Subversion Repositories HelenOS

Rev

Rev 1954 | Rev 2009 | 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. /**
  82.  * Each architecture decides what functions will be used to carry out
  83.  * address space operations such as creating or locking page tables.
  84.  */
  85. as_operations_t *as_operations = NULL;
  86.  
  87. /**
  88.  * Slab for as_t objects.
  89.  */
  90. static slab_cache_t *as_slab;
  91.  
  92. /** This lock protects inactive_as_with_asid_head list. It must be acquired before as_t mutex. */
  93. SPINLOCK_INITIALIZE(inactive_as_with_asid_lock);
  94.  
  95. /**
  96.  * This list contains address spaces that are not active on any
  97.  * processor and that have valid ASID.
  98.  */
  99. LIST_INITIALIZE(inactive_as_with_asid_head);
  100.  
  101. /** Kernel address space. */
  102. as_t *AS_KERNEL = NULL;
  103.  
  104. static int area_flags_to_page_flags(int aflags);
  105. static as_area_t *find_area_and_lock(as_t *as, uintptr_t va);
  106. static bool check_area_conflicts(as_t *as, uintptr_t va, size_t size, as_area_t *avoid_area);
  107. static void sh_info_remove_reference(share_info_t *sh_info);
  108.  
  109. static int as_constructor(void *obj, int flags)
  110. {
  111.     as_t *as = (as_t *) obj;
  112.     int rc;
  113.  
  114.     link_initialize(&as->inactive_as_with_asid_link);
  115.     mutex_initialize(&as->lock);   
  116.    
  117.     rc = as_constructor_arch(as, flags);
  118.    
  119.     return rc;
  120. }
  121.  
  122. static int as_destructor(void *obj)
  123. {
  124.     as_t *as = (as_t *) obj;
  125.  
  126.     return as_destructor_arch(as);
  127. }
  128.  
  129. /** Initialize address space subsystem. */
  130. void as_init(void)
  131. {
  132.     as_arch_init();
  133.    
  134.     as_slab = slab_cache_create("as_slab", sizeof(as_t), 0,
  135.         as_constructor, as_destructor, SLAB_CACHE_MAGDEFERRED);
  136.    
  137.     AS_KERNEL = as_create(FLAG_AS_KERNEL);
  138.     if (!AS_KERNEL)
  139.         panic("can't create kernel address space\n");
  140.    
  141. }
  142.  
  143. /** Create address space.
  144.  *
  145.  * @param flags Flags that influence way in wich the address space is created.
  146.  */
  147. as_t *as_create(int flags)
  148. {
  149.     as_t *as;
  150.  
  151.     as = (as_t *) slab_alloc(as_slab, 0);
  152.     (void) as_create_arch(as, 0);
  153.    
  154.     btree_create(&as->as_area_btree);
  155.    
  156.     if (flags & FLAG_AS_KERNEL)
  157.         as->asid = ASID_KERNEL;
  158.     else
  159.         as->asid = ASID_INVALID;
  160.    
  161.     as->refcount = 0;
  162.     as->cpu_refcount = 0;
  163.     as->page_table = page_table_create(flags);
  164.  
  165.     return as;
  166. }
  167.  
  168. /** Destroy adress space.
  169.  *
  170.  * When there are no tasks referencing this address space (i.e. its refcount is zero),
  171.  * the address space can be destroyed.
  172.  */
  173. void as_destroy(as_t *as)
  174. {
  175.     ipl_t ipl;
  176.     bool cond;
  177.  
  178.     ASSERT(as->refcount == 0);
  179.    
  180.     /*
  181.      * Since there is no reference to this area,
  182.      * it is safe not to lock its mutex.
  183.      */
  184.     ipl = interrupts_disable();
  185.     spinlock_lock(&inactive_as_with_asid_lock);
  186.     if (as->asid != ASID_INVALID && as != AS_KERNEL) {
  187.         if (as != AS && as->cpu_refcount == 0)
  188.             list_remove(&as->inactive_as_with_asid_link);
  189.         asid_put(as->asid);
  190.     }
  191.     spinlock_unlock(&inactive_as_with_asid_lock);
  192.  
  193.     /*
  194.      * Destroy address space areas of the address space.
  195.      * The B+tree must be walked carefully because it is
  196.      * also being destroyed.
  197.      */
  198.     for (cond = true; cond; ) {
  199.         btree_node_t *node;
  200.  
  201.         ASSERT(!list_empty(&as->as_area_btree.leaf_head));
  202.         node = list_get_instance(as->as_area_btree.leaf_head.next, btree_node_t, leaf_link);
  203.  
  204.         if ((cond = node->keys)) {
  205.             as_area_destroy(as, node->key[0]);
  206.         }
  207.     }
  208.  
  209.     btree_destroy(&as->as_area_btree);
  210.     page_table_destroy(as->page_table);
  211.  
  212.     interrupts_restore(ipl);
  213.    
  214.     slab_free(as_slab, as);
  215. }
  216.  
  217. /** Create address space area of common attributes.
  218.  *
  219.  * The created address space area is added to the target address space.
  220.  *
  221.  * @param as Target address space.
  222.  * @param flags Flags of the area memory.
  223.  * @param size Size of area.
  224.  * @param base Base address of area.
  225.  * @param attrs Attributes of the area.
  226.  * @param backend Address space area backend. NULL if no backend is used.
  227.  * @param backend_data NULL or a pointer to an array holding two void *.
  228.  *
  229.  * @return Address space area on success or NULL on failure.
  230.  */
  231. as_area_t *as_area_create(as_t *as, int flags, size_t size, uintptr_t base, int attrs,
  232.            mem_backend_t *backend, mem_backend_data_t *backend_data)
  233. {
  234.     ipl_t ipl;
  235.     as_area_t *a;
  236.    
  237.     if (base % PAGE_SIZE)
  238.         return NULL;
  239.  
  240.     if (!size)
  241.         return NULL;
  242.  
  243.     /* Writeable executable areas are not supported. */
  244.     if ((flags & AS_AREA_EXEC) && (flags & AS_AREA_WRITE))
  245.         return NULL;
  246.    
  247.     ipl = interrupts_disable();
  248.     mutex_lock(&as->lock);
  249.    
  250.     if (!check_area_conflicts(as, base, size, NULL)) {
  251.         mutex_unlock(&as->lock);
  252.         interrupts_restore(ipl);
  253.         return NULL;
  254.     }
  255.    
  256.     a = (as_area_t *) malloc(sizeof(as_area_t), 0);
  257.  
  258.     mutex_initialize(&a->lock);
  259.    
  260.     a->as = as;
  261.     a->flags = flags;
  262.     a->attributes = attrs;
  263.     a->pages = SIZE2FRAMES(size);
  264.     a->base = base;
  265.     a->sh_info = NULL;
  266.     a->backend = backend;
  267.     if (backend_data)
  268.         a->backend_data = *backend_data;
  269.     else
  270.         memsetb((uintptr_t) &a->backend_data, sizeof(a->backend_data), 0);
  271.  
  272.     btree_create(&a->used_space);
  273.    
  274.     btree_insert(&as->as_area_btree, base, (void *) a, NULL);
  275.  
  276.     mutex_unlock(&as->lock);
  277.     interrupts_restore(ipl);
  278.  
  279.     return a;
  280. }
  281.  
  282. /** Find address space area and change it.
  283.  *
  284.  * @param as Address space.
  285.  * @param address Virtual address belonging to the area to be changed. Must be page-aligned.
  286.  * @param size New size of the virtual memory block starting at address.
  287.  * @param flags Flags influencing the remap operation. Currently unused.
  288.  *
  289.  * @return Zero on success or a value from @ref errno.h otherwise.
