/* * linux/kernel/power/swsusp.c * * This file provides code to write suspend image to swap and read it back. * * Copyright (C) 1998-2001 Gabor Kuti * Copyright (C) 1998,2001-2005 Pavel Machek * * This file is released under the GPLv2. * * I'd like to thank the following people for their work: * * Pavel Machek : * Modifications, defectiveness pointing, being with me at the very beginning, * suspend to swap space, stop all tasks. Port to 2.4.18-ac and 2.5.17. * * Steve Doddi : * Support the possibility of hardware state restoring. * * Raph : * Support for preserving states of network devices and virtual console * (including X and svgatextmode) * * Kurt Garloff : * Straightened the critical function in order to prevent compilers from * playing tricks with local variables. * * Andreas Mohr * * Alex Badea : * Fixed runaway init * * Rafael J. Wysocki * Reworked the freeing of memory and the handling of swap * * More state savers are welcome. Especially for the scsi layer... * * For TODOs,FIXMEs also look in Documentation/power/swsusp.txt */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include "power.h" /* * Preferred image size in bytes (tunable via /sys/power/image_size). * When it is set to N, swsusp will do its best to ensure the image * size will not exceed N bytes, but if that is impossible, it will * try to create the smallest image possible. */ unsigned long image_size = 500 * 1024 * 1024; int in_suspend __nosavedata = 0; #ifdef CONFIG_HIGHMEM unsigned int count_highmem_pages(void); int restore_highmem(void); #else static inline int restore_highmem(void) { return 0; } static inline unsigned int count_highmem_pages(void) { return 0; } #endif /** * The following functions are used for tracing the allocated * swap pages, so that they can be freed in case of an error. */ struct swsusp_extent { struct rb_node node; unsigned long start; unsigned long end; }; static struct rb_root swsusp_extents = RB_ROOT; static int swsusp_extents_insert(unsigned long swap_offset) { struct rb_node **new = &(swsusp_extents.rb_node); struct rb_node *parent = NULL; struct swsusp_extent *ext; /* Figure out where to put the new node */ while (*new) { ext = container_of(*new, struct swsusp_extent, node); parent = *new; if (swap_offset < ext->start) { /* Try to merge */ if (swap_offset == ext->start - 1) { ext->start--; return 0; } new = &((*new)->rb_left); } else if (swap_offset > ext->end) { /* Try to merge */ if (swap_offset == ext->end + 1) { ext->end++; return 0; } new = &((*new)->rb_right); } else { /* It already is in the tree */ return -EINVAL; } } /* Add the new node and rebalance the tree. */ ext = kzalloc(sizeof(struct swsusp_extent), GFP_KERNEL); if (!ext) return -ENOMEM; ext->start = swap_offset; ext->end = swap_offset; rb_link_node(&ext->node, parent, new); rb_insert_color(&ext->node, &swsusp_extents); return 0; } /** * alloc_swapdev_block - allocate a swap page and register that it has * been allocated, so that it can be freed in case of an error. */ sector_t alloc_swapdev_block(int swap) { unsigned long offset; offset = swp_offset(get_swap_page_of_type(swap)); if (offset) { if (swsusp_extents_insert(offset)) swap_free(swp_entry(swap, offset)); else return swapdev_block(swap, offset); } return 0; } /** * free_all_swap_pages - free swap pages allocated for saving image data. * It also frees the extents used to register which swap entres had been * allocated. */ void free_all_swap_pages(int swap) { struct rb_node *node; while ((node = swsusp_extents.rb_node)) { struct swsusp_extent *ext; unsigned long offset; ext = container_of(node, struct swsusp_extent, node); rb_erase(node, &swsusp_extents); for (offset = ext->start; offset <= ext->end; offset++) swap_free(swp_entry(swap, offset)); kfree(ext); } } int swsusp_swap_in_use(void) { return (swsusp_extents.rb_node != NULL); } /** * swsusp_show_speed - print the time elapsed between two events represented by * @start and @stop * * @nr_pages - number