++);
-               s--;
-       }
-#endif
-
-       *cellp = p;
-       return r;
+       *cellp = p + s;
+       return of_read_number(p, s);
 }
 
+#ifdef CONFIG_PPC_PSERIES
+/*
+ * Interpret the ibm,dynamic-memory property in the
+ * /ibm,dynamic-reconfiguration-memory node.
+ * This contains a list of memory blocks along with NUMA affinity
+ * information.
+ */
+static int __init early_init_dt_scan_drconf_memory(unsigned long node)
+{
+       cell_t *dm, *ls, *usm;
+       unsigned long l, n, flags;
+       u64 base, size, lmb_size;
+       unsigned int is_kexec_kdump = 0, rngs;
+
+       ls = of_get_flat_dt_prop(node, "ibm,lmb-size", &l);
+       if (ls == NULL || l < dt_root_size_cells * sizeof(cell_t))
+               return 0;
+       lmb_size = dt_mem_next_cell(dt_root_size_cells, &ls);
+
+       dm = of_get_flat_dt_prop(node, "ibm,dynamic-memory", &l);
+       if (dm == NULL || l < sizeof(cell_t))
+               return 0;
+
+       n = *dm++;      /* number of entries */
+       if (l < (n * (dt_root_addr_cells + 4) + 1) * sizeof(cell_t))
+               return 0;
+
+       /* check if this is a kexec/kdump kernel. */
+       usm = of_get_flat_dt_prop(node, "linux,drconf-usable-memory",
+                                                &l);
+       if (usm != NULL)
+               is_kexec_kdump = 1;
+
+       for (; n != 0; --n) {
+               base = dt_mem_next_cell(dt_root_addr_cells, &dm);
+               flags = dm[3];
+               /* skip DRC index, pad, assoc. list index, flags */
+               dm += 4;
+               /* skip this block if the reserved bit is set in flags (0x80)
+                  or if the block is not assigned to this partition (0x8) */
+               if ((flags & 0x80) || !(flags & 0x8))
+                       continue;
+               size = lmb_size;
+               rngs = 1;
+               if (is_kexec_kdump) {
+                       /*
+                        * For each lmb in ibm,dynamic-memory, a corresponding
+                        * entry in linux,drconf-usable-memory property contains
+                        * a counter 'p' followed by 'p' (base, size) duple.
+                        * Now read the counter from
+                        * linux,drconf-usable-memory property
+                        */
+                       rngs = dt_mem_next_cell(dt_root_size_cells, &usm);
+                       if (!rngs) /* there are no (base, size) duple */
+                               continue;
+               }
+               do {
+                       if (is_kexec_kdump) {
+                               base = dt_mem_next_cell(dt_root_addr_cells,
+                                                        &usm);
+                               size = dt_mem_next_cell(dt_root_size_cells,
+                                                        &usm);
+                       }
+                       if (iommu_is_off) {
+                               if (base >= 0x80000000ul)
+                                       continue;
+                               if ((base + size) > 0x80000000ul)
+                                       size = 0x80000000ul - base;
+                       }
+                       lmb_add(base, size);
+               } while (--rngs);
+       }
+       lmb_dump_all();
+       return 0;
+}
+#else
+#define early_init_dt_scan_drconf_memory(node) 0
+#endif /* CONFIG_PPC_PSERIES */
 
 static int __init early_init_dt_scan_memory(unsigned long node,
                                            const char *uname, int depth, void *data)
@@ -1106,6 +970,11 @@ static int __init early_init_dt_scan_memory(unsigned long node,
        cell_t *reg, *endp;
        unsigned long l;
 
+       /* Look for the ibm,dynamic-reconfiguration-memory node */
+       if (depth == 1 &&
+           strcmp(uname, "ibm,dynamic-reconfiguration-memory") == 0)
+               return early_init_dt_scan_drconf_memory(node);
+
        /* We are scanning "memory" nodes only */
        if (type == NULL) {
                /*
@@ -1117,9 +986,9 @@ static int __init early_init_dt_scan_memory(unsigned long node,
        } else if (strcmp(type, "memory") != 0)
                return 0;
 
-       reg = (cell_t *)of_get_flat_dt_prop(node, "linux,usable-memory", &l);
+       reg = of_get_flat_dt_prop(node, "linux,usable-memory", &l);
        if (reg == NULL)
-               reg = (cell_t *)of_get_flat_dt_prop(node, "reg", &l);
+               reg = of_get_flat_dt_prop(node, "reg", &l);
        if (reg == NULL)
                return 0;
 
