diff -urN oldtree/Documentation/kernel-parameters.txt newtree/Documentation/kernel-parameters.txt --- oldtree/Documentation/kernel-parameters.txt 2006-10-05 15:26:55.000000000 -0400 +++ newtree/Documentation/kernel-parameters.txt 2006-10-06 17:28:20.000000000 -0400 @@ -81,6 +81,7 @@ SERIAL Serial support is enabled. SMP The kernel is an SMP kernel. SPARC Sparc architecture is enabled. + SUSPEND2 Suspend2 is enabled. SWSUSP Software suspend is enabled. TS Appropriate touchscreen support is enabled. USB USB support is enabled. @@ -1080,6 +1081,8 @@ noresume [SWSUSP] Disables resume and restores original swap space. + noresume2 [SUSPEND2] Disables resuming and restores original swap signature. + no-scroll [VGA] Disables scrollback. This is required for the Braillex ib80-piezo Braille reader made by F.H. Papenmeier (Germany). @@ -1382,6 +1385,11 @@ in units (needed only for swap files). See Documentation/power/swsusp-and-swap-files.txt + resume2= [SUSPEND2] Specify the storage device for Suspend2. + Format: :. + See Documentation/power/suspend2.txt for details of the + formats for available image writers. + rhash_entries= [KNL,NET] Set number of hash buckets for route cache diff -urN oldtree/Documentation/power/Suspend2-Changelog.txt newtree/Documentation/power/Suspend2-Changelog.txt --- oldtree/Documentation/power/Suspend2-Changelog.txt 1969-12-31 19:00:00.000000000 -0500 +++ newtree/Documentation/power/Suspend2-Changelog.txt 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,775 @@ +Suspend2 changelog. + +2.2.8 - 20 September 2006 + +- Add Changelog as Documentation/power/Suspend2-Changelog.txt +- Don't replace swsusp completely when the do-it-by-default option is on. + +2.2.7.6 - 15 September 2006 + +- Let swsusp work. The readding of BUG_ON(non suspend2 io) test + wasn't targetted enough. +- Fix the swsusp replacement code to work with initramfses. +- Explicitly case the sizeof(struct suspend_header) as int. One case + reported where the lack of a cast seems to have caused huge header + size values. +- Make the suspend_file routine parse_signature static. +- Fix handling of swapwriter code so that the filewriter can still work + with the new swsusp replacement functionality. + +2.2.7.5 - 5 September 2006 + +- Documentation updates +- Add support for beeping at signficant lowlevel suspending points, + to help in diagnosing resume failures. + suspend_beep=1 on commandline to activate. +- Add support for beeping while invoking driver suspend and resume + routines, to help in diagnosing resume failures. + Boot with device_beep=1 to activate. +- Add a timer_resume fix that I'm hoping to see John Stultz submit + upstream. +- Readd BUG_ON(non suspend2 io during suspend) in response to a + recent corruption report. +- Remove deflate fix. We were the only ones needing it, and I don't + recommend the use of glacially slow compression anyway. +- Remove PPC SOFTWARE_REPLACE_SLEEP configuration option. +- Better typing of pageflags variables. +- Use max_pfn rather than -1 to indicate the start/end of a scan. +- Remove LRU paranoia. No reports of it being hit. +- Consolidate Suspend2 specific kernel/power/*.h. +- Unlink LRU pages instead of avoiding vmscanning during suspend. +- Make Suspend2 a drop-in replacement for swsusp (compile time + option). This doesn't stop swsusp from working. Instead it lets you + * use resume= instead of resume2= (resume= takes precedence if set) + * use noresume instead of noresume2 + * use echo disk > /sys/power/state to activate suspend2. + By default suspend2 is configured to use lzf compression, if available, + and no encryption. + Overriding swsusp can by disabled via /sys/power/suspend2. +- Add further printks to help diagnose resume issues. +- Replace zone->spanned_pages with populated_zone() test. +- Some initial work on code to prove/ensure LRU pages aren't modified + while being written. Relies on DEBUG_PAGEALLOC, so shouldn't interfere + with normal use of this release, but it isn't working yet. x86 only. +- Some further cleanups from sysfs conversion. + +2.2.7.4 - 19 July 2006 + +- Update Documentation to refer to /sys/power/suspend2. +- Make invocation of pm_console_suspend/resume optional + and disabled by default. +- Add some printks for -EBUSY error messages. +- Rename proc.c to sysfs.c. +- Apply Rafael's suggestion of using the zone table info. +- Change .disabled to .enabled in the module structure. +- Change all references to proc to be sysfs. +- Add one second delay between powering down devices and + doing the acpi state enter/machine_power_off/machine_halt. +- Add result code flag for drivers failing to suspend. + +2.2.7.3 - 11 July 2006 + +- Add subdirectories for modules. +- Fix do_suspend and do_resume. + +2.2.7.2 - 11 July 2006 + +- Console suspend and resume fix. +- Initial sysfs support. /proc support retained for the interim. Note that the + sysfs layout is subject to change. I may make subdirectories for the modules + (compressor etc) and put their entries in there. +- Modifed debugging info for compressor and encryptor so it's clearer when the + name of the algorithm to be used hasn't been set. +- Try kernel_halt() if kernel_power_off() fails. + +2.2.7.1 - 6 July 2006 + +- Remove Linus patches. +- Remove some debugging code from atomic_copy.c +- Change quotes and remove leading spaces in Kconfig +- Shift resetting accounting of module header storage usage. +- Add a missing reset of suspend2_running. +- Remove some extra whitespace that was used when splitting the patches. +- Enable some code in dyn_pageflags.c that was disabled and correct a typo + in it. + +2.2.7 - 2 July 2006 + +- Fix test debugging patch to run on powering down devices instead of + when suspending them, and re-disable interrupts when we find something + to complain about. + +2.2.6.3 - 2 July 2006 + +- Add Bernard's 2.2.6.2 fix patch (thanks, Bernard!) +- Add test debugging patch to capture drivers that reenable irqs when they + ought not. + +2.2.6.2 - 1 July 2006 + +- Fix date on last release :) +- Fix freeze when eating memory that I forgot about before doing 2.2.6.1 +- Small commenting fix. +- Quieten pdflush when thawed. + +2.2.6.1 - 1 July 2006 + +- Cleanups: Whitespace/line wrapping, removal of asm-ppc/cpu_context.h, + old/unnecessary declarations in dyn_pageflags.h, freezer.h. Atomic_copy + prepare_pbe_list. +- Commenting: dyn_pageflags.h +- Rename: pageflags_space_needed -> suspend_* +- Add zones_in_use variable, to save recalculating how many zones are in use + all the time (ta Rafael). +- !(CONFIG_VT && VT_CONSOLE) compile fix from LKML. +- Fix SYSTEM_BOOTING->RUNNING, from old version of Linus' console patch. +- Quoting of KConfig options. Addresses warnings that have appeared in + post 2.6.17 git. +- Reverse early pm_ops->prepare and late ->finish invocations. +- Handle errors if bdev freezing kmalloc fails or suspend2_suspend fails. +- Correct error handling in suspend_main(). +- Remove block_dump debugging in file_writer. +- Re-enable populate_block_list printk condition. +- Disable pm_prepare_console and pm_restore_console calls for now. Still + seem to be problematic. +- Add MAINTAINERS file entry. + +2.2.6 - 23 June 2006 + +- No problems reported so far with 2.2.5.3, so I'm + renaming it to 2.2.6. + +2.2.5.3 - 22 June 2006 + +- Update to 2.6.17 +- Bring figuring out which pages to save closer to the + algorithm used by swsusp. +- Move allocation of image header storage last in updating + the image contents. +- Reset the header bytes used by each module when getting + the request so S3 powerdown doesn't only work once. +- Tidy header allocation code. +- Tidy process refreezing when updating the image metadata. +- Add debugging code to check that LRU pages are not modified + while the freezer is on. +- Do pm_state->prepare and finish right at the start and + end of a cycle. +- Add Ben Collin's patch for a trace data section for ppc + (needed for Linus' debugging patch, added in 2.2.5.2). +- Rework extra pageset1 pages allowance so that we have a + hardcoded minimum and default (now 500) that is more + easily modified (prepare_image.h). +- Take pm_sem when suspending and add PM_SEM failure bit + to the result code. +- Use console suspend/resume code. + +2.2.5.2 - 16 June 2006 + +- Remove x86_64 e820 code. +- Add #include to Linus' tracing code to fix compilation in + Ubuntu backport. +- Add display of number of extra pageset 1 pages used. +- Modify the code which tracks our position in storage so that + it sits at the start of a page and not the end. Simplifies + finding where to start when resuming. +- Add pdflush post-resume fix from LKML. + +2.2.5.1 - 15 June 2006 + +- Updated to 2.6.17-rc6. +- Fixed filewriter issue when header is not at the lowest block + number in the file. +- Moved refrigerator closer to the vanilla implementation. +- Removed per-workqueue freezing. +- Carefully audited header space usage and added debugging + support. +- Fixed compilation issues with sharing code with swsusp. +- Removed SUSPEND_TEST_BIO flag. +- Added paragraph to suspend2.txt. Pavel forked, not us :) +- Removed arm code. No recent updates. +- Added (temporary) e820 amd64 code. +- Fix BITMAP_FOR_EACH_SET to check for -1 instead of max_pfn. +- Add ifdefs around some highmem code. +- Fix some line length issues. +- Rename attempt_to_parse_resume_device to suspend_... +- Remove DEBUG_PAGEALLOC support. +- Prepare to power off with POWER_OFF, not SUSPEND_DISK. +- Max extra pages allowance should be INT_MAX, not 32768. + +2.2.5 - 21 April 2006 + +- Cleanups from ongoing preparation of the git tree. +- Correction to x86_64 page_is_ram function. +- New FREEZING_COMPLETE status variable, used in debugging the + new shrink_all_memory implementation. +- Improvements to calculations of header space required. More + work to be done. +- Fix compilation with !CONFIG_NET +- Some new image calculation paranoia. Shouldn't be needed, but + you can't always know everything that changes from one kernel + version to the next. +- Use Con Kolivas and Rafael Wysocki's shrink_all_memory rework, + with a couple of bug fixes of my own. In combination with this, + simplify the memory freeing in suspend2 to use a single pass. +- Fix suspend_running not being set at resume time, which caused + spontaneous reboots when the temporary pagedir allocator then + got unsafe pages. +- Remove setting swap info struct data during resume. +- Update hooks in the pagealloc routines that prevented memory + freeing while we're writing the image. + +2.2.4.1 - 13 April 2006 + +- Removed use of max_pfn to make life simpler on ARM. +- Fix line length in atomic_copy.c. +- Rename suspend_recalculate_stats to suspend_recalculate_image_contents +- Remove checksumming code. +- Apply Jens Gustedt's checks for which filesystems are mounted + on resume. +- Fix encryption where PAGE_SIZE % key length != 0. +- Move swsusp Makefile entries below Suspend2 again - it doesn't play + nicely with Suspend2, and so needs to have it's boot time routines + run later. noresume2 will also not work with noresume being compiled + in first. +- Rework the allocation of extra pages to use PageNosave instead of + PageAllocd and don't completely clear Pageset1Copy on every iteration + through count_data_pages. +- Remove the ConflictingPages bitmap. +- Remove a bogus comment that said we remember conflicting pages and + free them. +- Remove DEBUG_PAGEALLOC support. +- Add extra message for when powering off. +- Rework and cleanup code which determines which pages to save. +- Correction to ram_to_suspend function - it should halve the + amount calculated and take account of extra pagedir pages that have + been allocated. +- Correct handling of proc entries that use longs. This is particularly + relevant to the image_size_limit. It wasn't previously correctly set + to -1 or -2. (This would have broken the nocaches functionality). +- Move suspend_default_console_level to ui.c. +- Remove unused suspend_bio_task in block_io.c. +- Use progress granularity that was saved in the image header, even + if the userui helper program stored in the image header won't be + used because it was overridden by the initrd/ramfs script. +- Don't update the load average while the freezer is on. This fixes + an unintended sideeffect wherein mail services will refuse to + operate post-resume because the load average had been pushed up + by Suspend's intensive CPU use. + +2.2.4 - 2 April 2006 + +- Fixed what are hopefully the last outstanding issues with + the switchover to using the swsusp lowlevel code: + * Corrected compilation issues with the difference + combinations of swsusp and Suspend2. + * Make x86_64 work correctly - get_safe_page() was using + the swsusp code instead of Suspend2 code. + +2.2.3 - 1 April 2006 + +- Add better powerdown_methods documentation (thanks + Jens Gustedt). +- Further work addressing issues with using the swsusp + lowlevel code. +- Abort on resume if we fail to initialise the compression + transform used when suspending. +- Switch filewriter to use add_to_extent_chain instead of + append_extent_to_extent_chain and remove now the now + unused function. +- Fix compilation errors if swsusp and Suspend2 are both + disabled. +- Fix PPC compilation error with !rtas. + +2.2.2.1 - 28 March 2006 + +- Missed a part of the backport, inadvertently making + a requirement that you also select + CONFIG_SOFTWARE_SUSPEND. Fix that by making Suspend2 + select SOFTWARE_SUSPEND automatically. + +2.2.2 - 28 March 2006 + +- Backported to 2.6.16. +- Added a couple of printks for when we fail to write a + block of the image. Saw this in unusual circumstances + while testing, and need to seek to reproduce and fix. + +2.2.1.3 - 27 March 2006 +Prepared against current git as at 7.30pm AEST +- Further cleanups as I begin to prepare the git tree + in earnest. +- Highmem fixes I almost forgot for this release! + +2.2.1.2 - 23 March 2006 +Prepared against current git as at 11pm AEST. + +- Further cleanups from the switch to swsusp lowlevel + code. Should not longer require swsusp to be selected. + +2.2.1.1 - 23 March 2006 +Prepared against current git as at 4:20pm AEST. + +- Switch suspend2 to use the swsusp lowlevel code. +- Rework freeing of extra pages allocated for the + atomic copy. +- Misc cleanups. + +2.2.1 - 17 March 2006 + +- Remove debug_pagealloc support +- Documentation update +- Move device_suspend at resume time prior to setting + up temporary page tables (Lamarque Souza). + +2.2.0.8 - 15 March 2006 + +- Convert pageset sizes to longs. +- Convert [read|write]_header_chunk to + rw_header_chunk([READ|WRITE]...) and do similar + cleanups with other i/o routines. +- Remove some unused module structure members. +- Flatten the module ops. +- Fix a couple of harmless compile warnings. + +2.2.0.7 - 15 March 2006 + +- Further fixes from 2.2.0.6. Now tested with modifying + resume2= in an initrd and checking image_exists prior + to do the echo > do_resume. + +2.2.0.6 - 13 March 2006 + +- Split logic for determining whether we can suspend from + that for figuring out whether we can resume. This is + useful for the filewriter, where resume2= can be valid + (allowing us to resume) while filewriter_target isn't + set (prohibiting suspending). +- Update file headers. + +2.2.0.5 - 13 March 2006 + +- Oh dear. I somehow reversed the 2.2.0.3 fix in 2.2.0.4. + Let's try and get them both in the same patch, shall we? + +2.2.0.4 - 13 March 2006 + +- Fix missing incrementation of # extents freed in new + swap storage code. + +2.2.0.3 - 11 March 2006 + +- Add x86 compilation error fix I missed. + +2.2.0.2 - 11 March 2006 + +- Fix minimum free ram at 2000 pages rather than nr pages >> 7. +- Remove command line suspend_act, suspend_lvl and suspend_dbg + options. +- Fix issues with swap allocation that made suspend not work + properly with multiple swap storage at the same priority or + highly fragmented storage. +- Add Randy Dunlap's libata patches (sata). +- Removed time patches. +- Documentation updates +- Remove un-needed x86_64 fix processor context patch. +- Remove x86_64 mcheck_init invocation. +- Separate workqueue freezer, dynamic pageflags and + lzf patches. +- Remove obsolete configuration flags from the enums in suspend2.h. +- Correct various typos in headers. +- Various cleanups. +- Remove command line parameters suspend_act, suspend_dbg and + suspend_lvl. +- Remove HALT_ON debugging macro from pagedir.c. +- Add tags for my new patch splitting utility. + +2.2.0.1 - 15 February 2006 + +- Update to 2.6.16-rc3. +- Further cleanups, removing fragments that are no longer needed + and applying suggestions from code review. +- Remove Christoph's todo list code. Not because I don't like it, + but because it was originally designed to help SMP cleanness, + and cpu hotplug has taken over that need. + +2.2 - 24 January 2006 (Happy Birthday, Dad!) + +- Fix amd64 page_is_ram typo. +- Fix unintentional dependence on swsusp (amd64). +- Make swapwriter select swap support instead of depending + on it already being selected. +- Make a new common function for writing the last (partial) + page of a header. +- Remove temporary enabling of block dump when reading pageset1. +- Extend have_image to also display some of the details from + the header. +- Cleanup count_data_pages changes. +- Save and restore the block_dump setting when starting/finishing + a cycle. +- Add support for debugging the image header reading & writing. +- A few fixes to the filewriter. +- Cleanups in swapwriter and arch specific code. + +2.2-rc16 - 8 January 2006 + +- Small updates to documentation. +- Fix amd64 support. +- Remove no-unneeded modifications to fs/super.c & mtrr code. +- Cleanup dyn pageflags include files. +- Clean freezer header. +- Move swsusp to bottom of Makefile to make interoperability work better. +- Move pm_ops prepare/finish calls back -moving the calls to pre/post + hotplugging doesn't seem to be the right thing to do at the moment. +- Save zone info that's used during atomic restore in memory that can + be safely accessed during that period. +- Modify swsusp so it can be used with the new freezer without causing + a deadlock. +- Reinstate the ability to cancel a suspend part way through writing a + pageset. +- Fix some duplicated allowance for extra pd1 pages. +- Further work on new PageNoSave code. +- Increase max outstanding I/O to 2048. +- Remove some unnecessary forced inlining. + +2.2-rc15 - 26 December 2005 + +- Call the correct writer init routine (read/write were transposed!) +- Freeze filesystems in reverse order so loopback doesn't result in + a hang. +- Select HOTPLUG_CPU instead of depending on it (SMP). +- Further switchovers from assuming pfns start at 0 to using per zone + ranges. +- Lots of architecture specific fixes for PPC, x86_64 and ARM. +- Work on removing PageNosave dependency and changes outside kernel/power. +- Support new read-only section patches. +- Remove b44 patch. +- Cleanups (as always!). This time, we remove driver_model.[ch], and + remove some now unused MTRR modifications and lowlevel x86 code. +- Small modifications to lowlevel code routine. +- Reintroduce /proc/suspend2/no_pageset2 for testing. +- Compile fix for !SUSPEND2. +- Fixups to powerdown code (yes, again!). +- Bug fix to rw_init_plugins. +- Small rework of sanity checking. +- Make num_pcp_pages() numa compatible. +- Cleanup filesystem freezing code. +- More consolidation of swapwriter and filewriter code into bio.c. +- Attempt to parse resume device when checking if can suspend and + suspend is currently disabled. +- Rename noresume_setup to noresume2_setup so it works with swsusp + compiled in. +- Fix filewriter so it again tries to stat the file if name_to_dev_t + fails. + +2.2-rc14 - 2 December 2005 + +- Correct header_locations in documentation to headerlocations. +- Remove BUG_ON() in ll_rw_blk.c. +- Correction to nvidia agp suspend/resume support. +- Complete removal of syncthread code. +- Consolidate all of the page i/o code from swapwriter and + filewriter into one version, moved to block_io.c. +- Combine the io.c read & write pageset routines and make + separate routines of plugin setup/cleanup and the main loop. +- Fix for AMD64 reading wrong block for header at resume time + (and thus reporting 0.0.0 kernel version in the image). + +2.2-rc13 - 25 November 2005 + +- Add ARM support, thanks to Hiroki Kaminaga at Sony. +- Replace remounting with freezing and thawing filesystems. +- Remove now unneeded syncthread code (more to be done). +- Don't sys_sync after freezing filesystems. That way lies deadlocks. +- Beginnings of cleanups for writers. +- Make number of extra pagedir 1 pages allowed for tunable via a + new proc entry. Set it to 0 for autotuning. +- Fix an oops in the filewriter. +- Fix for typo in power off code (thanks, _damjan_). + +2.2-rc12 - 22 November 2005 + +- New test_bio proc debugging entry; don't actually write anything, + just say where the data would be put. +- Further filewriter fixes. It should be perfectly usable now, + but I would still recommend care and backups! +- Don't use smp_processor_id in lowlevel code. With cpuhotplug, we + know we're on processor zero. +- Fix some indentation in the lowlevel code. +- Touch softlockup watchdog when resuming. Shouldn't be needed and + doesn't seem to make any difference, but I'll seek consistency + with Pavel at this point. +- Show what processes enter and leave the fridge and when iff + PM Debug is enabled (wRAR). +- Modify remounting logic introduced in rc11 so that syslog etc + continue to work postresume. +- Remove some extra debugging code in the freezer. +- Prune the number of calls to attempt_to_parse so that we can + properly figure out whether suspending should be enabled. +- Don't unblank the console from the kernel code. Doesn't play + nicely with vesafb-tng. + +2.2-rc11 - 15 November 2005 + +- Remount filesystems readonly while suspending so as to ensure + xfs flushes all pending writes. This also means things should + be nice and clean if you fail to resume (for whatever reason). +- Fix highmem support that I broke with rc10. +- Fix typo in 2nd+ attempt at resuming message. + +2.2-rc10 - 14 November 2005 + +- Increased allowance for extra pagedir1 pages. This is primarily + for the new flgrx driver. If you don't use it, you'll see a + bigger drop in the total image size after the atomic copy. This + is nothing to be concerned about. +- Some debugging code has been temporarily modified to always + print, to help address some particular users' issues. +- More helpful debugging for AMD64 preemption problems. +- Rework of powerdown code. Powerdown method can now also be 0, + meaning Suspend won't try to use ACPI to powerdown. 