Subversion Repositories HelenOS

Rev

Rev 2071 | Rev 2089 | Go to most recent revision | Blame | Compare with Previous | Last modification | View Log | Download | RSS feed

  1. /*
  2.  * Copyright (c) 2001-2006 Jakub Jermar
  3.  * All rights reserved.
  4.  *
  5.  * Redistribution and use in source and binary forms, with or without
  6.  * modification, are permitted provided that the following conditions
  7.  * are met:
  8.  *
  9.  * - Redistributions of source code must retain the above copyright
  10.  *   notice, this list of conditions and the following disclaimer.
  11.  * - Redistributions in binary form must reproduce the above copyright
  12.  *   notice, this list of conditions and the following disclaimer in the
  13.  *   documentation and/or other materials provided with the distribution.
  14.  * - The name of the author may not be used to endorse or promote products
  15.  *   derived from this software without specific prior written permission.
  16.  *
  17.  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
  18.  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  19.  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  20.  * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
  21.  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  22.  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  23.  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  24.  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  25.  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  26.  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  27.  */
  28.  
  29. /** @addtogroup genericmm
  30.  * @{
  31.  */
  32.  
  33. /**
  34.  * @file
  35.  * @brief   Address space related functions.
  36.  *
  37.  * This file contains address space manipulation functions.
  38.  * Roughly speaking, this is a higher-level client of
  39.  * Virtual Address Translation (VAT) subsystem.
  40.  *
  41.  * Functionality provided by this file allows one to
  42.  * create address spaces and create, resize and share
  43.  * address space areas.
  44.  *
  45.  * @see page.c
  46.  *
  47.  */
  48.  
  49. #include <mm/as.h>
  50. #include <arch/mm/as.h>
  51. #include <mm/page.h>
  52. #include <mm/frame.h>
  53. #include <mm/slab.h>
  54. #include <mm/tlb.h>
  55. #include <arch/mm/page.h>
  56. #include <genarch/mm/page_pt.h>
  57. #include <genarch/mm/page_ht.h>
  58. #include <mm/asid.h>
  59. #include <arch/mm/asid.h>
  60. #include <synch/spinlock.h>
  61. #include <synch/mutex.h>
  62. #include <adt/list.h>
  63. #include <adt/btree.h>
  64. #include <proc/task.h>
  65. #include <proc/thread.h>
  66. #include <arch/asm.h>
  67. #include <panic.h>
  68. #include <debug.h>
  69. #include <print.h>
  70. #include <memstr.h>
  71. #include <macros.h>
  72. #include <arch.h>
  73. #include <errno.h>
  74. #include <config.h>
  75. #include <align.h>
  76. #include <arch/types.h>
  77. #include <typedefs.h>
  78. #include <syscall/copy.h>
  79. #include <arch/interrupt.h>
  80.  
  81. #ifdef CONFIG_VIRT_IDX_DCACHE
  82. #include <arch/mm/cache.h>
  83. #endif /* CONFIG_VIRT_IDX_DCACHE */
  84.  
  85. /**
  86.  * Each architecture decides what functions will be used to carry out
  87.  * address space operations such as creating or locking page tables.
  88.  */
  89. as_operations_t *as_operations = NULL;
  90.  
  91. /**
  92.  * Slab for as_t objects.
  93.  */
  94. static slab_cache_t *as_slab;
  95.  
  96. /**
  97.  * This lock protects inactive_as_with_asid_head list. It must be acquired
  98.  * before as_t mutex.
  99.  */
  100. SPINLOCK_INITIALIZE(inactive_as_with_asid_lock);
  101.  
  102. /**
  103.  * This list contains address spaces that are not active on any
  104.  * processor and that have valid ASID.
  105.  */
  106. LIST_INITIALIZE(inactive_as_with_asid_head);
  107.  
  108. /** Kernel address space. */
  109. as_t *AS_KERNEL = NULL;
  110.  
  111. static int area_flags_to_page_flags(int aflags);
  112. static as_area_t *find_area_and_lock(as_t *as, uintptr_t va);
  113. static bool check_area_conflicts(as_t *as, uintptr_t va, size_t size,
  114.     as_area_t *avoid_area);
  115. static void sh_info_remove_reference(share_info_t *sh_info);
  116.  
  117. static int as_constructor(void *obj, int flags)
  118. {
  119.     as_t *as = (as_t *) obj;
  120.     int rc;
  121.  
  122.     link_initialize(&as->inactive_as_with_asid_link);
  123.     mutex_initialize(&as->lock);   
  124.    
  125.     rc = as_constructor_arch(as, flags);
  126.    
  127.     return rc;
  128. }
  129.  
  130. static int as_destructor(void *obj)
  131. {
  132.     as_t *as = (as_t *) obj;
  133.  
  134.     return as_destructor_arch(as);
  135. }
  136.  
  137. /** Initialize address space subsystem. */
  138. void as_init(void)
  139. {
  140.     as_arch_init();
  141.    
  142.     as_slab = slab_cache_create("as_slab", sizeof(as_t), 0,
  143.         as_constructor, as_destructor, SLAB_CACHE_MAGDEFERRED);
  144.    
  145.     AS_KERNEL = as_create(FLAG_AS_KERNEL);
  146.     if (!AS_KERNEL)
  147.         panic("can't create kernel address space\n");
  148.    
  149. }
  150.  
  151. /** Create address space.
  152.  *
  153.  * @param flags Flags that influence way in wich the address space is created.
  154.  */
  155. as_t *as_create(int flags)
  156. {
  157.     as_t *as;
  158.  
  159.     as = (as_t *) slab_alloc(as_slab, 0);
  160.     (void) as_create_arch(as, 0);
  161.    
  162.     btree_create(&as->as_area_btree);
  163.    
  164.     if (flags & FLAG_AS_KERNEL)
  165.         as->asid = ASID_KERNEL;
  166.     else
  167.         as->asid = ASID_INVALID;
  168.    
  169.     as->refcount = 0;
  170.     as->cpu_refcount = 0;
  171.     as->page_table = page_table_create(flags);
  172.  
  173.     return as;
  174. }
  175.  
  176. /** Destroy adress space.
  177.  *
  178.  * When there are no tasks referencing this address space (i.e. its refcount is
  179.  * zero), the address space can be destroyed.
  180.  */
  181. void as_destroy(as_t *as)
  182. {
  183.     ipl_t ipl;
  184.     bool cond;
  185.  
  186.     ASSERT(as->refcount == 0);
  187.    
  188.     /*
  189.      * Since there is no reference to this area,
  190.      * it is safe not to lock its mutex.
  191.      */
  192.     ipl = interrupts_disable();
  193.     spinlock_lock(&inactive_as_with_asid_lock);
  194.     if (as->asid != ASID_INVALID && as != AS_KERNEL) {
  195.         if (as != AS && as->cpu_refcount == 0)
  196.             list_remove(&as->inactive_as_with_asid_link);
  197.         asid_put(as->asid);
  198.     }
  199.     spinlock_unlock(&inactive_as_with_asid_lock);
  200.  
  201.     /*
  202.      * Destroy address space areas of the address space.
  203.      * The B+tree must be walked carefully because it is
  204.      * also being destroyed.
  205.      */
  206.     for (cond = true; cond; ) {
  207.         btree_node_t *node;
  208.  
  209.         ASSERT(!list_empty(&as->as_area_btree.leaf_head));
  210.         node = list_get_instance(as->as_area_btree.leaf_head.next,
  211.             btree_node_t, leaf_link);
  212.  
  213.         if ((cond = node->keys)) {
  214.             as_area_destroy(as, node->key[0]);
  215.         }
  216.     }
  217.  
  218.     btree_destroy(&as->as_area_btree);
  219.     page_table_destroy(as->page_table);
  220.  
  221.     interrupts_restore(ipl);
  222.    
  223.     slab_free(as_slab, as);
  224. }
  225.  
  226. /** Create address space area of common attributes.
  227.  *
  228.  * The created address space area is added to the target address space.
  229.  *
  230.  * @param as Target address space.
  231.  * @param flags Flags of the area memory.
  232.  * @param size Size of area.
  233.  * @param base Base address of area.
  234.  * @param attrs Attributes of the area.
  235.  * @param backend Address space area backend. NULL if no backend is used.
  236.  * @param backend_data NULL or a pointer to an array holding two void *.