@@ -1129,14 +998,15 @@ static int __init early_init_dt_scan_memory(unsigned long node,
            uname, l, reg[0], reg[1], reg[2], reg[3]);
 
        while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) {
-               unsigned long base, size;
+               u64 base, size;
 
                base = dt_mem_next_cell(dt_root_addr_cells, &reg);
                size = dt_mem_next_cell(dt_root_size_cells, &reg);
 
                if (size == 0)
                        continue;
-               DBG(" - %lx ,  %lx\n", base, size);
+               DBG(" - %llx ,  %llx\n", (unsigned long long)base,
+                   (unsigned long long)size);
 #ifdef CONFIG_PPC64
                if (iommu_is_off) {
                        if (base >= 0x80000000ul)
@@ -1146,7 +1016,10 @@ static int __init early_init_dt_scan_memory(unsigned long node,
                }
 #endif
                lmb_add(base, size);
+
+               memstart_addr = min((u64)memstart_addr, base);
        }
+
        return 0;
 }
 
@@ -1154,9 +1027,23 @@ static void __init early_reserve_mem(void)
 {
        u64 base, size;
        u64 *reserve_map;
+       unsigned long self_base;
+       unsigned long self_size;
 
        reserve_map = (u64 *)(((unsigned long)initial_boot_params) +
                                        initial_boot_params->off_mem_rsvmap);
+
+       /* before we do anything, lets reserve the dt blob */
+       self_base = __pa((unsigned long)initial_boot_params);
+       self_size = initial_boot_params->totalsize;
+       lmb_reserve(self_base, self_size);
+
+#ifdef CONFIG_BLK_DEV_INITRD
+       /* then reserve the initrd, if any */
+       if (initrd_start && (initrd_end > initrd_start))
+               lmb_reserve(__pa(initrd_start), initrd_end - initrd_start);
+#endif /* CONFIG_BLK_DEV_INITRD */
+
 #ifdef CONFIG_PPC32
        /* 
         * Handle the case where we might be booting from an old kexec
@@ -1171,6 +1058,9 @@ static void __init early_reserve_mem(void)
                        size_32 = *(reserve_map_32++);
                        if (size_32 == 0)
                                break;
+                       /* skip if the reservation is for the blob */
+                       if (base_32 == self_base && size_32 == self_size)
+                               continue;
                        DBG("reserving: %x -> %x\n", base_32, size_32);
                        lmb_reserve(base_32, size_32);
                }
@@ -1185,20 +1075,108 @@ static void __init early_reserve_mem(void)
                DBG("reserving: %llx -> %llx\n", base, size);
                lmb_reserve(base, size);
        }
+}
 
-#if 0
-       DBG("memory reserved, lmbs :\n");
-       lmb_dump_all();
-#endif
+#ifdef CONFIG_PHYP_DUMP
+/**
+ * phyp_dump_calculate_reserve_size() - reserve variable boot area 5% or arg
+ *
+ * Function to find the largest size we need to reserve
+ * during early boot process.
+ *
+ * It either looks for boot param and returns that OR
+ * returns larger of 256 or 5% rounded down to multiples of 256MB.
+ *
+ */
+static inline unsigned long phyp_dump_calculate_reserve_size(void)
+{
+       unsigned long tmp;
+
+       if (phyp_dump_info->reserve_bootvar)
+               return phyp_dump_info->reserve_bootvar;
+
+       /* divide by 20 to get 5% of value */
+       tmp = lmb_end_of_DRAM();
+       do_div(tmp, 20);
+
+       /* round it down in multiples of 256 */
+       tmp = tmp & ~0x0FFFFFFFUL;
+
+       return (tmp > PHYP_DUMP_RMR_END ? tmp : PHYP_DUMP_RMR_END);
 }
 