1 & 2 are + also accepted values. You might find Standby (1) useful, but 2 + will probably only make the computer resume again immediately. +- Big rework of blocksize handling. Suspend2 is now tested with the + swap files and ordinary files on ext3, xfs and reiser with each + blocksize allowed by the filesystem type. Other filesystems should + work without a problem because the algorithm is basically the same + as that used for swapfiles. +- Initial version of discontiguous memory support. +- Replace /proc/software_suspend with /proc/suspend2. +- Make Suspend2 explicitly depend upon cpu hotplug for SMP. +- Leave interrupts enabled when powering down. +- Make storage manager testing proc entry depend on PM_DEBUG. +- Remove unnecessary check at the start of save image. +- Small cleanups in the swapwriter (swapfile management). + +2.2-rc9 - 5 November 2005 + +- Filewriter cleanups and bug fixes. +- Display stats when unable to allocate enough storage, to help diagnosing + this issue. +- Some cleanups to e820 table support. +- Various small fixes to PPC support. +- Modifed BIO submission check to be smarter (use new Suspend2 flag in BIO + flags). +- Add userspace storage manager (usm) support. Requires userspace program. + None are available yet. This change included a separation of the netlink + code from the userui code into a new file. +- Replace type * name with type *name everywhere. +- Made debugging code for driver suspend/resume calls more gentle + (delay/print instead of BUG_ON(). +- Don't mark image as resumed before when in KEEP_IMAGE made. +- Seek to address BUG_ONs being hit in marking a task as pageset1. +- Address some arch/config specific compilation warnings. +- Use kernel_power_off_prepare instead of device_shutdown when not entering + suspend to ram. +- Thaw kernel processes before beginning to eat memory. +- Make freezer quiet by default. +- Export freezer state for when ext3 (eg) built as a module. +- Add news_storage_manager to proc entry info so that entries that needs + access to storage for reading and/or writing can enable it as required. + There is also a new activate_storage entry that can be used in a script to + avoid having it repetitively enabled and disabled by a number of instructions. + echo 1 > /proc/suspend2/activate_storage + at the start of the sequence and + echo 0 > /proc/suspend2/activate_storage + and the end. +- Correct bits in result code so hibernate script gives the correct + interpretation again. +- Cleanup and simplify file management in the file writer. (Debugging printks + left in for this release). +- Poke blanked console when resuming. + +2.2-rc8 - October 3, 2005. + +REQUIRES hibernate script 1.12 or newer. + +- Changed name of proc directory to suspend2. +- Removed suspend_drivers_init/cleanup +- Cleaned up debug_info entry +- Remove atomic_restore conflict check +- Stop tracking max extents used. +- put_extent. Just BUG_ON(!extent). Or don't handle. +- put_extent_chain remove if (!this) return +- Removed references to chain optimisation info +- Remove error messages in _extent_chain (should + come from real cause of err) +- attempt_to_parse_resume_device should return whether successful +- Merge plugin read/write inits/cleanups into one loop. +- Merge read & write routines into one generic I/O loop? +- Sanity check kernel version - macros for digits? +- Rename free_pagedir_data to free_extra_pagedir_memory. Don't reset pageset + sizes there. +- Removed SetPageConflicting +- release_conflicting_pages: Delete. Not used. +- remove version info in proc.c +- Remove #ifdef CONFIG_SUSPEND2_DEFAULT_RESUME2 in suspend2 +- compress _bitmaps functions +- Remove #ifdef DEBUG_PAGE_ALLOC in suspend2.c +- can_suspend: if(DISABLED && !attempt_to_parse()) +- Remove keep_metadata proc entry +- Use (DISABLED && !attempt_to_parse() in suspend2_resume, too. +- Removed unused fields used in io_info struct +- flush_dcache_page needed? +- Use put_page to free pages. +- Some extraneous comments removed. +- Removed header_data comment +- /*^N *^N *^N* ^N* ^N */ +- Remove PRINTK in swapwriter +- Remove swapwriter file management? +- Cleaned up get_blocks_per_page in swapwriter +- Integrated get_path_for_swapfile to caller +- Parse signature - remove old pmdisk version support +- Removed unneeded extern int signature_check? (swapwriter:478) +- Cleaned up swapwriter_storage_allocated +- Remove space_requests = ((long) space_really_requested) << PAGE_SHIFT +- Remove allocate_header_space comment "was going to be in bytes" +- Simplified header space allocated calculation. +- Remove extra escape action in request_abort_suspend (not used and would + be a problem anyway). +- Fix compile error for AMD64 SMP. +- Fix error in the filewriter. + +2.2-rc7 - September 16 2005 + +- Further fixing of powerdown code. If !pm_ops, just use + kernel_power_off. +- Use return code when freezing failure occurs. +- Make userui disabling effective. + +2.2-rc6 - September 6 2005 (The Hamilton Release) + +- Fix for corruption caused when failing to power down. +- Fix for failing to power down. +- Separated refrigerator changes from Suspend2 for separate + merge. +- Misc other cleanups. + +2.2-rc5 - September 2 2005 +- Add more detailed debugging information regarding which + driver re-enables interrupts and recover. +- Handle wrong userui interface version by resetting userui + and continuing without it. +- Fix for powerdown issues. +- Ensure resumed before flag isn't unintentionally set when + writing header. +- Seek to address trace given in refrigerator code. +- Complain if can't find the userui process number given + for setting nofreeze. +- Fixed a bug in the swap allocation routine that would + result in swap being allocated but not freed. + +2.2-rc4 - August 30 2005 + +- Fixed bounding error in swapwriter device closing. This + had 'interesting' side effects. +- Moved storing the console log level used to before asking + the userui to cleanup, so that after suspending. + default_console_loglevel contains the last one used, and + not the original value. +- Fixed a mistake in include/asm-i386/suspend2.h that broke + compilation for gcc 4 users. +- Made the refrigerator ignore Stopped processes again. Mea + culpa. As always :| + +2.2-rc3 - August 29 2005 + +- Modified error message for setting the blocksize to give + some useful information. +- Fixed typo in todo_list_active calls in jfs_txnmgr.c +- Fixed size of blocksize storage in header. Thanks Ola! +- Merged in initramfs changes to filewriter. Added new state + flag for resuming from initramfs and call from do_mounts_rd.c. + Yet to test. +- Implemented communication of the netlink socket number to use + to the userui program when invoking it. (Changes to commit to + userui tree) +- Give more information and invalidate the image when we fail to + initialise a filter at resume. +- Clean up console when failing to resume. +- Make escape and toggling pause and single step take effect + immediately. +- Fixed log everything so that it does again. +- Expand userui message buffer from 80 to 255 characters. +- Fixed a silly bug when the header is not stored on the resume + block device. +- Modify freezer implementation to signal all processes once, + separate to waiting for them to enter the refrigerator. This + will be more robust and quicker if/when the todo list gets used + for other things. +- Modify userui version to 6. +- Move freeing of freezer notifier entry to post spinlock release. +- Fix compilation error when !SUSPEND2 & CRYPTO +- Fix missing ; in macro for !SUSPEND2 + +2.2-rc2 - August 24 2005 +- Apply LZF 64 bit from Anders Fugmann. +- Complete transition to todo list and fix up jffs2. +- Temporary fix for gcc 4, telling it lies that we're + not clobbering ebp. Something similar needed for other + archs? + +2.2-rc1 - August 23 2005 +- Fixed problem with swapwriter buffer offset not being reset + when beginning to write the image header. +- PPC fixes from Johannes Berg. +- Give a better message when can't initialise a compressor. +- Remove an unused variable from the pageset_sizes_result struct. +- Make Suspend still compile ok when networking is disabled. + Userui won't work (it uses netlink), but suspending will. + This will also be handy for embedded support. +- Say why we can't do userui when we can't open /dev/console + (entirely possible when resuming from an initrd). + +2.1.9.14 - August 22 2005 +- Fixes for previous patch - base tree was slightly corrupted. +- Also fixed agp driver cleanups - humble apologies to ATI and + Nvidia users. +- Small fixes for new bdev management code in swapwriter. + +2.1.9.13 - August 20 2005 +- Added support for Christoph Lameter's todo list changes. +- Reworked device setup at resume time. +- Began splitting up patches into many more changesets for + the (to be made) git tree. I'll now ship the patch as one + big patch. +- Miscellaneous other bug fixes and cleanups. + +2.1.9.12 - August 4 2005 +- Bug fixes for the refrigerator code +- Fixed bug in dyn_pageflags code that would cause an oops if + allocating pageflags ever failed. +- Changed block io code to use kblockd_flush. +- Reworking of semantics for workqueue calls (merge work). + +2.1.9.11 - July 30 2005 +- More cleanups. + +2.1.9.10 - July 23 2005 +- Lots more cleanups and little fixes. +- Changed workqueue freezer implementation from using an extra parameter + to using new routines with nofreeze in the name. Greatly reduces the + number of files we touch. +- Remove SYNCTHREAD flag from kernel threads. It really means nothing. + (They are frozen anyway, after real syncthreads). +- In the process of re-adding checksumming as a standalone patch. Not + to be submitted. +- Cleaned up LZF. Removed macros and ifdefs. +- Started pushing patches. +- Renamed arch specific functions to include _arch_. +- Removed all_settings after asking how many people use it. +- Fixed up copyback thread spawning so as to ensure that the stack used + is not in a conflicting page. We spwan new threads until one is ok. + Each thread spawned will keep spawning other threads (with a delay) + until one has a non conflicting page. This means you can end up with + a large number of threads quickly. We guarantee that only one will + continue on to the copyback. All others enter the freezer. + +2.1.9.9 - July 8 2005 +- Move suspend2_core directory contents up to kernel/power. +- Further work on x86_64 (more to be done before it can be tested). +- Lots of cleanups in response to premerge (I hope) comments. As part + of this, converted a number of lists of #defined bit indices to + enums. This will break saved all_settings files, so I bumped the + version. +- Added debugging info to swapwriter to try and diagnose set block size + oops some people are seeing. diff -urN oldtree/Documentation/power/internals.txt newtree/Documentation/power/internals.txt --- oldtree/Documentation/power/internals.txt 1969-12-31 19:00:00.000000000 -0500 +++ newtree/Documentation/power/internals.txt 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,238 @@ + Software Suspend 2.2 Internal Documentation. + Version 1 + +1. Introduction. + + Software Suspend 2.2 is an addition to the Linux Kernel, designed to + allow the user to quickly shutdown and quickly boot a computer, without + needing to close documents or programs. It is equivalent to the + hibernate facility in some laptops. This implementation, however, + requires no special BIOS or hardware support. + + The code in these files is based upon the original implementation + prepared by Gabor Kuti and additional work by Pavel Machek and a + host of others. This code has been substantially reworked by Nigel + Cunningham, again with the help and testing of many others, not the + least of whom is Michael Frank, At its heart, however, the operation is + essentially the same as Gabor's version. + +2. Overview of operation. + + The basic sequence of operations is as follows: + + a. Quiesce all other activity. + b. Ensure enough memory and storage space are available, and attempt + to free memory/storage if necessary. + c. Allocate the required memory and storage space. + d. Write the image. + e. Power down. + + There are a number of complicating factors which mean that things are + not as simple as the above would imply, however... + + o The activity of each process must be stopped at a point where it will + not be holding locks necessary for saving the image, or unexpectedly + restart operations due to something like a timeout and thereby make + our image inconsistent. + + o It is desirous that we sync outstanding I/O to disk before calculating + image statistics. This reduces corruption if one should suspend but + then not resume, and also makes later parts of the operation safer (see + below). + + o We need to get as close as we can to an atomic copy of the data. + Inconsistencies in the image will result in inconsistent memory contents at + resume time, and thus in instability of the system and/or file system + corruption. This would appear to imply a maximum image size of one half of + the amount of RAM, but we have a solution... (again, below). + + o In 2.6, we choose to play nicely with the other suspend-to-disk + implementations. + +3. Detailed description of internals. + + a. Quiescing activity. + + Safely quiescing the system is achieved using two methods. + + First, we note that the vast majority of processes don't need to run during + suspend. They can be 'frozen'. We therefore implement a refrigerator + routine, which processes enter and in which they remain until the cycle is + complete. Processes enter the refrigerator via try_to_freeze() invocations + at appropriate places. A process cannot be frozen in any old place. It + must not be holding locks that will be needed for writing the image or + freezing other processes. For this reason, userspace processes generally + enter the refrigerator via the signal handling code, and kernel threads at + the place in their event loops where they drop locks and yield to other + processes or sleep. + + The second part of our method for quisescing the system involves freezing + the filesystems. We use the standard freeze_bdev and thaw_bdev functions to + ensure that all of the user's data is synced to disk before we begin to + write the image. + + Quiescing the system works most quickly and reliably when we add one more + element to the algorithm: separating the freezing of userspace processes + from the freezing of kernel space processes, and doing the filesystem freeze + in between. The filesystem freeze needs to be done while kernel threads such + as kjournald can still run. At the same time, though, everything will be + less racy and run more quickly if we stop userspace submitting more I/O work + while we're trying to quiesce. + + Quiescing the system is therefore done in three steps: + - Freeze userspace + - Freeze filesystems + - Freeze kernel threads + + If we need to free memory, we thaw kernel threads and filesystems, but not + userspace. We can then free caches without worrying about deadlocks due to + swap files being on frozen filesystems or such like. + + b. Ensure enough memory & storage are available. + + We have a number of constraints to meet in order to be able to successfully + suspend and resume. + + First, the image will be written in two parts, described below. One of these + parts needs to have an atomic copy made, which of course implies a maximum + size of one half of the amount of system memory. The other part ('pageset') + is not atomically copied, and can therefore be as large or small as desired. + + Second, we have constraints on the amount of storage available. In these + calculations, we may also consider any compression that will be done. The + cryptoapi module allows the user to configure an expected compression ratio. + + Third, the user can specify an arbitrary limit on the image size, in + megabytes. This limit is treated as a soft limit, so that we don't fail the + attempt to suspend if we cannot meet this constraint. + + c. Allocate the required memory and storage space. + + Having done the initial freeze, we determine whether the above constraints + are met, and seek to allocate the metadata for the image. If the constraints + are not met, or we fail to allocate the required space for the metadata, we + seek to free the amount of memory that we calculate is needed and try again. + We allow up to four iterations of this loop before aborting the cycle. If we + do fail, it should only be because of a bug in Suspend's calculations. + + These steps are merged together in the prepare_image function, found in + prepare_image.c. The functions are merged because of the cyclical nature + of the problem of calculating how much memory and storage is needed. Since + the data structures containing the information about the image must + themselves take memory and use storage, the amount of memory and storage + required changes as we prepare the image. Since the changes are not large, + only one or two iterations will be required to achieve a solution. + + d. Write the image. + + We previously mentioned the need to create an atomic copy of the data, and + the half-of-memory limitation that is implied in this. This limitation is + circumvented by dividing the memory to be saved into two parts, called + pagesets. + + Pageset2 contains the page cache - the pages on the active and inactive + lists. These pages are saved first and reloaded last. While saving these + pages, the swapwriter module carefully ensures that the work of writing + the pages doesn't make the image inconsistent. No atomic copy of these + pages needs to be made. + + Writing the image requires memory, of course, and at this point we have + also not yet suspended the drivers. We therefore ensure that some memory is + available for both writing the image and letting drivers allocate memory + for saving state. A sysfs entry lets the user tune the amount of memory + kept free, since if DRI is enabled, this requirement can be quite large. + + Once pageset2 has been saved, we suspend the drivers and save the CPU + context before making an atomic copy of pageset1, resuming the drivers + and saving the atomic copy. After saving the two pagesets, we just need to + save our metadata before powering down. + + Having saved pageset2 pages, we can safely overwrite their contents with + the atomic copy of pageset1. This is how we manage to overcome the half of + memory limitation. Pageset2 is normally far larger than pageset1, and + pageset1 is normally much smaller than half of the memory, with the result + that pageset2 pages can be safely overwritten with the atomic copy of + pageset1. This is where we need to be careful about syncing, however. + Pageset2 will probably contain filesystem meta data. If this is overwritten + with pageset1 and then a sync occurs, the filesystem will be corrupted - + at least until resume time and another sync of the restored data. Since + there is a possibility that the user might not resume or (may it never be!) + that suspend might oops, we do our utmost to avoid syncing filesystems after + copying pageset1. + + e. Power down. + + Powering down uses standard kernel routines. Prior to this, however, we + suspend drivers again, ensuring that write caches are flushed. + +4. Data Structures. + +- For recording which pages are in each pageset, and where pageset1 will be copied to. Also need to be stored easily and quickly in the image header... pageflags. +- For recording which devices and blocks the image will be stored in ... extents. +- For recording the configuration in the image header and making enable/disabling/replacing parts simple... modules. +- For letting the user configure options... sysfs data structures. + + +4. The method of writing the image. + + Suspend2 contains an internal API which is designed to simplify the + implementation of new methods of transforming the image to be written and + writing the image itself. In early versions of Suspend2, compression support + was inlined in the image writing code, and the data structures and code for + managing swap were intertwined with the rest of the code. A number of people + had expressed interest in implementing image encryption, and alternative + methods of storing the image. This internal API makes that possible by + implementing 'modules'. + + A module is a single file which encapsulates the functionality needed + to transform a pageset of data (encryption or compression, for example), + or to write the pageset to a device. The former type of module is called + a 'page-transformer', the later a 'writer'. + + Modules are linked together in pipeline fashion. There may be zero or more + page transformers in a pipeline, and there is always exactly one writer. + The pipeline follows this pattern: + + --------------------------------- + | Suspend2 Core | + --------------------------------- + | + | + --------------------------------- + | Page transformer 1 | + --------------------------------- + | + | + --------------------------------- + | Page transformer 2 | + --------------------------------- + | + | + --------------------------------- + | Writer | + --------------------------------- + + During the writing of an image, the core code feeds pages one at a time + to the first module. This module performs whatever transformations it + implements on the incoming data, completely consuming the incoming data and + feeding output in a similar manner to the next module. A module may buffer + its output. + + During reading, the pipeline works in the reverse direction. The core code + calls the first module with the address of a buffer which should be filled. + (Note that the buffer size is always PAGE_SIZE at this time). This module + will in turn request data from the next module and so on down until the + writer is made to read from the stored image. + + Part of definition of the structure of a module thus looks like this: + + int (*rw_init) (int rw, int stream_number); + int (*rw_cleanup) (int rw); + int (*write_chunk) (struct page *buffer_page); + int (*read_chunk) (struct page *buffer_page, int sync); + + It should be noted that the _cleanup routine may be called before the + full stream of data has been read or written. While writing the image, + the user may (depending upon settings) choose to abort suspending, and + if we are in the midst of writing the last portion of the image, a portion + of the second pageset may be reread. diff -urN oldtree/Documentation/power/suspend2.txt newtree/Documentation/power/suspend2.txt --- oldtree/Documentation/power/suspend2.txt 1969-12-31 19:00:00.000000000 -0500 +++ newtree/Documentation/power/suspend2.txt 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,703 @@ + --- Suspend2, version 2.2 --- + +1. What is it? +2. Why would you want it? +3. What do you need to use it? +4. Why not just use the version already in the kernel? +5. How do you use it? +6. What do all those entries in /sys/power/suspend2 do? +7. How do you get support? +8. I think I've found a bug. What should I do? +9. When will XXX be supported? +10 How does it work? +11. Who wrote Suspend2? + +1. What is it? + + Imagine you're sitting at your computer, working away. For some reason, you + need to turn off your computer for a while - perhaps it's time to go home + for the day. When you come back to your computer next, you're going to want + to carry on where you left off. Now imagine that you could push a button and + have your computer store the contents of its memory to disk and power down. + Then, when you next start up your computer, it loads that image back into + memory and you can carry on from where you were, just as if you'd never + turned the computer off. Far less time to start up, no reopening + applications and finding what directory you put that file in yesterday. + That's what Suspend2 does. + + Suspend2 has a long heritage. It began life as work by Gabor Kuti, who, + with some help from Pavel Machek, got an early version going in 1999. The + project was then taken over by Florent Chabaud while still in alpha version + numbers. Nigel Cunningham came on the scene when Florent was unable to + continue, moving the project into betas, then 1.0, 2.0 and so on up to + the present 2.2 series. Pavel Machek's swsusp code, which was merged around + 2.5.17 retains the original name, and was essentially a fork of the beta + code until Rafael Wysocki came on the scene in 2005 and began to improve it + further. + +2. Why would you want it? + + Why wouldn't you want it? + + Being able to save the state of your system and quickly restore it improves + your productivity - you get a useful system in far less time than through + the normal boot process. + +3. What do you need to use it? + + a. Kernel Support. + + i) The Suspend2 patch. + + Suspend2 is part of the Linux Kernel. This version is not part of Linus's + 2.6 tree at the moment, so you will need to download the kernel source and + apply the latest patch. Having done that, enable the appropriate options in + make [menu|x]config (under Power Management Options), compile and install your + kernel. Suspend2 works with SMP, Highmem, preemption, x86-32, PPC and x86_64. + + Suspend2 patches are available from http://suspend2.net. + + ii) Compression and encryption support. + + Compression and encryption support are implemented via the + cryptoapi. You will therefore want to select any Cryptoapi transforms that + you want to use on your image from the Cryptoapi menu while configuring + your kernel. + + You can also tell Suspend to write it's image to an encrypted and/or + compressed filesystem/swap partition. In that case, you don't need to do + anything special for Suspend2 when it comes to kernel configuration. + + iii) Configuring other options. + + While you're configuring your kernel, try to configure as much as possible + to build as modules. We recommend this because there are a number of drivers + that are still in the process of implementing proper power management + support. In those cases, the best way to work around their current lack is + to build them as modules and remove the modules while suspending. You might + also bug the driver authors to get their support up to speed, or even help! + + b. Storage. + + i) Swap. + + Suspend2 can store the suspend image in your swap partition, a swap file or + a combination thereof. Whichever combination you choose, you will probably + want to create enough swap space to store the largest image you could have, + plus the space you'd normally use for swap. A good rule of thumb would be + to calculate the amount of swap you'd want without using Suspend2, and then + add the amount of memory you have. This swapspace can be arranged in any way + you'd like. It can be in one partition or file, or spread over a number. The + only requirement is that they be active when you start a suspend cycle. + + There is one exception to this requirement. Suspend2 has the ability to turn + on one swap file or partition at the start of suspending and turn it back off + at the end. If you want to ensure you have enough memory to store a image + when your memory is fully used, you might want to make one swap partition or + file for 'normal' use, and another for Suspend2 to activate & deactivate + automatically. (Further details below). + + ii) Normal files. + + Suspend2 includes a 'filewriter'. The filewriter can store your image in a + simple file. Since Linux has the idea of everything being a file, this is + more powerful than it initially sounds. If, for example, you were to set up + a network block device file, you could suspend to a network server. This has + been tested and works to a point, but nbd itself isn't stateless enough for + our purposes. + + Take extra care when setting up the filewriter. If you just type commands + without thinking and then try to suspend, you could cause irreversible + corruption on your filesystems! Make sure you have backups. + + Most people will only want to suspend to a local file. To achieve that, do + something along the lines of: + + echo "Suspend2" > /suspend-file + dd if=/dev/zero bs=1M count=512 >> suspend-file + + This will create a 512MB file called /suspend-file. To get Suspend2 to use + it: + + echo /suspend-file > /sys/power/suspend2/filewriter/filewriter_target + + Then + + cat /sys/power/suspend2/resume2 + + Put the results of this into your bootloader's configuration (see also step + C, below: + + ---EXAMPLE-ONLY-DON'T-COPY-AND-PASTE--- + # cat /sys/power/suspend2/resume2 + file:/dev/hda2:0x1e001 + + In this example, we would edit the append= line of our lilo.conf|menu.lst + so that it included: + + resume2=file:/dev/hda2:0x1e001 + ---EXAMPLE-ONLY-DON'T-COPY-AND-PASTE--- + + For those who are thinking 'Could I make the file sparse?', the answer is + 'No!'