  237.  *
  238.  * @return Address space area on success or NULL on failure.
  239.  */
  240. as_area_t *
  241. as_area_create(as_t *as, int flags, size_t size, uintptr_t base, int attrs,
  242.            mem_backend_t *backend, mem_backend_data_t *backend_data)
  243. {
  244.     ipl_t ipl;
  245.     as_area_t *a;
  246.    
  247.     if (base % PAGE_SIZE)
  248.         return NULL;
  249.  
  250.     if (!size)
  251.         return NULL;
  252.  
  253.     /* Writeable executable areas are not supported. */
  254.     if ((flags & AS_AREA_EXEC) && (flags & AS_AREA_WRITE))
  255.         return NULL;
  256.    
  257.     ipl = interrupts_disable();
  258.     mutex_lock(&as->lock);
  259.    
  260.     if (!check_area_conflicts(as, base, size, NULL)) {
  261.         mutex_unlock(&as->lock);
  262.         interrupts_restore(ipl);
  263.         return NULL;
  264.     }
  265.    
  266.     a = (as_area_t *) malloc(sizeof(as_area_t), 0);
  267.  
  268.     mutex_initialize(&a->lock);
  269.    
  270.     a->as = as;
  271.     a->flags = flags;
  272.     a->attributes = attrs;
  273.     a->pages = SIZE2FRAMES(size);
  274.     a->base = base;
  275.     a->sh_info = NULL;
  276.     a->backend = backend;
  277.     if (backend_data)
  278.         a->backend_data = *backend_data;
  279.     else
  280.         memsetb((uintptr_t) &a->backend_data, sizeof(a->backend_data),
  281.             0);
  282.  
  283.     btree_create(&a->used_space);
  284.    
  285.     btree_insert(&as->as_area_btree, base, (void *) a, NULL);
  286.  
  287.     mutex_unlock(&as->lock);
  288.     interrupts_restore(ipl);
  289.  
  290.     return a;
  291. }
  292.  
  293. /** Find address space area and change it.
  294.  *
  295.  * @param as Address space.
  296.  * @param address Virtual address belonging to the area to be changed. Must be
  297.  *     page-aligned.
  298.  * @param size New size of the virtual memory block starting at address.
  299.  * @param flags Flags influencing the remap operation. Currently unused.
  300.  *
  301.  * @return Zero on success or a value from @ref errno.h otherwise.
  302.  */
  303. int as_area_resize(as_t *as, uintptr_t address, size_t size, int flags)
  304. {
  305.     as_area_t *area;
  306.     ipl_t ipl;
  307.     size_t pages;
  308.    
  309.     ipl = interrupts_disable();
  310.     mutex_lock(&as->lock);
  311.    
  312.     /*
  313.      * Locate the area.
  314.      */
  315.     area = find_area_and_lock(as, address);
  316.     if (!area) {
  317.         mutex_unlock(&as->lock);
  318.         interrupts_restore(ipl);
  319.         return ENOENT;
  320.     }
  321.  
  322.     if (area->backend == &phys_backend) {
  323.         /*
  324.          * Remapping of address space areas associated
  325.          * with memory mapped devices is not supported.
  326.          */
  327.         mutex_unlock(&area->lock);
  328.         mutex_unlock(&as->lock);
  329.         interrupts_restore(ipl);
  330.         return ENOTSUP;
  331.     }
  332.     if (area->sh_info) {
  333.         /*
  334.          * Remapping of shared address space areas
  335.          * is not supported.
  336.          */
  337.         mutex_unlock(&area->lock);
  338.         mutex_unlock(&as->lock);
  339.         interrupts_restore(ipl);
  340.         return ENOTSUP;
  341.     }
  342.  
  343.     pages = SIZE2FRAMES((address - area->base) + size);
  344.     if (!pages) {
  345.         /*
  346.          * Zero size address space areas are not allowed.
  347.          */
  348.         mutex_unlock(&area->lock);
  349.         mutex_unlock(&as->lock);
  350.         interrupts_restore(ipl);
  351.         return EPERM;
  352.     }
  353.    
  354.     if (pages < area->pages) {
  355.         bool cond;
  356.         uintptr_t start_free = area->base + pages*PAGE_SIZE;
  357.  
  358.         /*
  359.          * Shrinking the area.
  360.          * No need to check for overlaps.
  361.          */
  362.  
  363.         /*
  364.          * Start TLB shootdown sequence.
  365.          */
  366.         tlb_shootdown_start(TLB_INVL_PAGES, AS->asid, area->base +
  367.             pages * PAGE_SIZE, area->pages - pages);
  368.  
  369.         /*
  370.          * Remove frames belonging to used space starting from
  371.          * the highest addresses downwards until an overlap with
  372.          * the resized address space area is found. Note that this
  373.          * is also the right way to remove part of the used_space
  374.          * B+tree leaf list.
  375.          */    
  376.         for (cond = true; cond;) {
  377.             btree_node_t *node;
  378.        
  379.             ASSERT(!list_empty(&area->used_space.leaf_head));
  380.             node =
  381.                 list_get_instance(area->used_space.leaf_head.prev,
  382.                 btree_node_t, leaf_link);
  383.             if ((cond = (bool) node->keys)) {
  384.                 uintptr_t b = node->key[node->keys - 1];
  385.                 count_t c =
  386.                     (count_t) node->value[node->keys - 1];
  387.                 int i = 0;
  388.            
  389.                 if (overlaps(b, c * PAGE_SIZE, area->base,
  390.                     pages*PAGE_SIZE)) {
  391.                    
  392.                     if (b + c * PAGE_SIZE <= start_free) {
  393.                         /*
  394.                          * The whole interval fits
  395.                          * completely in the resized
  396.                          * address space area.
  397.                          */
  398.                         break;
  399.                     }
  400.        
  401.                     /*
  402.                      * Part of the interval corresponding
  403.                      * to b and c overlaps with the resized
  404.                      * address space area.
  405.                      */
  406.        
  407.                     cond = false;   /* we are almost done */
  408.                     i = (start_free - b) >> PAGE_WIDTH;
  409.                     if (!used_space_remove(area, start_free,
  410.                         c - i))
  411.                         panic("Could not remove used "
  412.                             "space.\n");
  413.                 } else {
  414.                     /*
  415.                      * The interval of used space can be
  416.                      * completely removed.
  417.                      */
  418.                     if (!used_space_remove(area, b, c))
  419.                         panic("Could not remove used "
  420.                             "space.\n");
  421.                 }
  422.            
  423.                 for (; i < c; i++) {
  424.                     pte_t *pte;
  425.            
  426.                     page_table_lock(as, false);
  427.                     pte = page_mapping_find(as, b +
  428.                         i * PAGE_SIZE);
  429.                     ASSERT(pte && PTE_VALID(pte) &&
  430.                         PTE_PRESENT(pte));
  431.                     if (area->backend &&
  432.                         area->backend->frame_free) {
  433.                         area->backend->frame_free(area,
  434.                             b + i * PAGE_SIZE,
  435.                             PTE_GET_FRAME(pte));
  436.                     }
  437.                     page_mapping_remove(as, b +
  438.                         i * PAGE_SIZE);
  439.                     page_table_unlock(as, false);
  440.                 }
  441.             }
  442.         }
  443.  
  444.         /*
  445.          * Finish TLB shootdown sequence.
  446.          */
  447.         tlb_invalidate_pages(as->asid, area->base + pages * PAGE_SIZE,
  448.             area->pages - pages);
  449.         tlb_shootdown_finalize();
  450.        
  451.         /*
  452.          * Invalidate software translation caches (e.g. TSB on sparc64).
  453.          */
  454.         as_invalidate_translation_cache(as, area->base +
  455.             pages * PAGE_SIZE, area->pages - pages);
  456.     } else {
  457.         /*
  458.          * Growing the area.
  459.          * Check for overlaps with other address space areas.
  460.          */
  461.         if (!check_area_conflicts(as, address, pages * PAGE_SIZE,
  462.             area)) {
  463.             mutex_unlock(&area->lock);
  464.             mutex_unlock(&as->lock);       
  465.             interrupts_restore(ipl);
  466.             return EADDRNOTAVAIL;
  467.         }
  468.     }
  469.  
  470.     area->pages = pages;
  471.    