+/**
+ * phyp_dump_reserve_mem() - reserve all not-yet-dumped mmemory
+ *
+ * This routine may reserve memory regions in the kernel only
+ * if the system is supported and a dump was taken in last
+ * boot instance or if the hardware is supported and the
+ * scratch area needs to be setup. In other instances it returns
+ * without reserving anything. The memory in case of dump being
+ * active is freed when the dump is collected (by userland tools).
+ */
+static void __init phyp_dump_reserve_mem(void)
+{
+       unsigned long base, size;
+       unsigned long variable_reserve_size;
+
+       if (!phyp_dump_info->phyp_dump_configured) {
+               printk(KERN_ERR "Phyp-dump not supported on this hardware\n");
+               return;
+       }
+
+       if (!phyp_dump_info->phyp_dump_at_boot) {
+               printk(KERN_INFO "Phyp-dump disabled at boot time\n");
+               return;
+       }
+
+       variable_reserve_size = phyp_dump_calculate_reserve_size();
+
+       if (phyp_dump_info->phyp_dump_is_active) {
+               /* Reserve *everything* above RMR.Area freed by userland tools*/
+               base = variable_reserve_size;
+               size = lmb_end_of_DRAM() - base;
+
+               /* XXX crashed_ram_end is wrong, since it may be beyond
+                * the memory_limit, it will need to be adjusted. */
+               lmb_reserve(base, size);
+
+               phyp_dump_info->init_reserve_start = base;
+               phyp_dump_info->init_reserve_size = size;
+       } else {
+               size = phyp_dump_info->cpu_state_size +
+                       phyp_dump_info->hpte_region_size +
+                       variable_reserve_size;
+               base = lmb_end_of_DRAM() - size;
+               lmb_reserve(base, size);
+               phyp_dump_info->init_reserve_start = base;
+               phyp_dump_info->init_reserve_size = size;
+       }
+}
+#else
+static inline void __init phyp_dump_reserve_mem(void) {}
+#endif /* CONFIG_PHYP_DUMP  && CONFIG_PPC_RTAS */
+
+
 void __init early_init_devtree(void *params)
 {
-       DBG(" -> early_init_devtree()\n");
+       phys_addr_t limit;
+
+       DBG(" -> early_init_devtree(%p)\n", params);
 
        /* Setup flat device-tree pointer */
        initial_boot_params = params;
 
+#ifdef CONFIG_PPC_RTAS
+       /* Some machines might need RTAS info for debugging, grab it now. */
+       of_scan_flat_dt(early_init_dt_scan_rtas, NULL);
+#endif
+
+#ifdef CONFIG_PHYP_DUMP
+       /* scan tree to see if dump occured during last boot */
+       of_scan_flat_dt(early_init_dt_scan_phyp_dump, NULL);
+#endif
+
        /* Retrieve various informations from the /chosen node of the
         * device-tree, including the platform type, initrd location and
         * size, TCE reserve, and more ...
@@ -1209,17 +1187,42 @@ void __init early_init_devtree(void *params)
        lmb_init();
        of_scan_flat_dt(early_init_dt_scan_root, NULL);
        of_scan_flat_dt(early_init_dt_scan_memory, NULL);
-       lmb_enforce_memory_limit(memory_limit);
-       lmb_analyze();
 
-       DBG("Phys. mem: %lx\n", lmb_phys_mem_size());
+       /* Save command line for /proc/cmdline and then parse parameters */
+       strlcpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE);
+       parse_early_param();
 
        /* Reserve LMB regions used by kernel, initrd, dt, etc... */
        lmb_reserve(PHYSICAL_START, __pa(klimit) - PHYSICAL_START);
-#ifdef CONFIG_CRASH_DUMP
-       lmb_reserve(0, KDUMP_RESERVE_LIMIT);
-#endif
+       /* If relocatable, reserve first 32k for interrupt vectors etc. */
+       if (PHYSICAL_START > MEMORY_START)
+               lmb_reserve(MEMORY_START, 0x8000);
+       reserve_kdump_trampoline();
+       reserve_crashkernel();
        early_reserve_mem();
+       phyp_dump_reserve_mem();
+
+       limit = memory_limit;
+       if (! limit) {
+               phys_addr_t memsize;
+
+               /* Ensure that total memory size is page-aligned, because
+                * otherwise mark_bootmem() gets upset. */
+               lmb_analyze();
+               memsize = lmb_phys_mem_size();
+               if ((memsize & PAGE_MASK) != memsize)
+                       limit = memsize & PAGE_MASK;
+       }
+       lmb_enforce_memory_limit(limit);
+
+       lmb_analyze();
+       lmb_dump_all();
+
+       DBG("Phys. mem: %llx\n", lmb_phys_mem_size());
+
+       /* We may need to relocate the flat tree, do it now.
+        * FIXME .. and the initrd too? */
+       move_device_tree();
 
        DBG("Scanning CPUs ...\n");
 
@@ -1231,139 +1234,6 @@ void __init early_init_devtree(void *params)
        DBG(" <- early_init_devtree()\n");
 }
 