of pages processed between @start and @stop * @msg - introductory message to print */ void swsusp_show_speed(struct timeval *start, struct timeval *stop, unsigned nr_pages, char *msg) { s64 elapsed_centisecs64; int centisecs; int k; int kps; elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start); do_div(elapsed_centisecs64, NSEC_PER_SEC / 100); centisecs = elapsed_centisecs64; if (centisecs == 0) centisecs = 1; /* avoid div-by-zero */ k = nr_pages * (PAGE_SIZE / 1024); kps = (k * 100) / centisecs; printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k, centisecs / 100, centisecs % 100, kps / 1000, (kps % 1000) / 10); } /** * swsusp_shrink_memory - Try to free as much memory as needed * * ... but do not OOM-kill anyone * * Notice: all userland should be stopped before it is called, or * livelock is possible. */ #define SHRINK_BITE 10000 static inline unsigned long __shrink_memory(long tmp) { if (tmp > SHRINK_BITE) tmp = SHRINK_BITE; return shrink_all_memory(tmp); } int swsusp_shrink_memory(void) { long tmp; struct zone *zone; unsigned long pages = 0; unsigned int i = 0; char *p = "-\\|/"; struct timeval start, stop; printk("Shrinking memory... "); do_gettimeofday(&start); do { long size, highmem_size; highmem_size = count_highmem_pages(); size = count_data_pages() + PAGES_FOR_IO + SPARE_PAGES; tmp = size; size += highmem_size; for_each_zone (zone) if (populated_zone(zone)) { tmp += snapshot_additional_pages(zone); if (is_highmem(zone)) { highmem_size -= zone_page_state(zone, NR_FREE_PAGES); } else { tmp -= zone_page_state(zone, NR_FREE_PAGES); tmp += zone->lowmem_reserve[ZONE_NORMAL]; } } if (highmem_size < 0) highmem_size = 0; tmp += highmem_size; if (tmp > 0) { tmp = __shrink_memory(tmp); if (!tmp) return -ENOMEM; pages += tmp; } else if (size > image_size / PAGE_SIZE) { tmp = __shrink_memory(size - (image_size / PAGE_SIZE)); pages += tmp; } printk("\b%c", p[i++%4]); } while (tmp > 0); do_gettimeofday(&stop); printk("\bdone (%lu pages freed)\n", pages); swsusp_show_speed(&start, &stop, pages, "Freed"); return 0; } int swsusp_suspend(void) { int error; if ((error = arch_prepare_suspend())) return error; local_irq_disable(); /* At this point, device_suspend() has been called, but *not* * device_power_down(). We *must* device_power_down() now. * Otherwise, drivers for some devices (e.g. interrupt controllers) * become desynchronized with the actual state of the hardware * at resume time, and evil weirdness ensues. */ if ((error = device_power_down(PMSG_FREEZE))) { printk(KERN_ERR "Some devices failed to power down, aborting suspend\n"); goto Enable_irqs; } save_processor_state(); if ((error = swsusp_arch_suspend())) printk(KERN_ERR "Error %d suspending\n", error); /* Restore control flow magically appears here */ restore_processor_state(); /* NOTE: device_power_up() is just a resume() for devices * that suspended with irqs off ... no overall powerup. */ device_power_up(); Enable_irqs: local_irq_enable(); return error; } int swsusp_resume(void) { int error; local_irq_disable(); /* NOTE: device_power_down() is just a suspend() with irqs off; * it has no special "power things down" semantics */ if (device_power_down(PMSG_PRETHAW)) printk(KERN_ERR "Some devices failed to power down, very bad\n"); /* We'll ignore saved state, but this gets preempt count (etc) right */ save_processor_state(); error = restore_highmem(); if (!error) { error = swsusp_arch_resume(); /* The code below is only ever reached in case of a failure. * Otherwise execution continues at place where * swsusp_arch_suspend() was called */ BUG_ON(!error); /* This call to restore_highmem() undos the previous one */ restore_highmem(); } /* The only reason why swsusp_arch_resume() can fail is memory being * very tight, so we have to free it as soon as we can to avoid * subsequent failures */ swsusp_free(); restore_processor_state(); touch_softlockup_watchdog(); device_power_up(); local_irq_enable(); return error; }