. At the moment, there is no way for Suspend2 to fill in the holes in + a sparse file while suspending. In the longer term (post merge!), I'd like + to change things so that the file could be dynamically resized as needed. + Right now, however, that's not possible and not a priority. + + c. Bootloader configuration. + + Using Suspend2 also requires that you add an extra parameter to + your lilo.conf or equivalent. Here's an example for a swap partition: + + append="resume2=swap:/dev/hda1" + + This would tell Suspend2 that /dev/hda1 is a swap partition you + have. Suspend2 will use the swap signature of this partition as a + pointer to your data when you suspend. This means that (in this example) + /dev/hda1 doesn't need to be _the_ swap partition where all of your data + is actually stored. It just needs to be a swap partition that has a + valid signature. + + You don't need to have a swap partition for this purpose. Suspend2 + can also use a swap file, but usage is a little more complex. Having made + your swap file, turn it on and do + + cat /sys/power/suspend2/swapwriter/headerlocations + + (this assumes you've already compiled your kernel with Suspend2 + support and booted it). The results of the cat command will tell you + what you need to put in lilo.conf: + + For swap partitions like /dev/hda1, simply use resume2=/dev/hda1. + For swapfile `swapfile`, use resume2=swap:/dev/hda2:0x242d. + + If the swapfile changes for any reason (it is moved to a different + location, it is deleted and recreated, or the filesystem is + defragmented) then you will have to check + /sys/power/suspend2/swapwriter/headerlocations for a new resume_block value. + + Once you've compiled and installed the kernel and adjusted your bootloader + configuration, you should only need to reboot for the most basic part + of Suspend2 to be ready. + + If you only compile in the swapwriter, or only compile in the filewriter, + you don't need to add the "swap:" part of the resume2= parameters above. + resume2=/dev/hda2:0x242d will work just as well. + + d. The hibernate script. + + Since the driver model in 2.6 kernels is still being developed, you may need + to do more, however. Users of Suspend2 usually start the process via a script + which prepares for the suspend, tells the kernel to do its stuff and then + restore things afterwards. This script might involve: + + - Switching to a text console and back if X doesn't like the video card + status on resume. + - Un/reloading PCMCIA support since it doesn't play well with suspend. + + Note that you might not be able to unload some drivers if there are + processes using them. You might have to kill off processes that hold + devices open. Hint: if your X server accesses an USB mouse, doing a + 'chvt' to a text console releases the device and you can unload the + module. + + Check out the latest script (available on suspend2.net). + +4. Why not just use the version already in the kernel? + + The version in the vanilla kernel has a number of drawbacks. Among these: + - it has a maximum image size of 1/2 total memory. + - it doesn't allocate storage until after it has snapshotted memory. + This means that you can't be sure suspending will work until you + see it start to write the image. + - it performs all of it's I/O synchronously. + - it does not allow you to press escape to cancel a cycle + - it does not allow you to automatically swapon a file when + starting a cycle. + - it does not allow you to use multiple swap partitions. + - it does not allow you to use swapfiles. + - it does not allow you to use ordinary files. + - it just invalidates an image and continues to boot if you + accidentally boot the wrong kernel after suspending. + - it doesn't support any sort of nice display while suspending + - it is moving toward requiring that you have an initrd/initramfs + to ever have a hope of resuming (uswsusp). While uswsusp will + address some of the concerns above, it won't address all, and + will be more complicated to get set up. + +5. How do you use it? + + A suspend cycle can be started directly by doing: + + echo > /sys/power/suspend2/do_resume + + In practice, though, you'll probably want to use the hibernate script + to unload modules, configure the kernel the way you like it and so on. + In that case, you'd do (as root): + + hibernate + + See the hibernate script's man page for more details on the options it + takes. + + If you're using the text or splash user interface modules, one neat feature + of Suspend2 that you might find useful is that you can press Escape at any + time during suspending, and the process will be aborted. + + Due to the way suspend works, this means you'll have your system back and + perfectly usable almost instantly. The only exception is when it's at the + very end of writing the image. Then it will need to reload a small ( + usually 4-50MBs, depending upon the image characteristics) portion first. + + If you run into problems with resuming, adding the "noresume2" option to + the kernel command line will let you skip the resume step and recover your + system. + +6. What do all those entries in /sys/power/suspend2 do? + + /sys/power/suspend2 is the directory which contains files you can use to + tune and configure Suspend2 to your liking. The exact contents of + the directory will depend upon the version of Suspend2 you're + running and the options you selected at compile time. In the following + descriptions, names in brackets refer to compile time options. + (Note that they're all dependant upon you having selected CONFIG_SUSPEND2 + in the first place!). + + Since the values of these settings can open potential security risks, they + are usually accessible only to the root user. You can, however, enable a + compile time option which makes all of these files world-accessible. This + should only be done if you trust everyone with shell access to this + computer! + + - compression/algorithm + + Set the cryptoapi algorithm used for compressing the image. + + - compression/expected_compression + + These values allow you to set an expected compression ratio, which Software + Suspend will use in calculating whether it meets constraints on the image + size. If this expected compression ratio is not attained, the suspend will + abort, so it is wise to allow some spare. You can see what compression + ratio is achieved in the logs after suspending. + + - debug_info: + + This file returns information about your configuration that may be helpful + in diagnosing problems with suspending. + + - do_resume: + + When anything is written to this file suspend will attempt to read and + restore an image. If there is no image, it will return almost immediately. + If an image exists, the echo > will never return. Instead, the original + kernel context will be restored and the original echo > do_suspend will + return. + + - do_suspend: + + When anything is written to this file, the kernel side of Suspend2 will + begin to attempt to write an image to disk and power down. You'll normally + want to run the hibernate script instead, to get modules unloaded first. + + - driver_model_beeping + + Enable beeping when suspending and resuming the drivers. Might help with + determining where a problem in resuming occurs. + + - */enabled + + These option can be used to temporarily disable various parts of suspend. + + - encryption/* + + The iv, key, save_key_and_iv, mode and algorithm values allow you to + select a cryptoapi encryption algoritm, set the iv and key and whether + they are saved in the image header. Saving the iv and key in the image + header is of course less secure than having them on some external device, + such as a USB key. If you want to use a USB key, you'll need to write + some scripting in your initrd/ramfs to retrieve the key & iv from your + USB key and put them into the entries again prior to doing the echo to + do_resume. + + - extra_pages_allowance + + When Suspend2 does its atomic copy, it calls the driver model suspend + and resume methods. If you have DRI enabled with a driver such as fglrx, + this can result in the driver allocating a substantial amount of memory + for storing its state. Extra_pages_allowance tells suspend2 how much + extra memory it should ensure is available for those allocations. If + your attempts at suspending end with a message in dmesg indicating that + insufficient extra pages were allowed, you need to increase this value. + + - filewriter/target: + + Read this value to get the current setting. Write to it to point Suspend + at a new storage location for the filewriter. See above for details of how + to set up the filewriter. + + - freezer_test + + This entry can be used to get Suspend2 to just test the freezer without + actually doing a suspend cycle. It is useful for diagnosing freezing + issues. + + - image_exists: + + Can be used in a script to determine whether a valid image exists at the + location currently pointed to by resume2=. Returns up to three lines. + The first is whether an image exists (-1 for unsure, otherwise 0 or 1). + If an image eixsts, additional lines will return the machine and version. + Echoing anything to this entry removes any current image. + + - image_size_limit: + + The maximum size of suspend image written to disk, measured in megabytes + (1024*1024). + + - interface_version: + + The value returned by this file can be used by scripts and configuration + tools to determine what entries should be looked for. The value is + incremented whenever an entry in /sys/power/suspend2 is obsoleted or + added. + + - last_result: + + The result of the last suspend, as defined in + include/linux/suspend-debug.h with the values SUSPEND_ABORTED to + SUSPEND_KEPT_IMAGE. This is a bitmask. + + - log_everything (CONFIG_PM_DEBUG): + + Setting this option results in all messages printed being logged. Normally, + only a subset are logged, so as to not slow the process and not clutter the + logs. Useful for debugging. It can be toggled during a cycle by pressing + 'L'. + + - pause_between_steps (CONFIG_PM_DEBUG): + + This option is used during debugging, to make Suspend2 pause between + each step of the process. It is ignored when the nice display is on. + + - powerdown_method: + + Used to select a method by which Suspend2 should powerdown after writing the + image. Currently: + + 0: Don't use ACPI to power off. + 3: Attempt to enter Suspend-to-ram. + 4: Attempt to enter ACPI S4 mode. + 5: Attempt to power down via ACPI S5 mode. + + Note that these options are highly dependant upon your hardware & software: + + 3: When succesful, your machine suspends-to-ram instead of powering off. + The advantage of using this mode is that it doesn't matter whether your + battery has enough charge to make it through to your next resume. If it + lasts, you will simply resume from suspend to ram (and the image on disk + will be discarded). If the battery runs out, you will resume from disk + instead. The disadvantage is that it takes longer than a normal + suspend-to-ram to enter the state, since the suspend-to-disk image needs + to be written first. + 4/5: When successful, your machine will be off and comsume (almost) no power. + But it might still react to some external events like opening the lid or + trafic on a network or usb device. For the bios, resume is then the same + as warm boot, similar to a situation where you used the command `reboot' + to reboot your machine. If your machine has problems on warm boot or if + you want to protect your machine with the bios password, this is probably + not the right choice. Mode 4 may be necessary on some machines where ACPI + wake up methods need to be run to properly reinitialise hardware after a + suspend-to-disk cycle. + 0: Switch the machine completely off. The only possible wakeup is the power + button. For the bios, resume is then the same as a cold boot, in + particular you would have to provide your bios boot password if your + machine uses that feature for booting. + + - progressbar_granularity_limit: + + This option can be used to limit the granularity of the progress bar + displayed with a bootsplash screen. The value is the maximum number of + steps. That is, 10 will make the progress bar jump in 10% increments. + + - reboot: + + This option causes Suspend2 to reboot rather than powering down + at the end of saving an image. It can be toggled during a cycle by pressing + 'R'. + + - resume_commandline: + + This entry can be read after resuming to see the commandline that was used + when resuming began. You might use this to set up two bootloader entries + that are the same apart from the fact that one includes a extra append= + argument "at_work=1". You could then grep resume_commandline in your + post-resume scripts and configure networking (for example) differently + depending upon whether you're at home or work. resume_commandline can be + set to arbitrary text if you wish to remove sensitive contents. + + - swapwriter/swapfilename: + + This entry is used to specify the swapfile or partition that + Suspend2 will attempt to swapon/swapoff automatically. Thus, if + I normally use /dev/hda1 for swap, and want to use /dev/hda2 for specifically + for my suspend image, I would + + echo /dev/hda2 > /sys/power/suspend2/swapwriter/swapfile + + /dev/hda2 would then be automatically swapon'd and swapoff'd. Note that the + swapon and swapoff occur while other processes are frozen (including kswapd) + so this swap file will not be used up when attempting to free memory. The + parition/file is also given the highest priority, so other swapfiles/partitions + will only be used to save the image when this one is filled. + + The value of this file is used by headerlocations along with any currently + activated swapfiles/partitions. + + - swapwriter/headerlocations: + + This option tells you the resume2= options to use for swap devices you + currently have activated. It is particularly useful when you only want to + use a swap file to store your image. See above for further details. + + - toggle_process_nofreeze + + This entry can be used to toggle the NOFREEZE flag on a process, to allow it + to run during Suspending. It should be used with extreme caution. There are + strict limitations on what a process running during suspend can do. This is + really only intended for use by Suspend's helpers (userui in particular). + + - userui_program + + This entry is used to tell Suspend what userspace program to use for + providing a user interface while suspending. The program uses a netlink + socket to pass messages back and forward to the kernel, allowing all of the + functions formerly implemented in the kernel user interface components. + + - user_interface/debug_sections (CONFIG_PM_DEBUG): + + This value, together with the console log level, controls what debugging + information is displayed. The console log level determines the level of + detail, and this value determines what detail is displayed. This value is + a bit vector, and the meaning of the bits can be found in the kernel tree + in include/linux/suspend2.h. It can be overridden using the kernel's + command line option suspend_dbg. + + - user_interface/default_console_level (CONFIG_PM_DEBUG): + + This determines the value of the console log level at the start of a + suspend cycle. If debugging is compiled in, the console log level can be + changed during a cycle by pressing the digit keys. Meanings are: + + 0: Nice display. + 1: Nice display plus numerical progress. + 2: Errors only. + 3: Low level debugging info. + 4: Medium level debugging info. + 5: High level debugging info. + 6: Verbose debugging info. + + - user_interface/enable_escape: + + Setting this to "1" will enable you abort a suspend by + pressing escape, "0" (default) disables this feature. Note that enabling + this option means that you cannot initiate a suspend and then walk away + from your computer, expecting it to be secure. With feature disabled, + you can validly have this expectation once Suspend begins to write the + image to disk. (Prior to this point, it is possible that Suspend might + about because of failure to freeze all processes or because constraints + on its ability to save the image are not met). + + - version: + + The version of suspend you have compiled into the currently running kernel. + +7. How do you get support? + + Glad you asked. Suspend2 is being actively maintained and supported + by Nigel (the guy doing most of the kernel coding at the moment), Bernard + (who maintains the hibernate script and userspace user interface components) + and its users. + + Resources availble include HowTos, FAQs and a Wiki, all available via + suspend2.net. You can find the mailing lists there. + +8. I think I've found a bug. What should I do? + + By far and a way, the most common problems people have with suspend2 + related to drivers not having adequate power management support. In this + case, it is not a bug with suspend2, but we can still help you. As we + mentioned above, such issues can usually be worked around by building the + functionality as modules and unloading them while suspending. Please visit + the Wiki for up-to-date lists of known issues and work arounds. + + If this information doesn't help, try running: + + hibernate --bug-report + + ..and sending the output to the users mailing list. + + Good information on how to provide us with useful information from an + oops is found in the file REPORTING-BUGS, in the top level directory + of the kernel tree. If you get an oops, please especially note the + information about running what is printed on the screen through ksymoops. + The raw information is useless. + +9. When will XXX be supported? + + If there's a feature missing from Suspend2 that you'd like, feel free to + ask. We try to be obliging, within reason. + + Patches are welcome. Please send to the list. + +10. How does it work? + + Suspend2 does its work in a number of steps. + + a. Freezing system activity. + + The first main stage in suspending is to stop all other activity. This is + achieved in stages. Processes are considered in fours groups, which we will + describe in reverse order for clarity's sake: Threads with the PF_NOFREEZE + flag, kernel threads without this flag, userspace processes with the + PF_SYNCTHREAD flag and all other processes. The first set (PF_NOFREEZE) are + untouched by the refrigerator code. They are allowed to run during suspending + and resuming, and are used to support user interaction, storage access or the + like. Other kernel threads (those unneeded while suspending) are frozen last. + This leaves us with userspace processes that need to be frozen. When a + process enters one of the *_sync system calls, we set a PF_SYNCTHREAD flag on + that process for the duration of that call. Processes that have this flag are + frozen after processes without it, so that we can seek to ensure that dirty + data is synced to disk as quickly as possible in a situation where other + processes may be submitting writes at the same time. Freezing the processes + that are submitting data stops new I/O from being submitted. Syncthreads can + then cleanly finish their work. So the order is: + + - Userspace processes without PF_SYNCTHREAD or PF_NOFREEZE; + - Userspace processes with PF_SYNCTHREAD (they won't have NOFREEZE); + - Kernel processes without PF_NOFREEZE. + + b. Eating memory. + + For a successful suspend, you need to have enough disk space to store the + image and enough memory for the various limitations of Suspend2's + algorithm. You can also specify a maximum image size. In order to attain + to those constraints, Suspend2 may 'eat' memory. If, after freezing + processes, the constraints aren't met, Suspend2 will thaw all the + other processes and begin to eat memory until its calculations indicate + the constraints are met. It will then freeze processes again and recheck + its calculations. + + c. Allocation of storage. + + Next, Suspend2 allocates the storage that will be used to save + the image. + + The core of Suspend2 knows nothing about how or where pages are stored. We + therefore request the active writer (remember you might have compiled in + more than one!) to allocate enough storage for our expect image size. If + this request cannot be fulfilled, we eat more memory and try again. If it + is fulfiled, we seek to allocate additional storage, just in case our + expected compression ratio (if any) isn't achieved. This time, however, we + just continue if we can't allocate enough storage. + + If these calls to our writer change the characteristics of the image such + that we haven't allocated enough memory, we also loop. (The writer may well + need to allocate space for its storage information). + + d. Write the first part of the image. + + Suspend2 stores the image in two sets of pages called 'pagesets'. + Pageset 2 contains pages on the active and inactive lists; essentially + the page cache. Pageset 1 contains all other pages, including the kernel. + We use two pagesets for one important reason: We need to make an atomic copy + of the kernel to ensure consistency of the image. Without a second pageset, + that would limit us to an image that was at most half the amount of memory + available. Using two pagesets allows us to store a full image. Since pageset + 2 pages won't be needed in saving pageset 1, we first save pageset 2 pages. + We can then make our atomic copy of the remaining pages using both pageset 2 + pages and any other pages that are free. While saving both pagesets, we are + careful not to corrupt the image. Among other things, we use lowlevel block + I/O routines that don't change the pagecache contents. + + The next step, then, is writing pageset 2. + + e. Suspending drivers and storing processor context. + + Having written pageset2, Suspend2 calls the power management functions to + notify drivers of the suspend, and saves the processor state in preparation + for the atomic copy of memory we are about to make. + + f. Atomic copy. + + At this stage, everything else but the Suspend2 code is halted. Processes + are frozen or idling, drivers are quiesced and have stored (ideally and where + necessary) their configuration in memory we are about to atomically copy. + In our lowlevel architecture specific code, we have saved the CPU state. + We can therefore now do our atomic copy before resuming drivers etc. + + g. Save the atomic copy (pageset 1). + + Suspend can then write the atomic copy of the remaining pages. Since we + have copied the pages into other locations, we can continue to use the + normal block I/O routines without fear of corruption our image. + + f. Save the suspend header. + + Nearly there! We save our settings and other parameters needed for + reloading pageset 1 in a 'suspend header'. We also tell our writer to + serialise its data at this stage, so that it can reread the image at resume + time. Note that the writer can write this data in any format - in the case + of the swapwriter, for example, it splits header pages in 4092 byte blocks, + using the last four bytes to link pages of data together. This is completely + transparent to the core. + + g. Set the image header. + + Finally, we edit the header at our resume2= location. The signature is + changed by the writer to reflect the fact that an image exists, and to point + to the start of that data if necessary (swapwriter). + + h. Power down. + + Or reboot if we're debugging and the appropriate option is selected. + + Whew! + + Reloading the image. + -------------------- + + Reloading the image is essentially the reverse of all the above. We load + our copy of pageset 1, being careful to choose locations that aren't going + to be overwritten as we copy it back (We start very early in the boot + process, so there are no other processes to quiesce here). We then copy + pageset 1 back to its original location in memory and restore the process + context. We are now running with the original kernel. Next, we reload the + pageset 2 pages, free the memory and swap used by Suspend2, restore + the pageset header and restart processes. Sounds easy in comparison to + suspending, doesn't it! + + There is of course more to Suspend2 than this, but this explanation + should be a good start. If there's interest, I'll write further + documentation on range pages and the low level I/O. + +11. Who wrote Suspend2? + + (Answer based on the writings of Florent Chabaud, credits in files and + Nigel's limited knowledge; apologies to anyone missed out!) + + The main developers of Suspend2 have been... + + Gabor Kuti + Pavel Machek + Florent Chabaud + Bernard Blackham + Nigel Cunningham + + They have been aided in their efforts by a host of hundreds, if not thousands + of testers and people who have submitted bug fixes & suggestions. Of special + note are the efforts of Michael Frank, who had his computers repetitively + suspend and resume for literally tens of thousands of cycles and developed + scripts to stress the system and test Suspend2 far beyond the point + most of us (Nigel included!) would consider testing. His efforts have + contributed as much to Suspend2 as any of the names above. diff -urN oldtree/MAINTAINERS newtree/MAINTAINERS --- oldtree/MAINTAINERS 2006-10-05 15:26:55.000000000 -0400 +++ newtree/MAINTAINERS 2006-10-06 17:28:20.000000000 -0400 @@ -2882,6 +2882,13 @@ W: http://sammy.net/sun3/ S: Maintained +SUSPEND2 +P: Nigel Cunningham +M: nigel@suspend2.net +L: suspend2-devel@suspend2.net +W: http://suspend2.net +S: Maintained + SVGA HANDLING P: Martin Mares M: mj@ucw.cz diff -urN oldtree/arch/i386/mm/init.c newtree/arch/i386/mm/init.c --- oldtree/arch/i386/mm/init.