  290.  */
  291. int as_area_resize(as_t *as, uintptr_t address, size_t size, int flags)
  292. {
  293.     as_area_t *area;
  294.     ipl_t ipl;
  295.     size_t pages;
  296.    
  297.     ipl = interrupts_disable();
  298.     mutex_lock(&as->lock);
  299.    
  300.     /*
  301.      * Locate the area.
  302.      */
  303.     area = find_area_and_lock(as, address);
  304.     if (!area) {
  305.         mutex_unlock(&as->lock);
  306.         interrupts_restore(ipl);
  307.         return ENOENT;
  308.     }
  309.  
  310.     if (area->backend == &phys_backend) {
  311.         /*
  312.          * Remapping of address space areas associated
  313.          * with memory mapped devices is not supported.
  314.          */
  315.         mutex_unlock(&area->lock);
  316.         mutex_unlock(&as->lock);
  317.         interrupts_restore(ipl);
  318.         return ENOTSUP;
  319.     }
  320.     if (area->sh_info) {
  321.         /*
  322.          * Remapping of shared address space areas
  323.          * is not supported.
  324.          */
  325.         mutex_unlock(&area->lock);
  326.         mutex_unlock(&as->lock);
  327.         interrupts_restore(ipl);
  328.         return ENOTSUP;
  329.     }
  330.  
  331.     pages = SIZE2FRAMES((address - area->base) + size);
  332.     if (!pages) {
  333.         /*
  334.          * Zero size address space areas are not allowed.
  335.          */
  336.         mutex_unlock(&area->lock);
  337.         mutex_unlock(&as->lock);
  338.         interrupts_restore(ipl);
  339.         return EPERM;
  340.     }
  341.    
  342.     if (pages < area->pages) {
  343.         bool cond;
  344.         uintptr_t start_free = area->base + pages*PAGE_SIZE;
  345.  
  346.         /*
  347.          * Shrinking the area.
  348.          * No need to check for overlaps.
  349.          */
  350.  
  351.         /*
  352.          * Start TLB shootdown sequence.
  353.          */
  354.         tlb_shootdown_start(TLB_INVL_PAGES, AS->asid, area->base + pages*PAGE_SIZE, area->pages - pages);
  355.  
  356.         /*
  357.          * Remove frames belonging to used space starting from
  358.          * the highest addresses downwards until an overlap with
  359.          * the resized address space area is found. Note that this
  360.          * is also the right way to remove part of the used_space
  361.          * B+tree leaf list.
  362.          */    
  363.         for (cond = true; cond;) {
  364.             btree_node_t *node;
  365.        
  366.             ASSERT(!list_empty(&area->used_space.leaf_head));
  367.             node = list_get_instance(area->used_space.leaf_head.prev, btree_node_t, leaf_link);
  368.             if ((cond = (bool) node->keys)) {
  369.                 uintptr_t b = node->key[node->keys - 1];
  370.                 count_t c = (count_t) node->value[node->keys - 1];
  371.                 int i = 0;
  372.            
  373.                 if (overlaps(b, c*PAGE_SIZE, area->base, pages*PAGE_SIZE)) {
  374.                    
  375.                     if (b + c*PAGE_SIZE <= start_free) {
  376.                         /*
  377.                          * The whole interval fits completely
  378.                          * in the resized address space area.
  379.                          */
  380.                         break;
  381.                     }
  382.        
  383.                     /*
  384.                      * Part of the interval corresponding to b and c
  385.                      * overlaps with the resized address space area.
  386.                      */
  387.        
  388.                     cond = false;   /* we are almost done */
  389.                     i = (start_free - b) >> PAGE_WIDTH;
  390.                     if (!used_space_remove(area, start_free, c - i))
  391.                         panic("Could not remove used space.\n");
  392.                 } else {
  393.                     /*
  394.                      * The interval of used space can be completely removed.
  395.                      */
  396.                     if (!used_space_remove(area, b, c))
  397.                         panic("Could not remove used space.\n");
  398.                 }
  399.            
  400.                 for (; i < c; i++) {
  401.                     pte_t *pte;
  402.            
  403.                     page_table_lock(as, false);
  404.                     pte = page_mapping_find(as, b + i*PAGE_SIZE);
  405.                     ASSERT(pte && PTE_VALID(pte) && PTE_PRESENT(pte));
  406.                     if (area->backend && area->backend->frame_free) {
  407.                         area->backend->frame_free(area,
  408.                             b + i*PAGE_SIZE, PTE_GET_FRAME(pte));
  409.                     }
  410.                     page_mapping_remove(as, b + i*PAGE_SIZE);
  411.                     page_table_unlock(as, false);
  412.                 }
  413.             }
  414.         }
  415.  
  416.         /*
  417.          * Finish TLB shootdown sequence.
  418.          */
  419.         tlb_invalidate_pages(as->asid, area->base + pages*PAGE_SIZE, area->pages - pages);
  420.         tlb_shootdown_finalize();
  421.        
  422.         /*
  423.          * Invalidate software translation caches (e.g. TSB on sparc64).
  424.          */
  425.         as_invalidate_translation_cache(as, area->base + pages*PAGE_SIZE, area->pages - pages);
  426.     } else {
  427.         /*
  428.          * Growing the area.
  429.          * Check for overlaps with other address space areas.
  430.          */
  431.         if (!check_area_conflicts(as, address, pages * PAGE_SIZE, area)) {
  432.             mutex_unlock(&area->lock);
  433.             mutex_unlock(&as->lock);       
  434.             interrupts_restore(ipl);
  435.             return EADDRNOTAVAIL;
  436.         }
  437.     }
  438.  
  439.     area->pages = pages;
  440.    
  441.     mutex_unlock(&area->lock);
  442.     mutex_unlock(&as->lock);
  443.     interrupts_restore(ipl);
  444.  
  445.     return 0;
  446. }
  447.  
  448. /** Destroy address space area.
  449.  *
  450.  * @param as Address space.
  451.  * @param address Address withing the area to be deleted.
  452.  *
  453.  * @return Zero on success or a value from @ref errno.h on failure.
  454.  */
  455. int as_area_destroy(as_t *as, uintptr_t address)
  456. {
  457.     as_area_t *area;
  458.     uintptr_t base;
  459.     link_t *cur;
  460.     ipl_t ipl;
  461.  
  462.     ipl = interrupts_disable();
  463.     mutex_lock(&as->lock);
  464.  
  465.     area = find_area_and_lock(as, address);
  466.     if (!area) {
  467.         mutex_unlock(&as->lock);
  468.         interrupts_restore(ipl);
  469.         return ENOENT;
  470.     }
  471.  
  472.     base = area->base;
  473.  
  474.     /*
  475.      * Start TLB shootdown sequence.
  476.      */
  477.     tlb_shootdown_start(TLB_INVL_PAGES, as->asid, area->base, area->pages);
  478.  
  479.     /*
  480.      * Visit only the pages mapped by used_space B+tree.
  481.      */
  482.     for (cur = area->used_space.leaf_head.next; cur != &area->used_space.leaf_head; cur = cur->next) {
  483.         btree_node_t *node;
  484.         int i;
  485.        
  486.         node = list_get_instance(cur, btree_node_t, leaf_link);
  487.         for (i = 0; i < node->keys; i++) {
  488.             uintptr_t b = node->key[i];
  489.             count_t j;
  490.             pte_t *pte;
  491.            
