6 files changed, 111 insertions, 136 deletions

diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 696387ffe49c..2c7121d9bff1 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -27,8 +27,8 @@ config PM_DEBUG
         like suspend support.
 
 config SOFTWARE_SUSPEND
-        bool "Software Suspend (EXPERIMENTAL)"
-        depends on EXPERIMENTAL && PM && SWAP
+        bool "Software Suspend"
+        depends on EXPERIMENTAL && PM && SWAP && ((X86 && SMP) || ((FVR || PPC32 || X86) && !SMP))
         ---help---
           Enable the possibility of suspending the machine.
           It doesn't need APM.
@@ -72,3 +72,7 @@ config PM_STD_PARTITION
           suspended image to. It will simply pick the first available swap 
           device.
 
+config SUSPEND_SMP
+        bool
+        depends on HOTPLUG_CPU && X86 && PM
+        default y
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index fbdc634135a7..2f438d0eaa13 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -3,9 +3,9 @@ ifeq ($(CONFIG_PM_DEBUG),y)
 EXTRA_CFLAGS    +=      -DDEBUG
 endif
 
-swsusp-smp-$(CONFIG_SMP)        += smp.o
-
 obj-y                           := main.o process.o console.o pm.o
-obj-$(CONFIG_SOFTWARE_SUSPEND)  += swsusp.o $(swsusp-smp-y) disk.o
+obj-$(CONFIG_SOFTWARE_SUSPEND)  += swsusp.o disk.o
+
+obj-$(CONFIG_SUSPEND_SMP)       += smp.o
 
 obj-$(CONFIG_MAGIC_SYSRQ)       += poweroff.o
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 02b6764034dc..fb8de63c2919 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -117,8 +117,8 @@ static void finish(void)
 {
         device_resume();
         platform_finish();
-        enable_nonboot_cpus();
         thaw_processes();
+        enable_nonboot_cpus();
         pm_restore_console();
 }
 
@@ -131,28 +131,35 @@ static int prepare_processes(void)
 
         sys_sync();
 
+        disable_nonboot_cpus();
+
         if (freeze_processes()) {
                 error = -EBUSY;
-                return error;
+                goto thaw;
         }
 
         if (pm_disk_mode == PM_DISK_PLATFORM) {
                 if (pm_ops && pm_ops->prepare) {
                         if ((error = pm_ops->prepare(PM_SUSPEND_DISK)))
-                                return error;
+                                goto thaw;
                 }
         }
 
         /* Free memory before shutting down devices. */
         free_some_memory();
-
         return 0;
+thaw:
+        thaw_processes();
+        enable_nonboot_cpus();
+        pm_restore_console();
+        return error;
 }
 
 static void unprepare_processes(void)
 {
-        enable_nonboot_cpus();
+        platform_finish();
         thaw_processes();
+        enable_nonboot_cpus();
         pm_restore_console();
 }
 
@@ -160,15 +167,9 @@ static int prepare_devices(void)
 {
         int error;
 
-        disable_nonboot_cpus();
-        if ((error = device_suspend(PMSG_FREEZE))) {
+        if ((error = device_suspend(PMSG_FREEZE)))
                 printk("Some devices failed to suspend\n");
-                platform_finish();
-                enable_nonboot_cpus();
-                return error;
-        }
-
-        return 0;
+        return error;
 }
 
 /**
@@ -185,9 +186,9 @@ int pm_suspend_disk(void)
         int error;
 
         error = prepare_processes();
-        if (!error) {
-                error = prepare_devices();
-        }
+        if (error)
+                return error;
+        error = prepare_devices();
 
         if (error) {
                 unprepare_processes();
@@ -250,7 +251,7 @@ static int software_resume(void)
 
         if ((error = prepare_processes())) {
                 swsusp_close();
-                goto Cleanup;
+                goto Done;
         }
 
         pr_debug("PM: Reading swsusp image.\n");
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 4cdebc972ff2..c94cb9e95090 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -55,6 +55,13 @@ static int suspend_prepare(suspend_state_t state)
 
         pm_prepare_console();
 