  472.     mutex_unlock(&area->lock);
  473.     mutex_unlock(&as->lock);
  474.     interrupts_restore(ipl);
  475.  
  476.     return 0;
  477. }
  478.  
  479. /** Destroy address space area.
  480.  *
  481.  * @param as Address space.
  482.  * @param address Address withing the area to be deleted.
  483.  *
  484.  * @return Zero on success or a value from @ref errno.h on failure.
  485.  */
  486. int as_area_destroy(as_t *as, uintptr_t address)
  487. {
  488.     as_area_t *area;
  489.     uintptr_t base;
  490.     link_t *cur;
  491.     ipl_t ipl;
  492.  
  493.     ipl = interrupts_disable();
  494.     mutex_lock(&as->lock);
  495.  
  496.     area = find_area_and_lock(as, address);
  497.     if (!area) {
  498.         mutex_unlock(&as->lock);
  499.         interrupts_restore(ipl);
  500.         return ENOENT;
  501.     }
  502.  
  503.     base = area->base;
  504.  
  505.     /*
  506.      * Start TLB shootdown sequence.
  507.      */
  508.     tlb_shootdown_start(TLB_INVL_PAGES, as->asid, area->base, area->pages);
  509.  
  510.     /*
  511.      * Visit only the pages mapped by used_space B+tree.
  512.      */
  513.     for (cur = area->used_space.leaf_head.next;
  514.         cur != &area->used_space.leaf_head; cur = cur->next) {
  515.         btree_node_t *node;
  516.         int i;
  517.        
  518.         node = list_get_instance(cur, btree_node_t, leaf_link);
  519.         for (i = 0; i < node->keys; i++) {
  520.             uintptr_t b = node->key[i];
  521.             count_t j;
  522.             pte_t *pte;
  523.            
  524.             for (j = 0; j < (count_t) node->value[i]; j++) {
  525.                 page_table_lock(as, false);
  526.                 pte = page_mapping_find(as, b + j * PAGE_SIZE);
  527.                 ASSERT(pte && PTE_VALID(pte) &&
  528.                     PTE_PRESENT(pte));
  529.                 if (area->backend &&
  530.                     area->backend->frame_free) {
  531.                     area->backend->frame_free(area, b +
  532.                     j * PAGE_SIZE, PTE_GET_FRAME(pte));
  533.                 }
  534.                 page_mapping_remove(as, b + j * PAGE_SIZE);            
  535.                 page_table_unlock(as, false);
  536.             }
  537.         }
  538.     }
  539.  
  540.     /*
  541.      * Finish TLB shootdown sequence.
  542.      */
  543.     tlb_invalidate_pages(as->asid, area->base, area->pages);
  544.     tlb_shootdown_finalize();
  545.    
  546.     /*
  547.      * Invalidate potential software translation caches (e.g. TSB on
  548.      * sparc64).
  549.      */
  550.     as_invalidate_translation_cache(as, area->base, area->pages);
  551.    
  552.     btree_destroy(&area->used_space);
  553.  
  554.     area->attributes |= AS_AREA_ATTR_PARTIAL;
  555.    
  556.     if (area->sh_info)
  557.         sh_info_remove_reference(area->sh_info);
  558.        
  559.     mutex_unlock(&area->lock);
  560.  
  561.     /*
  562.      * Remove the empty area from address space.
  563.      */
  564.     btree_remove(&as->as_area_btree, base, NULL);
  565.    
  566.     free(area);
  567.    
  568.     mutex_unlock(&as->lock);
  569.     interrupts_restore(ipl);
  570.     return 0;
  571. }
  572.  
  573. /** Share address space area with another or the same address space.
  574.  *
  575.  * Address space area mapping is shared with a new address space area.
  576.  * If the source address space area has not been shared so far,
  577.  * a new sh_info is created. The new address space area simply gets the
  578.  * sh_info of the source area. The process of duplicating the
  579.  * mapping is done through the backend share function.
  580.  *
  581.  * @param src_as Pointer to source address space.
  582.  * @param src_base Base address of the source address space area.
  583.  * @param acc_size Expected size of the source area.
  584.  * @param dst_as Pointer to destination address space.
  585.  * @param dst_base Target base address.
  586.  * @param dst_flags_mask Destination address space area flags mask.
  587.  *
  588.  * @return Zero on success or ENOENT if there is no such task or if there is no
  589.  * such address space area, EPERM if there was a problem in accepting the area
  590.  * or ENOMEM if there was a problem in allocating destination address space
  591.  * area. ENOTSUP is returned if the address space area backend does not support
  592.  * sharing or if the kernel detects an attempt to create an illegal address
  593.  * alias.
  594.  */
  595. int as_area_share(as_t *src_as, uintptr_t src_base, size_t acc_size,
  596.           as_t *dst_as, uintptr_t dst_base, int dst_flags_mask)
  597. {
  598.     ipl_t ipl;
  599.     int src_flags;
  600.     size_t src_size;
  601.     as_area_t *src_area, *dst_area;
  602.     share_info_t *sh_info;
  603.     mem_backend_t *src_backend;
  604.     mem_backend_data_t src_backend_data;
  605.    
  606.     ipl = interrupts_disable();
  607.     mutex_lock(&src_as->lock);
  608.     src_area = find_area_and_lock(src_as, src_base);
  609.     if (!src_area) {
  610.         /*
  611.          * Could not find the source address space area.
  612.          */
  613.         mutex_unlock(&src_as->lock);
  614.         interrupts_restore(ipl);
  615.         return ENOENT;
  616.     }
  617.  
  618.     if (!src_area->backend || !src_area->backend->share) {
  619.         /*
  620.          * There is no backend or the backend does not
  621.          * know how to share the area.
  622.          */
  623.         mutex_unlock(&src_area->lock);
  624.         mutex_unlock(&src_as->lock);
  625.         interrupts_restore(ipl);
  626.         return ENOTSUP;
  627.     }
  628.    
  629.     src_size = src_area->pages * PAGE_SIZE;
  630.     src_flags = src_area->flags;
  631.     src_backend = src_area->backend;
  632.     src_backend_data = src_area->backend_data;
  633.  
  634.     /* Share the cacheable flag from the original mapping */
  635.     if (src_flags & AS_AREA_CACHEABLE)
  636.         dst_flags_mask |= AS_AREA_CACHEABLE;
  637.  
  638.     if (src_size != acc_size ||
  639.         (src_flags & dst_flags_mask) != dst_flags_mask) {
  640.         mutex_unlock(&src_area->lock);
  641.         mutex_unlock(&src_as->lock);
  642.         interrupts_restore(ipl);
  643.         return EPERM;
  644.     }
  645.  
  646. #ifdef CONFIG_VIRT_IDX_DCACHE
  647.     if (!(dst_flags_mask & AS_AREA_EXEC)) {
  648.         if (PAGE_COLOR(src_area->base) != PAGE_COLOR(dst_base)) {
  649.             /*
  650.              * Refuse to create an illegal address alias.
  651.              */
  652.             mutex_unlock(&src_area->lock);
  653.             mutex_unlock(&src_as->lock);
  654.             interrupts_restore(ipl);
  655.             return ENOTSUP;
  656.         }
  657.     }
  658. #endif /* CONFIG_VIRT_IDX_DCACHE */
  659.  
  660.     /*
  661.      * Now we are committed to sharing the area.
  662.      * First, prepare the area for sharing.
  663.      * Then it will be safe to unlock it.
  664.      */
  665.     sh_info = src_area->sh_info;
  666.     if (!sh_info) {
  667.         sh_info = (share_info_t *) malloc(sizeof(share_info_t), 0);
  668.         mutex_initialize(&sh_info->lock);
  669.         sh_info->refcount = 2;
  670.         btree_create(&sh_info->pagemap);
  671.         src_area->sh_info = sh_info;
  672.     } else {
  673.         mutex_lock(&sh_info->lock);
  674.         sh_info->refcount++;
  675.         mutex_unlock(&sh_info->lock);
  676.     }
  677.  
  678.     src_area->backend->share(src_area);
  679.  
  680.     mutex_unlock(&src_area->lock);
  681.     mutex_unlock(&src_as->lock);
  682.  
  683.     /*
  684.      * Create copy of the source address space area.
  685.      * The destination area is created with AS_AREA_ATTR_PARTIAL
  686.      * attribute set which prevents race condition with
  687.      * preliminary as_page_fault() calls.