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/i386/mm/init.c 2006-10-06 17:28:20.000000000 -0400 @@ -385,7 +385,7 @@ #endif } -#if defined(CONFIG_SOFTWARE_SUSPEND) || defined(CONFIG_ACPI_SLEEP) +#if defined(CONFIG_SUSPEND_SHARED) || defined(CONFIG_ACPI_SLEEP) /* * Swap suspend & friends need this for resume because things like the intel-agp * driver might have split up a kernel 4MB mapping. diff -urN oldtree/arch/i386/power/Makefile newtree/arch/i386/power/Makefile --- oldtree/arch/i386/power/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/i386/power/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -1,2 +1,2 @@ obj-$(CONFIG_PM) += cpu.o -obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o +obj-$(CONFIG_SUSPEND_SHARED) += swsusp.o diff -urN oldtree/arch/powerpc/kernel/Makefile newtree/arch/powerpc/kernel/Makefile --- oldtree/arch/powerpc/kernel/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/powerpc/kernel/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -36,7 +36,7 @@ obj-$(CONFIG_CRASH_DUMP) += crash_dump.o obj-$(CONFIG_6xx) += idle_6xx.o l2cr_6xx.o cpu_setup_6xx.o obj-$(CONFIG_TAU) += tau_6xx.o -obj32-$(CONFIG_SOFTWARE_SUSPEND) += swsusp_32.o +obj32-$(CONFIG_SUSPEND_SHARED) += swsusp_32.o obj32-$(CONFIG_MODULES) += module_32.o obj-$(CONFIG_E500) += perfmon_fsl_booke.o diff -urN oldtree/arch/x86_64/kernel/Makefile newtree/arch/x86_64/kernel/Makefile --- oldtree/arch/x86_64/kernel/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/x86_64/kernel/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -26,7 +26,7 @@ obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump.o obj-$(CONFIG_PM) += suspend.o -obj-$(CONFIG_SOFTWARE_SUSPEND) += suspend_asm.o +obj-$(CONFIG_SUSPEND_SHARED) += suspend_asm.o obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_IOMMU) += pci-gart.o aperture.o diff -urN oldtree/arch/x86_64/kernel/suspend.c newtree/arch/x86_64/kernel/suspend.c --- oldtree/arch/x86_64/kernel/suspend.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/x86_64/kernel/suspend.c 2006-10-06 17:28:20.000000000 -0400 @@ -140,7 +140,7 @@ } -#ifdef CONFIG_SOFTWARE_SUSPEND +#ifdef CONFIG_SUSPEND_SHARED /* Defined in arch/x86_64/kernel/suspend_asm.S */ extern int restore_image(void); @@ -209,6 +209,12 @@ return 0; } +#ifdef CONFIG_SUSPEND2 +void suspend2_beep(int suspending); +#else +#define suspend2_beep(num) do { } while(0) +#endif + int swsusp_arch_resume(void) { int error; @@ -216,7 +222,8 @@ /* We have got enough memory and from now on we cannot recover */ if ((error = set_up_temporary_mappings())) return error; + suspend2_beep(0); restore_image(); return 0; } -#endif /* CONFIG_SOFTWARE_SUSPEND */ +#endif /* CONFIG_SUSPEND_SHARED */ diff -urN oldtree/arch/x86_64/kernel/time.c newtree/arch/x86_64/kernel/time.c --- oldtree/arch/x86_64/kernel/time.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/arch/x86_64/kernel/time.c 2006-10-06 17:28:20.000000000 -0400 @@ -1064,9 +1064,9 @@ #endif } else vxtime.last_tsc = get_cycles_sync(); - write_sequnlock_irqrestore(&xtime_lock,flags); jiffies += sleep_length; monotonic_base += sleep_length * (NSEC_PER_SEC/HZ); + write_sequnlock_irqrestore(&xtime_lock,flags); touch_softlockup_watchdog(); return 0; } diff -urN oldtree/block/ll_rw_blk.c newtree/block/ll_rw_blk.c --- oldtree/block/ll_rw_blk.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/block/ll_rw_blk.c 2006-10-06 17:28:20.000000000 -0400 @@ -28,6 +28,8 @@ #include #include #include +#include +#include /* * for max sense size @@ -3242,6 +3244,10 @@ bdevname(bio->bi_bdev,b)); } + BUG_ON( test_suspend_state(SUSPEND_RUNNING) && /* Suspend2, that is */ + test_freezer_state(FREEZING_COMPLETE) && + !(bio->bi_flags & (1 << BIO_SUSPEND_DATA))); + generic_make_request(bio); } diff -urN oldtree/crypto/Kconfig newtree/crypto/Kconfig --- oldtree/crypto/Kconfig 2006-10-05 15:26:55.000000000 -0400 +++ newtree/crypto/Kconfig 2006-10-06 17:28:20.000000000 -0400 @@ -403,6 +403,13 @@ You will most probably want this if using IPSec. +config CRYPTO_LZF + tristate "LZF compression algorithm" + depends on CRYPTO + help + This is the LZF algorithm. It is especially useful for Suspend2, + because it achieves good compression quickly. + config CRYPTO_MICHAEL_MIC tristate "Michael MIC keyed digest algorithm" select CRYPTO_ALGAPI diff -urN oldtree/crypto/Makefile newtree/crypto/Makefile --- oldtree/crypto/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/crypto/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -40,5 +40,6 @@ obj-$(CONFIG_CRYPTO_DEFLATE) += deflate.o obj-$(CONFIG_CRYPTO_MICHAEL_MIC) += michael_mic.o obj-$(CONFIG_CRYPTO_CRC32C) += crc32c.o +obj-$(CONFIG_CRYPTO_LZF) += lzf.o obj-$(CONFIG_CRYPTO_TEST) += tcrypt.o diff -urN oldtree/crypto/lzf.c newtree/crypto/lzf.c --- oldtree/crypto/lzf.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/crypto/lzf.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,335 @@ +/* + * Cryptoapi LZF compression module. + * + * Copyright (c) 2004-2005 Nigel Cunningham + * + * based on the deflate.c file: + * + * Copyright (c) 2003 James Morris + * + * and upon the LZF compression module donated to the Suspend2 project with + * the following copyright: + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * Copyright (c) 2000-2003 Marc Alexander Lehmann + * + * Redistribution and use in source and binary forms, with or without modifica- + * tion, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * 3. The name of the author may not be used to endorse or promote products + * derived from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MER- + * CHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO + * EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPE- + * CIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, + * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTH- + * ERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Alternatively, the contents of this file may be used under the terms of + * the GNU General Public License version 2 (the "GPL"), in which case the + * provisions of the GPL are applicable instead of the above. If you wish to + * allow the use of your version of this file only under the terms of the + * GPL and not to allow others to use your version of this file under the + * BSD license, indicate your decision by deleting the provisions above and + * replace them with the notice and other provisions required by the GPL. If + * you do not delete the provisions above, a recipient may use your version + * of this file under either the BSD or the GPL. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +struct lzf_ctx { + void *hbuf; + unsigned int bufofs; +}; + +/* + * size of hashtable is (1 << hlog) * sizeof (char *) + * decompression is independent of the hash table size + * the difference between 15 and 14 is very small + * for small blocks (and 14 is also faster). + * For a low-memory configuration, use hlog == 13; + * For best compression, use 15 or 16. + */ +static const int hlog = 14; + +/* + * don't play with this unless you benchmark! + * decompression is not dependent on the hash function + * the hashing function might seem strange, just believe me + * it works ;) + */ +static inline u16 first(const u8 *p) +{ + return ((p[0]) << 8) + p[1]; +} + +static inline u16 next(u8 v, const u8 *p) +{ + return ((v) << 8) + p[2]; +} + +static inline u32 idx(unsigned int h) +{ + return (((h ^ (h << 5)) >> (3*8 - hlog)) + h*3) & ((1 << hlog) - 1); +} + +/* + * IDX works because it is very similar to a multiplicative hash, e.g. + * (h * 57321 >> (3*8 - hlog)) + * the next one is also quite good, albeit slow ;) + * (int)(cos(h & 0xffffff) * 1e6) + */ + +static const int max_lit = (1 << 5); +static const int max_off = (1 << 13); +static const int max_ref = ((1 << 8) + (1 << 3)); + +/* + * compressed format + * + * 000LLLLL ; literal + * LLLOOOOO oooooooo ; backref L + * 111OOOOO LLLLLLLL oooooooo ; backref L+7 + * + */ + +static void lzf_compress_exit(struct crypto_tfm *tfm) +{ + struct lzf_ctx *ctx = crypto_tfm_ctx(tfm); + + if (ctx->hbuf) { + vfree(ctx->hbuf); + ctx->hbuf = NULL; + } +} + +static int lzf_compress_init(struct crypto_tfm *tfm) +{ + struct lzf_ctx *ctx = crypto_tfm_ctx(tfm); + + /* Get LZF ready to go */ + ctx->hbuf = vmalloc_32((1 << hlog) * sizeof(char *)); + if (!ctx->hbuf) { + printk(KERN_WARNING + "Failed to allocate %ld bytes for lzf workspace\n", + (long) ((1 << hlog) * sizeof(char *))); + return -ENOMEM; + } + return 0; +} + +static int lzf_compress(struct crypto_tfm *tfm, const u8 *in_data, + unsigned int in_len, u8 *out_data, unsigned int *out_len) +{ + struct lzf_ctx *ctx = crypto_tfm_ctx(tfm); + const u8 **htab = ctx->hbuf; + const u8 **hslot; + const u8 *ip = in_data; + u8 *op = out_data; + const u8 *in_end = ip + in_len; + u8 *out_end = op + *out_len - 3; + const u8 *ref; + + unsigned int hval = first(ip); + unsigned long off; + int lit = 0; + + memset(htab, 0, sizeof(htab)); + + for (;;) { + if (ip < in_end - 2) { + hval = next(hval, ip); + hslot = htab + idx(hval); + ref = *hslot; + *hslot = ip; + + if ((off = ip - ref - 1) < max_off + && ip + 4 < in_end && ref > in_data + && *(u16 *) ref == *(u16 *) ip && ref[2] == ip[2] + ) { + /* match found at *ref++ */ + unsigned int len = 2; + unsigned int maxlen = in_end - ip - len; + maxlen = maxlen > max_ref ? max_ref : maxlen; + + do + len++; + while (len < maxlen && ref[len] == ip[len]); + + if (op + lit + 1 + 3 >= out_end) { + *out_len = PAGE_SIZE; + return 0; + } + + if (lit) { + *op++ = lit - 1; + lit = -lit; + do + *op++ = ip[lit]; + while (++lit); + } + + len -= 2; + ip++; + + if (len < 7) { + *op++ = (off >> 8) + (len << 5); + } else { + *op++ = (off >> 8) + (7 << 5); + *op++ = len - 7; + } + + *op++ = off; + + ip += len; + hval = first(ip); + hval = next(hval, ip); + htab[idx(hval)] = ip; + ip++; + continue; + } + } else if (ip == in_end) + break; + + /* one more literal byte we must copy */ + lit++; + ip++; + + if (lit == max_lit) { + if (op + 1 + max_lit >= out_end) { + *out_len = PAGE_SIZE; + return 0; + } + + *op++ = max_lit - 1; + memcpy(op, ip - max_lit, max_lit); + op += max_lit; + lit = 0; + } + } + + if (lit) { + if (op + lit + 1 >= out_end) { + *out_len = PAGE_SIZE; + return 0; + } + + *op++ = lit - 1; + lit = -lit; + do + *op++ = ip[lit]; + while (++lit); + } + + *out_len = op - out_data; + return 0; +} + +static int lzf_decompress(struct crypto_tfm *tfm, const u8 *src, + unsigned int slen, u8 *dst, unsigned int *dlen) +{ + u8 const *ip = src; + u8 *op = dst; + u8 const *const in_end = ip + slen; + u8 *const out_end = op + *dlen; + + do { + unsigned int ctrl = *ip++; + + if (ctrl < (1 << 5)) { /* literal run */ + ctrl++; + + if (op + ctrl > out_end) { + *dlen = PAGE_SIZE; + return 0; + } + memcpy(op, ip, ctrl); + op += ctrl; + ip += ctrl; + } else { /* back reference */ + + unsigned int len = ctrl >> 5; + + u8 *ref = op - ((ctrl & 0x1f) << 8) - 1; + + if (len == 7) + len += *ip++; + + ref -= *ip++; + + if (op + len + 2 > out_end) { + *dlen = PAGE_SIZE; + return 0; + } + + if (ref < (u8 *) dst) { + *dlen = PAGE_SIZE; + return 0; + } + + *op++ = *ref++; + *op++ = *ref++; + + do + *op++ = *ref++; + while (--len); + } + } + while (op < out_end && ip < in_end); + + *dlen = op - (u8 *) dst; + return 0; +} + +static struct crypto_alg alg = { + .cra_name = "lzf", + .cra_flags = CRYPTO_ALG_TYPE_COMPRESS, + .cra_ctxsize = 0, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(alg.cra_list), + .cra_init = lzf_compress_init, + .cra_exit = lzf_compress_exit, + .cra_u = { .compress = { + .coa_compress = lzf_compress, + .coa_decompress = lzf_decompress } } +}; + +static int __init init(void) +{ + return crypto_register_alg(&alg); +} + +static void __exit fini(void) +{ + crypto_unregister_alg(&alg); +} + +module_init(init); +module_exit(fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("LZF Compression Algorithm"); +MODULE_AUTHOR("Marc Alexander Lehmann & Nigel Cunningham"); diff -urN oldtree/drivers/base/power/resume.c newtree/drivers/base/power/resume.c --- oldtree/drivers/base/power/resume.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/drivers/base/power/resume.c 2006-10-06 17:28:20.000000000 -0400 @@ -36,6 +36,7 @@ } if (dev->bus && dev->bus->resume) { dev_dbg(dev,"resuming\n"); + device_beep(BEEP_RESUMING); error = dev->bus->resume(dev); } if (dev->class && dev->class->resume) { diff -urN oldtree/drivers/base/power/suspend.c newtree/drivers/base/power/suspend.c --- oldtree/drivers/base/power/suspend.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/drivers/base/power/suspend.c 2006-10-06 17:28:20.000000000 -0400 @@ -74,6 +74,7 @@ ? ", may wakeup" : "" ); + device_beep(BEEP_SUSPENDING); error = dev->class->suspend(dev, state); suspend_report_result(dev->class->suspend, error); } @@ -86,6 +87,7 @@ ? ", may wakeup" : "" ); + device_beep(BEEP_SUSPENDING); error = dev->bus->suspend(dev, state); suspend_report_result(dev->bus->suspend, error); } @@ -111,6 +113,7 @@ ? ", may wakeup" : "" ); + device_beep(BEEP_SUSPENDING); error = dev->bus->suspend_late(dev, state); suspend_report_result(dev->bus->suspend_late, error); } @@ -152,6 +155,7 @@ printk("Suspending device %s\n", kobject_name(&dev->kobj)); + device_beep(BEEP_SUSPENDING); error = suspend_device(dev, state); down(&dpm_list_sem); @@ -197,6 +201,7 @@ struct list_head * entry = dpm_off.prev; dev = to_device(entry); + device_beep(BEEP_SUSPENDING); error = suspend_device_late(dev, state); if (error) goto Error; diff -urN oldtree/drivers/base/sys.c newtree/drivers/base/sys.c --- oldtree/drivers/base/sys.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/drivers/base/sys.c 2006-10-06 17:28:20.000000000 -0400 @@ -21,6 +21,8 @@ #include #include #include +#include +#include #include #include "base.h" @@ -30,6 +32,7 @@ #define to_sysdev(k) container_of(k, struct sys_device, kobj) #define to_sysdev_attr(a) container_of(a, struct sysdev_attribute, attr) +int __nosavedata driver_model_beeping = 0; static ssize_t sysdev_show(struct kobject * kobj, struct attribute * attr, char * buffer) @@ -340,25 +343,83 @@ up(&sysdev_drivers_lock); } +static void device_do_beep(unsigned int count) +{ + /* enable counter 2 */ + outb_p(inb_p(0x61) | 3, 0x61); + /* set command for counter 2, 2 byte write */ + outb_p(0xB6, 0x43); + /* select desired HZ */ + outb_p(count & 0xff, 0x42); + outb((count >> 8) & 0xff, 0x42); + + mdelay(100); + + /* disable counter 2 */ + outb(inb_p(0x61) & 0xFC, 0x61); +} + +/* + * Beep in groups of three, with different frequencies so the individual + * beeps can be distinguished. In addition, frequency rises when suspending + * and drops when resuming. + * + * Yes, I know freq isn't really _the_ frequency. + */ + +#define SUSPEND_FREQ_START 2000 +#define FREQ_CHANGE -500 +#define GROUP_SIZE 3 +#define RESUME_FREQ_START (SUSPEND_FREQ_START + (GROUP_SIZE - 1) * FREQ_CHANGE) + +void device_beep(int suspending) +{ + static int last_action = 0, this_freq = SUSPEND_FREQ_START; + int limit = suspending ? RESUME_FREQ_START : SUSPEND_FREQ_START; + + if (!driver_model_beeping) + return; + + if (suspending != last_action) + this_freq = suspending ? SUSPEND_FREQ_START : RESUME_FREQ_START; + + device_do_beep(this_freq); + + if (this_freq == limit) { + /* Time for a gap & freq reset */ + mdelay(300); + this_freq = suspending ? SUSPEND_FREQ_START : RESUME_FREQ_START; + } else + this_freq += (suspending ? FREQ_CHANGE : -FREQ_CHANGE); + + last_action = suspending; +} + static void __sysdev_resume(struct sys_device *dev) { struct sysdev_class *cls = dev->cls; struct sysdev_driver *drv; /* First, call the class-specific one */ - if (cls->resume) + if (cls->resume) { + device_beep(BEEP_RESUMING); cls->resume(dev); + } /* Call auxillary drivers next. */ list_for_each_entry(drv, &cls->drivers, entry) { - if (drv->resume) + if (drv->resume) { + device_beep(BEEP_RESUMING); drv->resume(dev); + } } /* Call global drivers. */ list_for_each_entry(drv, &sysdev_drivers, entry) { - if (drv->resume) + if (drv->resume) { + device_beep(BEEP_RESUMING); drv->resume(dev); + } } } @@ -396,6 +457,7 @@ /* Call global drivers first. */ list_for_each_entry(drv, &sysdev_drivers, entry) { if (drv->suspend) { + device_beep(BEEP_SUSPENDING); ret = drv->suspend(sysdev, state); if (ret) goto gbl_driver; @@ -405,6 +467,7 @@ /* Call auxillary drivers next. */ list_for_each_entry(drv, &cls->drivers, entry) { if (drv->suspend) { + device_beep(BEEP_SUSPENDING); ret = drv->suspend(sysdev, state); if (ret) goto aux_driver; @@ -413,6 +476,7 @@ /* Now call the generic one */ if (cls->suspend) { + device_beep(BEEP_SUSPENDING); ret = cls->suspend(sysdev, state); if (ret) goto cls_driver; @@ -423,7 +487,8 @@ /* resume current sysdev */ cls_driver: drv = NULL; - printk(KERN_ERR "Class suspend failed for %s\n", + printk(KERN_ERR "Class suspend (%p) failed for %s\n", + cls->suspend, kobject_name(&sysdev->kobj)); aux_driver: @@ -433,8 +498,10 @@ list_for_each_entry(err_drv, &cls->drivers, entry) { if (err_drv == drv) break; - if (err_drv->resume) + if (err_drv->resume) { + device_beep(BEEP_RESUMING); err_drv->resume(sysdev); + } } drv = NULL; @@ -445,8 +512,10 @@ list_for_each_entry(err_drv, &sysdev_drivers, entry) { if (err_drv == drv) break; - if (err_drv->resume) + if (err_drv->resume) { + device_beep(BEEP_RESUMING); err_drv->resume(sysdev); + } } /* resume other sysdevs in current class */ list_for_each_entry(err_dev, &cls->kset.list, kobj.entry) { @@ -505,5 +574,13 @@ return subsystem_register(&system_subsys); } +static int device_beep_setup(char *str) +{ + driver_model_beeping = !!simple_strtol(str, NULL, 0); + return 1; +} + +__setup("device_beep=", device_beep_setup); + EXPORT_SYMBOL_GPL(sysdev_register); EXPORT_SYMBOL_GPL(sysdev_unregister); diff -urN oldtree/include/linux/bio.h newtree/include/linux/bio.h --- oldtree/include/linux/bio.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/include/linux/bio.h 2006-10-06 17:28:20.000000000 -0400 @@ -124,6 +124,7 @@ #define BIO_BOUNCED 5 /* bio is a bounce bio */ #define BIO_USER_MAPPED 6 /* contains user pages */ #define BIO_EOPNOTSUPP 7 /* not supported */ +#define BIO_SUSPEND_DATA 8 #define bio_flagged(bio, flag) ((bio)->bi_flags & (1 << (flag))) /* diff -urN oldtree/include/linux/device.h newtree/include/linux/device.h --- oldtree/include/linux/device.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/include/linux/device.h 2006-10-06 17:28:20.000000000 -0400 @@ -470,4 +470,9 @@ MODULE_ALIAS("char-major-" __stringify(major) "-" __stringify(minor)) #define MODULE_ALIAS_CHARDEV_MAJOR(major) \ MODULE_ALIAS("char-major-" __stringify(major) "-*") + +/* Beep debugging support */ +#define BEEP_SUSPENDING 0 +#define BEEP_RESUMING 1 +extern void device_beep(int suspending); #endif /* _DEVICE_H_ */ diff -urN oldtree/include/linux/dyn_pageflags.h newtree/include/linux/dyn_pageflags.h --- oldtree/include/linux/dyn_pageflags.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/include/linux/dyn_pageflags.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,67 @@ +/* + * include/linux/dyn_pageflags.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It implements support for dynamically allocated bitmaps that are + * used for temporary or infrequently used pageflags, in lieu of + * bits in the struct page flags entry. + */ + +#ifndef DYN_PAGEFLAGS_H +#define DYN_PAGEFLAGS_H + +#include + +typedef unsigned long *** dyn_pageflags_t; + +#if BITS_PER_LONG == 32 +#define UL_SHIFT 5 +#else +#if BITS_PER_LONG == 64 +#define UL_SHIFT 6 +#else +#error Bits per long not 32 or 64? +#endif +#endif + +#define BIT_NUM_MASK (sizeof(unsigned long) * 8 - 1) +#define PAGE_NUM_MASK (~((1 << (PAGE_SHIFT + 3)) - 1)) +#define UL_NUM_MASK (~(BIT_NUM_MASK | PAGE_NUM_MASK)) + +/* + * PAGENUMBER gives the index of the page within the zone. + * PAGEINDEX gives the index of the unsigned long within that page. + * PAGEBIT gives the index of the bit within the unsigned long. + */ +#define BITS_PER_PAGE (PAGE_SIZE << 3) +#define PAGENUMBER(zone_offset) ((int) (zone_offset >> (PAGE_SHIFT + 3))) +#define PAGEINDEX(zone_offset) ((int) ((zone_offset & UL_NUM_MASK) >> UL_SHIFT)) +#define PAGEBIT(zone_offset) ((int) (zone_offset & BIT_NUM_MASK)) + +#define PAGE_UL_PTR(bitmap, zone_num, zone_pfn) \ + ((bitmap[zone_num][PAGENUMBER(zone_pfn)])+PAGEINDEX(zone_pfn)) + +#define BITMAP_FOR_EACH_SET(bitmap, counter) \ + for (counter = get_next_bit_on(bitmap, max_pfn); counter < max_pfn; \ + counter = get_next_bit_on(bitmap, counter)) + +extern void clear_dyn_pageflags(dyn_pageflags_t pagemap); +extern int allocate_dyn_pageflags(dyn_pageflags_t *pagemap); +extern void free_dyn_pageflags(dyn_pageflags_t *pagemap); +extern unsigned long get_next_bit_on(dyn_pageflags_t bitmap, unsigned long counter); + +extern int test_dynpageflag(dyn_pageflags_t *bitmap, struct page *page); +extern void set_dynpageflag(dyn_pageflags_t *bitmap, struct page *page); +extern void clear_dynpageflag(dyn_pageflags_t *bitmap, struct page *page); +#endif + +/* + * With the above macros defined, you can do... + * #define PagePageset1(page) (test_dynpageflag(&pageset1_map, page)) + * #define SetPagePageset1(page) (set_dynpageflag(&pageset1_map, page)) + * #define ClearPagePageset1(page) (clear_dynpageflag(&pageset1_map, page)) + */ + diff -urN oldtree/include/linux/freezer.h newtree/include/linux/freezer.h --- oldtree/include/linux/freezer.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/include/linux/freezer.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,27 @@ +/* Freezer declarations */ + +#define FREEZER_ON 0 +#define ABORT_FREEZING 1 +#define FREEZING_COMPLETE 2 + +#define FREEZER_KERNEL_THREADS 0 +#define FREEZER_ALL_THREADS 1 + +#ifdef CONFIG_PM +extern unsigned long freezer_state; + +#define test_freezer_state(bit) test_bit(bit, &freezer_state) +#define set_freezer_state(bit) set_bit(bit, &freezer_state) +#define clear_freezer_state(bit) clear_bit(bit, &freezer_state) + +#define freezer_is_on() (test_freezer_state(FREEZER_ON)) + +#else + +#define test_freezer_state(bit) (0) +#define set_freezer_state(bit) do { } while(0) +#define clear_freezer_state(bit) do { } while(0) + +#define freezer_is_on() (0) + +#endif diff -urN oldtree/include/linux/kernel.h newtree/include/linux/kernel.h --- oldtree/include/linux/kernel.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/include/linux/kernel.h 2006-10-06 17:28:20.000000000 -0400 @@ -114,6 +114,8 @@ __attribute__ ((format (printf, 2, 0))); extern int snprintf(char * buf, size_t size, const char * fmt, ...) __attribute__ ((format (printf, 3, 4))); +extern int snprintf_used(char *buffer, int buffer_size, + const char *fmt, ...); extern int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) __attribute__ ((format (printf, 3, 0))); extern int scnprintf(char * buf, size_t size, const char * fmt, ...) diff -urN oldtree/include/linux/netlink.h newtree/include/linux/netlink.h --- oldtree/include/linux/netlink.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/include/linux/netlink.h 2006-10-06 17:28:20.000000000 -0400 @@ -23,6 +23,8 @@ #define NETLINK_GENERIC 16 /* leave room for NETLINK_DM (DM Events) */ #define NETLINK_SCSITRANSPORT 18 /* SCSI Transports */ +#define NETLINK_SUSPEND2_USERUI 19 /* For suspend2's userui */ +#define NETLINK_SUSPEND2_USM 20 /* For suspend2's userui */ #define MAX_LINKS 32 diff -urN oldtree/include/linux/sched.h newtree/include/linux/sched.h --- oldtree/include/linux/sched.h 2006-10-06 16:28:30.000000000 -0400 +++ newtree/include/linux/sched.h 2006-10-06 17:28:20.000000000 -0400 @@ -1747,7 +1747,7 @@ extern void refrigerator(void); extern int freeze_processes(void); -extern void thaw_processes(void); +extern void thaw_processes(int which_threads); static inline int try_to_freeze(void) { @@ -1766,7 +1766,7 @@ static inline void refrigerator(void) {} static inline int freeze_processes(void) { BUG(); return 0; } -static inline void thaw_processes(void) {} +static inline void thaw_processes(int which_threads) {} static inline int try_to_freeze(void) { return 0; } diff -urN oldtree/include/linux/suspend.h newtree/include/linux/suspend.h --- oldtree/include/linux/suspend.