+        disable_nonboot_cpus();
+
+        if (num_online_cpus() != 1) {
+                error = -EPERM;
+                goto Enable_cpu;
+        }
+
         if (freeze_processes()) {
                 error = -EAGAIN;
                 goto Thaw;
@@ -75,6 +82,8 @@ static int suspend_prepare(suspend_state_t state)
                 pm_ops->finish(state);
  Thaw:
         thaw_processes();
+ Enable_cpu:
+        enable_nonboot_cpus();
         pm_restore_console();
         return error;
 }
@@ -113,6 +122,7 @@ static void suspend_finish(suspend_state_t state)
         if (pm_ops && pm_ops->finish)
                 pm_ops->finish(state);
         thaw_processes();
+        enable_nonboot_cpus();
         pm_restore_console();
 }
 
@@ -150,12 +160,6 @@ static int enter_state(suspend_state_t state)
                 goto Unlock;
         }
 
-        /* Suspend is hard to get right on SMP. */
-        if (num_online_cpus() != 1) {
-                error = -EPERM;
-                goto Unlock;
-        }
-
         pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]);
         if ((error = suspend_prepare(state)))
                 goto Unlock;
diff --git a/kernel/power/smp.c b/kernel/power/smp.c
index 457c2302ed42..bbe23079c62c 100644
--- a/kernel/power/smp.c
+++ b/kernel/power/smp.c
@@ -13,73 +13,52 @@
 #include <linux/interrupt.h>
 #include <linux/suspend.h>
 #include <linux/module.h>
+#include <linux/cpu.h>
 #include <asm/atomic.h>
 #include <asm/tlbflush.h>
 
-static atomic_t cpu_counter, freeze;
-
-
-static void smp_pause(void * data)
-{
-        struct saved_context ctxt;
-        __save_processor_state(&ctxt);
-        printk("Sleeping in:\n");
-        dump_stack();
-        atomic_inc(&cpu_counter);
-        while (atomic_read(&freeze)) {
-                /* FIXME: restore takes place at random piece inside this.
-                   This should probably be written in assembly, and
-                   preserve general-purpose registers, too
-
-                   What about stack? We may need to move to new stack here.
-
-                   This should better be ran with interrupts disabled.
-                 */
-                cpu_relax();
-                barrier();
-        }
-        atomic_dec(&cpu_counter);
-        __restore_processor_state(&ctxt);
-}
-
-static cpumask_t oldmask;
+/* This is protected by pm_sem semaphore */
+static cpumask_t frozen_cpus;
 
 void disable_nonboot_cpus(void)
 {
-        oldmask = current->cpus_allowed;
-        set_cpus_allowed(current, cpumask_of_cpu(0));
-        printk("Freezing CPUs (at %d)", raw_smp_processor_id());
-        current->state = TASK_INTERRUPTIBLE;
-        schedule_timeout(HZ);
-        printk("...");
-        BUG_ON(raw_smp_processor_id() != 0);
-
-        /* FIXME: for this to work, all the CPUs must be running
-         * "idle" thread (or we deadlock). Is that guaranteed? */
+        int cpu, error;
 
-        atomic_set(&cpu_counter, 0);
-        atomic_set(&freeze, 1);
-        smp_call_function(smp_pause, NULL, 0, 0);
-        while (atomic_read(&cpu_counter) < (num_online_cpus() - 1)) {
-                cpu_relax();
-                barrier();
+        error = 0;
+        cpus_clear(frozen_cpus);
+        printk("Freezing cpus ...\n");
+        for_each_online_cpu(cpu) {
+                if (cpu == 0)
+                        continue;
+                error = cpu_down(cpu);
+                if (!error) {
+                        cpu_set(cpu, frozen_cpus);
+                        printk("CPU%d is down\n", cpu);
+                        continue;
+                }
+                printk("Error taking cpu %d down: %d\n", cpu, error);
         }
-        printk("ok\n");
+        BUG_ON(smp_processor_id() != 0);
+        if (error)
+                panic("cpus not sleeping");
 }
 
 void enable_nonboot_cpus(void)
 {
-        printk("Restarting CPUs");
-        atomic_set(&freeze, 0);
-        while (atomic_read(&cpu_counter)) {
-                cpu_relax();
-                barrier();
-        }
-        printk("...");
-        set_cpus_allowed(current, oldmask);
-        schedule();
-        printk("ok\n");
+        int cpu, error;
 