  492.             for (j = 0; j < (count_t) node->value[i]; j++) {
  493.                 page_table_lock(as, false);
  494.                 pte = page_mapping_find(as, b + j*PAGE_SIZE);
  495.                 ASSERT(pte && PTE_VALID(pte) && PTE_PRESENT(pte));
  496.                 if (area->backend && area->backend->frame_free) {
  497.                     area->backend->frame_free(area,
  498.                         b + j*PAGE_SIZE, PTE_GET_FRAME(pte));
  499.                 }
  500.                 page_mapping_remove(as, b + j*PAGE_SIZE);              
  501.                 page_table_unlock(as, false);
  502.             }
  503.         }
  504.     }
  505.  
  506.     /*
  507.      * Finish TLB shootdown sequence.
  508.      */
  509.     tlb_invalidate_pages(as->asid, area->base, area->pages);
  510.     tlb_shootdown_finalize();
  511.    
  512.     /*
  513.      * Invalidate potential software translation caches (e.g. TSB on sparc64).
  514.      */
  515.     as_invalidate_translation_cache(as, area->base, area->pages);
  516.    
  517.     btree_destroy(&area->used_space);
  518.  
  519.     area->attributes |= AS_AREA_ATTR_PARTIAL;
  520.    
  521.     if (area->sh_info)
  522.         sh_info_remove_reference(area->sh_info);
  523.        
  524.     mutex_unlock(&area->lock);
  525.  
  526.     /*
  527.      * Remove the empty area from address space.
  528.      */
  529.     btree_remove(&as->as_area_btree, base, NULL);
  530.    
  531.     free(area);
  532.    
  533.     mutex_unlock(&as->lock);
  534.     interrupts_restore(ipl);
  535.     return 0;
  536. }
  537.  
  538. /** Share address space area with another or the same address space.
  539.  *
  540.  * Address space area mapping is shared with a new address space area.
  541.  * If the source address space area has not been shared so far,
  542.  * a new sh_info is created. The new address space area simply gets the
  543.  * sh_info of the source area. The process of duplicating the
  544.  * mapping is done through the backend share function.
  545.  *
  546.  * @param src_as Pointer to source address space.
  547.  * @param src_base Base address of the source address space area.
  548.  * @param acc_size Expected size of the source area.
  549.  * @param dst_as Pointer to destination address space.
  550.  * @param dst_base Target base address.
  551.  * @param dst_flags_mask Destination address space area flags mask.
  552.  *
  553.  * @return Zero on success or ENOENT if there is no such task or if there is no
  554.  * such address space area, EPERM if there was a problem in accepting the area
  555.  * or ENOMEM if there was a problem in allocating destination address space
  556.  * area. ENOTSUP is returned if the address space area backend does not support
  557.  * sharing. It can be also returned if the architecture uses virtually indexed
  558.  * caches and the source and destination areas start at pages with different
  559.  * page colors.
  560.  */
  561. int as_area_share(as_t *src_as, uintptr_t src_base, size_t acc_size,
  562.           as_t *dst_as, uintptr_t dst_base, int dst_flags_mask)
  563. {
  564.     ipl_t ipl;
  565.     int src_flags;
  566.     size_t src_size;
  567.     as_area_t *src_area, *dst_area;
  568.     share_info_t *sh_info;
  569.     mem_backend_t *src_backend;
  570.     mem_backend_data_t src_backend_data;
  571.    
  572.     ipl = interrupts_disable();
  573.     mutex_lock(&src_as->lock);
  574.     src_area = find_area_and_lock(src_as, src_base);
  575.     if (!src_area) {
  576.         /*
  577.          * Could not find the source address space area.
  578.          */
  579.         mutex_unlock(&src_as->lock);
  580.         interrupts_restore(ipl);
  581.         return ENOENT;
  582.     }
  583.    
  584. #if 0   /* disable the check for now */
  585. #ifdef CONFIG_VIRT_IDX_CACHE
  586.     if (PAGE_COLOR(src_area->base) != PAGE_COLOR(dst_base)) {
  587.         /*
  588.          * Refuse to create illegal address alias.
  589.          */
  590.         mutex_unlock(&src_area->lock);
  591.         mutex_unlock(&src_as->lock);
  592.         interrupts_restore(ipl);
  593.         return ENOTSUP;
  594.     }
  595. #endif /* CONFIG_VIRT_IDX_CACHE */
  596. #endif
  597.  
  598.     if (!src_area->backend || !src_area->backend->share) {
  599.         /*
  600.          * There is no backend or the backend does not
  601.          * know how to share the area.
  602.          */
  603.         mutex_unlock(&src_area->lock);
  604.         mutex_unlock(&src_as->lock);
  605.         interrupts_restore(ipl);
  606.         return ENOTSUP;
  607.     }
  608.    
  609.     src_size = src_area->pages * PAGE_SIZE;
  610.     src_flags = src_area->flags;
  611.     src_backend = src_area->backend;
  612.     src_backend_data = src_area->backend_data;
  613.  
  614.     /* Share the cacheable flag from the original mapping */
  615.     if (src_flags & AS_AREA_CACHEABLE)
  616.         dst_flags_mask |= AS_AREA_CACHEABLE;
  617.  
  618.     if (src_size != acc_size || (src_flags & dst_flags_mask) != dst_flags_mask) {
  619.         mutex_unlock(&src_area->lock);
  620.         mutex_unlock(&src_as->lock);
  621.         interrupts_restore(ipl);
  622.         return EPERM;
  623.     }
  624.  
  625.     /*
  626.      * Now we are committed to sharing the area.
  627.      * First, prepare the area for sharing.
  628.      * Then it will be safe to unlock it.
  629.      */
  630.     sh_info = src_area->sh_info;
  631.     if (!sh_info) {
  632.         sh_info = (share_info_t *) malloc(sizeof(share_info_t), 0);
  633.         mutex_initialize(&sh_info->lock);
  634.         sh_info->refcount = 2;
  635.         btree_create(&sh_info->pagemap);
  636.         src_area->sh_info = sh_info;
  637.     } else {
  638.         mutex_lock(&sh_info->lock);
  639.         sh_info->refcount++;
  640.         mutex_unlock(&sh_info->lock);
  641.     }
  642.  
  643.     src_area->backend->share(src_area);
  644.  
  645.     mutex_unlock(&src_area->lock);
  646.     mutex_unlock(&src_as->lock);
  647.  
  648.     /*
  649.      * Create copy of the source address space area.
  650.      * The destination area is created with AS_AREA_ATTR_PARTIAL
  651.      * attribute set which prevents race condition with
  652.      * preliminary as_page_fault() calls.
  653.      * The flags of the source area are masked against dst_flags_mask
  654.      * to support sharing in less privileged mode.
  655.      */
  656.     dst_area = as_area_create(dst_as, dst_flags_mask, src_size, dst_base,
  657.                   AS_AREA_ATTR_PARTIAL, src_backend, &src_backend_data);
  658.     if (!dst_area) {
  659.         /*
  660.          * Destination address space area could not be created.
  661.          */
  662.         sh_info_remove_reference(sh_info);
  663.        
  664.         interrupts_restore(ipl);
  665.         return ENOMEM;
  666.     }
  667.    
  668.     /*
  669.      * Now the destination address space area has been
  670.      * fully initialized. Clear the AS_AREA_ATTR_PARTIAL
  671.      * attribute and set the sh_info.
  672.      */
  673.     mutex_lock(&dst_area->lock);
  674.     dst_area->attributes &= ~AS_AREA_ATTR_PARTIAL;
  675.     dst_area->sh_info = sh_info;
  676.     mutex_unlock(&dst_area->lock);
  677.    
  678.     interrupts_restore(ipl);
  679.    
  680.     return 0;
  681. }
  682.  
  683. /** Check access mode for address space area.
  684.  *
  685.  * The address space area must be locked prior to this call.
  686.  *
  687.  * @param area Address space area.
  688.  * @param access Access mode.
  689.  *
  690.  * @return False if access violates area's permissions, true otherwise.