  688.      * The flags of the source area are masked against dst_flags_mask
  689.      * to support sharing in less privileged mode.
  690.      */
  691.     dst_area = as_area_create(dst_as, dst_flags_mask, src_size, dst_base,
  692.         AS_AREA_ATTR_PARTIAL, src_backend, &src_backend_data);
  693.     if (!dst_area) {
  694.         /*
  695.          * Destination address space area could not be created.
  696.          */
  697.         sh_info_remove_reference(sh_info);
  698.        
  699.         interrupts_restore(ipl);
  700.         return ENOMEM;
  701.     }
  702.  
  703.     /*
  704.      * Now the destination address space area has been
  705.      * fully initialized. Clear the AS_AREA_ATTR_PARTIAL
  706.      * attribute and set the sh_info.
  707.      */
  708.     mutex_lock(&dst_as->lock); 
  709.     mutex_lock(&dst_area->lock);
  710.     dst_area->attributes &= ~AS_AREA_ATTR_PARTIAL;
  711.     dst_area->sh_info = sh_info;
  712.     mutex_unlock(&dst_area->lock);
  713.     mutex_unlock(&dst_as->lock);   
  714.  
  715.     interrupts_restore(ipl);
  716.    
  717.     return 0;
  718. }
  719.  
  720. /** Check access mode for address space area.
  721.  *
  722.  * The address space area must be locked prior to this call.
  723.  *
  724.  * @param area Address space area.
  725.  * @param access Access mode.
  726.  *
  727.  * @return False if access violates area's permissions, true otherwise.
  728.  */
  729. bool as_area_check_access(as_area_t *area, pf_access_t access)
  730. {
  731.     int flagmap[] = {
  732.         [PF_ACCESS_READ] = AS_AREA_READ,
  733.         [PF_ACCESS_WRITE] = AS_AREA_WRITE,
  734.         [PF_ACCESS_EXEC] = AS_AREA_EXEC
  735.     };
  736.  
  737.     if (!(area->flags & flagmap[access]))
  738.         return false;
  739.    
  740.     return true;
  741. }
  742.  
  743. /** Handle page fault within the current address space.
  744.  *
  745.  * This is the high-level page fault handler. It decides
  746.  * whether the page fault can be resolved by any backend
  747.  * and if so, it invokes the backend to resolve the page
  748.  * fault.
  749.  *
  750.  * Interrupts are assumed disabled.
  751.  *
  752.  * @param page Faulting page.
  753.  * @param access Access mode that caused the fault (i.e. read/write/exec).
  754.  * @param istate Pointer to interrupted state.
  755.  *
  756.  * @return AS_PF_FAULT on page fault, AS_PF_OK on success or AS_PF_DEFER if the
  757.  *     fault was caused by copy_to_uspace() or copy_from_uspace().
  758.  */
  759. int as_page_fault(uintptr_t page, pf_access_t access, istate_t *istate)
  760. {
  761.     pte_t *pte;
  762.     as_area_t *area;
  763.    
  764.     if (!THREAD)
  765.         return AS_PF_FAULT;
  766.        
  767.     ASSERT(AS);
  768.  
  769.     mutex_lock(&AS->lock);
  770.     area = find_area_and_lock(AS, page);   
  771.     if (!area) {
  772.         /*
  773.          * No area contained mapping for 'page'.
  774.          * Signal page fault to low-level handler.
  775.          */
  776.         mutex_unlock(&AS->lock);
  777.         goto page_fault;
  778.     }
  779.  
  780.     if (area->attributes & AS_AREA_ATTR_PARTIAL) {
  781.         /*
  782.          * The address space area is not fully initialized.
  783.          * Avoid possible race by returning error.
  784.          */
  785.         mutex_unlock(&area->lock);
  786.         mutex_unlock(&AS->lock);
  787.         goto page_fault;       
  788.     }
  789.  
  790.     if (!area->backend || !area->backend->page_fault) {
  791.         /*
  792.          * The address space area is not backed by any backend
  793.          * or the backend cannot handle page faults.
  794.          */
  795.         mutex_unlock(&area->lock);
  796.         mutex_unlock(&AS->lock);
  797.         goto page_fault;       
  798.     }
  799.  
  800.     page_table_lock(AS, false);
  801.    
  802.     /*
  803.      * To avoid race condition between two page faults
  804.      * on the same address, we need to make sure
  805.      * the mapping has not been already inserted.
  806.      */
  807.     if ((pte = page_mapping_find(AS, page))) {
  808.         if (PTE_PRESENT(pte)) {
  809.             if (((access == PF_ACCESS_READ) && PTE_READABLE(pte)) ||
  810.                 (access == PF_ACCESS_WRITE && PTE_WRITABLE(pte)) ||
  811.                 (access == PF_ACCESS_EXEC && PTE_EXECUTABLE(pte))) {
  812.                 page_table_unlock(AS, false);
  813.                 mutex_unlock(&area->lock);
  814.                 mutex_unlock(&AS->lock);
  815.                 return AS_PF_OK;
  816.             }
  817.         }
  818.     }
  819.    
  820.     /*
  821.      * Resort to the backend page fault handler.
  822.      */
  823.     if (area->backend->page_fault(area, page, access) != AS_PF_OK) {
  824.         page_table_unlock(AS, false);
  825.         mutex_unlock(&area->lock);
  826.         mutex_unlock(&AS->lock);
  827.         goto page_fault;
  828.     }
  829.    
  830.     page_table_unlock(AS, false);
  831.     mutex_unlock(&area->lock);
  832.     mutex_unlock(&AS->lock);
  833.     return AS_PF_OK;
  834.  
  835. page_fault:
  836.     if (THREAD->in_copy_from_uspace) {
  837.         THREAD->in_copy_from_uspace = false;
  838.         istate_set_retaddr(istate,
  839.             (uintptr_t) &memcpy_from_uspace_failover_address);
  840.     } else if (THREAD->in_copy_to_uspace) {
  841.         THREAD->in_copy_to_uspace = false;
  842.         istate_set_retaddr(istate,
  843.             (uintptr_t) &memcpy_to_uspace_failover_address);
  844.     } else {
  845.         return AS_PF_FAULT;
  846.     }
  847.  
  848.     return AS_PF_DEFER;
  849. }
  850.  
  851. /** Switch address spaces.
  852.  *
  853.  * Note that this function cannot sleep as it is essentially a part of
  854.  * scheduling. Sleeping here would lead to deadlock on wakeup.
  855.  *
  856.  * @param old Old address space or NULL.
  857.  * @param new New address space.
  858.  */
  859. void as_switch(as_t *old, as_t *new)
  860. {
  861.     ipl_t ipl;
  862.     bool needs_asid = false;
  863.    
  864.     ipl = interrupts_disable();
  865.     spinlock_lock(&inactive_as_with_asid_lock);
  866.  
  867.     /*
  868.      * First, take care of the old address space.
  869.      */
  870.     if (old) {
  871.         mutex_lock_active(&old->lock);
  872.         ASSERT(old->cpu_refcount);
  873.         if((--old->cpu_refcount == 0) && (old != AS_KERNEL)) {
  874.             /*
  875.              * The old address space is no longer active on
  876.              * any processor. It can be appended to the
  877.              * list of inactive address spaces with assigned
  878.              * ASID.
  879.              */
  880.              ASSERT(old->asid != ASID_INVALID);
  881.              list_append(&old->inactive_as_with_asid_link,
  882.                  &inactive_as_with_asid_head);
  883.         }
  884.         mutex_unlock(&old->lock);
  885.  
  886.         /*
  887.          * Perform architecture-specific tasks when the address space
  888.          * is being removed from the CPU.
  889.          */
  890.         as_deinstall_arch(old);
  891.     }
  892.  
  893.     /*
  894.      * Second, prepare the new address space.
  895.      */
  896.     mutex_lock_active(&new->lock);
  897.     if ((new->cpu_refcount++ == 0) && (new != AS_KERNEL)) {
  898.         if (new->asid != ASID_INVALID) {
  899.             list_remove(&new->inactive_as_with_asid_link);
  900.         } else {
  901.             /*
  902.              * Defer call to asid_get() until new->lock is released.
  903.              */
  904.             needs_asid = true;
  905.         }
  906.     }
  907.     SET_PTL0_ADDRESS(new->page_table);
  908.     mutex_unlock(&new->lock);
  909.  