-#undef printk
-
-int
-prom_n_addr_cells(struct device_node* np)
-{
-       int* ip;
-       do {
-               if (np->parent)
-                       np = np->parent;
-               ip = (int *) get_property(np, "#address-cells", NULL);
-               if (ip != NULL)
-                       return *ip;
-       } while (np->parent);
-       /* No #address-cells property for the root node, default to 1 */
-       return 1;
-}
-EXPORT_SYMBOL(prom_n_addr_cells);
-
-int
-prom_n_size_cells(struct device_node* np)
-{
-       int* ip;
-       do {
-               if (np->parent)
-                       np = np->parent;
-               ip = (int *) get_property(np, "#size-cells", NULL);
-               if (ip != NULL)
-                       return *ip;
-       } while (np->parent);
-       /* No #size-cells property for the root node, default to 1 */
-       return 1;
-}
-EXPORT_SYMBOL(prom_n_size_cells);
-
-/**
- * Work out the sense (active-low level / active-high edge)
- * of each interrupt from the device tree.
- */
-void __init prom_get_irq_senses(unsigned char *senses, int off, int max)
-{
-       struct device_node *np;
-       int i, j;
-
-       /* default to level-triggered */
-       memset(senses, IRQ_SENSE_LEVEL | IRQ_POLARITY_NEGATIVE, max - off);
-
-       for (np = allnodes; np != 0; np = np->allnext) {
-               for (j = 0; j < np->n_intrs; j++) {
-                       i = np->intrs[j].line;
-                       if (i >= off && i < max)
-                               senses[i-off] = np->intrs[j].sense;
-               }
-       }
-}
-
-/**
- * Construct and return a list of the device_nodes with a given name.
- */
-struct device_node *find_devices(const char *name)
-{
-       struct device_node *head, **prevp, *np;
-
-       prevp = &head;
-       for (np = allnodes; np != 0; np = np->allnext) {
-               if (np->name != 0 && strcasecmp(np->name, name) == 0) {
-                       *prevp = np;
-                       prevp = &np->next;
-               }
-       }
-       *prevp = NULL;
-       return head;
-}
-EXPORT_SYMBOL(find_devices);
-
-/**
- * Construct and return a list of the device_nodes with a given type.
- */
-struct device_node *find_type_devices(const char *type)
-{
-       struct device_node *head, **prevp, *np;
-
-       prevp = &head;
-       for (np = allnodes; np != 0; np = np->allnext) {
-               if (np->type != 0 && strcasecmp(np->type, type) == 0) {
-                       *prevp = np;
-                       prevp = &np->next;
-               }
-       }
-       *prevp = NULL;
-       return head;
-}
-EXPORT_SYMBOL(find_type_devices);
-
-/**
- * Returns all nodes linked together
- */
-struct device_node *find_all_nodes(void)
-{
-       struct device_node *head, **prevp, *np;
-
-       prevp = &head;
-       for (np = allnodes; np != 0; np = np->allnext) {
-               *prevp = np;
-               prevp = &np->next;
-       }
-       *prevp = NULL;
-       return head;
-}
-EXPORT_SYMBOL(find_all_nodes);
-
-/** Checks if the given "compat" string matches one of the strings in
- * the device's "compatible" property
- */
-int device_is_compatible(struct device_node *device, const char *compat)
-{
-       const char* cp;
-       int cplen, l;
-
-       cp = (char *) get_property(device, "compatible", &cplen);
-       if (cp == NULL)
-               return 0;
-       while (cplen > 0) {
-               if (strncasecmp(cp, compat, strlen(compat)) == 0)
-                       return 1;
-               l = strlen(cp) + 1;
-               cp += l;
-               cplen -= l;
-       }
-
-       return 0;
-}
-EXPORT_SYMBOL(device_is_compatible);
-
 
 /**
  * Indicates whether the root node has a given value in its
@@ -1376,51 +1246,13 @@ int machine_is_compatible(const char *compat)
 
        root = of_find_node_by_path("/");
        if (root) {
-               rc = device_is_compatible(root, compat);
+               rc = of_device_is_compatible(root, compat);
                of_node_put(root);
        }
        return rc;
 }
 EXPORT_SYMBOL(machine_is_compatible);
 
-/**
- * Construct and return a list of the device_nodes with a given type
- * and compatible property.
- */
-struct device_node *find_compatible_devices(const char *type,
-                                           const char *compat)
-{
-       struct device_node *head, **prevp, *np;
-
-       prevp = &head;
-       for (np = allnodes; np != 0; np = np->allnext) {
-               if (type != NULL
-                   && !(np->type != 0 && strcasecmp(np->type, type) == 0))
-                       continue;
-               if (device_is_compatible(np, compat)) {
-                       *prevp = np;
-                       prevp = &np->next;
-               }
-       }
-       *prevp = NULL;
-       return head;
-}
-EXPORT_SYMBOL(find_compatible_devices);
-
-/**
- * Find the device_node with a given full_name.
- */
-struct device_node *find_path_device(const char *path)
-{
-       struct device_node *np;
-
-       for (np = allnodes; np != 0; np = np->allnext)
-               if (np->full_name != 0 && strcasecmp(np->full_name, path) == 0)
-                       return np;
-       return NULL;
-}
-EXPORT_SYMBOL(find_path_device);
-
 /*******
  *
  * New implementation of the OF "find" APIs, return a refcounted
@@ -1433,141 +1265,56 @@ EXPORT_SYMBOL(find_path_device);
  *******/
 