+        printk("Thawing cpus ...\n");
+        for_each_cpu_mask(cpu, frozen_cpus) {
+                error = smp_prepare_cpu(cpu);
+                if (!error)
+                        error = cpu_up(cpu);
+                if (!error) {
+                        printk("CPU%d is up\n", cpu);
+                        continue;
+                }
+                printk("Error taking cpu %d up: %d\n", cpu, error);
+                panic("Not enough cpus");
+        }
+        cpus_clear(frozen_cpus);
 }
 
-
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 53f9f8720ee4..c285fc5a2320 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -10,12 +10,12 @@
  * This file is released under the GPLv2.
  *
  * I'd like to thank the following people for their work:
- * 
+ *
  * Pavel Machek <pavel@ucw.cz>:
  * 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 <dirk@loth.demon.co.uk>: 
+ * Steve Doddi <dirk@loth.demon.co.uk>:
  * Support the possibility of hardware state restoring.
  *
  * Raph <grey.havens@earthling.net>:
@@ -84,11 +84,11 @@ extern char resume_file[];
 static unsigned int nr_copy_pages __nosavedata = 0;
 
 /* Suspend pagedir is allocated before final copy, therefore it
-   must be freed after resume 
+   must be freed after resume
 
    Warning: this is evil. There are actually two pagedirs at time of
    resume. One is "pagedir_save", which is empty frame allocated at
-   time of suspend, that must be freed. Second is "pagedir_nosave", 
+   time of suspend, that must be freed. Second is "pagedir_nosave",
    allocated at time of resume, that travels through memory not to
    collide with anything.
 
@@ -132,7 +132,7 @@ static int mark_swapfiles(swp_entry_t prev)
 {
         int error;
 
-        rw_swap_page_sync(READ, 
+        rw_swap_page_sync(READ,
                           swp_entry(root_swap, 0),
                           virt_to_page((unsigned long)&swsusp_header));
         if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
@@ -140,7 +140,7 @@ static int mark_swapfiles(swp_entry_t prev)
                 memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
                 memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
                 swsusp_header.swsusp_info = prev;
-                error = rw_swap_page_sync(WRITE, 
+                error = rw_swap_page_sync(WRITE,
                                           swp_entry(root_swap, 0),
                                           virt_to_page((unsigned long)
                                                        &swsusp_header));
@@ -174,22 +174,22 @@ static int is_resume_device(const struct swap_info_struct *swap_info)
 static int swsusp_swap_check(void) /* This is called before saving image */
 {
         int i, len;
-        
+
         len=strlen(resume_file);
         root_swap = 0xFFFF;
-        
+
         swap_list_lock();
-        for(i=0; i<MAX_SWAPFILES; i++) {
+        for (i=0; i<MAX_SWAPFILES; i++) {
                 if (swap_info[i].flags == 0) {
                         swapfile_used[i]=SWAPFILE_UNUSED;
                 } else {
-                        if(!len) {
+                        if (!len) {
                                 printk(KERN_WARNING "resume= option should be used to set suspend device" );
-                                if(root_swap == 0xFFFF) {
+                                if (root_swap == 0xFFFF) {
                                         swapfile_used[i] = SWAPFILE_SUSPEND;
                                         root_swap = i;
                                 } else
-                                        swapfile_used[i] = SWAPFILE_IGNORED;                              
+                                        swapfile_used[i] = SWAPFILE_IGNORED;
                         } else {
                                 /* we ignore all swap devices that are not the resume_file */
                                 if (is_resume_device(&swap_info[i])) {
@@ -209,15 +209,15 @@ static int swsusp_swap_check(void) /* This is called before saving image */
  * This is called after saving image so modification
  * will be lost after resume... and that's what we want.
  * we make the device unusable. A new call to
- * lock_swapdevices can unlock the devices. 
+ * lock_swapdevices can unlock the devices.
  */
 static void lock_swapdevices(void)
 {
         int i;
 