  910.     if (needs_asid) {
  911.         /*
  912.          * Allocation of new ASID was deferred
  913.          * until now in order to avoid deadlock.
  914.          */
  915.         asid_t asid;
  916.        
  917.         asid = asid_get();
  918.         mutex_lock_active(&new->lock);
  919.         new->asid = asid;
  920.         mutex_unlock(&new->lock);
  921.     }
  922.     spinlock_unlock(&inactive_as_with_asid_lock);
  923.     interrupts_restore(ipl);
  924.    
  925.     /*
  926.      * Perform architecture-specific steps.
  927.      * (e.g. write ASID to hardware register etc.)
  928.      */
  929.     as_install_arch(new);
  930.    
  931.     AS = new;
  932. }
  933.  
  934. /** Convert address space area flags to page flags.
  935.  *
  936.  * @param aflags Flags of some address space area.
  937.  *
  938.  * @return Flags to be passed to page_mapping_insert().
  939.  */
  940. int area_flags_to_page_flags(int aflags)
  941. {
  942.     int flags;
  943.  
  944.     flags = PAGE_USER | PAGE_PRESENT;
  945.    
  946.     if (aflags & AS_AREA_READ)
  947.         flags |= PAGE_READ;
  948.        
  949.     if (aflags & AS_AREA_WRITE)
  950.         flags |= PAGE_WRITE;
  951.    
  952.     if (aflags & AS_AREA_EXEC)
  953.         flags |= PAGE_EXEC;
  954.    
  955.     if (aflags & AS_AREA_CACHEABLE)
  956.         flags |= PAGE_CACHEABLE;
  957.        
  958.     return flags;
  959. }
  960.  
  961. /** Compute flags for virtual address translation subsytem.
  962.  *
  963.  * The address space area must be locked.
  964.  * Interrupts must be disabled.
  965.  *
  966.  * @param a Address space area.
  967.  *
  968.  * @return Flags to be used in page_mapping_insert().
  969.  */
  970. int as_area_get_flags(as_area_t *a)
  971. {
  972.     return area_flags_to_page_flags(a->flags);
  973. }
  974.  
  975. /** Create page table.
  976.  *
  977.  * Depending on architecture, create either address space
  978.  * private or global page table.
  979.  *
  980.  * @param flags Flags saying whether the page table is for kernel address space.
  981.  *
  982.  * @return First entry of the page table.
  983.  */
  984. pte_t *page_table_create(int flags)
  985. {
  986.         ASSERT(as_operations);
  987.         ASSERT(as_operations->page_table_create);
  988.  
  989.         return as_operations->page_table_create(flags);
  990. }
  991.  
  992. /** Destroy page table.
  993.  *
  994.  * Destroy page table in architecture specific way.
  995.  *
  996.  * @param page_table Physical address of PTL0.
  997.  */
  998. void page_table_destroy(pte_t *page_table)
  999. {
  1000.         ASSERT(as_operations);
  1001.         ASSERT(as_operations->page_table_destroy);
  1002.  
  1003.         as_operations->page_table_destroy(page_table);
  1004. }
  1005.  
  1006. /** Lock page table.
  1007.  *
  1008.  * This function should be called before any page_mapping_insert(),
  1009.  * page_mapping_remove() and page_mapping_find().
  1010.  *
  1011.  * Locking order is such that address space areas must be locked
  1012.  * prior to this call. Address space can be locked prior to this
  1013.  * call in which case the lock argument is false.
  1014.  *
  1015.  * @param as Address space.
  1016.  * @param lock If false, do not attempt to lock as->lock.
  1017.  */
  1018. void page_table_lock(as_t *as, bool lock)
  1019. {
  1020.     ASSERT(as_operations);
  1021.     ASSERT(as_operations->page_table_lock);
  1022.  
  1023.     as_operations->page_table_lock(as, lock);
  1024. }
  1025.  
  1026. /** Unlock page table.
  1027.  *
  1028.  * @param as Address space.
  1029.  * @param unlock If false, do not attempt to unlock as->lock.
  1030.  */
  1031. void page_table_unlock(as_t *as, bool unlock)
  1032. {
  1033.     ASSERT(as_operations);
  1034.     ASSERT(as_operations->page_table_unlock);
  1035.  
  1036.     as_operations->page_table_unlock(as, unlock);
  1037. }
  1038.  
  1039.  
  1040. /** Find address space area and lock it.
  1041.  *
  1042.  * The address space must be locked and interrupts must be disabled.
  1043.  *
  1044.  * @param as Address space.
  1045.  * @param va Virtual address.
  1046.  *
  1047.  * @return Locked address space area containing va on success or NULL on
  1048.  *     failure.
  1049.  */
  1050. as_area_t *find_area_and_lock(as_t *as, uintptr_t va)
  1051. {
  1052.     as_area_t *a;
  1053.     btree_node_t *leaf, *lnode;
  1054.     int i;
  1055.    
  1056.     a = (as_area_t *) btree_search(&as->as_area_btree, va, &leaf);
  1057.     if (a) {
  1058.         /* va is the base address of an address space area */
  1059.         mutex_lock(&a->lock);
  1060.         return a;
  1061.     }
  1062.    
  1063.     /*
  1064.      * Search the leaf node and the righmost record of its left neighbour
  1065.      * to find out whether this is a miss or va belongs to an address
  1066.      * space area found there.
  1067.      */
  1068.    
  1069.     /* First, search the leaf node itself. */
  1070.     for (i = 0; i < leaf->keys; i++) {
  1071.         a = (as_area_t *) leaf->value[i];
  1072.         mutex_lock(&a->lock);
  1073.         if ((a->base <= va) && (va < a->base + a->pages * PAGE_SIZE)) {
  1074.             return a;
  1075.         }
  1076.         mutex_unlock(&a->lock);
  1077.     }
  1078.  
  1079.     /*
  1080.      * Second, locate the left neighbour and test its last record.
  1081.      * Because of its position in the B+tree, it must have base < va.
  1082.      */
  1083.     lnode = btree_leaf_node_left_neighbour(&as->as_area_btree, leaf);
  1084.     if (lnode) {
  1085.         a = (as_area_t *) lnode->value[lnode->keys - 1];
  1086.         mutex_lock(&a->lock);
  1087.         if (va < a->base + a->pages * PAGE_SIZE) {
  1088.             return a;
  1089.         }
  1090.         mutex_unlock(&a->lock);
  1091.     }
  1092.  
  1093.     return NULL;
  1094. }
  1095.  
  1096. /** Check area conflicts with other areas.
  1097.  *
  1098.  * The address space must be locked and interrupts must be disabled.
  1099.  *
  1100.  * @param as Address space.
  1101.  * @param va Starting virtual address of the area being tested.
  1102.  * @param size Size of the area being tested.
  1103.  * @param avoid_area Do not touch this area.
  1104.  *
  1105.  * @return True if there is no conflict, false otherwise.
  1106.  */
  1107. bool check_area_conflicts(as_t *as, uintptr_t va, size_t size,
  1108.               as_area_t *avoid_area)
  1109. {
  1110.     as_area_t *a;
  1111.     btree_node_t *leaf, *node;
  1112.     int i;
  1113.    
  1114.     /*
  1115.      * We don't want any area to have conflicts with NULL page.
  1116.      */
  1117.     if (overlaps(va, size, NULL, PAGE_SIZE))
  1118.         return false;
  1119.    
  1120.     /*
  1121.      * The leaf node is found in O(log n), where n is proportional to
  1122.      * the number of address space areas belonging to as.
  1123.      * The check for conflicts is then attempted on the rightmost
  1124.      * record in the left neighbour, the leftmost record in the right
  1125.      * neighbour and all records in the leaf node itself.
  1126.      */
  1127.    
  1128.     if ((a = (as_area_t *) btree_search(&as->as_area_btree, va, &leaf))) {
  1129.         if (a != avoid_area)
  1130.             return false;
  1131.     }
  1132.    