 /**
- *     of_find_node_by_name - Find a node by its "name" property
- *     @from:  The node to start searching from or NULL, the node
- *             you pass will not be searched, only the next one
- *             will; typically, you pass what the previous call
- *             returned. of_node_put() will be called on it
- *     @name:  The name string to match against
- *
- *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
- */
-struct device_node *of_find_node_by_name(struct device_node *from,
-       const char *name)
-{
-       struct device_node *np;
-
-       read_lock(&devtree_lock);
-       np = from ? from->allnext : allnodes;
-       for (; np != NULL; np = np->allnext)
-               if (np->name != NULL && strcasecmp(np->name, name) == 0
-                   && of_node_get(np))
-                       break;
-       if (from)
-               of_node_put(from);
-       read_unlock(&devtree_lock);
-       return np;
-}
-EXPORT_SYMBOL(of_find_node_by_name);
-
-/**
- *     of_find_node_by_type - Find a node by its "device_type" property
- *     @from:  The node to start searching from or NULL, the node
- *             you pass will not be searched, only the next one
- *             will; typically, you pass what the previous call
- *             returned. of_node_put() will be called on it
- *     @name:  The type string to match against
+ *     of_find_node_by_phandle - Find a node given a phandle
+ *     @handle:        phandle of the node to find
  *
  *     Returns a node pointer with refcount incremented, use
  *     of_node_put() on it when done.
  */
-struct device_node *of_find_node_by_type(struct device_node *from,
-       const char *type)
+struct device_node *of_find_node_by_phandle(phandle handle)
 {
        struct device_node *np;
 
        read_lock(&devtree_lock);
-       np = from ? from->allnext : allnodes;
-       for (; np != 0; np = np->allnext)
-               if (np->type != 0 && strcasecmp(np->type, type) == 0
-                   && of_node_get(np))
+       for (np = allnodes; np != 0; np = np->allnext)
+               if (np->linux_phandle == handle)
                        break;
-       if (from)
-               of_node_put(from);
+       of_node_get(np);
        read_unlock(&devtree_lock);
        return np;
 }
-EXPORT_SYMBOL(of_find_node_by_type);
+EXPORT_SYMBOL(of_find_node_by_phandle);
 
 /**
- *     of_find_compatible_node - Find a node based on type and one of the
- *                                tokens in its "compatible" property
- *     @from:          The node to start searching from or NULL, the node
- *                     you pass will not be searched, only the next one
- *                     will; typically, you pass what the previous call
- *                     returned. of_node_put() will be called on it
- *     @type:          The type string to match "device_type" or NULL to ignore
- *     @compatible:    The string to match to one of the tokens in the device
- *                     "compatible" list.
+ *     of_find_next_cache_node - Find a node's subsidiary cache
+ *     @np:    node of type "cpu" or "cache"
  *
  *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
+ *     of_node_put() on it when done.  Caller should hold a reference
+ *     to np.
  */
-struct device_node *of_find_compatible_node(struct device_node *from,
-       const char *type, const char *compatible)
+struct device_node *of_find_next_cache_node(struct device_node *np)
 {
-       struct device_node *np;
+       struct device_node *child;
+       const phandle *handle;
 
-       read_lock(&devtree_lock);
-       np = from ? from->allnext : allnodes;
-       for (; np != 0; np = np->allnext) {
-               if (type != NULL
-                   && !(np->type != 0 && strcasecmp(np->type, type) == 0))
-                       continue;
-               if (device_is_compatible(np, compatible) && of_node_get(np))
-                       break;
-       }
-       if (from)
-               of_node_put(from);
-       read_unlock(&devtree_lock);
-       return np;
-}
-EXPORT_SYMBOL(of_find_compatible_node);
+       handle = of_get_property(np, "l2-cache", NULL);
+       if (!handle)
+               handle = of_get_property(np, "next-level-cache", NULL);
 
-/**
- *     of_find_node_by_path - Find a node matching a full OF path
- *     @path:  The full path to match
- *
- *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
- */
-struct device_node *of_find_node_by_path(const char *path)
-{
-       struct device_node *np = allnodes;
+       if (handle)
+               return of_find_node_by_phandle(*handle);
 