         swap_list_lock();
-        for(i = 0; i< MAX_SWAPFILES; i++)
-                if(swapfile_used[i] == SWAPFILE_IGNORED) {
+        for (i = 0; i< MAX_SWAPFILES; i++)
+                if (swapfile_used[i] == SWAPFILE_IGNORED) {
                         swap_info[i].flags ^= 0xFF;
                 }
         swap_list_unlock();
@@ -229,7 +229,7 @@ static void lock_swapdevices(void)
  *      @loc:   Place to store the entry we used.
  *
  *      Allocate a new swap entry and 'sync' it. Note we discard -EIO
- *      errors. That is an artifact left over from swsusp. It did not 
+ *      errors. That is an artifact left over from swsusp. It did not
  *      check the return of rw_swap_page_sync() at all, since most pages
  *      written back to swap would return -EIO.
  *      This is a partial improvement, since we will at least return other
@@ -241,7 +241,7 @@ static int write_page(unsigned long addr, swp_entry_t * loc)
         int error = 0;
 
         entry = get_swap_page();
-        if (swp_offset(entry) && 
+        if (swp_offset(entry) &&
             swapfile_used[swp_type(entry)] == SWAPFILE_SUSPEND) {
                 error = rw_swap_page_sync(WRITE, entry,
                                           virt_to_page(addr));
@@ -257,7 +257,7 @@ static int write_page(unsigned long addr, swp_entry_t * loc)
 /**
  *      data_free - Free the swap entries used by the saved image.
  *
- *      Walk the list of used swap entries and free each one. 
+ *      Walk the list of used swap entries and free each one.
  *      This is only used for cleanup when suspend fails.
  */
 static void data_free(void)
@@ -290,7 +290,7 @@ static int data_write(void)
                 mod = 1;
 
         printk( "Writing data to swap (%d pages)...     ", nr_copy_pages );
-        for_each_pbe(p, pagedir_nosave) {
+        for_each_pbe (p, pagedir_nosave) {
                 if (!(i%mod))
                         printk( "\b\b\b\b%3d%%", i / mod );
                 if ((error = write_page(p->address, &(p->swap_address))))
@@ -335,7 +335,7 @@ static int close_swap(void)
 
         dump_info();
         error = write_page((unsigned long)&swsusp_info, &entry);
-        if (!error) { 
+        if (!error) {
                 printk( "S" );
                 error = mark_swapfiles(entry);
                 printk( "|\n" );
@@ -370,7 +370,7 @@ static int write_pagedir(void)
         struct pbe * pbe;
 
         printk( "Writing pagedir...");
-        for_each_pb_page(pbe, pagedir_nosave) {
+        for_each_pb_page (pbe, pagedir_nosave) {
                 if ((error = write_page((unsigned long)pbe, &swsusp_info.pagedir[n++])))
                         return error;
         }
@@ -472,7 +472,7 @@ static int save_highmem(void)
         int res = 0;
 
         pr_debug("swsusp: Saving Highmem\n");
-        for_each_zone(zone) {
+        for_each_zone (zone) {
                 if (is_highmem(zone))
                         res = save_highmem_zone(zone);
                 if (res)
@@ -547,7 +547,7 @@ static void count_data_pages(void)
 
         nr_copy_pages = 0;
 
-        for_each_zone(zone) {
+        for_each_zone (zone) {
                 if (is_highmem(zone))
                         continue;
                 mark_free_pages(zone);
@@ -562,9 +562,9 @@ static void copy_data_pages(void)
         struct zone *zone;
         unsigned long zone_pfn;
         struct pbe * pbe = pagedir_nosave;
-        
+
         pr_debug("copy_data_pages(): pages to copy: %d\n", nr_copy_pages);
-        for_each_zone(zone) {
+        for_each_zone (zone) {
                 if (is_highmem(zone))
                         continue;
                 mark_free_pages(zone);
@@ -702,7 +702,7 @@ static void free_image_pages(void)
 {
         struct pbe * p;
 
-        for_each_pbe(p, pagedir_save) {
+        for_each_pbe (p, pagedir_save) {
                 if (p->address) {
                         ClearPageNosave(virt_to_page(p->address));
                         free_page(p->address);
@@ -719,7 +719,7 @@ static int alloc_image_pages(void)
 {
         struct pbe * p;
 