  1133.     /* First, check the two border cases. */
  1134.     if ((node = btree_leaf_node_left_neighbour(&as->as_area_btree, leaf))) {
  1135.         a = (as_area_t *) node->value[node->keys - 1];
  1136.         mutex_lock(&a->lock);
  1137.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1138.             mutex_unlock(&a->lock);
  1139.             return false;
  1140.         }
  1141.         mutex_unlock(&a->lock);
  1142.     }
  1143.     node = btree_leaf_node_right_neighbour(&as->as_area_btree, leaf);
  1144.     if (node) {
  1145.         a = (as_area_t *) node->value[0];
  1146.         mutex_lock(&a->lock);
  1147.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1148.             mutex_unlock(&a->lock);
  1149.             return false;
  1150.         }
  1151.         mutex_unlock(&a->lock);
  1152.     }
  1153.    
  1154.     /* Second, check the leaf node. */
  1155.     for (i = 0; i < leaf->keys; i++) {
  1156.         a = (as_area_t *) leaf->value[i];
  1157.    
  1158.         if (a == avoid_area)
  1159.             continue;
  1160.    
  1161.         mutex_lock(&a->lock);
  1162.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1163.             mutex_unlock(&a->lock);
  1164.             return false;
  1165.         }
  1166.         mutex_unlock(&a->lock);
  1167.     }
  1168.  
  1169.     /*
  1170.      * So far, the area does not conflict with other areas.
  1171.      * Check if it doesn't conflict with kernel address space.
  1172.      */  
  1173.     if (!KERNEL_ADDRESS_SPACE_SHADOWED) {
  1174.         return !overlaps(va, size,
  1175.             KERNEL_ADDRESS_SPACE_START,
  1176.             KERNEL_ADDRESS_SPACE_END - KERNEL_ADDRESS_SPACE_START);
  1177.     }
  1178.  
  1179.     return true;
  1180. }
  1181.  
  1182. /** Return size of the address space area with given base.  */
  1183. size_t as_get_size(uintptr_t base)
  1184. {
  1185.     ipl_t ipl;
  1186.     as_area_t *src_area;
  1187.     size_t size;
  1188.  
  1189.     ipl = interrupts_disable();
  1190.     src_area = find_area_and_lock(AS, base);
  1191.     if (src_area){
  1192.         size = src_area->pages * PAGE_SIZE;
  1193.         mutex_unlock(&src_area->lock);
  1194.     } else {
  1195.         size = 0;
  1196.     }
  1197.     interrupts_restore(ipl);
  1198.     return size;
  1199. }
  1200.  
  1201. /** Mark portion of address space area as used.
  1202.  *
  1203.  * The address space area must be already locked.
  1204.  *
  1205.  * @param a Address space area.
  1206.  * @param page First page to be marked.
  1207.  * @param count Number of page to be marked.
  1208.  *
  1209.  * @return 0 on failure and 1 on success.
  1210.  */
  1211. int used_space_insert(as_area_t *a, uintptr_t page, count_t count)
  1212. {
  1213.     btree_node_t *leaf, *node;
  1214.     count_t pages;
  1215.     int i;
  1216.  
  1217.     ASSERT(page == ALIGN_DOWN(page, PAGE_SIZE));
  1218.     ASSERT(count);
  1219.  
  1220.     pages = (count_t) btree_search(&a->used_space, page, &leaf);
  1221.     if (pages) {
  1222.         /*
  1223.          * We hit the beginning of some used space.
  1224.          */
  1225.         return 0;
  1226.     }
  1227.  
  1228.     if (!leaf->keys) {
  1229.         btree_insert(&a->used_space, page, (void *) count, leaf);
  1230.         return 1;
  1231.     }
  1232.  
  1233.     node = btree_leaf_node_left_neighbour(&a->used_space, leaf);
  1234.     if (node) {
  1235.         uintptr_t left_pg = node->key[node->keys - 1];
  1236.         uintptr_t right_pg = leaf->key[0];
  1237.         count_t left_cnt = (count_t) node->value[node->keys - 1];
  1238.         count_t right_cnt = (count_t) leaf->value[0];
  1239.        
  1240.         /*
  1241.          * Examine the possibility that the interval fits
  1242.          * somewhere between the rightmost interval of
  1243.          * the left neigbour and the first interval of the leaf.
  1244.          */
  1245.          
  1246.         if (page >= right_pg) {
  1247.             /* Do nothing. */
  1248.         } else if (overlaps(page, count * PAGE_SIZE, left_pg,
  1249.             left_cnt * PAGE_SIZE)) {
  1250.             /* The interval intersects with the left interval. */
  1251.             return 0;
  1252.         } else if (overlaps(page, count * PAGE_SIZE, right_pg,
  1253.             right_cnt * PAGE_SIZE)) {
  1254.             /* The interval intersects with the right interval. */
  1255.             return 0;          
  1256.         } else if ((page == left_pg + left_cnt * PAGE_SIZE) &&
  1257.             (page + count * PAGE_SIZE == right_pg)) {
  1258.             /*
  1259.              * The interval can be added by merging the two already
  1260.              * present intervals.
  1261.              */
  1262.             node->value[node->keys - 1] += count + right_cnt;
  1263.             btree_remove(&a->used_space, right_pg, leaf);
  1264.             return 1;
  1265.         } else if (page == left_pg + left_cnt * PAGE_SIZE) {
  1266.             /*
  1267.              * The interval can be added by simply growing the left
  1268.              * interval.
  1269.              */
  1270.             node->value[node->keys - 1] += count;
  1271.             return 1;
  1272.         } else if (page + count * PAGE_SIZE == right_pg) {
  1273.             /*
  1274.              * The interval can be addded by simply moving base of
  1275.              * the right interval down and increasing its size
  1276.              * accordingly.
  1277.              */
  1278.             leaf->value[0] += count;
  1279.             leaf->key[0] = page;
  1280.             return 1;
  1281.         } else {
  1282.             /*
  1283.              * The interval is between both neigbouring intervals,
  1284.              * but cannot be merged with any of them.
  1285.              */
  1286.             btree_insert(&a->used_space, page, (void *) count,
  1287.                 leaf);
  1288.             return 1;
  1289.         }
  1290.     } else if (page < leaf->key[0]) {
  1291.         uintptr_t right_pg = leaf->key[0];
  1292.         count_t right_cnt = (count_t) leaf->value[0];
  1293.    
  1294.         /*
  1295.          * Investigate the border case in which the left neighbour does
  1296.          * not exist but the interval fits from the left.
  1297.          */
  1298.          
  1299.         if (overlaps(page, count * PAGE_SIZE, right_pg,
  1300.             right_cnt * PAGE_SIZE)) {
  1301.             /* The interval intersects with the right interval. */
  1302.             return 0;
  1303.         } else if (page + count * PAGE_SIZE == right_pg) {
  1304.             /*
  1305.              * The interval can be added by moving the base of the
  1306.              * right interval down and increasing its size
  1307.              * accordingly.
  1308.              */
  1309.             leaf->key[0] = page;
  1310.             leaf->value[0] += count;
  1311.             return 1;
  1312.         } else {
  1313.             /*
  1314.              * The interval doesn't adjoin with the right interval.
  1315.              * It must be added individually.
  1316.              */
  1317.             btree_insert(&a->used_space, page, (void *) count,
  1318.                 leaf);
  1319.             return 1;
  1320.         }
  1321.     }
  1322.  
  1323.     node = btree_leaf_node_right_neighbour(&a->used_space, leaf);
  1324.     if (node) {
  1325.         uintptr_t left_pg = leaf->key[leaf->keys - 1];
  1326.         uintptr_t right_pg = node->key[0];
  1327.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1];
  1328.         count_t right_cnt = (count_t) node->value[0];
  1329.        
  1330.         /*
  1331.          * Examine the possibility that the interval fits
  1332.          * somewhere between the leftmost interval of
  1333.          * the right neigbour and the last interval of the leaf.
  1334.          */
  1335.  
  1336.         if (page < left_pg) {
  1337.             /* Do nothing. */
  1338.         } else if (overlaps(page, count * PAGE_SIZE, left_pg,
  1339.             left_cnt * PAGE_SIZE)) {
  1340.             /* The interval intersects with the left interval. */
  1341.             return 0;
  1342.         } else if (overlaps(page, count * PAGE_SIZE, right_pg,
  1343.             right_cnt * PAGE_SIZE)) {
  1344.             /* The interval intersects with the right interval. */
  1345.             return 0;          
  1346.         } else if ((page == left_pg + left_cnt * PAGE_SIZE) &&
  1347.             (page + count * PAGE_SIZE == right_pg)) {
  1348.             /*
  1349.              * The interval can be added by merging the two already
  1350.              * present intervals.