  691.  */
  692. bool as_area_check_access(as_area_t *area, pf_access_t access)
  693. {
  694.     int flagmap[] = {
  695.         [PF_ACCESS_READ] = AS_AREA_READ,
  696.         [PF_ACCESS_WRITE] = AS_AREA_WRITE,
  697.         [PF_ACCESS_EXEC] = AS_AREA_EXEC
  698.     };
  699.  
  700.     if (!(area->flags & flagmap[access]))
  701.         return false;
  702.    
  703.     return true;
  704. }
  705.  
  706. /** Handle page fault within the current address space.
  707.  *
  708.  * This is the high-level page fault handler. It decides
  709.  * whether the page fault can be resolved by any backend
  710.  * and if so, it invokes the backend to resolve the page
  711.  * fault.
  712.  *
  713.  * Interrupts are assumed disabled.
  714.  *
  715.  * @param page Faulting page.
  716.  * @param access Access mode that caused the fault (i.e. read/write/exec).
  717.  * @param istate Pointer to interrupted state.
  718.  *
  719.  * @return AS_PF_FAULT on page fault, AS_PF_OK on success or AS_PF_DEFER if the
  720.  *     fault was caused by copy_to_uspace() or copy_from_uspace().
  721.  */
  722. int as_page_fault(uintptr_t page, pf_access_t access, istate_t *istate)
  723. {
  724.     pte_t *pte;
  725.     as_area_t *area;
  726.    
  727.     if (!THREAD)
  728.         return AS_PF_FAULT;
  729.        
  730.     ASSERT(AS);
  731.  
  732.     mutex_lock(&AS->lock);
  733.     area = find_area_and_lock(AS, page);   
  734.     if (!area) {
  735.         /*
  736.          * No area contained mapping for 'page'.
  737.          * Signal page fault to low-level handler.
  738.          */
  739.         mutex_unlock(&AS->lock);
  740.         goto page_fault;
  741.     }
  742.  
  743.     if (area->attributes & AS_AREA_ATTR_PARTIAL) {
  744.         /*
  745.          * The address space area is not fully initialized.
  746.          * Avoid possible race by returning error.
  747.          */
  748.         mutex_unlock(&area->lock);
  749.         mutex_unlock(&AS->lock);
  750.         goto page_fault;       
  751.     }
  752.  
  753.     if (!area->backend || !area->backend->page_fault) {
  754.         /*
  755.          * The address space area is not backed by any backend
  756.          * or the backend cannot handle page faults.
  757.          */
  758.         mutex_unlock(&area->lock);
  759.         mutex_unlock(&AS->lock);
  760.         goto page_fault;       
  761.     }
  762.  
  763.     page_table_lock(AS, false);
  764.    
  765.     /*
  766.      * To avoid race condition between two page faults
  767.      * on the same address, we need to make sure
  768.      * the mapping has not been already inserted.
  769.      */
  770.     if ((pte = page_mapping_find(AS, page))) {
  771.         if (PTE_PRESENT(pte)) {
  772.             if (((access == PF_ACCESS_READ) && PTE_READABLE(pte)) ||
  773.                 (access == PF_ACCESS_WRITE && PTE_WRITABLE(pte)) ||
  774.                 (access == PF_ACCESS_EXEC && PTE_EXECUTABLE(pte))) {
  775.                 page_table_unlock(AS, false);
  776.                 mutex_unlock(&area->lock);
  777.                 mutex_unlock(&AS->lock);
  778.                 return AS_PF_OK;
  779.             }
  780.         }
  781.     }
  782.    
  783.     /*
  784.      * Resort to the backend page fault handler.
  785.      */
  786.     if (area->backend->page_fault(area, page, access) != AS_PF_OK) {
  787.         page_table_unlock(AS, false);
  788.         mutex_unlock(&area->lock);
  789.         mutex_unlock(&AS->lock);
  790.         goto page_fault;
  791.     }
  792.    
  793.     page_table_unlock(AS, false);
  794.     mutex_unlock(&area->lock);
  795.     mutex_unlock(&AS->lock);
  796.     return AS_PF_OK;
  797.  
  798. page_fault:
  799.     if (THREAD->in_copy_from_uspace) {
  800.         THREAD->in_copy_from_uspace = false;
  801.         istate_set_retaddr(istate, (uintptr_t) &memcpy_from_uspace_failover_address);
  802.     } else if (THREAD->in_copy_to_uspace) {
  803.         THREAD->in_copy_to_uspace = false;
  804.         istate_set_retaddr(istate, (uintptr_t) &memcpy_to_uspace_failover_address);
  805.     } else {
  806.         return AS_PF_FAULT;
  807.     }
  808.  
  809.     return AS_PF_DEFER;
  810. }
  811.  
  812. /** Switch address spaces.
  813.  *
  814.  * Note that this function cannot sleep as it is essentially a part of
  815.  * scheduling. Sleeping here would lead to deadlock on wakeup.
  816.  *
  817.  * @param old Old address space or NULL.
  818.  * @param new New address space.
  819.  */
  820. void as_switch(as_t *old, as_t *new)
  821. {
  822.     ipl_t ipl;
  823.     bool needs_asid = false;
  824.    
  825.     ipl = interrupts_disable();
  826.     spinlock_lock(&inactive_as_with_asid_lock);
  827.  
  828.     /*
  829.      * First, take care of the old address space.
  830.      */
  831.     if (old) {
  832.         mutex_lock_active(&old->lock);
  833.         ASSERT(old->cpu_refcount);
  834.         if((--old->cpu_refcount == 0) && (old != AS_KERNEL)) {
  835.             /*
  836.              * The old address space is no longer active on
  837.              * any processor. It can be appended to the
  838.              * list of inactive address spaces with assigned
  839.              * ASID.
  840.              */
  841.              ASSERT(old->asid != ASID_INVALID);
  842.              list_append(&old->inactive_as_with_asid_link, &inactive_as_with_asid_head);
  843.         }
  844.         mutex_unlock(&old->lock);
  845.  
  846.         /*
  847.          * Perform architecture-specific tasks when the address space
  848.          * is being removed from the CPU.
  849.          */
  850.         as_deinstall_arch(old);
  851.     }
  852.  
  853.     /*
  854.      * Second, prepare the new address space.
  855.      */
  856.     mutex_lock_active(&new->lock);
  857.     if ((new->cpu_refcount++ == 0) && (new != AS_KERNEL)) {
  858.         if (new->asid != ASID_INVALID)
  859.             list_remove(&new->inactive_as_with_asid_link);
  860.         else
  861.             needs_asid = true;  /* defer call to asid_get() until new->lock is released */
  862.     }
  863.     SET_PTL0_ADDRESS(new->page_table);
  864.     mutex_unlock(&new->lock);
  865.  
  866.     if (needs_asid) {
  867.         /*
  868.          * Allocation of new ASID was deferred
  869.          * until now in order to avoid deadlock.
  870.          */
  871.         asid_t asid;
  872.        
  873.         asid = asid_get();
  874.         mutex_lock_active(&new->lock);
  875.         new->asid = asid;
  876.         mutex_unlock(&new->lock);
  877.     }
  878.     spinlock_unlock(&inactive_as_with_asid_lock);
  879.     interrupts_restore(ipl);
  880.    
  881.     /*
  882.      * Perform architecture-specific steps.
  883.      * (e.g. write ASID to hardware register etc.)
  884.      */
  885.     as_install_arch(new);
  886.    
  887.     AS = new;
  888. }
  889.  
  890. /** Convert address space area flags to page flags.
  891.  *
  892.  * @param aflags Flags of some address space area.
  893.  *
  894.  * @return Flags to be passed to page_mapping_insert().