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/include/linux/suspend.h 2006-10-06 17:28:20.000000000 -0400 @@ -8,6 +8,7 @@ #include #include #include +#include /* page backup entry */ struct pbe { @@ -24,26 +25,24 @@ /* kernel/power/swsusp.c */ extern int software_suspend(void); -#if defined(CONFIG_VT) && defined(CONFIG_VT_CONSOLE) extern int pm_prepare_console(void); extern void pm_restore_console(void); -#else -static inline int pm_prepare_console(void) { return 0; } -static inline void pm_restore_console(void) {} -#endif /* defined(CONFIG_VT) && defined(CONFIG_VT_CONSOLE) */ +extern int freeze_processes(void); +extern void thaw_processes(int which_threads); + #else static inline int software_suspend(void) { printk("Warning: fake suspend called\n"); return -ENOSYS; } +static inline int freeze_processes(void) { return 0; } +static inline void thaw_processes(int which_threads) { } #endif /* CONFIG_PM */ void save_processor_state(void); void restore_processor_state(void); struct saved_context; -void __save_processor_state(struct saved_context *ctxt); -void __restore_processor_state(struct saved_context *ctxt); unsigned long get_safe_page(gfp_t gfp_mask); /* diff -urN oldtree/include/linux/suspend2.h newtree/include/linux/suspend2.h --- oldtree/include/linux/suspend2.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/include/linux/suspend2.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,183 @@ +#ifndef _LINUX_SUSPEND2_H +#define _LINUX_SUSPEND2_H + +#include + +/* arch/i386/mm/init.c */ +extern char __nosave_begin, __nosave_end; + +#define SECTOR_SIZE 512 + +/* kernel/power/main.c */ +extern unsigned long suspend_result; + +/* kernel/power/process.c */ +extern unsigned long suspend_debug_state; + +/* arch/i386/power/suspend2.c */ +extern unsigned long suspend_action; +extern int suspend_io_time[2][2]; + +extern dyn_pageflags_t pageset1_map; +extern dyn_pageflags_t pageset1_copy_map; + +#ifdef CONFIG_PM_DEBUG +#define test_debug_state(bit) (test_bit(bit, &suspend_debug_state)) +#else +#define test_debug_state(bit) (0) +#endif + +#define test_result_state(bit) (test_bit(bit, &suspend_result)) + +/* Debug sections - if debugging compiled in */ +enum { + SUSPEND_ANY_SECTION, + SUSPEND_EAT_MEMORY, + SUSPEND_IO, + SUSPEND_HEADER, + SUSPEND_WRITER, + SUSPEND_MEMORY, +}; + +/* debugging levels. */ +#define SUSPEND_STATUS 0 +#define SUSPEND_ERROR 2 +#define SUSPEND_LOW 3 +#define SUSPEND_MEDIUM 4 +#define SUSPEND_HIGH 5 +#define SUSPEND_VERBOSE 6 + +/* Configuration flags */ +enum { + SUSPEND_REBOOT, + SUSPEND_PAUSE, + SUSPEND_SLOW, + SUSPEND_LOGALL, + SUSPEND_CAN_CANCEL, + SUSPEND_KEEP_IMAGE, + SUSPEND_FREEZER_TEST, + SUSPEND_SINGLESTEP, + SUSPEND_PAUSE_NEAR_PAGESET_END, + SUSPEND_TEST_FILTER_SPEED, + SUSPEND_TEST_BIO, + SUSPEND_NO_PAGESET2, + SUSPEND_PM_PREPARE_CONSOLE, + SUSPEND_IGNORE_ROOTFS, + SUSPEND_REPLACE_SWSUSP, + SUSPEND_RETRY_RESUME, +}; + +#ifdef CONFIG_SUSPEND2 +#define test_action_state(bit) (test_bit(bit, &suspend_action)) +#define set_action_state(bit) (test_and_set_bit(bit, &suspend_action)) +#define clear_action_state(bit) (test_and_clear_bit(bit, &suspend_action)) +#else +#define test_action_state(bit) (0) +#endif + +extern void __suspend_message(unsigned long section, unsigned long level, int log_normally, + const char *fmt, ...); + +#ifdef CONFIG_PM_DEBUG +#define suspend_message(sn, lev, log, fmt, a...) \ +do { \ + if (!sn || test_debug_state(sn)) \ + __suspend_message(sn, lev, log, fmt, ##a); \ +} while(0) +#else /* CONFIG_PM_DEBUG */ +#define suspend_message(sn, lev, log, fmt, a...) \ +do { \ + if (lev == SUSPEND_STATUS) \ + __suspend_message(sn, lev, log, fmt, ##a); \ +} while(0) +#endif /* CONFIG_PM_DEBUG */ + +/* Suspend 2 */ + +enum { + SUSPEND_CAN_SUSPEND, + SUSPEND_CAN_RESUME, + SUSPEND_RUNNING, + SUSPEND_RESUME_DEVICE_OK, + SUSPEND_NORESUME_SPECIFIED, + SUSPEND_SANITY_CHECK_PROMPT, + SUSPEND_PAGESET2_NOT_LOADED, + SUSPEND_CONTINUE_REQ, + SUSPEND_RESUMED_BEFORE, + SUSPEND_RESUME_NOT_DONE, + SUSPEND_BOOT_TIME, + SUSPEND_NOW_RESUMING, + SUSPEND_IGNORE_LOGLEVEL, + SUSPEND_TRYING_TO_RESUME, + SUSPEND_TRY_RESUME_RD, +}; + +/* --------------------------------------------------------------------- */ +#ifdef CONFIG_SUSPEND2 + +/* Used in init dir files */ +extern unsigned long suspend_state; + +extern void suspend2_try_resume(void); +extern int suspend_early_boot_message + (int can_erase_image, int default_answer, char *warning_reason, ...); +extern unsigned long suspend_update_status (unsigned long value, unsigned long maximum, + const char *fmt, ...); +extern void suspend_prepare_status (int clearbar, const char *fmt, ...); + +#define test_suspend_state(bit) \ + (test_bit(bit, &suspend_state)) + +#define clear_suspend_state(bit) \ + (clear_bit(bit, &suspend_state)) + +#define set_suspend_state(bit) \ + (set_bit(bit, &suspend_state)) + +extern void suspend2_try_suspend(int have_pmsem); + +/* --------------------------------------------------------------------- */ +#else +/* --------------------------------------------------------------------- */ + +#define suspend_state (0) +#define clear_suspend_state(bit) do { } while (0) +#define test_suspend_state(bit) (0) +#define set_suspend_state(bit) do { } while(0) + +#define suspend2_try_resume() do { } while(0) +static inline int suspend_early_boot_message(int a, int b, char *c, ...) { return 0; } +static inline unsigned long suspend_update_status(unsigned long value, unsigned long maximum, + const char *fmt, ...) +{ + return maximum; +} +#define suspend_prepare_status(a, ...) do { } while(0) + +#endif /* CONFIG_SUSPEND2 */ + +#define test_and_set_suspend_state(bit) \ + (test_and_set_bit(bit, &suspend_state)) + +#define get_suspend_state() (suspend_state) + +#define restore_suspend_state(saved_state) \ + do { suspend_state = saved_state; } while(0) + +#if defined(CONFIG_SUSPEND2) && defined(CONFIG_ACPI) +#include +static inline int may_try_suspend2(u32 state) +{ + if (state == ACPI_STATE_S4) { + suspend2_try_suspend(0); + return 1; + } + return 0; +} +#else +static inline int may_try_suspend2(u32 state) +{ + return 0; +} +#endif +#endif /* _LINUX_SUSPEND2_H */ diff -urN oldtree/include/linux/swap.h newtree/include/linux/swap.h --- oldtree/include/linux/swap.h 2006-10-06 14:45:31.000000000 -0400 +++ newtree/include/linux/swap.h 2006-10-06 17:28:20.000000000 -0400 @@ -370,5 +370,10 @@ #define disable_swap_token() do { } while(0) #endif /* CONFIG_SWAP */ + +/* For Suspend2 - unlink LRU pages while saving separately */ +void unlink_lru_lists(void); +void relink_lru_lists(void); + #endif /* __KERNEL__*/ #endif /* _LINUX_SWAP_H */ diff -urN oldtree/init/do_mounts.c newtree/init/do_mounts.c --- oldtree/init/do_mounts.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/init/do_mounts.c 2006-10-06 17:28:20.000000000 -0400 @@ -26,6 +26,19 @@ dev_t ROOT_DEV; +#ifdef CONFIG_SOFTWARE_SUSPEND +extern int software_resume(void); +#else +#ifdef CONFIG_SUSPEND2 +extern void suspend2_try_resume(void); +static int software_resume(void) +{ + suspend2_try_resume(); + return 0; +} +#endif +#endif + static int __init load_ramdisk(char *str) { rd_doload = simple_strtol(str,NULL,0) & 3; @@ -139,11 +152,16 @@ char s[32]; char *p; dev_t res = 0; - int part; + int part, mount_result; #ifdef CONFIG_SYSFS int mkdir_err = sys_mkdir("/sys", 0700); - if (sys_mount("sysfs", "/sys", "sysfs", 0, NULL) < 0) + /* + * When changing resume2 parameter for Software Suspend, sysfs may + * already be mounted. + */ + mount_result = sys_mount("sysfs", "/sys", "sysfs", 0, NULL); + if (mount_result < 0 && mount_result != -EBUSY) goto out; #endif @@ -195,7 +213,8 @@ res = try_name(s, part); done: #ifdef CONFIG_SYSFS - sys_umount("/sys", 0); + if (mount_result >= 0) + sys_umount("/sys", 0); out: if (!mkdir_err) sys_rmdir("/sys"); @@ -434,9 +453,27 @@ is_floppy = MAJOR(ROOT_DEV) == FLOPPY_MAJOR; + /* Suspend2: + * By this point, suspend_early_init has been called to initialise our + * sysfs interface. If modules are built in, they have registered (all + * of the above via initcalls). + * + * We have not yet looked to see if an image exists, however. If we + * have an initrd, it is expected that the user will have set it up + * to echo > /sys/power/suspend2/do_resume and thus initiate any + * resume. If they don't do that, we do it immediately after the initrd + * is finished (major issues if they mount filesystems rw from the + * initrd! - they are warned. If there's no usable initrd, we do our + * check next. + */ if (initrd_load()) goto out; +#ifdef CONFIG_SUSPEND2 + if (test_suspend_state(SUSPEND_RESUME_NOT_DONE)) + software_resume(); +#endif + if (is_floppy && rd_doload && rd_load_disk(0)) ROOT_DEV = Root_RAM0; diff -urN oldtree/init/do_mounts_initrd.c newtree/init/do_mounts_initrd.c --- oldtree/init/do_mounts_initrd.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/init/do_mounts_initrd.c 2006-10-06 17:28:20.000000000 -0400 @@ -6,6 +6,7 @@ #include #include #include +#include #include "do_mounts.h" @@ -57,10 +58,17 @@ current->flags |= PF_NOFREEZE; pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD); if (pid > 0) { - while (pid != sys_wait4(-1, NULL, 0, NULL)) + while (pid != sys_wait4(-1, NULL, 0, NULL)) { yield(); + try_to_freeze(); + } } + if (test_suspend_state(SUSPEND_RESUME_NOT_DONE)) + printk(KERN_ERR "Suspend2: Initrd lacks echo > /sys/power/suspend2/do_resume.\n"); + clear_suspend_state(SUSPEND_BOOT_TIME); + current->flags &= ~PF_NOFREEZE; + /* move initrd to rootfs' /old */ sys_fchdir(old_fd); sys_mount("/", ".", NULL, MS_MOVE, NULL); diff -urN oldtree/init/main.c newtree/init/main.c --- oldtree/init/main.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/init/main.c 2006-10-06 17:28:20.000000000 -0400 @@ -742,7 +742,9 @@ /* * check if there is an early userspace init. If yes, let it do all - * the work + * the work. For suspend2, we assume that it will do the right thing + * with regard to trying to resume at the right place. When that + * happens, the BOOT_TIME flag will be cleared. */ if (!ramdisk_execute_command) diff -urN oldtree/kernel/kmod.c newtree/kernel/kmod.c --- oldtree/kernel/kmod.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/kmod.c 2006-10-06 17:28:20.000000000 -0400 @@ -34,6 +34,7 @@ #include #include #include +#include #include extern int max_threads; @@ -280,6 +281,11 @@ if (path[0] == '\0') return 0; + if (freezer_is_on()) { + printk(KERN_WARNING "Freezer is on. Refusing to start %s.\n", path); + return -EBUSY; + } + queue_work(khelper_wq, &work); wait_for_completion(&done); return sub_info.retval; diff -urN oldtree/kernel/power/Kconfig newtree/kernel/power/Kconfig --- oldtree/kernel/power/Kconfig 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/Kconfig 2006-10-06 17:28:20.000000000 -0400 @@ -130,3 +130,85 @@ bool depends on HOTPLUG_CPU && X86 && PM default y + +config SUSPEND2_CRYPTO + bool + depends on SUSPEND2 && CRYPTO + default y + +menuconfig SUSPEND2 + bool "Suspend2" + depends on PM + select DYN_PAGEFLAGS + select HOTPLUG_CPU if SMP + ---help--- + Suspend2 is the 'new and improved' suspend support. + + See the Suspend2 home page (suspend2.net) + for FAQs, HOWTOs and other documentation. + + comment "Image Storage (you need at least one writer)" + depends on SUSPEND2 + + config SUSPEND2_FILEWRITER + bool "File Writer" + depends on SUSPEND2 + ---help--- + This option enables support for storing an image in a + simple file. This should be possible, but we're still + testing it. + + config SUSPEND2_SWAPWRITER + bool "Swap Writer" + depends on SUSPEND2 + select SWAP + ---help--- + This option enables support for storing an image in your + swap space. + + comment "General Options" + depends on SUSPEND2 + + config SUSPEND2_DEFAULT_RESUME2 + string "Default resume device name" + depends on SUSPEND2 + ---help--- + You normally need to add a resume2= parameter to your lilo.conf or + equivalent. With this option properly set, the kernel has a value + to default. No damage will be done if the value is invalid. + + config SUSPEND2_KEEP_IMAGE + bool "Allow Keep Image Mode" + depends on SUSPEND2 + ---help--- + This option allows you to keep and image and reuse it. It is intended + __ONLY__ for use with systems where all filesystems are mounted read- + only (kiosks, for example). To use it, compile this option in and boot + normally. Set the KEEP_IMAGE flag in /proc/suspend2 and suspend. + When you resume, the image will not be removed. You will be unable to turn + off swap partitions (assuming you are using the swap writer), but future + suspends simply do a power-down. The image can be updated using the + kernel command line parameter suspend_act= to turn off the keep image + bit. Keep image mode is a little less user friendly on purpose - it + should not be used without thought! + + config SUSPEND2_REPLACE_SWSUSP + bool "Replace swsusp by default" + default y + depends on SUSPEND2 + ---help--- + Suspend2 can replace swsusp. This option makes that the default state, + requiring you to echo 0 > /sys/power/suspend2/replace_swsusp if you want + to use the vanilla kernel functionality. Note that your initrd/ramfs will + need to do this before trying to resume, too. + With overriding swsusp enabled, Suspend2 will use both the resume= and + noresume commandline options _and_ the resume2= and noresume2 ones (for + compatibility). resume= takes precedence over resume2=. Echoing disk + to /sys/power/state will start a Suspend2 cycle. If resume= doesn't + specify a writer and both the swapwriter and filewriter are compiled in, + the swapwriter will be used by default. + +config SUSPEND_SHARED + bool + depends on SUSPEND2 || SOFTWARE_SUSPEND + default y diff -urN oldtree/kernel/power/Makefile newtree/kernel/power/Makefile --- oldtree/kernel/power/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -5,6 +5,26 @@ obj-y := main.o process.o console.o obj-$(CONFIG_PM_LEGACY) += pm.o -obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o snapshot.o swap.o user.o +obj-$(CONFIG_SUSPEND_SHARED) += snapshot.o + +# Order is important for compression and encryption - we +# compress before encrypting. + +suspend_core-objs := io.o pagedir.o prepare_image.o \ + extent.o suspend.o modules.o \ + pageflags.o ui.o sysfs.o \ + power_off.o atomic_copy.o + +#ifdef CONFIG_NET +suspend_core-objs += storage.o netlink.o +#endif + +obj-$(CONFIG_SUSPEND2) += suspend_core.o +obj-$(CONFIG_SUSPEND2_CRYPTO) += compression.o encryption.o + +obj-$(CONFIG_SUSPEND2_SWAPWRITER) += suspend_block_io.o suspend_swap.o +obj-$(CONFIG_SUSPEND2_FILEWRITER) += suspend_block_io.o suspend_file.o + +obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o swap.o user.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff -urN oldtree/kernel/power/atomic_copy.c newtree/kernel/power/atomic_copy.c --- oldtree/kernel/power/atomic_copy.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/atomic_copy.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,543 @@ +/* + * kernel/power/atomic_copy.c + * + * Copyright 2004-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * Routines for doing the atomic save/restore. + */ + +#include +#include +#include +#include +#include +#include +#include "suspend.h" +#include "storage.h" +#include "power_off.h" +#include "ui.h" +#include "power.h" +#include "io.h" +#include "prepare_image.h" +#include "pageflags.h" +#include "extent.h" + +static unsigned long state1 __nosavedata = 0; +static unsigned long state2 __nosavedata = 0; +static int state3 __nosavedata = 0; +static int io_speed_save[2][2] __nosavedata; +__nosavedata char suspend_resume_commandline[COMMAND_LINE_SIZE]; + +#define SUSPEND_FREQ_START 2000 +#define FREQ_CHANGE -100 +#define GROUP_SIZE 10 +#define RESUME_FREQ_START (SUSPEND_FREQ_START + (GROUP_SIZE - 1) * FREQ_CHANGE) + +static int __nosavedata suspend_beeping = 0; +static int __nosavedata this_freq = SUSPEND_FREQ_START; + +extern void suspend_power_down(void); +extern int swsusp_resume(void); +extern int suspend2_in_suspend __nosavedata; +int extra_pd1_pages_used; + +#ifdef CONFIG_HIGHMEM +static dyn_pageflags_t __nosavedata origmap; +static dyn_pageflags_t __nosavedata copymap; +static unsigned long __nosavedata origoffset; +static unsigned long __nosavedata copyoffset; +static __nosavedata int o_zone_num, c_zone_num; + +struct zone_data { + unsigned long start_pfn; + unsigned long end_pfn; + int is_highmem; +}; + +static __nosavedata struct zone_data *zone_nosave; +static __nosavedata unsigned long boot_max_pfn; + +/** + * suspend_init_nosave_zone_table: Set up nosave copy of zone table data. + * + * Zone information might be overwritten during the copy back, so we copy + * the fields we need to a non-conflicting page and use it. + **/ +static void suspend_init_nosave_zone_table(void) +{ + struct zone *zone; + + zone_nosave = (struct zone_data *) suspend_get_nonconflicting_page(); + + BUG_ON(!zone_nosave); + + /* It should all fit in one page. */ + BUG_ON(ZONEID_MASK > (PAGE_SIZE / sizeof(struct zone_data))); + + memset((char *) zone_nosave, 0, PAGE_SIZE); + + for_each_zone(zone) { + if (populated_zone(zone)) { + int zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + zone_nosave[zone_num].start_pfn = zone->zone_start_pfn; + zone_nosave[zone_num].end_pfn = zone->zone_start_pfn + + zone->spanned_pages - 1; + zone_nosave[zone_num].is_highmem = is_highmem(zone); + } + } + + boot_max_pfn = max_pfn; +} + +/** + * __suspend_get_next_bit_on: Atomic-copy safe location of next bit on in a bitmap. + * + * @bitmap: The bitmap we are traversing. + * @zone_num: Which zone we are in. + * @counter: The current pfn. + * + * A version of __get_next_bit_on that can be used when doing the atomic + * copy. While doing it, we can't rely on the zone information being in + * a constant location. + **/ +static unsigned long __suspend_get_next_bit_on(dyn_pageflags_t bitmap, + int *zone_num, unsigned long counter) +{ + unsigned long *ul_ptr = NULL; + int reset_ul_ptr = 1, pagebit; + BUG_ON(*zone_num == (1 << ZONEID_SHIFT)); + + if (counter == boot_max_pfn) { + *zone_num = 0; + while (!zone_nosave[*zone_num].end_pfn) + (*zone_num)++; + counter = zone_nosave[*zone_num].start_pfn - 1; + } + + do { + counter++; + if (counter > zone_nosave[*zone_num].end_pfn) { + (*zone_num)++; + while (*zone_num < (1 << ZONEID_SHIFT) && + !zone_nosave[*zone_num].end_pfn) + (*zone_num)++; + + if (*zone_num == (1 << ZONEID_SHIFT)) + return boot_max_pfn; + counter = zone_nosave[*zone_num].start_pfn; + reset_ul_ptr = 1; + } else + if (!(counter & BIT_NUM_MASK)) + reset_ul_ptr = 1; + + pagebit = PAGEBIT(counter); + + if (reset_ul_ptr) { + reset_ul_ptr = 0; + ul_ptr = PAGE_UL_PTR(bitmap, *zone_num, + (counter - zone_nosave[*zone_num].start_pfn)); + } + + if (!(*ul_ptr & ~((1 << pagebit) - 1))) { + counter += BITS_PER_LONG - pagebit - 1; + continue; + } + } while(!test_bit(pagebit, ul_ptr)); + + return counter; +} + +/* + * copyback_high: Restore highmem pages. + * + * Iterate through the source and destination bitmaps, restoring + * highmem pages that were atomically copied. + */ +void copyback_high(void) +{ + unsigned long *origpage; + unsigned long *copypage; + + origoffset = __suspend_get_next_bit_on(origmap, &o_zone_num, boot_max_pfn); + copyoffset = __suspend_get_next_bit_on(copymap, &c_zone_num, boot_max_pfn); + + while (o_zone_num < (1 << ZONEID_SHIFT)) { + if (zone_nosave[o_zone_num].is_highmem) { + origpage = (unsigned long *) kmap_atomic(pfn_to_page(origoffset), KM_USER1); + copypage = (unsigned long *) __va(copyoffset << PAGE_SHIFT); + + memcpy(origpage, copypage, PAGE_SIZE); + + kunmap_atomic(origpage, KM_USER1); + } + + origoffset = __suspend_get_next_bit_on(origmap, &o_zone_num, origoffset); + copyoffset = __suspend_get_next_bit_on(copymap, &c_zone_num, copyoffset); + } +} +#else +void copyback_high(void) { } +#endif + +static void suspend_do_beep(unsigned int count) +{ + /* enable counter 2 */ + outb_p(inb_p(0x61) | 3, 0x61); + /* set command for counter 2, 2 byte write */ + outb_p(0xB6, 0x43); + /* select desired HZ */ + outb_p(count & 0xff, 0x42); + outb((count >> 8) & 0xff, 0x42); + + mdelay(100); + + /* disable counter 2 */ + outb(inb_p(0x61) & 0xFC, 0x61); +} + +/* + * Beep in groups of three, with different frequencies so the individual + * beeps can be distinguished. In addition, frequency rises when suspending + * and drops when resuming. + * + * Yes, I know freq isn't really _the_ frequency. + */ + +void suspend2_beep(int suspending) +{ + static int last_action = 0; + int limit = suspending ? RESUME_FREQ_START : SUSPEND_FREQ_START; + + if (!suspend_beeping) + return; + + if (suspending != last_action) + this_freq = suspending ? SUSPEND_FREQ_START : RESUME_FREQ_START; + + suspend_do_beep(this_freq); + + if (this_freq == limit) { + /* Time for a gap & freq reset */ + mdelay(300); + this_freq = suspending ? SUSPEND_FREQ_START : RESUME_FREQ_START; + } else + this_freq += (suspending ? FREQ_CHANGE : -FREQ_CHANGE); + + last_action = suspending; +} + +/* + * prepare_suspend2_pbe_list + * + * Prepare pageset2 pages for doing the atomic copy. If necessary, + * we allocate extra pages. + * + */ + +void prepare_suspend2_pbe_list(void) +{ + unsigned long orig_pfn, copy_pfn; + int i = 1; + struct pbe *this_pbe = NULL, *last_pbe = NULL; + + orig_pfn = copy_pfn = max_pfn; + + restore_pblist = NULL; + + do { + if (!this_pbe || + ((((unsigned long) this_pbe) & (PAGE_SIZE - 1)) + + 2 * sizeof(struct pbe)) > PAGE_SIZE) { + /* Get the next page for pbes */ + this_pbe = (struct pbe *) suspend_get_nonconflicting_page(); + BUG_ON(!this_pbe); + BUG_ON(PagePageset1(virt_to_page(this_pbe))); + } else + this_pbe++; + + do { + orig_pfn = get_next_bit_on(pageset1_map, orig_pfn); + if (orig_pfn == max_pfn) + return; + copy_pfn = get_next_bit_on(pageset1_copy_map, copy_pfn); + } while (PageHighMem(pfn_to_page(orig_pfn))); + + if (!last_pbe) + restore_pblist = this_pbe; + else + last_pbe->next = this_pbe; + + last_pbe = this_pbe; + this_pbe->orig_address = (unsigned long) page_address(pfn_to_page(orig_pfn)); + this_pbe->address = (unsigned long) page_address(pfn_to_page(copy_pfn)); + this_pbe->next = NULL; /* get_nonconflicting_page doesn't get zeroed pages */ + + i++; + + } while (1); +} + +/* + * copyback_post: Post atomic-restore actions. + * + * After doing the atomic restore, we have a few more things to do: + * 1) We want to retain some values across the restore, so we now copy + * these from the nosave variables to the normal ones. + * 2) Set the status flags. + * 3) Resume devices. + * 4) Get userui to redraw. + * 5) Reread the page cache. + */ + +void copyback_post(void) +{ + int loop; + + suspend2_beep(0); + + suspend_action = state1; + suspend_debug_state = state2; + console_loglevel = state3; + + for (loop = 0; loop < 4; loop++) + suspend_io_time[loop/2][loop%2] = + io_speed_save[loop/2][loop%2]; + + set_suspend_state(SUSPEND_NOW_RESUMING); + set_suspend_state(SUSPEND_PAGESET2_NOT_LOADED); + + if (pm_ops && pm_ops->finish && suspend_powerdown_method > 3) + pm_ops->finish(suspend_powerdown_method); + + if (suspend_activate_storage(1)) + panic("Failed to reactivate our storage."); + + userui_redraw(); + + suspend_cond_pause(1, "About to reload secondary pagedir."); + + if (read_pageset2(0)) + panic("Unable to successfully reread the page cache."); + + clear_suspend_state(SUSPEND_PAGESET2_NOT_LOADED); + + suspend_prepare_status(DONT_CLEAR_BAR, "Cleaning up..."); +} + +/* + * suspend_post_context_save: Steps after saving the cpu context. + * + * Steps taken after saving the CPU state to make the actual + * atomic copy. + * + * Called from swsusp_save in snapshot.c. + */ + +int suspend_post_context_save(void) +{ + int old_ps1_size = pagedir1.pageset_size; + int old_ps2_size = pagedir2.pageset_size; + + BUG_ON(!irqs_disabled()); + + suspend2_beep(1); + + suspend_recalculate_image_contents(1); + + extra_pd1_pages_used = pagedir1.pageset_size - old_ps1_size; + + if ((pagedir1.pageset_size - old_ps1_size) > extra_pd1_pages_allowance) { + abort_suspend("Pageset1 has grown by %d pages. " + "extra_pages_allowance is currently only %d.