-       read_lock(&devtree_lock);
-       for (; np != 0; np = np->allnext) {
-               if (np->full_name != 0 && strcasecmp(np->full_name, path) == 0
-                   && of_node_get(np))
-                       break;
-       }
-       read_unlock(&devtree_lock);
-       return np;
-}
-EXPORT_SYMBOL(of_find_node_by_path);
-
-/**
- *     of_find_node_by_phandle - Find a node given a phandle
- *     @handle:        phandle of the node to find
- *
- *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
- */
-struct device_node *of_find_node_by_phandle(phandle handle)
-{
-       struct device_node *np;
+       /* OF on pmac has nodes instead of properties named "l2-cache"
+        * beneath CPU nodes.
+        */
+       if (!strcmp(np->type, "cpu"))
+               for_each_child_of_node(np, child)
+                       if (!strcmp(child->type, "cache"))
+                               return child;
 
-       read_lock(&devtree_lock);
-       for (np = allnodes; np != 0; np = np->allnext)
-               if (np->linux_phandle == handle)
-                       break;
-       if (np)
-               of_node_get(np);
-       read_unlock(&devtree_lock);
-       return np;
+       return NULL;
 }
-EXPORT_SYMBOL(of_find_node_by_phandle);
 
 /**
  *     of_find_all_nodes - Get next node in global list
@@ -1586,60 +1333,13 @@ struct device_node *of_find_all_nodes(struct device_node *prev)
        for (; np != 0; np = np->allnext)
                if (of_node_get(np))
                        break;
-       if (prev)
-               of_node_put(prev);
+       of_node_put(prev);
        read_unlock(&devtree_lock);
        return np;
 }
 EXPORT_SYMBOL(of_find_all_nodes);
 
 /**
- *     of_get_parent - Get a node's parent if any
- *     @node:  Node to get parent
- *
- *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
- */
-struct device_node *of_get_parent(const struct device_node *node)
-{
-       struct device_node *np;
-
-       if (!node)
-               return NULL;
-
-       read_lock(&devtree_lock);
-       np = of_node_get(node->parent);
-       read_unlock(&devtree_lock);
-       return np;
-}
-EXPORT_SYMBOL(of_get_parent);
-
-/**
- *     of_get_next_child - Iterate a node childs
- *     @node:  parent node
- *     @prev:  previous child of the parent node, or NULL to get first
- *
- *     Returns a node pointer with refcount incremented, use
- *     of_node_put() on it when done.
- */
-struct device_node *of_get_next_child(const struct device_node *node,
-       struct device_node *prev)
-{
-       struct device_node *next;
-
-       read_lock(&devtree_lock);
-       next = prev ? prev->sibling : node->child;
-       for (; next != 0; next = next->sibling)
-               if (of_node_get(next))
-                       break;
-       if (prev)
-               of_node_put(prev);
-       read_unlock(&devtree_lock);
-       return next;
-}
-EXPORT_SYMBOL(of_get_next_child);
-
-/**
  *     of_node_get - Increment refcount of a node
  *     @node:  Node to inc refcount, NULL is supported to
  *             simplify writing of callers
@@ -1671,8 +1371,17 @@ static void of_node_release(struct kref *kref)
        struct device_node *node = kref_to_device_node(kref);
        struct property *prop = node->properties;
 
-       if (!OF_IS_DYNAMIC(node))
+       /* We should never be releasing nodes that haven't been detached. */
+       if (!of_node_check_flag(node, OF_DETACHED)) {
+               printk("WARNING: Bad of_node_put() on %s\n", node->full_name);
+               dump_stack();
+               kref_init(&node->kref);
                return;
+       }
+
+       if (!of_node_check_flag(node, OF_DYNAMIC))
+               return;
+
        while (prop) {
                struct property *next = prop->next;
                kfree(prop->name);
@@ -1685,7 +1394,6 @@ static void of_node_release(struct kref *kref)
                        node->deadprops = NULL;
                }
        }
-       kfree(node->intrs);
        kfree(node->full_name);
        kfree(node->data);
        kfree(node);
@@ -1709,12 +1417,14 @@ EXPORT_SYMBOL(of_node_put);
  */
 void of_attach_node(struct device_node *np)
 {
-       write_lock(&devtree_lock);
+       unsigned long flags;
+
+       write_lock_irqsave(&devtree_lock, flags);
        np->sibling = np->parent->child;
        np->allnext = allnodes;
        np->parent->child = np;
        allnodes = np;
-       write_unlock(&devtree_lock);
+       write_unlock_irqrestore(&devtree_lock, flags);
 }
 