  1351.              * */
  1352.             leaf->value[leaf->keys - 1] += count + right_cnt;
  1353.             btree_remove(&a->used_space, right_pg, node);
  1354.             return 1;
  1355.         } else if (page == left_pg + left_cnt * PAGE_SIZE) {
  1356.             /*
  1357.              * The interval can be added by simply growing the left
  1358.              * interval.
  1359.              * */
  1360.             leaf->value[leaf->keys - 1] +=  count;
  1361.             return 1;
  1362.         } else if (page + count * PAGE_SIZE == right_pg) {
  1363.             /*
  1364.              * The interval can be addded by simply moving base of
  1365.              * the right interval down and increasing its size
  1366.              * accordingly.
  1367.              */
  1368.             node->value[0] += count;
  1369.             node->key[0] = page;
  1370.             return 1;
  1371.         } else {
  1372.             /*
  1373.              * The interval is between both neigbouring intervals,
  1374.              * but cannot be merged with any of them.
  1375.              */
  1376.             btree_insert(&a->used_space, page, (void *) count,
  1377.                 leaf);
  1378.             return 1;
  1379.         }
  1380.     } else if (page >= leaf->key[leaf->keys - 1]) {
  1381.         uintptr_t left_pg = leaf->key[leaf->keys - 1];
  1382.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1];
  1383.    
  1384.         /*
  1385.          * Investigate the border case in which the right neighbour
  1386.          * does not exist but the interval fits from the right.
  1387.          */
  1388.          
  1389.         if (overlaps(page, count * PAGE_SIZE, left_pg,
  1390.             left_cnt * PAGE_SIZE)) {
  1391.             /* The interval intersects with the left interval. */
  1392.             return 0;
  1393.         } else if (left_pg + left_cnt * PAGE_SIZE == page) {
  1394.             /*
  1395.              * The interval can be added by growing the left
  1396.              * interval.
  1397.              */
  1398.             leaf->value[leaf->keys - 1] += count;
  1399.             return 1;
  1400.         } else {
  1401.             /*
  1402.              * The interval doesn't adjoin with the left interval.
  1403.              * It must be added individually.
  1404.              */
  1405.             btree_insert(&a->used_space, page, (void *) count,
  1406.                 leaf);
  1407.             return 1;
  1408.         }
  1409.     }
  1410.    
  1411.     /*
  1412.      * Note that if the algorithm made it thus far, the interval can fit
  1413.      * only between two other intervals of the leaf. The two border cases
  1414.      * were already resolved.
  1415.      */
  1416.     for (i = 1; i < leaf->keys; i++) {
  1417.         if (page < leaf->key[i]) {
  1418.             uintptr_t left_pg = leaf->key[i - 1];
  1419.             uintptr_t right_pg = leaf->key[i];
  1420.             count_t left_cnt = (count_t) leaf->value[i - 1];
  1421.             count_t right_cnt = (count_t) leaf->value[i];
  1422.  
  1423.             /*
  1424.              * The interval fits between left_pg and right_pg.
  1425.              */
  1426.  
  1427.             if (overlaps(page, count * PAGE_SIZE, left_pg,
  1428.                 left_cnt * PAGE_SIZE)) {
  1429.                 /*
  1430.                  * The interval intersects with the left
  1431.                  * interval.
  1432.                  */
  1433.                 return 0;
  1434.             } else if (overlaps(page, count * PAGE_SIZE, right_pg,
  1435.                 right_cnt * PAGE_SIZE)) {
  1436.                 /*
  1437.                  * The interval intersects with the right
  1438.                  * interval.
  1439.                  */
  1440.                 return 0;          
  1441.             } else if ((page == left_pg + left_cnt * PAGE_SIZE) &&
  1442.                 (page + count * PAGE_SIZE == right_pg)) {
  1443.                 /*
  1444.                  * The interval can be added by merging the two
  1445.                  * already present intervals.
  1446.                  */
  1447.                 leaf->value[i - 1] += count + right_cnt;
  1448.                 btree_remove(&a->used_space, right_pg, leaf);
  1449.                 return 1;
  1450.             } else if (page == left_pg + left_cnt * PAGE_SIZE) {
  1451.                 /*
  1452.                  * The interval can be added by simply growing
  1453.                  * the left interval.
  1454.                  */
  1455.                 leaf->value[i - 1] += count;
  1456.                 return 1;
  1457.             } else if (page + count * PAGE_SIZE == right_pg) {
  1458.                 /*
  1459.                      * The interval can be addded by simply moving
  1460.                  * base of the right interval down and
  1461.                  * increasing its size accordingly.
  1462.                  */
  1463.                 leaf->value[i] += count;
  1464.                 leaf->key[i] = page;
  1465.                 return 1;
  1466.             } else {
  1467.                 /*
  1468.                  * The interval is between both neigbouring
  1469.                  * intervals, but cannot be merged with any of
  1470.                  * them.
  1471.                  */
  1472.                 btree_insert(&a->used_space, page,
  1473.                     (void *) count, leaf);
  1474.                 return 1;
  1475.             }
  1476.         }
  1477.     }
  1478.  
  1479.     panic("Inconsistency detected while adding %d pages of used space at "
  1480.         "%p.\n", count, page);
  1481. }
  1482.  
  1483. /** Mark portion of address space area as unused.
  1484.  *
  1485.  * The address space area must be already locked.
  1486.  *
  1487.  * @param a Address space area.
  1488.  * @param page First page to be marked.
  1489.  * @param count Number of page to be marked.
  1490.  *
  1491.  * @return 0 on failure and 1 on success.
  1492.  */
  1493. int used_space_remove(as_area_t *a, uintptr_t page, count_t count)
  1494. {
  1495.     btree_node_t *leaf, *node;
  1496.     count_t pages;
  1497.     int i;
  1498.  
  1499.     ASSERT(page == ALIGN_DOWN(page, PAGE_SIZE));
  1500.     ASSERT(count);
  1501.  
  1502.     pages = (count_t) btree_search(&a->used_space, page, &leaf);
  1503.     if (pages) {
  1504.         /*
  1505.          * We are lucky, page is the beginning of some interval.
  1506.          */
  1507.         if (count > pages) {
  1508.             return 0;
  1509.         } else if (count == pages) {
  1510.             btree_remove(&a->used_space, page, leaf);
  1511.             return 1;
  1512.         } else {
  1513.             /*
  1514.              * Find the respective interval.
  1515.              * Decrease its size and relocate its start address.
  1516.              */
  1517.             for (i = 0; i < leaf->keys; i++) {
  1518.                 if (leaf->key[i] == page) {
  1519.                     leaf->key[i] += count * PAGE_SIZE;
  1520.                     leaf->value[i] -= count;
  1521.                     return 1;
  1522.                 }
  1523.             }
  1524.             goto error;
  1525.         }
  1526.     }
  1527.  
  1528.     node = btree_leaf_node_left_neighbour(&a->used_space, leaf);
  1529.     if (node && page < leaf->key[0]) {
  1530.         uintptr_t left_pg = node->key[node->keys - 1];
  1531.         count_t left_cnt = (count_t) node->value[node->keys - 1];
  1532.  
  1533.         if (overlaps(left_pg, left_cnt * PAGE_SIZE, page,
  1534.             count * PAGE_SIZE)) {
  1535.             if (page + count * PAGE_SIZE ==
  1536.                 left_pg + left_cnt * PAGE_SIZE) {
  1537.                 /*
  1538.                  * The interval is contained in the rightmost
  1539.                  * interval of the left neighbour and can be
  1540.                  * removed by updating the size of the bigger
  1541.                  * interval.
  1542.                  */
  1543.                 node->value[node->keys - 1] -= count;
  1544.                 return 1;
  1545.             } else if (page + count * PAGE_SIZE <
  1546.                 left_pg + left_cnt*PAGE_SIZE) {
  1547.                 count_t new_cnt;
  1548.                
  1549.                 /*
  1550.                  * The interval is contained in the rightmost
  1551.                  * interval of the left neighbour but its
  1552.                  * removal requires both updating the size of
  1553.                  * the original interval and also inserting a
  1554.                  * new interval.