\n", + pagedir1.pageset_size - old_ps1_size, + extra_pd1_pages_allowance); + return -1; + } + + BUG_ON(old_ps2_size != pagedir2.pageset_size); + + BUG_ON(!irqs_disabled()); + + if (!test_action_state(SUSPEND_TEST_FILTER_SPEED) && + !test_action_state(SUSPEND_TEST_BIO)) + suspend_copy_pageset1(); + + + suspend2_beep(1); + + return 0; +} + +/* suspend_copy_pageset1: Do the atomic copy of pageset1. + * + * Make the atomic copy of pageset1. We can't use copy_page (as we once did) + * because we can't be sure what side effects it has. On my old Duron, with + * 3DNOW, kernel_fpu_begin increments preempt count, making our preempt + * count at resume time 4 instead of 3. + * + * We don't want to call kmap_atomic unconditionally because it has the side + * effect of incrementing the preempt count, which will leave it one too high + * post resume (the page containing the preempt count will be copied after + * its incremented. This is essentially the same problem. + */ + +void suspend_copy_pageset1(void) +{ + int i; + unsigned long source_index, dest_index; + + source_index = get_next_bit_on(pageset1_map, max_pfn); + dest_index = get_next_bit_on(pageset1_copy_map, max_pfn); + + for (i = 0; i < pagedir1.pageset_size; i++) { + unsigned long *origvirt, *copyvirt; + struct page *origpage; + int loop = (PAGE_SIZE / sizeof(unsigned long)) - 1; + + origpage = pfn_to_page(source_index); + + if (PageHighMem(origpage)) + origvirt = kmap_atomic(origpage, KM_USER0); + else + origvirt = page_address(origpage); + + copyvirt = (unsigned long *) page_address(pfn_to_page(dest_index)); + + while (loop >= 0) { + *(copyvirt + loop) = *(origvirt + loop); + loop--; + } + + if (PageHighMem(origpage)) + kunmap_atomic(origvirt, KM_USER0); + + source_index = get_next_bit_on(pageset1_map, source_index); + dest_index = get_next_bit_on(pageset1_copy_map, dest_index); + } +} + +int suspend2_suspend(void) +{ + int error; + + suspend2_beep(1); + + if (test_action_state(SUSPEND_PM_PREPARE_CONSOLE)) + pm_prepare_console(); + + 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))) { + set_result_state(SUSPEND_DEVICE_REFUSED); + set_result_state(SUSPEND_ABORTED); + printk(KERN_ERR "Some devices failed to power down, aborting suspend\n"); + goto enable_irqs; + } + + suspend2_beep(1); + + save_processor_state(); + if ((error = swsusp_arch_suspend())) + printk(KERN_ERR "Error %d suspending\n", error); + /* Restore control flow appears here */ + restore_processor_state(); + + suspend2_beep(1); + + if (!suspend2_in_suspend) + copyback_high(); + device_power_up(); +enable_irqs: + local_irq_enable(); + if (test_action_state(SUSPEND_PM_PREPARE_CONSOLE)) + pm_restore_console(); + + suspend2_beep(1); + + return error; +} + +/* + * suspend_atomic_restore + * + * Get ready to do the atomic restore. This part gets us into the same + * state we are in prior to do calling do_suspend2_lowlevel while + * suspending: hotunplugging secondary cpus and freeze processes, + * before starting the thread that will do the restore. + */ +int suspend_atomic_restore(void) +{ + int error, loop; + + suspend2_beep(0); + + suspend_prepare_status(DONT_CLEAR_BAR, "Atomic restore preparation"); + prepare_suspend2_pbe_list(); + + suspend2_beep(0); + + if (test_action_state(SUSPEND_PM_PREPARE_CONSOLE)) + pm_prepare_console(); + + disable_nonboot_cpus(); + + if ((error = device_suspend(PMSG_FREEZE))) { + printk("Some devices failed to suspend\n"); + if (test_action_state(SUSPEND_PM_PREPARE_CONSOLE)) + pm_restore_console(); + BUG(); + } + + suspend2_beep(0); + +#ifdef CONFIG_HIGHMEM + origmap = pageset1_map; + copymap = pageset1_copy_map; + suspend_init_nosave_zone_table(); +#endif + + state1 = suspend_action; + state2 = suspend_debug_state; + state3 = console_loglevel; + + for (loop = 0; loop < 4; loop++) + io_speed_save[loop/2][loop%2] = + suspend_io_time[loop/2][loop%2]; + memcpy(suspend_resume_commandline, saved_command_line, COMMAND_LINE_SIZE); + + mb(); + + local_irq_disable(); + if (device_power_down(PMSG_FREEZE)) { + printk(KERN_ERR "Some devices failed to power down. Very bad.\n"); + BUG(); + } + + /* We'll ignore saved state, but this gets preempt count (etc) right */ + save_processor_state(); + suspend2_beep(0); + error = swsusp_arch_resume(); + /* Code below is only ever reached in case of failure. Otherwise + * execution continues at place where swsusp_arch_suspend was called. + */ + BUG(); + return 1; +} + +static int suspend_beep_setup(char *str) +{ + suspend_beeping = !!simple_strtol(str, NULL, 0); + return 1; +} + +__setup("suspend_beep=", suspend_beep_setup); + diff -urN oldtree/kernel/power/block_io.h newtree/kernel/power/block_io.h --- oldtree/kernel/power/block_io.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/block_io.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,55 @@ +/* + * kernel/power/block_io.h + * + * Copyright 2004-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * This file contains declarations for functions exported from + * block_io.c, which contains low level io functions. + */ + +#include +#include "extent.h" + +struct suspend_bdev_info { + struct block_device *bdev; + dev_t dev_t; + int bmap_shift; + int blocks_per_page; +}; + +/* + * Our exported interface so the swapwriter and filewriter don't + * need these functions duplicated. + */ +struct suspend_bio_ops { + int (*bdev_page_io) (int rw, struct block_device *bdev, long pos, + struct page *page); + void (*check_io_stats) (void); + void (*reset_io_stats) (void); + void (*finish_all_io) (void); + int (*prepare_readahead) (int index); + void (*cleanup_readahead) (int index); + struct page ** readahead_pages; + int (*readahead_ready) (int readahead_index); + int (*forward_one_page) (void); + void (*set_extra_page_forward) (void); + void (*set_devinfo) (struct suspend_bdev_info *info); + int (*read_chunk) (struct page *buffer_page, int sync); + int (*write_chunk) (struct page *buffer_page); + int (*rw_header_chunk) (int rw, struct suspend_module_ops *owner, + char *buffer, int buffer_size); + int (*write_header_chunk_finish) (void); + int (*rw_init) (int rw, int stream_number); + int (*rw_cleanup) (int rw); +}; + +extern struct suspend_bio_ops suspend_bio_ops; + +extern char *suspend_writer_buffer; +extern int suspend_writer_buffer_posn; +extern int suspend_read_fd; +extern struct extent_iterate_saved_state suspend_writer_posn_save[3]; +extern struct extent_iterate_state suspend_writer_posn; +extern int suspend_header_bytes_used; diff -urN oldtree/kernel/power/compression.c newtree/kernel/power/compression.c --- oldtree/kernel/power/compression.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/compression.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,557 @@ +/* + * kernel/power/compression.c + * + * Copyright (C) 2003-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * This file contains data compression routines for suspend, + * using cryptoapi. + * + */ + +#include +#include +#include +#include +#include + +#include "suspend.h" +#include "modules.h" +#include "sysfs.h" +#include "io.h" + +#define S2C_WRITE 0 +#define S2C_READ 1 + +static int suspend_expected_compression = 0; + +static struct suspend_module_ops suspend_compression_ops; +static struct suspend_module_ops *next_driver; + +static char suspend_compressor_name[32] = "lzf"; +static struct crypto_tfm *suspend_compressor_transform; + +static u8 *local_buffer = NULL; +static u8 *page_buffer = NULL; +static unsigned int bufofs; + +static int position = 0; + +/* ---- Local buffer management ---- */ + +/* + * suspend_compress_allocate_local_buffer + * + * Allocates a page of memory for buffering output. + * Int: Zero if successful, -ENONEM otherwise. + */ +static int suspend_compress_allocate_local_buffer(void) +{ + if (!local_buffer) { + local_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + if (!local_buffer) { + printk(KERN_ERR + "Failed to allocate the local buffer for " + "suspend2 compression driver.\n"); + return -ENOMEM; + } + } + + if (!page_buffer) { + page_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + if (!page_buffer) { + printk(KERN_ERR + "Failed to allocate the page buffer for " + "suspend2 compression driver.\n"); + return -ENOMEM; + } + } + + return 0; +} + +/* + * suspend_compress_free_local_buffer + * + * Frees memory allocated for buffering output. + */ +static inline void suspend_compress_free_local_buffer(void) +{ + if (local_buffer) + free_page((unsigned long) local_buffer); + + local_buffer = NULL; + + if (page_buffer) + free_page((unsigned long) page_buffer); + + page_buffer = NULL; +} + +/* + * suspend_compress_cleanup + * + * Frees memory allocated for our labours. + */ +static void suspend_compress_cleanup(void) +{ + if (suspend_compressor_transform) { + crypto_free_tfm(suspend_compressor_transform); + suspend_compressor_transform = NULL; + } +} + +/* + * suspend_crypto_prepare + * + * Prepare to do some work by allocating buffers and transforms. + * Returns: Int: Zero. Even if we can't set up compression, we still + * seek to suspend. + */ +static int suspend_compress_crypto_prepare(void) +{ + if (!*suspend_compressor_name) { + printk("Suspend2: Compression enabled but no compressor name set.\n"); + suspend_compression_ops.enabled = 0; + return 0; + } + + if (!(suspend_compressor_transform = crypto_alloc_tfm(suspend_compressor_name, 0))) { + printk("Suspend2: Failed to initialise the %s compression transform.\n", + suspend_compressor_name); + suspend_compression_ops.enabled = 0; + return 0; + } + + return 0; +} + +/* + * suspend_compress_write_cleanup(): Write unflushed data and free workspace. + * + * Returns: Result of writing last page. + */ +static int suspend_compress_rw_cleanup(int rw) +{ + int ret = 0; + + if (rw == WRITE && suspend_compressor_transform) + ret = next_driver->write_chunk(virt_to_page(local_buffer)); + + suspend_compress_cleanup(); + suspend_compress_free_local_buffer(); + + return ret; +} + +/* + * suspend_compress_rw_init() + * @stream_number: Ignored. + * + * Allocate buffers and prepare to compress data. + * Returns: Zero on success, -ENOMEM if unable to vmalloc. + */ +static int suspend_compress_rw_init(int rw, int stream_number) +{ + int result; + + next_driver = suspend_get_next_filter(&suspend_compression_ops); + + if (!next_driver) { + printk("Compression Driver: Argh! Nothing follows me in" + " the pipeline!"); + return -ECHILD; + } + + if ((result = suspend_compress_crypto_prepare() || + !suspend_compression_ops.enabled)) + return result; + + if ((result = suspend_compress_allocate_local_buffer())) + return result; + + if (rw == READ) + bufofs = PAGE_SIZE; + else { + /* Only reset the stats if starting to write an image */ + if (stream_number == 2) + bytes_in = bytes_out = 0; + + bufofs = 0; + } + + position = 0; + + return 0; +} + +/* + * suspend_compress_write() + * @u8*: Output buffer to be written. + * @unsigned int: Length of buffer. + * + * Helper function for write_chunk. Write the compressed data. + * Return: Int. Result to be passed back to caller. + */ +static int suspend_compress_write (u8 *buffer, unsigned int len) +{ + int ret; + + bytes_out += len; + + while (len + bufofs > PAGE_SIZE) { + unsigned int chunk = PAGE_SIZE - bufofs; + memcpy (local_buffer + bufofs, buffer, chunk); + buffer += chunk; + len -= chunk; + bufofs = 0; + if ((ret = next_driver->write_chunk(virt_to_page(local_buffer))) < 0) + return ret; + } + memcpy (local_buffer + bufofs, buffer, len); + bufofs += len; + return 0; +} + +/* + * suspend_compress_write_chunk() + * + * Compress a page of data, buffering output and passing on filled + * pages to the next module in the pipeline. + * + * Buffer_page: Pointer to a buffer of size PAGE_SIZE, containing + * data to be compressed. + * + * Returns: 0 on success. Otherwise the error is that returned by later + * modules, -ECHILD if we have a broken pipeline or -EIO if + * zlib errs. + */ +static int suspend_compress_write_chunk(struct page *buffer_page) +{ + int ret; + unsigned int len; + u16 len_written; + char *buffer_start; + + if (!suspend_compressor_transform) + return next_driver->write_chunk(buffer_page); + + buffer_start = kmap(buffer_page); + + bytes_in += PAGE_SIZE; + + len = PAGE_SIZE; + + ret = crypto_comp_compress(suspend_compressor_transform, + buffer_start, PAGE_SIZE, + page_buffer, &len); + + if (ret) { + printk("Compression failed.\n"); + goto failure; + } + + len_written = (u16) len; + + if ((ret = suspend_compress_write((u8 *)&len_written, 2)) >= 0) { + if ((ret = suspend_compress_write((u8 *) &position, sizeof(position)))) + return -EIO; + if (len < PAGE_SIZE) { /* some compression */ + position += len; + ret = suspend_compress_write(page_buffer, len); + } else { + ret = suspend_compress_write(buffer_start, PAGE_SIZE); + position += PAGE_SIZE; + } + } + position += 2 + sizeof(int); + + +failure: + kunmap(buffer_page); + return ret; +} + +/* + * suspend_compress_read() + * @buffer: u8 *. Address of the buffer. + * @len: unsigned int. Length. + * + * Description: Read data into compression buffer. + * Returns: int: Result of reading the image chunk. + */ +static int suspend_compress_read (u8 *buffer, unsigned int len) +{ + int ret; + + while (len + bufofs > PAGE_SIZE) { + unsigned int chunk = PAGE_SIZE - bufofs; + memcpy(buffer, local_buffer + bufofs, chunk); + buffer += chunk; + len -= chunk; + bufofs = 0; + if ((ret = next_driver->read_chunk( + virt_to_page(local_buffer), SUSPEND_SYNC)) < 0) { + return ret; + } + } + memcpy (buffer, local_buffer + bufofs, len); + bufofs += len; + return 0; +} + +/* + * suspend_compress_read_chunk() + * @buffer_page: struct page *. Pointer to a buffer of size PAGE_SIZE. + * @sync: int. Whether the previous module (or core) wants its data synchronously. + * + * Retrieve data from later modules and decompress it until the input buffer + * is filled. + * Zero if successful. Error condition from me or from downstream on failure. + */ +static int suspend_compress_read_chunk(struct page *buffer_page, int sync) +{ + int ret, position_saved; + unsigned int len; + u16 len_written; + char *buffer_start; + + if (!suspend_compressor_transform) + return next_driver->read_chunk(buffer_page, SUSPEND_ASYNC); + + /* + * All our reads must be synchronous - we can't decompress + * data that hasn't been read yet. + */ + + buffer_start = kmap(buffer_page); + + if ((ret = suspend_compress_read ((u8 *)&len_written, 2)) >= 0) { + len = (unsigned int) len_written; + ret = suspend_compress_read((u8 *) &position_saved, sizeof(position_saved)); + if (ret) + return ret; + + if (position != position_saved) { + printk("Position saved (%d) != position I'm at now (%d).\n", + position_saved, position); + BUG_ON(1); + } + if (len >= PAGE_SIZE) { /* uncompressed */ + ret = suspend_compress_read(buffer_start, PAGE_SIZE); + if (ret) + return ret; + + position += PAGE_SIZE; + } else { /* compressed */ + if ((ret = suspend_compress_read(page_buffer, len)) >= 0) { + int outlen = PAGE_SIZE; + /* Important note. + * + * For Deflate, decompression return values may represent + * errors. Deflate complains when everything is alright, so + * we ignore the errors unless the number of output bytes is + * not PAGE_SIZE. + */ + crypto_comp_decompress(suspend_compressor_transform, + page_buffer, len, + buffer_start, &outlen); + if (outlen != PAGE_SIZE) { + printk("Decompression yielded %d bytes instead of %ld.\n", outlen, PAGE_SIZE); + ret = -EIO; + } else + ret = 0; + } + position += len; + } + position += 2 + sizeof(int); + } else + printk("Compress_read returned %d.", ret); + kunmap(buffer_page); + return ret; +} + +/* + * suspend_compress_print_debug_stats + * @buffer: Pointer to a buffer into which the debug info will be printed. + * @size: Size of the buffer. + * + * Print information to be recorded for debugging purposes into a buffer. + * Returns: Number of characters written to the buffer. + */ + +static int suspend_compress_print_debug_stats(char *buffer, int size) +{ + int pages_in = bytes_in >> PAGE_SHIFT, + pages_out = bytes_out >> PAGE_SHIFT; + int len; + + /* Output the compression ratio achieved. */ + if (*suspend_compressor_name) + len = snprintf_used(buffer, size, "- Compressor is '%s'.\n", + suspend_compressor_name); + else + len = snprintf_used(buffer, size, "- Compressor is not set.\n"); + + if (pages_in) + len+= snprintf_used(buffer+len, size - len, + " Compressed %ld bytes into %ld (%d percent compression).\n", + bytes_in, bytes_out, (pages_in - pages_out) * 100 / pages_in); + return len; +} + +/* + * suspend_compress_compression_memory_needed + * + * Tell the caller how much memory we need to operate during suspend/resume. + * Returns: Unsigned long. Maximum number of bytes of memory required for + * operation. + */ +static unsigned long suspend_compress_memory_needed(void) +{ + return 2 * PAGE_SIZE; +} + +static unsigned long suspend_compress_storage_needed(void) +{ + return 4 * sizeof(unsigned long) + strlen(suspend_compressor_name) + 1; +} + +/* + * suspend_compress_save_config_info + * @buffer: Pointer to a buffer of size PAGE_SIZE. + * + * Save informaton needed when reloading the image at resume time. + * Returns: Number of bytes used for saving our data. + */ +static int suspend_compress_save_config_info(char *buffer) +{ + int namelen = strlen(suspend_compressor_name) + 1; + int total_len; + + *((unsigned long *) buffer) = bytes_in; + *((unsigned long *) (buffer + 1 * sizeof(unsigned long))) = bytes_out; + *((unsigned long *) (buffer + 2 * sizeof(unsigned long))) = + suspend_expected_compression; + *((unsigned long *) (buffer + 3 * sizeof(unsigned long))) = namelen; + strncpy(buffer + 4 * sizeof(unsigned long), suspend_compressor_name, + namelen); + total_len = 4 * sizeof(unsigned long) + namelen; + return total_len; +} + +/* suspend_compress_load_config_info + * @buffer: Pointer to the start of the data. + * @size: Number of bytes that were saved. + * + * Description: Reload information needed for decompressing the image at + * resume time. + */ +static void suspend_compress_load_config_info(char *buffer, int size) +{ + int namelen; + + bytes_in = *((unsigned long *) buffer); + bytes_out = *((unsigned long *) (buffer + 1 * sizeof(unsigned long))); + suspend_expected_compression = *((unsigned long *) (buffer + 2 * + sizeof(unsigned long))); + namelen = *((unsigned long *) (buffer + 3 * sizeof(unsigned long))); + strncpy(suspend_compressor_name, buffer + 4 * sizeof(unsigned long), + namelen); + return; +} + +/* + * suspend_expected_compression_ratio + * + * Description: Returns the expected ratio between data passed into this module + * and the amount of data output when writing. + * Returns: 100 if the module is disabled. Otherwise the value set by the + * user via our sysfs entry. + */ + +int suspend_expected_compression_ratio(void) +{ + if (!suspend_compression_ops.enabled) + return 100; + else + return 100 - suspend_expected_compression; +} + +/* + * data for our sysfs entries. + */ +static struct suspend_sysfs_data sysfs_params[] = { + { + SUSPEND2_ATTR("expected_compression", SYSFS_RW), + SYSFS_INT(&suspend_expected_compression, 0, 99) + }, + + { + SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&suspend_compression_ops.enabled, 0, 1) + }, + + { + SUSPEND2_ATTR("algorithm", SYSFS_RW), + SYSFS_STRING(suspend_compressor_name, 31, SYSFS_SM_NOT_NEEDED) + } +}; + +/* + * Ops structure. + */ +static struct suspend_module_ops suspend_compression_ops = { + .type = FILTER_MODULE, + .name = "Suspend2 Compressor", + .module = THIS_MODULE, + .memory_needed = suspend_compress_memory_needed, + .print_debug_info = suspend_compress_print_debug_stats, + .save_config_info = suspend_compress_save_config_info, + .load_config_info = suspend_compress_load_config_info, + .storage_needed = suspend_compress_storage_needed, + + .rw_init = suspend_compress_rw_init, + .rw_cleanup = suspend_compress_rw_cleanup, + + .write_chunk = suspend_compress_write_chunk, + .read_chunk = suspend_compress_read_chunk, +}; + +/* ---- Registration ---- */ + +static __init int suspend_compress_load(void) +{ + int result; + int i, numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + printk("Suspend2 Compression Driver loading.\n"); + if (!(result = suspend_register_module(&suspend_compression_ops))) { + struct kobject *kobj = make_suspend2_sysdir("compression"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, &sysfs_params[i]); + } else + printk("Suspend2 Compression Driver unable to register!\n"); + return result; +} + +#ifdef MODULE +static __exit void suspend_compress_unload(void) +{ + printk("Suspend2 Compression Driver unloading.\n"); + for (i=0; i< numfiles; i++) + suspend_unregister_sysfs_file(&sysfs_params[i]); + suspend_unregister_module(&suspend_compression_ops); +} + +module_init(suspend_compress_load); +module_exit(suspend_compress_unload); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nigel Cunningham"); +MODULE_DESCRIPTION("Compression Support for Suspend2"); +#else +late_initcall(suspend_compress_load); +#endif diff -urN oldtree/kernel/power/disk.c newtree/kernel/power/disk.c --- oldtree/kernel/power/disk.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/disk.c 2006-10-06 17:28:20.000000000 -0400 @@ -10,6 +10,7 @@ */ #include +#include #include #include #include @@ -22,7 +23,7 @@ #include #include "power.h" - +#include "suspend.h" static int noresume = 0; char resume_file[256] = CONFIG_PM_STD_PARTITION; @@ -95,7 +96,7 @@ if (!(error = swsusp_shrink_memory())) return 0; thaw: - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); enable_cpus: enable_nonboot_cpus(); pm_restore_console(); @@ -105,7 +106,7 @@ static void unprepare_processes(void) { platform_finish(); - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); enable_nonboot_cpus(); pm_restore_console(); } @@ -123,6 +124,16 @@ { int error; +#ifdef CONFIG_SUSPEND2 + if (test_action_state(SUSPEND_REPLACE_SWSUSP)) { + if (suspend_start_anything(1)) + return -EBUSY; + suspend_main(1); + suspend_finish_anything(1); + return 0; + } +#endif + error = prepare_processes(); if (error) return error; @@ -186,10 +197,20 @@ * */ -static int software_resume(void) +int software_resume(void) { int error; +#ifdef CONFIG_SUSPEND2 + /* + * We can't know (until an image header - if any - is loaded), whether + * we did override swsusp. We therefore ensure that both are tried. + */ + if (test_action_state(SUSPEND_REPLACE_SWSUSP)) + printk("Replacing swsusp.\n"); + suspend2_try_resume(); +#endif + down(&pm_sem); if (!swsusp_resume_device) { if (!strlen(resume_file)) { @@ -257,9 +278,6 @@ return 0; } -late_initcall(software_resume); - - static const char * const pm_disk_modes[] = { [PM_DISK_FIRMWARE] = "firmware", [PM_DISK_PLATFORM] = "platform", @@ -440,6 +458,7 @@ static int __init noresume_setup(char *str) { noresume = 1; + set_suspend_state(SUSPEND_NORESUME_SPECIFIED); return 1; } diff -urN oldtree/kernel/power/encryption.c newtree/kernel/power/encryption.c --- oldtree/kernel/power/encryption.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/encryption.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,494 @@ +/* + * kernel/power/encryption.c + * + * Copyright (C) 2003-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * This file contains data encryption routines for suspend, + * using cryptoapi transforms. + * + * ToDo: + * - Apply min/max_keysize the cipher changes. + * - Test. + */ + +#include +#include +#include +#include +#include +#include + +#include "suspend.h" +#include "modules.h" +#include "sysfs.h" +#include "io.h" + +#define S2C_WRITE 0 +#define S2C_READ 1 + +static struct suspend_module_ops suspend_encryption_ops; +static struct suspend_module_ops *next_driver; + +static char suspend_encryptor_name[32]; +static struct crypto_blkcipher *suspend_encryptor_transform; +static struct blkcipher_desc suspend_encryptor_desc; +static char suspend_encryptor_key[256]; +static int suspend_key_len; +static char suspend_encryptor_iv[256]; +static int suspend_encryptor_save_key_and_iv; + +static u8 *page_buffer = NULL; +static unsigned int bufofs; + +static struct scatterlist suspend_crypt_sg[PAGE_SIZE/8]; + +/* ---- Local buffer management ---- */ + +/* allocate_local_buffer + * + * Description: Allocates a page of memory for buffering output. + * Returns: Int: Zero if successful, -ENONEM otherwise. + */ +static int allocate_local_buffer(void) +{ + if (!page_buffer) { + int i, remainder; + + page_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + if (!page_buffer) { + printk(KERN_ERR + "Failed to allocate the page buffer for " + "suspend2 encryption driver.\n"); + return -ENOMEM; + } + + for (i=0; i < (PAGE_SIZE / suspend_key_len); i++) { + suspend_crypt_sg[i].page = virt_to_page(page_buffer); + suspend_crypt_sg[i].offset = suspend_key_len * i; + suspend_crypt_sg[i].length = suspend_key_len; + } + + remainder = PAGE_SIZE % suspend_key_len; + + if (remainder) { + suspend_crypt_sg[i].page = virt_to_page(page_buffer); + suspend_crypt_sg[i].offset = suspend_key_len * i; + suspend_crypt_sg[i].length = remainder; + } + } + + return 0; +} + +/* free_local_buffer + * + * Description: Frees memory allocated for buffering output. + */ +static void free_local_buffer(void) +{ + if (page_buffer) + free_page((unsigned long) page_buffer); + + page_buffer = NULL; +} + +/* suspend_encrypt_rw_cleanup + * + * Description: Frees memory allocated for our labours. + */ +static int suspend_encrypt_rw_cleanup(int rw) +{ + if (suspend_encryptor_transform) { + crypto_free_blkcipher(suspend_encryptor_transform); + suspend_encryptor_transform = NULL; + } + + free_local_buffer(); + + return 0; +} + +/* suspend_crypto_prepare + * + * Description: Prepare to do some work by allocating buffers and transforms. + * Returns: Int: Zero if successful, 1 otherwise. + */ +static int suspend_encrypt_rw_prepare(int rw) +{ + int iv_len; + + if (!*suspend_encryptor_name) { + printk("Suspend2: Encryptor enabled but no name set.\n"); + suspend_encryption_ops.enabled = 0; + return 1; + } + + suspend_encryptor_desc.tfm = suspend_encryptor_transform; + suspend_encryptor_desc.flags = 0; + + suspend_encryptor_transform = crypto_alloc_blkcipher( + suspend_encryptor_name, 0, 0); + if (!suspend_encryptor_transform) { + printk("Suspend2: Failed to initialise the encryption " + "transform (%s).\n", + suspend_encryptor_name); + suspend_encryption_ops.enabled = 0; + return 1; + } + + if (rw == READ) + bufofs = PAGE_SIZE; + else + bufofs = 0; + + crypto_blkcipher_clear_flags(suspend_encryptor_transform, 0); + + suspend_key_len = strlen(suspend_encryptor_key); + + if (crypto_blkcipher_setkey(suspend_encryptor_transform, suspend_encryptor_key, + suspend_key_len)) { + printk("%d is an invalid key length for cipher %s.