 /*
@@ -1722,13 +1432,16 @@ void of_attach_node(struct device_node *np)
  * a reference to the node.  The memory associated with the node
  * is not freed until its refcount goes to zero.
  */
-void of_detach_node(const struct device_node *np)
+void of_detach_node(struct device_node *np)
 {
        struct device_node *parent;
+       unsigned long flags;
 
-       write_lock(&devtree_lock);
+       write_lock_irqsave(&devtree_lock, flags);
 
        parent = np->parent;
+       if (!parent)
+               goto out_unlock;
 
        if (allnodes == np)
                allnodes = np->allnext;
@@ -1752,29 +1465,31 @@ void of_detach_node(const struct device_node *np)
                prevsib->sibling = np->sibling;
        }
 
-       write_unlock(&devtree_lock);
+       of_node_set_flag(np, OF_DETACHED);
+
+out_unlock:
+       write_unlock_irqrestore(&devtree_lock, flags);
 }
 
 #ifdef CONFIG_PPC_PSERIES
 /*
  * Fix up the uninitialized fields in a new device node:
- * name, type, n_addrs, addrs, n_intrs, intrs, and pci-specific fields
- *
- * A lot of boot-time code is duplicated here, because functions such
- * as finish_node_interrupts, interpret_pci_props, etc. cannot use the
- * slab allocator.
- *
- * This should probably be split up into smaller chunks.
+ * name, type and pci-specific fields
  */
 
 static int of_finish_dynamic_node(struct device_node *node)
 {
        struct device_node *parent = of_get_parent(node);
        int err = 0;
-       phandle *ibm_phandle;
+       const phandle *ibm_phandle;
 
-       node->name = get_property(node, "name", NULL);
-       node->type = get_property(node, "device_type", NULL);
+       node->name = of_get_property(node, "name", NULL);
+       node->type = of_get_property(node, "device_type", NULL);
+
+       if (!node->name)
+               node->name = "<NULL>";
+       if (!node->type)
+               node->type = "<NULL>";
 
        if (!parent) {
                err = -ENODEV;
@@ -1784,12 +1499,11 @@ static int of_finish_dynamic_node(struct device_node *node)
        /* We don't support that function on PowerMac, at least
         * not yet
         */
-       if (_machine == PLATFORM_POWERMAC)
+       if (machine_is(powermac))
                return -ENODEV;
 
        /* fix up new node's linux_phandle field */
-       if ((ibm_phandle = (unsigned int *)get_property(node,
-                                                       "ibm,phandle", NULL)))
+       if ((ibm_phandle = of_get_property(node, "ibm,phandle", NULL)))
                node->linux_phandle = *ibm_phandle;
 
 out:
@@ -1805,8 +1519,6 @@ static int prom_reconfig_notifier(struct notifier_block *nb,
        switch (action) {
        case PSERIES_RECONFIG_ADD:
                err = of_finish_dynamic_node(node);
-               if (!err)
-                       finish_node(node, NULL, 0);
                if (err < 0) {
                        printk(KERN_ERR "finish_node returned %d\n", err);
                        err = NOTIFY_BAD;
@@ -1831,55 +1543,27 @@ static int __init prom_reconfig_setup(void)
 __initcall(prom_reconfig_setup);
 #endif
 
-struct property *of_find_property(struct device_node *np, const char *name,
-                                 int *lenp)
-{
-       struct property *pp;
-
-       read_lock(&devtree_lock);
-       for (pp = np->properties; pp != 0; pp = pp->next)
-               if (strcmp(pp->name, name) == 0) {
-                       if (lenp != 0)
-                               *lenp = pp->length;
-                       break;
-               }
-       read_unlock(&devtree_lock);
-
-       return pp;
-}
-
-/*
- * Find a property with a given name for a given node
- * and return the value.
- */
-unsigned char *get_property(struct device_node *np, const char *name,
-                           int *lenp)
-{
-       struct property *pp = of_find_property(np,name,lenp);
-       return pp ? pp->value : NULL;
-}
-EXPORT_SYMBOL(get_property);
-
 /*
  * Add a property to a node
  */
 int prom_add_property(struct device_node* np, struct property* prop)
 {
        struct property **next;
+       unsigned long flags;
 
        prop->next = NULL;      
-       write_lock(&devtree_lock);
+       write_lock_irqsave(&devtree_lock, flags);
        next = &np->properties;
        while (*next) {
                if (strcmp(prop->name, (*next)->name) == 0) {
                        /* duplicate ! don't insert it */
-                       write_unlock(&devtree_lock);
+                       write_unlock_irqrestore(&devtree_lock, flags);
                        return -1;
                }
                next = &(*next)->next;
        }
        *next = prop;
-       write_unlock(&devtree_lock);
+       write_unlock_irqrestore(&devtree_lock, flags);
 