  895.  */
  896. int area_flags_to_page_flags(int aflags)
  897. {
  898.     int flags;
  899.  
  900.     flags = PAGE_USER | PAGE_PRESENT;
  901.    
  902.     if (aflags & AS_AREA_READ)
  903.         flags |= PAGE_READ;
  904.        
  905.     if (aflags & AS_AREA_WRITE)
  906.         flags |= PAGE_WRITE;
  907.    
  908.     if (aflags & AS_AREA_EXEC)
  909.         flags |= PAGE_EXEC;
  910.    
  911.     if (aflags & AS_AREA_CACHEABLE)
  912.         flags |= PAGE_CACHEABLE;
  913.        
  914.     return flags;
  915. }
  916.  
  917. /** Compute flags for virtual address translation subsytem.
  918.  *
  919.  * The address space area must be locked.
  920.  * Interrupts must be disabled.
  921.  *
  922.  * @param a Address space area.
  923.  *
  924.  * @return Flags to be used in page_mapping_insert().
  925.  */
  926. int as_area_get_flags(as_area_t *a)
  927. {
  928.     return area_flags_to_page_flags(a->flags);
  929. }
  930.  
  931. /** Create page table.
  932.  *
  933.  * Depending on architecture, create either address space
  934.  * private or global page table.
  935.  *
  936.  * @param flags Flags saying whether the page table is for kernel address space.
  937.  *
  938.  * @return First entry of the page table.
  939.  */
  940. pte_t *page_table_create(int flags)
  941. {
  942.         ASSERT(as_operations);
  943.         ASSERT(as_operations->page_table_create);
  944.  
  945.         return as_operations->page_table_create(flags);
  946. }
  947.  
  948. /** Destroy page table.
  949.  *
  950.  * Destroy page table in architecture specific way.
  951.  *
  952.  * @param page_table Physical address of PTL0.
  953.  */
  954. void page_table_destroy(pte_t *page_table)
  955. {
  956.         ASSERT(as_operations);
  957.         ASSERT(as_operations->page_table_destroy);
  958.  
  959.         as_operations->page_table_destroy(page_table);
  960. }
  961.  
  962. /** Lock page table.
  963.  *
  964.  * This function should be called before any page_mapping_insert(),
  965.  * page_mapping_remove() and page_mapping_find().
  966.  *
  967.  * Locking order is such that address space areas must be locked
  968.  * prior to this call. Address space can be locked prior to this
  969.  * call in which case the lock argument is false.
  970.  *
  971.  * @param as Address space.
  972.  * @param lock If false, do not attempt to lock as->lock.
  973.  */
  974. void page_table_lock(as_t *as, bool lock)
  975. {
  976.     ASSERT(as_operations);
  977.     ASSERT(as_operations->page_table_lock);
  978.  
  979.     as_operations->page_table_lock(as, lock);
  980. }
  981.  
  982. /** Unlock page table.
  983.  *
  984.  * @param as Address space.
  985.  * @param unlock If false, do not attempt to unlock as->lock.
  986.  */
  987. void page_table_unlock(as_t *as, bool unlock)
  988. {
  989.     ASSERT(as_operations);
  990.     ASSERT(as_operations->page_table_unlock);
  991.  
  992.     as_operations->page_table_unlock(as, unlock);
  993. }
  994.  
  995.  
  996. /** Find address space area and lock it.
  997.  *
  998.  * The address space must be locked and interrupts must be disabled.
  999.  *
  1000.  * @param as Address space.
  1001.  * @param va Virtual address.
  1002.  *
  1003.  * @return Locked address space area containing va on success or NULL on failure.
  1004.  */
  1005. as_area_t *find_area_and_lock(as_t *as, uintptr_t va)
  1006. {
  1007.     as_area_t *a;
  1008.     btree_node_t *leaf, *lnode;
  1009.     int i;
  1010.    
  1011.     a = (as_area_t *) btree_search(&as->as_area_btree, va, &leaf);
  1012.     if (a) {
  1013.         /* va is the base address of an address space area */
  1014.         mutex_lock(&a->lock);
  1015.         return a;
  1016.     }
  1017.    
  1018.     /*
  1019.      * Search the leaf node and the righmost record of its left neighbour
  1020.      * to find out whether this is a miss or va belongs to an address
  1021.      * space area found there.
  1022.      */
  1023.    
  1024.     /* First, search the leaf node itself. */
  1025.     for (i = 0; i < leaf->keys; i++) {
  1026.         a = (as_area_t *) leaf->value[i];
  1027.         mutex_lock(&a->lock);
  1028.         if ((a->base <= va) && (va < a->base + a->pages * PAGE_SIZE)) {
  1029.             return a;
  1030.         }
  1031.         mutex_unlock(&a->lock);
  1032.     }
  1033.  
  1034.     /*
  1035.      * Second, locate the left neighbour and test its last record.
  1036.      * Because of its position in the B+tree, it must have base < va.
  1037.      */
  1038.     if ((lnode = btree_leaf_node_left_neighbour(&as->as_area_btree, leaf))) {
  1039.         a = (as_area_t *) lnode->value[lnode->keys - 1];
  1040.         mutex_lock(&a->lock);
  1041.         if (va < a->base + a->pages * PAGE_SIZE) {
  1042.             return a;
  1043.         }
  1044.         mutex_unlock(&a->lock);
  1045.     }
  1046.  
  1047.     return NULL;
  1048. }
  1049.  
  1050. /** Check area conflicts with other areas.
  1051.  *
  1052.  * The address space must be locked and interrupts must be disabled.
  1053.  *
  1054.  * @param as Address space.
  1055.  * @param va Starting virtual address of the area being tested.
  1056.  * @param size Size of the area being tested.
  1057.  * @param avoid_area Do not touch this area.
  1058.  *
  1059.  * @return True if there is no conflict, false otherwise.
  1060.  */
  1061. bool check_area_conflicts(as_t *as, uintptr_t va, size_t size, as_area_t *avoid_area)
  1062. {
  1063.     as_area_t *a;
  1064.     btree_node_t *leaf, *node;
  1065.     int i;
  1066.    
  1067.     /*
  1068.      * We don't want any area to have conflicts with NULL page.
  1069.      */
  1070.     if (overlaps(va, size, NULL, PAGE_SIZE))
  1071.         return false;
  1072.    
  1073.     /*
  1074.      * The leaf node is found in O(log n), where n is proportional to
  1075.      * the number of address space areas belonging to as.
  1076.      * The check for conflicts is then attempted on the rightmost
  1077.      * record in the left neighbour, the leftmost record in the right
  1078.      * neighbour and all records in the leaf node itself.
  1079.      */
  1080.    
  1081.     if ((a = (as_area_t *) btree_search(&as->as_area_btree, va, &leaf))) {
  1082.         if (a != avoid_area)
  1083.             return false;
  1084.     }
  1085.    
  1086.     /* First, check the two border cases. */
  1087.     if ((node = btree_leaf_node_left_neighbour(&as->as_area_btree, leaf))) {
  1088.         a = (as_area_t *) node->value[node->keys - 1];
  1089.         mutex_lock(&a->lock);
  1090.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1091.             mutex_unlock(&a->lock);
  1092.             return false;
  1093.         }
  1094.         mutex_unlock(&a->lock);
  1095.     }
  1096.     if ((node = btree_leaf_node_right_neighbour(&as->as_area_btree, leaf))) {
  1097.         a = (as_area_t *) node->value[0];
  1098.         mutex_lock(&a->lock);
  1099.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1100.             mutex_unlock(&a->lock);
  1101.             return false;
  1102.         }
  1103.         mutex_unlock(&a->lock);
  1104.     }
  1105.    