\n", + suspend_key_len, + suspend_encryptor_name); + return 1; + } + + iv_len = crypto_blkcipher_ivsize(suspend_encryptor_transform); + + if (iv_len) + crypto_blkcipher_set_iv(suspend_encryptor_transform, + suspend_encryptor_iv, + iv_len); + + return 0; +} + +/* ---- Exported functions ---- */ + +/* suspend_encrypt_write_chunk() + * + * Description: Encrypt a page of data, buffering output and passing on + * filled pages to the next module in the pipeline. + * Arguments: Buffer_page: Pointer to a buffer of size PAGE_SIZE, + * containing data to be encrypted. + * Returns: 0 on success. Otherwise the error is that returned by later + * modules, -ECHILD if we have a broken pipeline or -EIO if + * zlib errs. + */ +static int suspend_encrypt_write_chunk(struct page *buffer_page) +{ + int ret; + unsigned int len; + u16 len_written; + char *buffer_start; + + if (!suspend_encryptor_transform) + return next_driver->write_chunk(buffer_page); + + buffer_start = kmap(buffer_page); + memcpy(page_buffer, buffer_start, PAGE_SIZE); + kunmap(buffer_page); + + bytes_in += PAGE_SIZE; + + len = PAGE_SIZE; + + ret = crypto_blkcipher_encrypt(&suspend_encryptor_desc, + suspend_crypt_sg, suspend_crypt_sg, PAGE_SIZE); + + if (ret) { + printk("Encryption failed.\n"); + return -EIO; + } + + len_written = (u16) len; + + ret = next_driver->write_chunk(virt_to_page(page_buffer)); + + return ret; +} + +/* rw_init() + * + * Description: Prepare to read a new stream of data. + * Arguments: int: Section of image about to be read. + * Returns: int: Zero on success, error number otherwise. + */ +static int suspend_encrypt_rw_init(int rw, int stream_number) +{ + int result; + + next_driver = suspend_get_next_filter(&suspend_encryption_ops); + + if (!next_driver) { + printk("Encryption Driver: Argh! I'm at the end of the pipeline!"); + return -ECHILD; + } + + if ((result = suspend_encrypt_rw_prepare(rw))) { + set_result_state(SUSPEND_ENCRYPTION_SETUP_FAILED); + suspend_encrypt_rw_cleanup(rw); + return result; + } + + if ((result = allocate_local_buffer())) + return result; + + if (rw == WRITE && stream_number == 2) + bytes_in = bytes_out = 0; + + bufofs = (rw == READ) ? PAGE_SIZE : 0; + + return 0; +} + +/* suspend_encrypt_read_chunk() + * + * Description: Retrieve data from later modules and deencrypt it until the + * input buffer is filled. + * Arguments: Buffer_start: Pointer to a buffer of size PAGE_SIZE. + * Sync: Whether the previous module (or core) wants its + * data synchronously. + * Returns: Zero if successful. Error condition from me or from downstream + * on failure. + */ +static int suspend_encrypt_read_chunk(struct page *buffer_page, int sync) +{ + int ret; + char *buffer_start; + + if (!suspend_encryptor_transform) + return next_driver->read_chunk(buffer_page, sync); + + /* + * All our reads must be synchronous - we can't deencrypt + * data that hasn't been read yet. + */ + + if ((ret = next_driver->read_chunk( + virt_to_page(page_buffer), SUSPEND_SYNC)) < 0) { + printk("Failed to read an encrypted block.\n"); + return ret; + } + + ret = crypto_blkcipher_decrypt(&suspend_encryptor_desc, + suspend_crypt_sg, suspend_crypt_sg, PAGE_SIZE); + + if (ret) + printk("Decrypt function returned %d.\n", ret); + + buffer_start = kmap(buffer_page); + memcpy(buffer_start, page_buffer, PAGE_SIZE); + kunmap(buffer_page); + return ret; +} + +/* suspend_encrypt_print_debug_stats + * + * Description: Print information to be recorded for debugging purposes into a + * buffer. + * Arguments: buffer: Pointer to a buffer into which the debug info will be + * printed. + * size: Size of the buffer. + * Returns: Number of characters written to the buffer. + */ +static int suspend_encrypt_print_debug_stats(char *buffer, int size) +{ + int len; + + if (*suspend_encryptor_name) + len = snprintf_used(buffer, size, "- Encryptor is '%s'.\n", + suspend_encryptor_name); + else + len = snprintf_used(buffer, size, "- Encryptor is not set.\n"); + return len; +} + +/* encryption_memory_needed + * + * Description: Tell the caller how much memory we need to operate during + * suspend/resume. + * Returns: Unsigned long. Maximum number of bytes of memory required for + * operation. + */ +static unsigned long suspend_encrypt_memory_needed(void) +{ + return PAGE_SIZE; +} + +static unsigned long suspend_encrypt_storage_needed(void) +{ + return 3 + strlen(suspend_encryptor_name) + + (suspend_encryptor_save_key_and_iv ? + (4 + strlen(suspend_encryptor_key) + + strlen(suspend_encryptor_iv)) : 0); +} + +/* suspend_encrypt_save_config_info + * + * Description: Save informaton needed when reloading the image at resume time. + * Arguments: Buffer: Pointer to a buffer of size PAGE_SIZE. + * Returns: Number of bytes used for saving our data. + */ +static int suspend_encrypt_save_config_info(char *buffer) +{ + int buf_offset, str_size; + + str_size = strlen(suspend_encryptor_name); + *buffer = (char) str_size; + strncpy(buffer + 1, suspend_encryptor_name, str_size + 1); + buf_offset = str_size + 2; + + *(buffer + buf_offset) = (char) suspend_encryptor_save_key_and_iv; + buf_offset++; + + if (suspend_encryptor_save_key_and_iv) { + + str_size = strlen(suspend_encryptor_key); + *(buffer + buf_offset) = (char) str_size; + strncpy(buffer + buf_offset + 1, suspend_encryptor_key, str_size + 1); + + buf_offset+= str_size + 2; + + str_size = strlen(suspend_encryptor_iv); + *(buffer + buf_offset) = (char) str_size; + strncpy(buffer + buf_offset + 1, suspend_encryptor_iv, str_size + 1); + + buf_offset += str_size + 2; + } + + return buf_offset; +} + +/* suspend_encrypt_load_config_info + * + * Description: Reload information needed for deencrypting the image at + * resume time. + * Arguments: Buffer: Pointer to the start of the data. + * Size: Number of bytes that were saved. + */ +static void suspend_encrypt_load_config_info(char *buffer, int size) +{ + int buf_offset, str_size; + + str_size = (int) *buffer; + strncpy(suspend_encryptor_name, buffer + 1, str_size + 1); + buf_offset = str_size + 2; + + suspend_encryptor_save_key_and_iv = (int) *(buffer + buf_offset); + buf_offset++; + + if (suspend_encryptor_save_key_and_iv) { + str_size = (int) *(buffer + buf_offset); + strncpy(suspend_encryptor_key, buffer + buf_offset + 1, str_size + 1); + + buf_offset+= str_size + 2; + + str_size = (int) *(buffer + buf_offset); + strncpy(suspend_encryptor_iv, buffer + buf_offset + 1, str_size + 1); + + buf_offset += str_size + 2; + } else { + *suspend_encryptor_key = 0; + *suspend_encryptor_iv = 0; + } + + if (buf_offset != size) { + printk("Suspend Encryptor config info size mismatch (%d != %d): settings ignored.\n", + buf_offset, size); + *suspend_encryptor_key = 0; + *suspend_encryptor_iv = 0; + } + return; +} + +/* + * data for our sysfs entries. + */ +static struct suspend_sysfs_data sysfs_params[] = { + { + SUSPEND2_ATTR("algorithm", SYSFS_RW), + SYSFS_STRING(suspend_encryptor_name, 31, SYSFS_SM_NOT_NEEDED) + }, + + { + SUSPEND2_ATTR("save_key_and_iv", SYSFS_RW), + SYSFS_INT(&suspend_encryptor_save_key_and_iv, 0, 1) + }, + + { + SUSPEND2_ATTR("key", SYSFS_RW), + SYSFS_STRING(suspend_encryptor_key, 255, SYSFS_SM_NOT_NEEDED) + }, + + { + SUSPEND2_ATTR("iv", SYSFS_RW), + SYSFS_STRING(suspend_encryptor_iv, 255, SYSFS_SM_NOT_NEEDED) + }, + + { + SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&suspend_encryption_ops.enabled, 0, 1) + }, + +}; + +/* + * Ops structure. + */ + +static struct suspend_module_ops suspend_encryption_ops = { + .type = FILTER_MODULE, + .name = "Encryptor", + .module = THIS_MODULE, + .memory_needed = suspend_encrypt_memory_needed, + .print_debug_info = suspend_encrypt_print_debug_stats, + .save_config_info = suspend_encrypt_save_config_info, + .load_config_info = suspend_encrypt_load_config_info, + .storage_needed = suspend_encrypt_storage_needed, + + .rw_init = suspend_encrypt_rw_init, + .rw_cleanup = suspend_encrypt_rw_cleanup, + + .write_chunk = suspend_encrypt_write_chunk, + .read_chunk = suspend_encrypt_read_chunk, +}; + +/* ---- Registration ---- */ + +static __init int suspend_encrypt_load(void) +{ + int result; + int i, numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + printk("Suspend2 Encryption Driver loading.\n"); + if (!(result = suspend_register_module(&suspend_encryption_ops))) { + struct kobject *kobj = make_suspend2_sysdir("encryption"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, &sysfs_params[i]); + suspend_encryption_ops.enabled = 0; + } else + printk("Suspend2 Encryption Driver unable to register!\n"); + return result; +} + +#ifdef MODULE +static __exit void suspend_compress_unload(void) +{ + printk("Suspend2 Encryption Driver unloading.\n"); + for (i=0; i< numfiles; i++) + suspend_unregister_sysfs_file(&sysfs_params[i]); + suspend_unregister_module(&suspend_encryption_ops); +} + +module_init(suspend_encrypt_load); +module_exit(suspend_encrypt_unload); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nigel Cunningham"); +MODULE_DESCRIPTION("Encryption Support for Suspend2"); +#else +late_initcall(suspend_encrypt_load); +#endif diff -urN oldtree/kernel/power/extent.c newtree/kernel/power/extent.c --- oldtree/kernel/power/extent.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/extent.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,309 @@ +/* + * kernel/power/extent.c + * + * (C) 2003-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * These functions encapsulate the manipulation of storage metadata. For + * pageflags, we use dynamically allocated bitmaps. + */ + +#include +#include +#include "modules.h" +#include "extent.h" +#include "ui.h" + +int suspend_extents_allocated = 0; + +/* suspend_get_extent + * + * Returns a free extent. May fail, returning NULL instead. + */ +static struct extent *suspend_get_extent(void) +{ + struct extent *result; + + if (!(result = kmalloc(sizeof(struct extent), GFP_ATOMIC))) + return NULL; + + suspend_extents_allocated++; + result->minimum = result->maximum = 0; + result->next = NULL; + + return result; +} + +/* suspend_put_extent. + * + * Frees an extent. Assumes unlinking is done by the caller. + */ +void suspend_put_extent(struct extent *extent) +{ + BUG_ON(!extent); + + kfree(extent); + suspend_extents_allocated--; +} + +/* suspend_put_extent_chain. + * + * Frees a whole chain of extents. + */ +void suspend_put_extent_chain(struct extent_chain *chain) +{ + struct extent *this; + + this = chain->first; + + while(this) { + struct extent *next = this->next; + kfree(this); + chain->frees++; + suspend_extents_allocated --; + this = next; + } + + BUG_ON(chain->frees != chain->allocs); + chain->first = chain->last = chain->last_touched = NULL; + chain->size = chain->allocs = chain->frees = 0; +} + +/* + * suspend_add_to_extent_chain + * + * Add an extent to an existing chain. + */ +int suspend_add_to_extent_chain(struct extent_chain *chain, + unsigned long minimum, unsigned long maximum) +{ + struct extent *new_extent = NULL, *start_at; + + /* Find the right place in the chain */ + start_at = (chain->last_touched && + (chain->last_touched->minimum < minimum)) ? + chain->last_touched : NULL; + + if (!start_at && chain->first && chain->first->minimum < minimum) + start_at = chain->first; + + while (start_at && start_at->next && start_at->next->minimum < minimum) + start_at = start_at->next; + + if (start_at && start_at->maximum == (minimum - 1)) { + start_at->maximum = maximum; + + /* Merge with the following one? */ + if (start_at->next && + start_at->maximum + 1 == start_at->next->minimum) { + struct extent *to_free = start_at->next; + start_at->maximum = start_at->next->maximum; + start_at->next = start_at->next->next; + chain->frees++; + suspend_put_extent(to_free); + } + + chain->last_touched = start_at; + chain->size+= (maximum - minimum + 1); + + return 0; + } + + new_extent = suspend_get_extent(); + if (!new_extent) { + printk("Error unable to append a new extent to the chain.\n"); + return 2; + } + + chain->allocs++; + chain->size+= (maximum - minimum + 1); + new_extent->minimum = minimum; + new_extent->maximum = maximum; + new_extent->next = NULL; + + chain->last_touched = new_extent; + + if (start_at) { + struct extent *next = start_at->next; + start_at->next = new_extent; + new_extent->next = next; + if (!next) + chain->last = new_extent; + } else { + if (chain->first) { + new_extent->next = chain->first; + chain->first = new_extent; + } else + chain->last = chain->first = new_extent; + } + + return 0; +} + +/* suspend_serialise_extent_chain + * + * Write a chain in the image. + */ +int suspend_serialise_extent_chain(struct suspend_module_ops *owner, + struct extent_chain *chain) +{ + struct extent *this; + int ret, i = 0; + + if ((ret = suspend_active_writer->rw_header_chunk(WRITE, owner, + (char *) chain, + 3 * sizeof(int)))) + return ret; + + this = chain->first; + while (this) { + if ((ret = suspend_active_writer->rw_header_chunk(WRITE, owner, + (char *) this, + 2 * sizeof(unsigned long)))) + return ret; + this = this->next; + i++; + } + + if (i != (chain->allocs - chain->frees)) { + printk(KERN_EMERG "Saved %d extents but chain metadata says there should be %d-%d.\n", + i, chain->allocs, chain->frees); + BUG(); + } + + return ret; +} + +/* suspend_load_extent_chain + * + * Read back a chain saved in the image. + */ +int suspend_load_extent_chain(struct extent_chain *chain) +{ + struct extent *this, *last = NULL; + int i, ret; + + if (!(ret = suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) chain, + 3 * sizeof(int)))) + return ret; + + for (i = 0; i < (chain->allocs - chain->frees); i++) { + this = kmalloc(sizeof(struct extent), GFP_ATOMIC); + BUG_ON(!this); /* Shouldn't run out of memory trying this! */ + this->next = NULL; + if (!(ret = suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) this, 2 * sizeof(unsigned long)))) + return ret; + if (last) + last->next = this; + else + chain->first = this; + last = this; + } + chain->last = last; + return ret; +} + +/* suspend_extent_state_next + * + * Given a state, progress to the next valid entry. We may begin in an + * invalid state, as we do when invoked after extent_state_goto_start below. + * + * When using compression and expected_compression > 0, we allocate fewer + * swap entries, so we can validly run out of data to return. + */ +unsigned long suspend_extent_state_next(struct extent_iterate_state *state) +{ + if (state->current_chain > state->num_chains) + return 0; + + if (state->current_extent) { + if (state->current_offset == state->current_extent->maximum) { + if (state->current_extent->next) { + state->current_extent = state->current_extent->next; + state->current_offset = state->current_extent->minimum; + } else { + state->current_extent = NULL; + state->current_offset = 0; + } + } else + state->current_offset++; + } + + while(!state->current_extent) { + int chain_num = ++(state->current_chain); + + if (chain_num > state->num_chains) + return 0; + + state->current_extent = (state->chains + chain_num)->first; + + if (!state->current_extent) + continue; + + state->current_offset = state->current_extent->minimum; + } + + return state->current_offset; +} + +/* suspend_extent_state_goto_start + * + * Find the first valid value in a group of chains. + */ +void suspend_extent_state_goto_start(struct extent_iterate_state *state) +{ + state->current_chain = -1; + state->current_extent = NULL; + state->current_offset = 0; +} + +/* suspend_extent_start_save + * + * Given a state and a struct extent_state_store, save the crreutn + * position in a format that can be used with relocated chains (at + * resume time). + */ +void suspend_extent_state_save(struct extent_iterate_state *state, + struct extent_iterate_saved_state *saved_state) +{ + struct extent *extent; + + saved_state->chain_num = state->current_chain; + saved_state->extent_num = 0; + saved_state->offset = state->current_offset; + + if (saved_state->chain_num == -1) + return; + + extent = (state->chains + state->current_chain)->first; + + while (extent != state->current_extent) { + saved_state->extent_num++; + extent = extent->next; + } +} + +/* suspend_extent_start_restore + * + * Restore the position saved by extent_state_save. + */ +void suspend_extent_state_restore(struct extent_iterate_state *state, + struct extent_iterate_saved_state *saved_state) +{ + int posn = saved_state->extent_num; + + if (saved_state->chain_num == -1) { + suspend_extent_state_goto_start(state); + return; + } + + state->current_chain = saved_state->chain_num; + state->current_extent = (state->chains + state->current_chain)->first; + state->current_offset = saved_state->offset; + + while (posn--) + state->current_extent = state->current_extent->next; +} diff -urN oldtree/kernel/power/extent.h newtree/kernel/power/extent.h --- oldtree/kernel/power/extent.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/extent.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,82 @@ +/* + * kernel/power/extent.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It contains declarations related to extents. Extents are + * suspend's method of storing some of the metadata for the image. + * See extent.c for more info. + * + */ + +#include "modules.h" + +#ifndef EXTENT_H +#define EXTENT_H +struct extent_chain { + int size; /* size of the extent ie sum (max-min+1) */ + int allocs, frees; + char *name; + struct extent *first, *last, *last_touched; +}; + +/* + * We rely on extents not fitting evenly into a page. + * The last four bytes are used to store the number + * of the page, to make saving & reloading pages simpler. + */ +struct extent { + unsigned long minimum, maximum; + struct extent *next; +}; + +struct extent_iterate_state { + struct extent_chain *chains; + int num_chains; + int current_chain; + struct extent *current_extent; + unsigned long current_offset; +}; + +struct extent_iterate_saved_state { + int chain_num; + int extent_num; + unsigned long offset; +}; + +#define suspend_extent_state_eof(state) ((state)->num_chains < (state)->current_chain) + +#define suspend_extent_for_each(extent_chain, extentpointer, value) \ +if ((extent_chain)->first) \ + for ((extentpointer) = (extent_chain)->first, (value) = \ + (extentpointer)->minimum; \ + ((extentpointer) && ((extentpointer)->next || (value) <= \ + (extentpointer)->maximum)); \ + (((value) == (extentpointer)->maximum) ? \ + ((extentpointer) = (extentpointer)->next, (value) = \ + ((extentpointer) ? (extentpointer)->minimum : 0)) : \ + (value)++)) + +extern int suspend_extents_allocated; +void suspend_put_extent_chain(struct extent_chain *chain); +int suspend_add_to_extent_chain(struct extent_chain *chain, + unsigned long minimum, unsigned long maximum); +int suspend_serialise_extent_chain(struct suspend_module_ops *owner, + struct extent_chain *chain); +int suspend_load_extent_chain(struct extent_chain *chain); + +/* swap_entry_to_extent_val & extent_val_to_swap_entry: + * We are putting offset in the low bits so consecutive swap entries + * make consecutive extent values */ +#define swap_entry_to_extent_val(swp_entry) (swp_entry.val) +#define extent_val_to_swap_entry(val) (swp_entry_t) { (val) } + +void suspend_extent_state_save(struct extent_iterate_state *state, + struct extent_iterate_saved_state *saved_state); +void suspend_extent_state_restore(struct extent_iterate_state *state, + struct extent_iterate_saved_state *saved_state); +void suspend_extent_state_goto_start(struct extent_iterate_state *state); +unsigned long suspend_extent_state_next(struct extent_iterate_state *state); +#endif diff -urN oldtree/kernel/power/io.c newtree/kernel/power/io.c --- oldtree/kernel/power/io.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/io.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,1014 @@ +/* + * kernel/power/io.c + * + * Copyright (C) 1998-2001 Gabor Kuti + * Copyright (C) 1998,2001,2002 Pavel Machek + * Copyright (C) 2002-2003 Florent Chabaud + * Copyright (C) 2002-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It contains high level IO routines for suspending. + * + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "suspend.h" +#include "modules.h" +#include "pageflags.h" +#include "io.h" +#include "ui.h" +#include "storage.h" +#include "prepare_image.h" +#include "extent.h" + +/* suspend_attempt_to_parse_resume_device + * + * Can we suspend, using the current resume2= parameter? + */ +int suspend_attempt_to_parse_resume_device(void) +{ + struct list_head *writer; + struct suspend_module_ops *this_writer; + int result, returning = 0; + + if (suspend_activate_storage(0)) + return 0; + + suspend_active_writer = NULL; + clear_suspend_state(SUSPEND_RESUME_DEVICE_OK); + clear_suspend_state(SUSPEND_CAN_RESUME); + clear_result_state(SUSPEND_ABORTED); + + if (!suspend_num_writers) { + printk(name_suspend "No writers have been registered. Suspending will be disabled.\n"); + goto cleanup; + } + + if (!resume2_file[0]) { + printk(name_suspend "Resume2 parameter is empty. Suspending will be disabled.\n"); + goto cleanup; + } + + list_for_each(writer, &suspend_writers) { + this_writer = list_entry(writer, struct suspend_module_ops, type_list); + + /* + * Not sure why you'd want to disable a writer, but + * we should honour the flag if we're providing it + */ + if (!this_writer->enabled) + continue; + + result = this_writer->parse_sig_location( + resume2_file, (suspend_num_writers == 1)); + + switch (result) { + case -EINVAL: + /* + * For this writer, but not a valid + * configuration. Error already printed. + */ + + goto cleanup; + + case 0: + /* + * For this writer and valid. + */ + + suspend_active_writer = this_writer; + + set_suspend_state(SUSPEND_RESUME_DEVICE_OK); + set_suspend_state(SUSPEND_CAN_RESUME); + printk(name_suspend "Resuming enabled.\n"); + + returning = 1; + goto cleanup; + } + } + printk(name_suspend "No matching enabled writer found. Resuming disabled.\n"); +cleanup: + suspend_deactivate_storage(0); + return returning; +} + +void attempt_to_parse_resume_device2(void) +{ + suspend_prepare_usm(); + suspend_attempt_to_parse_resume_device(); + suspend_cleanup_usm(); +} + +/* noresume_reset_modules + * + * Description: When we read the start of an image, modules (and especially the + * active writer) might need to reset data structures if we decide + * to invalidate the image rather than resuming from it. + */ + +static void noresume_reset_modules(void) +{ + struct suspend_module_ops *this_filter; + + list_for_each_entry(this_filter, &suspend_filters, type_list) { + if (this_filter->noresume_reset) + this_filter->noresume_reset(); + } + + if (suspend_active_writer && suspend_active_writer->noresume_reset) + suspend_active_writer->noresume_reset(); +} + +/* fill_suspend_header() + * + * Description: Fill the suspend header structure. + * Arguments: struct suspend_header: Header data structure to be filled. + */ + +static void fill_suspend_header(struct suspend_header *sh) +{ + int i; + + memset((char *)sh, 0, sizeof(*sh)); + + sh->version_code = LINUX_VERSION_CODE; + sh->num_physpages = num_physpages; + sh->orig_mem_free = suspend_orig_mem_free; + memcpy(&sh->uts, init_utsname(), sizeof(struct new_utsname)); + sh->page_size = PAGE_SIZE; + sh->pagedir = pagedir1; + sh->pageset_2_size = pagedir2.pageset_size; + sh->param0 = suspend_result; + sh->param1 = suspend_action; + sh->param2 = suspend_debug_state; + sh->param3 = console_loglevel; + sh->root_fs = current->fs->rootmnt->mnt_sb->s_dev; + for (i = 0; i < 4; i++) + sh->io_time[i/2][i%2] = + suspend_io_time[i/2][i%2]; +} + +/* + * rw_init_modules + * + * Iterate over modules, preparing the ones that will be used to read or write + * data. + */ +static int rw_init_modules(int rw, int which) +{ + struct suspend_module_ops *this_module; + /* Initialise page transformers */ + list_for_each_entry(this_module, &suspend_filters, type_list) { + if (!this_module->enabled) + continue; + if (this_module->rw_init && + this_module->rw_init(rw, which)) { + abort_suspend("Failed to initialise the %s filter.", + this_module->name); + return 1; + } + } + + /* Initialise writer */ + if (suspend_active_writer->rw_init(rw, which)) { + abort_suspend("Failed to initialise the writer."); + if (!rw) + suspend_active_writer->invalidate_image(); + return 1; + } + + /* Initialise other modules */ + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled) + continue; + if ((this_module->type == FILTER_MODULE) || + (this_module->type == WRITER_MODULE)) + continue; + if (this_module->rw_init && this_module->rw_init(rw, which)) { + set_result_state(SUSPEND_ABORTED); + return 1; + } + } + + return 0; +} + +/* + * rw_cleanup_modules + * + * Cleanup components after reading or writing a set of pages. + * Only the writer may fail. + */ +static int rw_cleanup_modules(int rw) +{ + struct suspend_module_ops *this_module; + int result = 0; + + /* Cleanup other modules */ + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled) + continue; + if ((this_module->type == FILTER_MODULE) || + (this_module->type == WRITER_MODULE)) + continue; + if (this_module->rw_cleanup) + result |= this_module->rw_cleanup(rw); + } + + /* Flush data and cleanup */ + list_for_each_entry(this_module, &suspend_filters, type_list) { + if (!this_module->enabled) + continue; + if (this_module->rw_cleanup) + result |= this_module->rw_cleanup(rw); + } + + result |= suspend_active_writer->rw_cleanup(rw); + + return result; +} + +/* + * do_rw_loop + * + * The main I/O loop for reading or writing pages. + */ +static int do_rw_loop(int write, int finish_at, dyn_pageflags_t *pageflags, + int base, int barmax) +{ + int pfn, pc, step = 1, nextupdate = 0, i; + int result; + struct suspend_module_ops *first_filter = suspend_get_next_filter(NULL); + char *buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + pfn = get_next_bit_on(*pageflags, max_pfn); + + pc = finish_at / 5; + + /* Read the pages */ + for (i=0; i< finish_at; i++) { + struct page *page = pfn_to_page(pfn); + + /* Status */ + if ((i+base) >= nextupdate) + nextupdate = suspend_update_status(i+base, barmax, + " %d/%d MB ", MB(base+i+1), MB(barmax)); + + if ((i + 1) == pc) { + printk("%d%%...", 20 * step); + step++; + pc = finish_at * step / 5; + } + + if (write) + result = first_filter->write_chunk(page); + else + result = first_filter->read_chunk(page, SUSPEND_ASYNC); + + if (result) { + if (write) { + printk("Write chunk returned %d.