  1555.                  */
  1556.                 new_cnt = ((left_pg + left_cnt * PAGE_SIZE) -
  1557.                     (page + count*PAGE_SIZE)) >> PAGE_WIDTH;
  1558.                 node->value[node->keys - 1] -= count + new_cnt;
  1559.                 btree_insert(&a->used_space, page +
  1560.                     count * PAGE_SIZE, (void *) new_cnt, leaf);
  1561.                 return 1;
  1562.             }
  1563.         }
  1564.         return 0;
  1565.     } else if (page < leaf->key[0]) {
  1566.         return 0;
  1567.     }
  1568.    
  1569.     if (page > leaf->key[leaf->keys - 1]) {
  1570.         uintptr_t left_pg = leaf->key[leaf->keys - 1];
  1571.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1];
  1572.  
  1573.         if (overlaps(left_pg, left_cnt * PAGE_SIZE, page,
  1574.             count * PAGE_SIZE)) {
  1575.             if (page + count * PAGE_SIZE ==
  1576.                 left_pg + left_cnt * PAGE_SIZE) {
  1577.                 /*
  1578.                  * The interval is contained in the rightmost
  1579.                  * interval of the leaf and can be removed by
  1580.                  * updating the size of the bigger interval.
  1581.                  */
  1582.                 leaf->value[leaf->keys - 1] -= count;
  1583.                 return 1;
  1584.             } else if (page + count * PAGE_SIZE < left_pg +
  1585.                 left_cnt * PAGE_SIZE) {
  1586.                 count_t new_cnt;
  1587.                
  1588.                 /*
  1589.                  * The interval is contained in the rightmost
  1590.                  * interval of the leaf but its removal
  1591.                  * requires both updating the size of the
  1592.                  * original interval and also inserting a new
  1593.                  * interval.
  1594.                  */
  1595.                 new_cnt = ((left_pg + left_cnt * PAGE_SIZE) -
  1596.                     (page + count * PAGE_SIZE)) >> PAGE_WIDTH;
  1597.                 leaf->value[leaf->keys - 1] -= count + new_cnt;
  1598.                 btree_insert(&a->used_space, page +
  1599.                     count * PAGE_SIZE, (void *) new_cnt, leaf);
  1600.                 return 1;
  1601.             }
  1602.         }
  1603.         return 0;
  1604.     }  
  1605.    
  1606.     /*
  1607.      * The border cases have been already resolved.
  1608.      * Now the interval can be only between intervals of the leaf.
  1609.      */
  1610.     for (i = 1; i < leaf->keys - 1; i++) {
  1611.         if (page < leaf->key[i]) {
  1612.             uintptr_t left_pg = leaf->key[i - 1];
  1613.             count_t left_cnt = (count_t) leaf->value[i - 1];
  1614.  
  1615.             /*
  1616.              * Now the interval is between intervals corresponding
  1617.              * to (i - 1) and i.
  1618.              */
  1619.             if (overlaps(left_pg, left_cnt * PAGE_SIZE, page,
  1620.                 count * PAGE_SIZE)) {
  1621.                 if (page + count * PAGE_SIZE ==
  1622.                     left_pg + left_cnt*PAGE_SIZE) {
  1623.                     /*
  1624.                      * The interval is contained in the
  1625.                      * interval (i - 1) of the leaf and can
  1626.                      * be removed by updating the size of
  1627.                      * the bigger interval.
  1628.                      */
  1629.                     leaf->value[i - 1] -= count;
  1630.                     return 1;
  1631.                 } else if (page + count * PAGE_SIZE <
  1632.                     left_pg + left_cnt * PAGE_SIZE) {
  1633.                     count_t new_cnt;
  1634.                
  1635.                     /*
  1636.                      * The interval is contained in the
  1637.                      * interval (i - 1) of the leaf but its
  1638.                      * removal requires both updating the
  1639.                      * size of the original interval and
  1640.                      * also inserting a new interval.
  1641.                      */
  1642.                     new_cnt = ((left_pg +
  1643.                         left_cnt * PAGE_SIZE) -
  1644.                         (page + count * PAGE_SIZE)) >>
  1645.                         PAGE_WIDTH;
  1646.                     leaf->value[i - 1] -= count + new_cnt;
  1647.                     btree_insert(&a->used_space, page +
  1648.                         count * PAGE_SIZE, (void *) new_cnt,
  1649.                         leaf);
  1650.                     return 1;
  1651.                 }
  1652.             }
  1653.             return 0;
  1654.         }
  1655.     }
  1656.  
  1657. error:
  1658.     panic("Inconsistency detected while removing %d pages of used space "
  1659.         "from %p.\n", count, page);
  1660. }
  1661.  
  1662. /** Remove reference to address space area share info.
  1663.  *
  1664.  * If the reference count drops to 0, the sh_info is deallocated.
  1665.  *
  1666.  * @param sh_info Pointer to address space area share info.
  1667.  */
  1668. void sh_info_remove_reference(share_info_t *sh_info)
  1669. {
  1670.     bool dealloc = false;
  1671.  
  1672.     mutex_lock(&sh_info->lock);
  1673.     ASSERT(sh_info->refcount);
  1674.     if (--sh_info->refcount == 0) {
  1675.         dealloc = true;
  1676.         link_t *cur;
  1677.        
  1678.         /*
  1679.          * Now walk carefully the pagemap B+tree and free/remove
  1680.          * reference from all frames found there.
  1681.          */
  1682.         for (cur = sh_info->pagemap.leaf_head.next;
  1683.             cur != &sh_info->pagemap.leaf_head; cur = cur->next) {
  1684.             btree_node_t *node;
  1685.             int i;
  1686.            
  1687.             node = list_get_instance(cur, btree_node_t, leaf_link);
  1688.             for (i = 0; i < node->keys; i++)
  1689.                 frame_free((uintptr_t) node->value[i]);
  1690.         }
  1691.        
  1692.     }
  1693.     mutex_unlock(&sh_info->lock);
  1694.    
  1695.     if (dealloc) {
  1696.         btree_destroy(&sh_info->pagemap);
  1697.         free(sh_info);
  1698.     }
  1699. }
  1700.  
  1701. /*
  1702.  * Address space related syscalls.
  1703.  */
  1704.  
  1705. /** Wrapper for as_area_create(). */
  1706. unative_t sys_as_area_create(uintptr_t address, size_t size, int flags)
  1707. {
  1708.     if (as_area_create(AS, flags | AS_AREA_CACHEABLE, size, address,
  1709.         AS_AREA_ATTR_NONE, &anon_backend, NULL))
  1710.         return (unative_t) address;
  1711.     else
  1712.         return (unative_t) -1;
  1713. }
  1714.  
  1715. /** Wrapper for as_area_resize(). */
  1716. unative_t sys_as_area_resize(uintptr_t address, size_t size, int flags)
  1717. {
  1718.     return (unative_t) as_area_resize(AS, address, size, 0);
  1719. }
  1720.  
  1721. /** Wrapper for as_area_destroy(). */
  1722. unative_t sys_as_area_destroy(uintptr_t address)
  1723. {
  1724.     return (unative_t) as_area_destroy(AS, address);
  1725. }
  1726.  
  1727. /** Print out information about address space.
  1728.  *
  1729.  * @param as Address space.
  1730.  */
  1731. void as_print(as_t *as)
  1732. {
  1733.     ipl_t ipl;
  1734.    
  1735.     ipl = interrupts_disable();
  1736.     mutex_lock(&as->lock);
  1737.    
  1738.     /* print out info about address space areas */
  1739.     link_t *cur;
  1740.     for (cur = as->as_area_btree.leaf_head.next;
  1741.         cur != &as->as_area_btree.leaf_head; cur = cur->next) {
  1742.         btree_node_t *node;
  1743.        
  1744.         node = list_get_instance(cur, btree_node_t, leaf_link);
  1745.        
  1746.         int i;
  1747.         for (i = 0; i < node->keys; i++) {
  1748.             as_area_t *area = node->value[i];
  1749.        
  1750.             mutex_lock(&area->lock);
  1751.             printf("as_area: %p, base=%p, pages=%d (%p - %p)\n",
  1752.                 area, area->base, area->pages, area->base,
  1753.                 area->base + area->pages*PAGE_SIZE);
  1754.             mutex_unlock(&area->lock);
  1755.         }
  1756.     }
  1757.    
  1758.     mutex_unlock(&as->lock);
  1759.     interrupts_restore(ipl);
  1760. }
  1761.  
  1762. /** @}
  1763.  */
  1764.