  1106.     /* Second, check the leaf node. */
  1107.     for (i = 0; i < leaf->keys; i++) {
  1108.         a = (as_area_t *) leaf->value[i];
  1109.    
  1110.         if (a == avoid_area)
  1111.             continue;
  1112.    
  1113.         mutex_lock(&a->lock);
  1114.         if (overlaps(va, size, a->base, a->pages * PAGE_SIZE)) {
  1115.             mutex_unlock(&a->lock);
  1116.             return false;
  1117.         }
  1118.         mutex_unlock(&a->lock);
  1119.     }
  1120.  
  1121.     /*
  1122.      * So far, the area does not conflict with other areas.
  1123.      * Check if it doesn't conflict with kernel address space.
  1124.      */  
  1125.     if (!KERNEL_ADDRESS_SPACE_SHADOWED) {
  1126.         return !overlaps(va, size,
  1127.             KERNEL_ADDRESS_SPACE_START, KERNEL_ADDRESS_SPACE_END-KERNEL_ADDRESS_SPACE_START);
  1128.     }
  1129.  
  1130.     return true;
  1131. }
  1132.  
  1133. /** Return size of the address space area with given base.  */
  1134. size_t as_get_size(uintptr_t base)
  1135. {
  1136.     ipl_t ipl;
  1137.     as_area_t *src_area;
  1138.     size_t size;
  1139.  
  1140.     ipl = interrupts_disable();
  1141.     src_area = find_area_and_lock(AS, base);
  1142.     if (src_area){
  1143.         size = src_area->pages * PAGE_SIZE;
  1144.         mutex_unlock(&src_area->lock);
  1145.     } else {
  1146.         size = 0;
  1147.     }
  1148.     interrupts_restore(ipl);
  1149.     return size;
  1150. }
  1151.  
  1152. /** Mark portion of address space area as used.
  1153.  *
  1154.  * The address space area must be already locked.
  1155.  *
  1156.  * @param a Address space area.
  1157.  * @param page First page to be marked.
  1158.  * @param count Number of page to be marked.
  1159.  *
  1160.  * @return 0 on failure and 1 on success.
  1161.  */
  1162. int used_space_insert(as_area_t *a, uintptr_t page, count_t count)
  1163. {
  1164.     btree_node_t *leaf, *node;
  1165.     count_t pages;
  1166.     int i;
  1167.  
  1168.     ASSERT(page == ALIGN_DOWN(page, PAGE_SIZE));
  1169.     ASSERT(count);
  1170.  
  1171.     pages = (count_t) btree_search(&a->used_space, page, &leaf);
  1172.     if (pages) {
  1173.         /*
  1174.          * We hit the beginning of some used space.
  1175.          */
  1176.         return 0;
  1177.     }
  1178.  
  1179.     if (!leaf->keys) {
  1180.         btree_insert(&a->used_space, page, (void *) count, leaf);
  1181.         return 1;
  1182.     }
  1183.  
  1184.     node = btree_leaf_node_left_neighbour(&a->used_space, leaf);
  1185.     if (node) {
  1186.         uintptr_t left_pg = node->key[node->keys - 1], right_pg = leaf->key[0];
  1187.         count_t left_cnt = (count_t) node->value[node->keys - 1], right_cnt = (count_t) leaf->value[0];
  1188.        
  1189.         /*
  1190.          * Examine the possibility that the interval fits
  1191.          * somewhere between the rightmost interval of
  1192.          * the left neigbour and the first interval of the leaf.
  1193.          */
  1194.          
  1195.         if (page >= right_pg) {
  1196.             /* Do nothing. */
  1197.         } else if (overlaps(page, count*PAGE_SIZE, left_pg, left_cnt*PAGE_SIZE)) {
  1198.             /* The interval intersects with the left interval. */
  1199.             return 0;
  1200.         } else if (overlaps(page, count*PAGE_SIZE, right_pg, right_cnt*PAGE_SIZE)) {
  1201.             /* The interval intersects with the right interval. */
  1202.             return 0;          
  1203.         } else if ((page == left_pg + left_cnt*PAGE_SIZE) && (page + count*PAGE_SIZE == right_pg)) {
  1204.             /* The interval can be added by merging the two already present intervals. */
  1205.             node->value[node->keys - 1] += count + right_cnt;
  1206.             btree_remove(&a->used_space, right_pg, leaf);
  1207.             return 1;
  1208.         } else if (page == left_pg + left_cnt*PAGE_SIZE) {
  1209.             /* The interval can be added by simply growing the left interval. */
  1210.             node->value[node->keys - 1] += count;
  1211.             return 1;
  1212.         } else if (page + count*PAGE_SIZE == right_pg) {
  1213.             /*
  1214.              * The interval can be addded by simply moving base of the right
  1215.              * interval down and increasing its size accordingly.
  1216.              */
  1217.             leaf->value[0] += count;
  1218.             leaf->key[0] = page;
  1219.             return 1;
  1220.         } else {
  1221.             /*
  1222.              * The interval is between both neigbouring intervals,
  1223.              * but cannot be merged with any of them.
  1224.              */
  1225.             btree_insert(&a->used_space, page, (void *) count, leaf);
  1226.             return 1;
  1227.         }
  1228.     } else if (page < leaf->key[0]) {
  1229.         uintptr_t right_pg = leaf->key[0];
  1230.         count_t right_cnt = (count_t) leaf->value[0];
  1231.    
  1232.         /*
  1233.          * Investigate the border case in which the left neighbour does not
  1234.          * exist but the interval fits from the left.
  1235.          */
  1236.          
  1237.         if (overlaps(page, count*PAGE_SIZE, right_pg, right_cnt*PAGE_SIZE)) {
  1238.             /* The interval intersects with the right interval. */
  1239.             return 0;
  1240.         } else if (page + count*PAGE_SIZE == right_pg) {
  1241.             /*
  1242.              * The interval can be added by moving the base of the right interval down
  1243.              * and increasing its size accordingly.
  1244.              */
  1245.             leaf->key[0] = page;
  1246.             leaf->value[0] += count;
  1247.             return 1;
  1248.         } else {
  1249.             /*
  1250.              * The interval doesn't adjoin with the right interval.
  1251.              * It must be added individually.
  1252.              */
  1253.             btree_insert(&a->used_space, page, (void *) count, leaf);
  1254.             return 1;
  1255.         }
  1256.     }
  1257.  
  1258.     node = btree_leaf_node_right_neighbour(&a->used_space, leaf);
  1259.     if (node) {
  1260.         uintptr_t left_pg = leaf->key[leaf->keys - 1], right_pg = node->key[0];
  1261.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1], right_cnt = (count_t) node->value[0];
  1262.        
  1263.         /*
  1264.          * Examine the possibility that the interval fits
  1265.          * somewhere between the leftmost interval of
  1266.          * the right neigbour and the last interval of the leaf.
  1267.          */
  1268.  
  1269.         if (page < left_pg) {
  1270.             /* Do nothing. */
  1271.         } else if (overlaps(page, count*PAGE_SIZE, left_pg, left_cnt*PAGE_SIZE)) {
  1272.             /* The interval intersects with the left interval. */
  1273.             return 0;
  1274.         } else if (overlaps(page, count*PAGE_SIZE, right_pg, right_cnt*PAGE_SIZE)) {
  1275.             /* The interval intersects with the right interval. */
  1276.             return 0;          
  1277.         } else if ((page == left_pg + left_cnt*PAGE_SIZE) && (page + count*PAGE_SIZE == right_pg)) {
  1278.             /* The interval can be added by merging the two already present intervals. */
  1279.             leaf->value[leaf->keys - 1] += count + right_cnt;
  1280.             btree_remove(&a->used_space, right_pg, node);
  1281.             return 1;
  1282.         } else if (page == left_pg + left_cnt*PAGE_SIZE) {
  1283.             /* The interval can be added by simply growing the left interval. */
  1284.             leaf->value[leaf->keys - 1] +=  count;
  1285.             return 1;
  1286.         } else if (page + count*PAGE_SIZE == right_pg) {
  1287.             /*
  1288.              * The interval can be addded by simply moving base of the right
  1289.              * interval down and increasing its size accordingly.