 #ifdef CONFIG_PROC_DEVICETREE
        /* try to add to proc as well if it was initialized */
@@ -1899,9 +1583,10 @@ int prom_add_property(struct device_node* np, struct property* prop)
 int prom_remove_property(struct device_node *np, struct property *prop)
 {
        struct property **next;
+       unsigned long flags;
        int found = 0;
 
-       write_lock(&devtree_lock);
+       write_lock_irqsave(&devtree_lock, flags);
        next = &np->properties;
        while (*next) {
                if (*next == prop) {
@@ -1914,7 +1599,7 @@ int prom_remove_property(struct device_node *np, struct property *prop)
                }
                next = &(*next)->next;
        }
-       write_unlock(&devtree_lock);
+       write_unlock_irqrestore(&devtree_lock, flags);
 
        if (!found)
                return -ENODEV;
@@ -1940,9 +1625,10 @@ int prom_update_property(struct device_node *np,
                         struct property *oldprop)
 {
        struct property **next;
+       unsigned long flags;
        int found = 0;
 
-       write_lock(&devtree_lock);
+       write_lock_irqsave(&devtree_lock, flags);
        next = &np->properties;
        while (*next) {
                if (*next == oldprop) {
@@ -1956,7 +1642,7 @@ int prom_update_property(struct device_node *np,
                }
                next = &(*next)->next;
        }
-       write_unlock(&devtree_lock);
+       write_unlock_irqrestore(&devtree_lock, flags);
 
        if (!found)
                return -ENODEV;
@@ -1970,29 +1656,67 @@ int prom_update_property(struct device_node *np,
        return 0;
 }
 
-#ifdef CONFIG_KEXEC
-/* We may have allocated the flat device tree inside the crash kernel region
- * in prom_init. If so we need to move it out into regular memory. */
-void kdump_move_device_tree(void)
+
+/* Find the device node for a given logical cpu number, also returns the cpu
+ * local thread number (index in ibm,interrupt-server#s) if relevant and
+ * asked for (non NULL)
+ */
+struct device_node *of_get_cpu_node(int cpu, unsigned int *thread)
 {
-       unsigned long start, end;
-       struct boot_param_header *new;
+       int hardid;
+       struct device_node *np;
 
-       start = __pa((unsigned long)initial_boot_params);
-       end = start + initial_boot_params->totalsize;
+       hardid = get_hard_smp_processor_id(cpu);
 
-       if (end < crashk_res.start || start > crashk_res.end)
-               return;
+       for_each_node_by_type(np, "cpu") {
+               const u32 *intserv;
+               unsigned int plen, t;
+
+               /* Check for ibm,ppc-interrupt-server#s. If it doesn't exist
+                * fallback to "reg" property and assume no threads
+                */
+               intserv = of_get_property(np, "ibm,ppc-interrupt-server#s",
+                               &plen);
+               if (intserv == NULL) {
+                       const u32 *reg = of_get_property(np, "reg", NULL);
+                       if (reg == NULL)
+                               continue;
+                       if (*reg == hardid) {
+                               if (thread)
+                                       *thread = 0;
+                               return np;
+                       }
+               } else {
+                       plen /= sizeof(u32);
+                       for (t = 0; t < plen; t++) {
+                               if (hardid == intserv[t]) {
+                                       if (thread)
+                                               *thread = t;
+                                       return np;
+                               }
+                       }
+               }
+       }
+       return NULL;
+}
+EXPORT_SYMBOL(of_get_cpu_node);
 
-       new = (struct boot_param_header*)
-               __va(lmb_alloc(initial_boot_params->totalsize, PAGE_SIZE));
+#if defined(CONFIG_DEBUG_FS) && defined(DEBUG)
+static struct debugfs_blob_wrapper flat_dt_blob;
 
-       memcpy(new, initial_boot_params, initial_boot_params->totalsize);
+static int __init export_flat_device_tree(void)
+{
+       struct dentry *d;
 
-       initial_boot_params = new;
+       flat_dt_blob.data = initial_boot_params;
+       flat_dt_blob.size = initial_boot_params->totalsize;
 
-       DBG("Flat device tree blob moved to %p\n", initial_boot_params);
+       d = debugfs_create_blob("flat-device-tree", S_IFREG | S_IRUSR,
+                               powerpc_debugfs_root, &flat_dt_blob);
+       if (!d)
+               return 1;
 
-       /* XXX should we unreserve the old DT? */
+       return 0;
 }
-#endif /* CONFIG_KEXEC */
+__initcall(export_flat_device_tree);
+#endif