-        for_each_pbe(p, pagedir_save) {
+        for_each_pbe (p, pagedir_save) {
                 p->address = get_zeroed_page(GFP_ATOMIC | __GFP_COLD);
                 if (!p->address)
                         return -ENOMEM;
@@ -740,7 +740,7 @@ void swsusp_free(void)
 /**
  *      enough_free_mem - Make sure we enough free memory to snapshot.
  *
- *      Returns TRUE or FALSE after checking the number of available 
+ *      Returns TRUE or FALSE after checking the number of available
  *      free pages.
  */
 
@@ -758,11 +758,11 @@ static int enough_free_mem(void)
 /**
  *      enough_swap - Make sure we have enough swap to save the image.
  *
- *      Returns TRUE or FALSE after checking the total amount of swap 
+ *      Returns TRUE or FALSE after checking the total amount of swap
  *      space avaiable.
  *
  *      FIXME: si_swapinfo(&i) returns all swap devices information.
- *      We should only consider resume_device. 
+ *      We should only consider resume_device.
  */
 
 static int enough_swap(void)
@@ -781,18 +781,18 @@ static int swsusp_alloc(void)
 {
         int error;
 
+        pagedir_nosave = NULL;
+        nr_copy_pages = calc_nr(nr_copy_pages);
+
         pr_debug("suspend: (pages needed: %d + %d free: %d)\n",
                  nr_copy_pages, PAGES_FOR_IO, nr_free_pages());
 
-        pagedir_nosave = NULL;
         if (!enough_free_mem())
                 return -ENOMEM;
 
         if (!enough_swap())
                 return -ENOSPC;
 
-        nr_copy_pages = calc_nr(nr_copy_pages);
-
         if (!(pagedir_save = alloc_pagedir(nr_copy_pages))) {
                 printk(KERN_ERR "suspend: Allocating pagedir failed.\n");
                 return -ENOMEM;
@@ -827,8 +827,8 @@ static int suspend_prepare_image(void)
         error = swsusp_alloc();
         if (error)
                 return error;
-        
-        /* During allocating of suspend pagedir, new cold pages may appear. 
+
+        /* During allocating of suspend pagedir, new cold pages may appear.
          * Kill them.
          */
         drain_local_pages();
@@ -929,21 +929,6 @@ int swsusp_resume(void)
         return error;
 }
 
-/* More restore stuff */
-
-/*
- * Returns true if given address/order collides with any orig_address 
- */
-static int does_collide_order(unsigned long addr, int order)
-{
-        int i;
-        
-        for (i=0; i < (1<<order); i++)
-                if (!PageNosaveFree(virt_to_page(addr + i * PAGE_SIZE)))
-                        return 1;
-        return 0;
-}
-
 /**
  *      On resume, for storing the PBE list and the image,
  *      we can only use memory pages that do not conflict with the pages
@@ -973,7 +958,7 @@ static unsigned long get_usable_page(unsigned gfp_mask)
         unsigned long m;
 
         m = get_zeroed_page(gfp_mask);
-        while (does_collide_order(m, 0)) {
+        while (!PageNosaveFree(virt_to_page(m))) {
                 eat_page((void *)m);
                 m = get_zeroed_page(gfp_mask);
                 if (!m)
@@ -1045,7 +1030,7 @@ static struct pbe * swsusp_pagedir_relocate(struct pbe *pblist)
 
         /* Set page flags */
 
-        for_each_zone(zone) {
+        for_each_zone (zone) {
                 for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
                         SetPageNosaveFree(pfn_to_page(zone_pfn +
                                         zone->zone_start_pfn));
@@ -1061,7 +1046,7 @@ static struct pbe * swsusp_pagedir_relocate(struct pbe *pblist)
         /* Relocate colliding pages */
 
         for_each_pb_page (pbpage, pblist) {
-                if (does_collide_order((unsigned long)pbpage, 0)) {
+                if (!PageNosaveFree(virt_to_page((unsigned long)pbpage))) {
                         m = (void *)get_usable_page(GFP_ATOMIC | __GFP_COLD);
                         if (!m) {
                                 error = -ENOMEM;
@@ -1193,8 +1178,10 @@ static const char * sanity_check(void)
                 return "version";
         if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
                 return "machine";
+#if 0
         if(swsusp_info.cpus != num_online_cpus())
                 return "number of cpus";
+#endif
         return NULL;
 }