  1290.              */
  1291.             node->value[0] += count;
  1292.             node->key[0] = page;
  1293.             return 1;
  1294.         } else {
  1295.             /*
  1296.              * The interval is between both neigbouring intervals,
  1297.              * but cannot be merged with any of them.
  1298.              */
  1299.             btree_insert(&a->used_space, page, (void *) count, leaf);
  1300.             return 1;
  1301.         }
  1302.     } else if (page >= leaf->key[leaf->keys - 1]) {
  1303.         uintptr_t left_pg = leaf->key[leaf->keys - 1];
  1304.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1];
  1305.    
  1306.         /*
  1307.          * Investigate the border case in which the right neighbour does not
  1308.          * exist but the interval fits from the right.
  1309.          */
  1310.          
  1311.         if (overlaps(page, count*PAGE_SIZE, left_pg, left_cnt*PAGE_SIZE)) {
  1312.             /* The interval intersects with the left interval. */
  1313.             return 0;
  1314.         } else if (left_pg + left_cnt*PAGE_SIZE == page) {
  1315.             /* The interval can be added by growing the left interval. */
  1316.             leaf->value[leaf->keys - 1] += count;
  1317.             return 1;
  1318.         } else {
  1319.             /*
  1320.              * The interval doesn't adjoin with the left interval.
  1321.              * It must be added individually.
  1322.              */
  1323.             btree_insert(&a->used_space, page, (void *) count, leaf);
  1324.             return 1;
  1325.         }
  1326.     }
  1327.    
  1328.     /*
  1329.      * Note that if the algorithm made it thus far, the interval can fit only
  1330.      * between two other intervals of the leaf. The two border cases were already
  1331.      * resolved.
  1332.      */
  1333.     for (i = 1; i < leaf->keys; i++) {
  1334.         if (page < leaf->key[i]) {
  1335.             uintptr_t left_pg = leaf->key[i - 1], right_pg = leaf->key[i];
  1336.             count_t left_cnt = (count_t) leaf->value[i - 1], right_cnt = (count_t) leaf->value[i];
  1337.  
  1338.             /*
  1339.              * The interval fits between left_pg and right_pg.
  1340.              */
  1341.  
  1342.             if (overlaps(page, count*PAGE_SIZE, left_pg, left_cnt*PAGE_SIZE)) {
  1343.                 /* The interval intersects with the left interval. */
  1344.                 return 0;
  1345.             } else if (overlaps(page, count*PAGE_SIZE, right_pg, right_cnt*PAGE_SIZE)) {
  1346.                 /* The interval intersects with the right interval. */
  1347.                 return 0;          
  1348.             } else if ((page == left_pg + left_cnt*PAGE_SIZE) && (page + count*PAGE_SIZE == right_pg)) {
  1349.                 /* The interval can be added by merging the two already present intervals. */
  1350.                 leaf->value[i - 1] += count + right_cnt;
  1351.                 btree_remove(&a->used_space, right_pg, leaf);
  1352.                 return 1;
  1353.             } else if (page == left_pg + left_cnt*PAGE_SIZE) {
  1354.                 /* The interval can be added by simply growing the left interval. */
  1355.                 leaf->value[i - 1] += count;
  1356.                 return 1;
  1357.             } else if (page + count*PAGE_SIZE == right_pg) {
  1358.                 /*
  1359.                      * The interval can be addded by simply moving base of the right
  1360.                  * interval down and increasing its size accordingly.
  1361.                  */
  1362.                 leaf->value[i] += count;
  1363.                 leaf->key[i] = page;
  1364.                 return 1;
  1365.             } else {
  1366.                 /*
  1367.                  * The interval is between both neigbouring intervals,
  1368.                  * but cannot be merged with any of them.
  1369.                  */
  1370.                 btree_insert(&a->used_space, page, (void *) count, leaf);
  1371.                 return 1;
  1372.             }
  1373.         }
  1374.     }
  1375.  
  1376.     panic("Inconsistency detected while adding %d pages of used space at %p.\n", count, page);
  1377. }
  1378.  
  1379. /** Mark portion of address space area as unused.
  1380.  *
  1381.  * The address space area must be already locked.
  1382.  *
  1383.  * @param a Address space area.
  1384.  * @param page First page to be marked.
  1385.  * @param count Number of page to be marked.
  1386.  *
  1387.  * @return 0 on failure and 1 on success.
  1388.  */
  1389. int used_space_remove(as_area_t *a, uintptr_t page, count_t count)
  1390. {
  1391.     btree_node_t *leaf, *node;
  1392.     count_t pages;
  1393.     int i;
  1394.  
  1395.     ASSERT(page == ALIGN_DOWN(page, PAGE_SIZE));
  1396.     ASSERT(count);
  1397.  
  1398.     pages = (count_t) btree_search(&a->used_space, page, &leaf);
  1399.     if (pages) {
  1400.         /*
  1401.          * We are lucky, page is the beginning of some interval.
  1402.          */
  1403.         if (count > pages) {
  1404.             return 0;
  1405.         } else if (count == pages) {
  1406.             btree_remove(&a->used_space, page, leaf);
  1407.             return 1;
  1408.         } else {
  1409.             /*
  1410.              * Find the respective interval.
  1411.              * Decrease its size and relocate its start address.
  1412.              */
  1413.             for (i = 0; i < leaf->keys; i++) {
  1414.                 if (leaf->key[i] == page) {
  1415.                     leaf->key[i] += count*PAGE_SIZE;
  1416.                     leaf->value[i] -= count;
  1417.                     return 1;
  1418.                 }
  1419.             }
  1420.             goto error;
  1421.         }
  1422.     }
  1423.  
  1424.     node = btree_leaf_node_left_neighbour(&a->used_space, leaf);
  1425.     if (node && page < leaf->key[0]) {
  1426.         uintptr_t left_pg = node->key[node->keys - 1];
  1427.         count_t left_cnt = (count_t) node->value[node->keys - 1];
  1428.  
  1429.         if (overlaps(left_pg, left_cnt*PAGE_SIZE, page, count*PAGE_SIZE)) {
  1430.             if (page + count*PAGE_SIZE == left_pg + left_cnt*PAGE_SIZE) {
  1431.                 /*
  1432.                  * The interval is contained in the rightmost interval
  1433.                  * of the left neighbour and can be removed by
  1434.                  * updating the size of the bigger interval.
  1435.                  */
  1436.                 node->value[node->keys - 1] -= count;
  1437.                 return 1;
  1438.             } else if (page + count*PAGE_SIZE < left_pg + left_cnt*PAGE_SIZE) {
  1439.                 count_t new_cnt;
  1440.                
  1441.                 /*
  1442.                  * The interval is contained in the rightmost interval
  1443.                  * of the left neighbour but its removal requires
  1444.                  * both updating the size of the original interval and
  1445.                  * also inserting a new interval.
  1446.                  */
  1447.                 new_cnt = ((left_pg + left_cnt*PAGE_SIZE) - (page + count*PAGE_SIZE)) >> PAGE_WIDTH;
  1448.                 node->value[node->keys - 1] -= count + new_cnt;
  1449.                 btree_insert(&a->used_space, page + count*PAGE_SIZE, (void *) new_cnt, leaf);
  1450.                 return 1;
  1451.             }
  1452.         }
  1453.         return 0;
  1454.     } else if (page < leaf->key[0]) {
  1455.         return 0;
  1456.     }
  1457.    
  1458.     if (page > leaf->key[leaf->keys - 1]) {
  1459.         uintptr_t left_pg = leaf->key[leaf->keys - 1];
  1460.         count_t left_cnt = (count_t) leaf->value[leaf->keys - 1];
  1461.  
  1462.         if (overlaps(left_pg, left_cnt*PAGE_SIZE, page, count*PAGE_SIZE)) {
  1463.             if (page + count*PAGE_SIZE == left_pg + left_cnt*PAGE_SIZE) {
  1464.                 /*
  1465.                  * The interval is contained in the rightmost interval
  1466.                  * of the leaf and can be removed by updating the size
  1467.                  * of the bigger interval.
  1468.                  */
  1469.                 leaf->value[leaf->keys - 1] -= count;
  1470.                 return 1;
  1471.             } else if (page + count*PAGE_SIZE < left_pg + left_cnt*PAGE_SIZE) {
  1472.                 count_t new_cnt;
  1473.                
  1474.                 /*
  1475.                  * The interval is contained in the rightmost interval
  1476.                  * of the leaf but its removal requires both updating
  1477.                  * the size of the original interval and
  1478.                  * also inserting a new interval.
  1479.                  */
  1480.                 new_cnt = ((left_pg + left_cnt*PAGE_SIZE) - (page + count*PAGE_SIZE)) >> PAGE_WIDTH;
  1481.                 leaf->value[leaf->keys - 1] -= count + new_cnt;
  1482.                 btree_insert(&a->used_space, page + count*PAGE_SIZE, (void *) new_cnt, leaf);
  1483.                 return 1;
  1484.             }
  1485.         }
  1486.         return 0;
  1487.     }  
  1488.    
  1489.     /*
  1490.      * The border cases have been already resolved.
  1491.      * Now the interval can be only between intervals of the leaf.
  1492.      */
  1493.     for (i = 1; i < leaf->keys - 1; i++) {
  1494.         if (page < leaf->key[i]) {
  1495.             uintptr_t left_pg = leaf->key[i - 1];
  1496.             count_t left_cnt = (count_t) leaf->value[i - 1];
  1497.  
  1498.             /*
  1499.              * Now the interval is between intervals corresponding to (i - 1) and i.
  1500.              */
  1501.             if (overlaps(left_pg, left_cnt*PAGE_SIZE, page, count*PAGE_SIZE)) {
  1502.                 if (page + count*PAGE_SIZE == left_pg + left_cnt*PAGE_SIZE) {
  1503.                     /*
  1504.                     * The interval is contained in the interval (i - 1)
  1505.                      * of the leaf and can be removed by updating the size
  1506.                      * of the bigger interval.
  1507.                      */
  1508.                     leaf->value[i - 1] -= count;
  1509.                     return 1;
  1510.                 } else if (page + count*PAGE_SIZE < left_pg + left_cnt*PAGE_SIZE) {
  1511.                     count_t new_cnt;
  1512.                
  1513.                     /*
  1514.                      * The interval is contained in the interval (i - 1)
  1515.                      * of the leaf but its removal requires both updating
  1516.                      * the size of the original interval and
  1517.                      * also inserting a new interval.
  1518.                      */
  1519.                     new_cnt = ((left_pg + left_cnt*PAGE_SIZE) - (page + count*PAGE_SIZE)) >> PAGE_WIDTH;
  1520.                     leaf->value[i - 1] -= count + new_cnt;
  1521.                     btree_insert(&a->used_space, page + count*PAGE_SIZE, (void *) new_cnt, leaf);
  1522.                     return 1;
  1523.                 }
  1524.             }
  1525.             return 0;
  1526.         }
  1527.     }
  1528.  
  1529. error:
  1530.     panic("Inconsistency detected while removing %d pages of used space from %p.\n", count, page);
  1531. }
  1532.  
  1533. /** Remove reference to address space area share info.
  1534.  *
  1535.  * If the reference count drops to 0, the sh_info is deallocated.
  1536.  *
  1537.  * @param sh_info Pointer to address space area share info.
  1538.  */
  1539. void sh_info_remove_reference(share_info_t *sh_info)
  1540. {
  1541.     bool dealloc = false;
  1542.  
  1543.     mutex_lock(&sh_info->lock);
  1544.     ASSERT(sh_info->refcount);
  1545.     if (--sh_info->refcount == 0) {
  1546.         dealloc = true;
  1547.         link_t *cur;
  1548.        
  1549.         /*
  1550.          * Now walk carefully the pagemap B+tree and free/remove
  1551.          * reference from all frames found there.
  1552.          */
  1553.         for (cur = sh_info->pagemap.leaf_head.next; cur != &sh_info->pagemap.leaf_head; cur = cur->next) {
  1554.             btree_node_t *node;
  1555.             int i;
  1556.            
  1557.             node = list_get_instance(cur, btree_node_t, leaf_link);
  1558.             for (i = 0; i < node->keys; i++)
  1559.                 frame_free((uintptr_t) node->value[i]);
  1560.         }
  1561.        
  1562.     }
  1563.     mutex_unlock(&sh_info->lock);
  1564.    
  1565.     if (dealloc) {
  1566.         btree_destroy(&sh_info->pagemap);
  1567.         free(sh_info);
  1568.     }
  1569. }
  1570.  
  1571. /*
  1572.  * Address space related syscalls.
  1573.  */
  1574.  
  1575. /** Wrapper for as_area_create(). */
  1576. unative_t sys_as_area_create(uintptr_t address, size_t size, int flags)
  1577. {
  1578.     if (as_area_create(AS, flags | AS_AREA_CACHEABLE, size, address, AS_AREA_ATTR_NONE, &anon_backend, NULL))
  1579.         return (unative_t) address;
  1580.     else
  1581.         return (unative_t) -1;
  1582. }
  1583.  
  1584. /** Wrapper for as_area_resize(). */
  1585. unative_t sys_as_area_resize(uintptr_t address, size_t size, int flags)
  1586. {
  1587.     return (unative_t) as_area_resize(AS, address, size, 0);
  1588. }
  1589.  
  1590. /** Wrapper for as_area_destroy(). */
  1591. unative_t sys_as_area_destroy(uintptr_t address)
  1592. {
  1593.     return (unative_t) as_area_destroy(AS, address);
  1594. }
  1595.  
  1596. /** Print out information about address space.
  1597.  *
  1598.  * @param as Address space.
  1599.  */
  1600. void as_print(as_t *as)
  1601. {
  1602.     ipl_t ipl;
  1603.    
  1604.     ipl = interrupts_disable();
  1605.     mutex_lock(&as->lock);
  1606.    
  1607.     /* print out info about address space areas */
  1608.     link_t *cur;
  1609.     for (cur = as->as_area_btree.leaf_head.next; cur != &as->as_area_btree.leaf_head; cur = cur->next) {
  1610.         btree_node_t *node = list_get_instance(cur, btree_node_t, leaf_link);
  1611.        
  1612.         int i;
  1613.         for (i = 0; i < node->keys; i++) {
  1614.             as_area_t *area = node->value[i];
  1615.        
  1616.             mutex_lock(&area->lock);
  1617.             printf("as_area: %p, base=%p, pages=%d (%p - %p)\n",
  1618.                 area, area->base, area->pages, area->base, area->base + area->pages*PAGE_SIZE);
  1619.             mutex_unlock(&area->lock);
  1620.         }
  1621.     }
  1622.    
  1623.     mutex_unlock(&as->lock);
  1624.     interrupts_restore(ipl);
  1625. }
  1626.  
  1627. /** @}
  1628.  */
  1629.