\n", result); + abort_suspend("Failed to write a chunk of the " + "image."); + free_page((unsigned long) buffer); + return result; + } else + panic("Failed to read chunk %d/%d of the image. (%d)", + i, finish_at, result); + } + + /* Interactivity*/ + suspend_cond_pause(0, NULL); + + if (test_result_state(SUSPEND_ABORTED) && write) { + free_page((unsigned long) buffer); + return 1; + } + + /* Prepare next */ + pfn = get_next_bit_on(*pageflags, pfn); + } + + printk("done.\n"); + + suspend_update_status(base + finish_at, barmax, " %d/%d MB ", + MB(base + finish_at), MB(barmax)); + free_page((unsigned long) buffer); + return 0; +} + +/* write_pageset() + * + * Description: Write a pageset to disk. + * Arguments: pagedir: Pointer to the pagedir to be saved. + * whichtowrite: Controls what debugging output is printed. + * Returns: Zero on success or -1 on failure. + */ + +int write_pageset(struct pagedir *pagedir, int whichtowrite) +{ + int finish_at, base = 0, start_time, end_time; + int barmax = pagedir1.pageset_size + pagedir2.pageset_size; + long error = 0; + dyn_pageflags_t *pageflags; + + /* + * Even if there is nothing to read or write, the writer + * may need the init/cleanup for it's housekeeping. (eg: + * Pageset1 may start where pageset2 ends when writing). + */ + finish_at = pagedir->pageset_size; + + if (whichtowrite == 1) { + suspend_prepare_status(DONT_CLEAR_BAR, + "Writing kernel & process data..."); + base = pagedir2.pageset_size; + if (test_action_state(SUSPEND_TEST_FILTER_SPEED) || + test_action_state(SUSPEND_TEST_BIO)) + pageflags = &pageset1_map; + else + pageflags = &pageset1_copy_map; + } else { + suspend_prepare_status(CLEAR_BAR, "Writing caches..."); + pageflags = &pageset2_map; + bytes_in = bytes_out = 0; + } + + start_time = jiffies; + + if (!rw_init_modules(1, whichtowrite)) + error = do_rw_loop(1, finish_at, pageflags, base, barmax); + + if (rw_cleanup_modules(WRITE)) { + abort_suspend("Failed to cleanup after writing."); + error = 1; + } + + /* Statistics */ + end_time = jiffies; + + if ((end_time - start_time) && (!test_result_state(SUSPEND_ABORTED))) { + suspend_io_time[0][0] += finish_at, + suspend_io_time[0][1] += (end_time - start_time); + } + + return error; +} + +/* read_pageset() + * + * Description: Read a pageset from disk. + * Arguments: pagedir: Pointer to the pagedir to be saved. + * whichtowrite: Controls what debugging output is printed. + * overwrittenpagesonly: Whether to read the whole pageset or + * only part. + * Returns: Zero on success or -1 on failure. + */ + +static int read_pageset(struct pagedir *pagedir, int whichtoread, + int overwrittenpagesonly) +{ + int result = 0, base = 0, start_time, end_time; + int finish_at = pagedir->pageset_size; + int barmax = pagedir1.pageset_size + pagedir2.pageset_size; + dyn_pageflags_t *pageflags; + + if (whichtoread == 1) { + suspend_prepare_status(CLEAR_BAR, + "Reading kernel & process data..."); + pageflags = &pageset1_copy_map; + } else { + suspend_prepare_status(DONT_CLEAR_BAR, "Reading caches..."); + if (overwrittenpagesonly) + barmax = finish_at = min(pagedir1.pageset_size, + pagedir2.pageset_size); + else { + base = pagedir1.pageset_size; + } + pageflags = &pageset2_map; + } + + start_time = jiffies; + + if (rw_init_modules(0, whichtoread)) { + suspend_active_writer->invalidate_image(); + result = 1; + } else + result = do_rw_loop(0, finish_at, pageflags, base, barmax); + + if (rw_cleanup_modules(READ)) { + abort_suspend("Failed to cleanup after reading."); + result = 1; + } + + /* Statistics */ + end_time=jiffies; + + if ((end_time - start_time) && (!test_result_state(SUSPEND_ABORTED))) { + suspend_io_time[1][0] += finish_at, + suspend_io_time[1][1] += (end_time - start_time); + } + + return result; +} + +/* write_module_configs() + * + * Description: Store the configuration for each module in the image header. + * Returns: Int: Zero on success, Error value otherwise. + */ +static int write_module_configs(void) +{ + struct suspend_module_ops *this_module; + char *buffer = (char *) get_zeroed_page(GFP_ATOMIC); + int len, index = 1; + struct suspend_module_header suspend_module_header; + + if (!buffer) { + printk("Failed to allocate a buffer for saving " + "module configuration info.\n"); + return -ENOMEM; + } + + /* + * We have to know which data goes with which module, so we at + * least write a length of zero for a module. Note that we are + * also assuming every module's config data takes <= PAGE_SIZE. + */ + + /* For each module (in registration order) */ + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled || !this_module->storage_needed || + (this_module->type == WRITER_MODULE && + suspend_active_writer != this_module)) + continue; + + /* Get the data from the module */ + len = 0; + if (this_module->save_config_info) + len = this_module->save_config_info(buffer); + + /* Save the details of the module */ + suspend_module_header.enabled = this_module->enabled; + suspend_module_header.type = this_module->type; + suspend_module_header.index = index++; + strncpy(suspend_module_header.name, this_module->name, + sizeof(suspend_module_header.name)); + suspend_active_writer->rw_header_chunk(WRITE, + this_module, + (char *) &suspend_module_header, + sizeof(suspend_module_header)); + + /* Save the size of the data and any data returned */ + suspend_active_writer->rw_header_chunk(WRITE, + this_module, + (char *) &len, sizeof(int)); + if (len) + suspend_active_writer->rw_header_chunk( + WRITE, this_module, buffer, len); + } + + /* Write a blank header to terminate the list */ + suspend_module_header.name[0] = '\0'; + suspend_active_writer->rw_header_chunk(WRITE, + NULL, + (char *) &suspend_module_header, + sizeof(suspend_module_header)); + + free_page((unsigned long) buffer); + return 0; +} + +/* read_module_configs() + * + * Description: Reload module configurations from the image header. + * Returns: Int. Zero on success, error value otherwise. + */ + +static int read_module_configs(void) +{ + struct suspend_module_ops *this_module; + char *buffer = (char *) get_zeroed_page(GFP_ATOMIC); + int len, result = 0; + struct suspend_module_header suspend_module_header; + + if (!buffer) { + printk("Failed to allocate a buffer for reloading module " + "configuration info.\n"); + return -ENOMEM; + } + + /* All modules are initially disabled. That way, if we have a module + * loaded now that wasn't loaded when we suspended, it won't be used + * in trying to read the data. + */ + list_for_each_entry(this_module, &suspend_modules, module_list) + this_module->enabled = 0; + + /* Get the first module header */ + result = suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) &suspend_module_header, sizeof(suspend_module_header)); + if (!result) { + printk("Failed to read the next module header.\n"); + free_page((unsigned long) buffer); + return -EINVAL; + } + + /* For each module (in registration order) */ + while (suspend_module_header.name[0]) { + + /* Find the module */ + this_module = suspend_find_module_given_name(suspend_module_header.name); + + if (!this_module) { + /* + * Is it used? Only need to worry about filters. The active + * writer must be loaded! + */ + if (suspend_module_header.enabled) { + suspend_early_boot_message(1, SUSPEND_CONTINUE_REQ, + "It looks like we need module %s for " + "reading the image but it hasn't been " + "registered.\n", + suspend_module_header.name); + if (!(test_suspend_state(SUSPEND_CONTINUE_REQ))) { + suspend_active_writer->invalidate_image(); + free_page((unsigned long) buffer); + return -EINVAL; + } + } else + printk("Module %s configuration data found, but the module " + "hasn't registered. Looks like it was disabled, so " + "we're ignoring it's data.", + suspend_module_header.name); + } + + /* Get the length of the data (if any) */ + result = suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) &len, sizeof(int)); + if (!result) { + printk("Failed to read the length of the module %s's" + " configuration data.\n", + suspend_module_header.name); + free_page((unsigned long) buffer); + return -EINVAL; + } + + /* Read any data and pass to the module (if we found one) */ + if (len) { + suspend_active_writer->rw_header_chunk(READ, NULL, + buffer, len); + if (this_module) { + if (!this_module->save_config_info) { + printk("Huh? Module %s appears to have " + "a save_config_info, but not a " + "load_config_info function!\n", + this_module->name); + } else + this_module->load_config_info(buffer, len); + } + } + + if (this_module) { + /* Now move this module to the tail of its lists. This + * will put it in order. Any new modules will end up at + * the top of the lists. They should have been set to + * disabled when loaded (people will normally not edit + * an initrd to load a new module and then suspend + * without using it!). + */ + + suspend_move_module_tail(this_module); + + /* + * We apply the disabled state; modules don't need to + * save whether they were disabled and if they do, we + * override them anyway. + */ + this_module->enabled = suspend_module_header.enabled; + } + + /* Get the next module header */ + result = suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) &suspend_module_header, + sizeof(suspend_module_header)); + + if (!result) { + printk("Failed to read the next module header.\n"); + free_page((unsigned long) buffer); + return -EINVAL; + } + + } + + free_page((unsigned long) buffer); + return 0; +} + +/* write_image_header() + * + * Description: Write the image header after write the image proper. + * Returns: Int. Zero on success or -1 on failure. + */ + +int write_image_header(void) +{ + int ret; + int total = pagedir1.pageset_size + pagedir2.pageset_size+2; + char *header_buffer = NULL; + + /* Now prepare to write the header */ + if ((ret = suspend_active_writer->write_header_init())) { + abort_suspend("Active writer's write_header_init" + " function failed."); + goto write_image_header_abort; + } + + /* Get a buffer */ + header_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + if (!header_buffer) { + abort_suspend("Out of memory when trying to get page " + "for header!"); + goto write_image_header_abort; + } + + /* Write suspend header */ + fill_suspend_header((struct suspend_header *) header_buffer); + suspend_active_writer->rw_header_chunk(WRITE, NULL, + header_buffer, sizeof(struct suspend_header)); + + free_page((unsigned long) header_buffer); + + /* Write module configurations */ + if ((ret = write_module_configs())) { + abort_suspend("Failed to write module configs."); + goto write_image_header_abort; + } + + save_dyn_pageflags(pageset1_map); + + /* Flush data and let writer cleanup */ + if (suspend_active_writer->write_header_cleanup()) { + abort_suspend("Failed to cleanup writing header."); + goto write_image_header_abort_no_cleanup; + } + + if (test_result_state(SUSPEND_ABORTED)) + goto write_image_header_abort_no_cleanup; + + suspend_message(SUSPEND_IO, SUSPEND_VERBOSE, 1, "|\n"); + suspend_update_status(total, total, NULL); + + return 0; + +write_image_header_abort: + suspend_active_writer->write_header_cleanup(); +write_image_header_abort_no_cleanup: + return -1; +} + +/* sanity_check() + * + * Description: Perform a few checks, seeking to ensure that the kernel being + * booted matches the one suspended. They need to match so we can + * be _sure_ things will work. It is not absolutely impossible for + * resuming from a different kernel to work, just not assured. + * Arguments: Struct suspend_header. The header which was saved at suspend + * time. + */ +static char *sanity_check(struct suspend_header *sh) +{ + if (sh->version_code != LINUX_VERSION_CODE) + return "Incorrect kernel version."; + + if (sh->num_physpages != num_physpages) + return "Incorrect memory size."; + + if (strncmp(sh->uts.sysname, init_utsname()->sysname, 65)) + return "Incorrect system type."; + + if (strncmp(sh->uts.release, init_utsname()->release, 65)) + return "Incorrect release."; + + if (strncmp(sh->uts.version, init_utsname()->version, 65)) + return "Right kernel version but wrong build number."; + + if (strncmp(sh->uts.machine, init_utsname()->machine, 65)) + return "Incorrect machine type."; + + if (sh->page_size != PAGE_SIZE) + return "Incorrect PAGE_SIZE."; + + if (!test_action_state(SUSPEND_IGNORE_ROOTFS)) { + const struct super_block *sb; + list_for_each_entry(sb, &super_blocks, s_list) { + if ((!(sb->s_flags & MS_RDONLY)) && + (sb->s_type->fs_flags & FS_REQUIRES_DEV)) + return "Device backed fs has been mounted " + "rw prior to resume."; + } + } + + return 0; +} + +/* __read_pageset1 + * + * Description: Test for the existence of an image and attempt to load it. + * Returns: Int. Zero if image found and pageset1 successfully loaded. + * Error if no image found or loaded. + */ +static int __read_pageset1(void) +{ + int i, result = 0; + char *header_buffer = (char *) get_zeroed_page(GFP_ATOMIC), + *sanity_error = NULL; + struct suspend_header *suspend_header; + + if (!header_buffer) { + printk("Unable to allocate a page for reading the signature.\n"); + return -ENOMEM; + } + + /* Check for an image */ + if (!(result = suspend_active_writer->image_exists())) { + result = -ENODATA; + noresume_reset_modules(); + printk(name_suspend "No image found.\n"); + goto out; + } + + /* Check for noresume command line option */ + if (test_suspend_state(SUSPEND_NORESUME_SPECIFIED)) { + suspend_active_writer->invalidate_image(); + result = -EINVAL; + noresume_reset_modules(); + printk(name_suspend "Noresume: Invalidated image.\n"); + goto out; + } + + /* Check whether we've resumed before */ + if (test_suspend_state(SUSPEND_RESUMED_BEFORE)) { + int resumed_before_default = 0; + if (test_suspend_state(SUSPEND_RETRY_RESUME)) + resumed_before_default = SUSPEND_CONTINUE_REQ; + suspend_early_boot_message(1, resumed_before_default, NULL); + clear_suspend_state(SUSPEND_RETRY_RESUME); + if (!(test_suspend_state(SUSPEND_CONTINUE_REQ))) { + suspend_active_writer->invalidate_image(); + result = -EINVAL; + noresume_reset_modules(); + printk(name_suspend "Tried to resume before: Invalidated image.\n"); + goto out; + } + } + + clear_suspend_state(SUSPEND_CONTINUE_REQ); + + /* + * Prepare the active writer for reading the image header. The + * activate writer might read its own configuration. + * + * NB: This call may never return because there might be a signature + * for a different image such that we warn the user and they choose + * to reboot. (If the device ids look erroneous (2.4 vs 2.6) or the + * location of the image might be unavailable if it was stored on a + * network connection. + */ + + if ((result = suspend_active_writer->read_header_init())) { + noresume_reset_modules(); + printk(name_suspend "Failed to initialise, reading the image header.\n"); + goto out; + } + + /* Read suspend header */ + if ((result = suspend_active_writer->rw_header_chunk(READ, NULL, + header_buffer, sizeof(struct suspend_header))) < 0) { + noresume_reset_modules(); + printk(name_suspend "Failed to read the image signature.\n"); + goto out; + } + + suspend_header = (struct suspend_header *) header_buffer; + + /* + * NB: This call may also result in a reboot rather than returning. + */ + + if ((sanity_error = sanity_check(suspend_header)) && + suspend_early_boot_message(1, SUSPEND_CONTINUE_REQ, sanity_error)) { + suspend_active_writer->invalidate_image(); + result = -EINVAL; + noresume_reset_modules(); + printk(name_suspend "Sanity check failed.\n"); + goto out; + } + + /* + * We have an image and it looks like it will load okay. + */ + + /* Get metadata from header. Don't override commandline parameters. + * + * We don't need to save the image size limit because it's not used + * during resume and will be restored with the image anyway. + */ + + suspend_orig_mem_free = suspend_header->orig_mem_free; + memcpy((char *) &pagedir1, + (char *) &suspend_header->pagedir, sizeof(pagedir1)); + suspend_result = suspend_header->param0; + suspend_action = suspend_header->param1; + suspend_debug_state = suspend_header->param2; + console_loglevel = suspend_header->param3; + clear_suspend_state(SUSPEND_IGNORE_LOGLEVEL); + pagedir2.pageset_size = suspend_header->pageset_2_size; + for (i = 0; i < 4; i++) + suspend_io_time[i/2][i%2] = + suspend_header->io_time[i/2][i%2]; + + /* Read module configurations */ + if ((result = read_module_configs())) { + noresume_reset_modules(); + pagedir1.pageset_size = + pagedir2.pageset_size = 0; + printk(name_suspend "Failed to read Suspend module configurations.\n"); + goto out; + } + + suspend_prepare_console(); + + suspend_cond_pause(1, "About to read original pageset1 locations."); + + /* + * Read original pageset1 locations. These are the addresses we can't + * use for the data to be restored. + */ + + allocate_dyn_pageflags(&pageset1_map); + load_dyn_pageflags(pageset1_map); + + set_suspend_state(SUSPEND_NOW_RESUMING); + + /* Relocate it so that it's not overwritten while we're using it to + * copy the original contents back */ + relocate_dyn_pageflags(&pageset1_map); + + allocate_dyn_pageflags(&pageset1_copy_map); + relocate_dyn_pageflags(&pageset1_copy_map); + + /* Clean up after reading the header */ + if ((result = suspend_active_writer->read_header_cleanup())) { + noresume_reset_modules(); + printk(name_suspend "Failed to cleanup after reading the image header.\n"); + goto out_reset_console; + } + + suspend_cond_pause(1, "About to read pagedir."); + + /* + * Get the addresses of pages into which we will load the kernel to + * be copied back + */ + if (suspend_get_pageset1_load_addresses()) { + result = -ENOMEM; + noresume_reset_modules(); + printk(name_suspend "Failed to get load addresses for pageset1.\n"); + goto out_reset_console; + } + + /* Read the original kernel back */ + suspend_cond_pause(1, "About to read pageset 1."); + + if (read_pageset(&pagedir1, 1, 0)) { + suspend_prepare_status(CLEAR_BAR, "Failed to read pageset 1."); + result = -EPERM; + noresume_reset_modules(); + printk(name_suspend "Failed to get load pageset1.\n"); + goto out_reset_console; + } + + suspend_cond_pause(1, "About to restore original kernel."); + result = 0; + + if (!test_action_state(SUSPEND_KEEP_IMAGE) && + suspend_active_writer->mark_resume_attempted) + suspend_active_writer->mark_resume_attempted(); + +out: + free_page((unsigned long) header_buffer); + return result; + +out_reset_console: + free_dyn_pageflags(&pageset1_map); + free_dyn_pageflags(&pageset1_copy_map); + suspend_cleanup_console(); + goto out; +} + +/* read_pageset1() + * + * Description: Attempt to read the header and pageset1 of a suspend image. + * Handle the outcome, complaining where appropriate. + */ + +int read_pageset1(void) +{ + int error; + + error = __read_pageset1(); + + switch (error) { + case 0: + case -ENODATA: + case -EINVAL: /* non fatal error */ + return error; + case -EIO: + printk(KERN_CRIT name_suspend "I/O error\n"); + break; + case -ENOENT: + printk(KERN_CRIT name_suspend "No such file or directory\n"); + break; + case -EPERM: + printk(KERN_CRIT name_suspend "Sanity check error\n"); + break; + default: + printk(KERN_CRIT name_suspend "Error %d resuming\n", + error); + break; + } + abort_suspend("Error %d in read_pageset1",error); + return error; +} + +/* + * get_have_image_data() + */ +char *get_have_image_data(void) +{ + char *output_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + struct suspend_header *suspend_header; + + if (!output_buffer) { + printk("Output buffer null.\n"); + return NULL; + } + + /* Check for an image */ + if (!suspend_active_writer->image_exists() || + suspend_active_writer->read_header_init() || + suspend_active_writer->rw_header_chunk(READ, NULL, + output_buffer, sizeof(struct suspend_header)) != + sizeof(struct suspend_header)) { + sprintf(output_buffer, "0\n"); + goto out; + } + + suspend_header = (struct suspend_header *) output_buffer; + + sprintf(output_buffer, "1\n%s\n%s\n", + suspend_header->uts.machine, + suspend_header->uts.version); + + /* Check whether we've resumed before */ + if (test_suspend_state(SUSPEND_RESUMED_BEFORE)) + strcat(output_buffer, "Resumed before.\n"); + +out: + noresume_reset_modules(); + return output_buffer; +} + +/* read_pageset2() + * + * Description: Read in part or all of pageset2 of an image, depending upon + * whether we are suspending and have only overwritten a portion + * with pageset1 pages, or are resuming and need to read them + * all. + * Arguments: Int. Boolean. Read only pages which would have been + * overwritten by pageset1? + * Returns: Int. Zero if no error, otherwise the error value. + */ +int read_pageset2(int overwrittenpagesonly) +{ + int result = 0; + + if (!pagedir2.pageset_size) + return 0; + + result = read_pageset(&pagedir2, 2, overwrittenpagesonly); + + suspend_update_status(100, 100, NULL); + suspend_cond_pause(1, "Pagedir 2 read."); + + return result; +} diff -urN oldtree/kernel/power/io.h newtree/kernel/power/io.h --- oldtree/kernel/power/io.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/io.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,46 @@ +/* + * kernel/power/io.h + * + * Copyright (C) 2005-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It contains high level IO routines for suspending. + * + */ + +#include +#include "pagedir.h" + +/* Non-module data saved in our image header */ +struct suspend_header { + u32 version_code; + unsigned long num_physpages; + unsigned long orig_mem_free; + struct new_utsname uts; + int num_cpus; + int page_size; + int pageset_2_size; + int param0; + int param1; + int param2; + int param3; + int progress0; + int progress1; + int progress2; + int progress3; + int io_time[2][2]; + struct pagedir pagedir; + dev_t root_fs; +}; + +extern int write_pageset(struct pagedir *pagedir, int whichtowrite); +extern int write_image_header(void); +extern int read_pageset1(void); +extern int read_pageset2(int overwrittenpagesonly); + +extern int suspend_attempt_to_parse_resume_device(void); +extern void attempt_to_parse_resume_device2(void); +extern dev_t name_to_dev_t(char *line); +extern __nosavedata unsigned long bytes_in, bytes_out; +extern char *get_have_image_data(void); diff -urN oldtree/kernel/power/main.c newtree/kernel/power/main.c --- oldtree/kernel/power/main.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/main.c 2006-10-06 17:28:20.000000000 -0400 @@ -9,6 +9,7 @@ */ #include +#include #include #include #include @@ -95,7 +96,7 @@ if (pm_ops->finish) pm_ops->finish(state); Thaw: - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); Enable_cpu: enable_nonboot_cpus(); pm_restore_console(); @@ -134,7 +135,7 @@ { device_resume(); resume_console(); - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); enable_nonboot_cpus(); if (pm_ops && pm_ops->finish) pm_ops->finish(state); @@ -147,7 +148,7 @@ static const char * const pm_states[PM_SUSPEND_MAX] = { [PM_SUSPEND_STANDBY] = "standby", [PM_SUSPEND_MEM] = "mem", -#ifdef CONFIG_SOFTWARE_SUSPEND +#if defined(CONFIG_SOFTWARE_SUSPEND) || defined(CONFIG_SUSPEND2) [PM_SUSPEND_DISK] = "disk", #endif }; diff -urN oldtree/kernel/power/modules.c newtree/kernel/power/modules.c --- oldtree/kernel/power/modules.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/modules.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,333 @@ +/* + * kernel/power/modules.c + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + */ + +#include +#include +#include "suspend.h" +#include "modules.h" +#include "ui.h" + +struct list_head suspend_filters, suspend_writers, suspend_modules; +struct suspend_module_ops *suspend_active_writer; +static int suspend_num_filters; +int suspend_num_writers, suspend_num_modules; + +/* + * suspend_header_storage_for_modules + * + * Returns the amount of space needed to store configuration + * data needed by the modules prior to copying back the original + * kernel. We can exclude data for pageset2 because it will be + * available anyway once the kernel is copied back. + */ +unsigned long suspend_header_storage_for_modules(void) +{ + struct suspend_module_ops *this_module; + unsigned long bytes = 0; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled || + (this_module->type == WRITER_MODULE && + suspend_active_writer != this_module)) + continue; + if (this_module->storage_needed) { + int this = this_module->storage_needed() + + sizeof(struct suspend_module_header) + + sizeof(int); + this_module->header_requested = this; + bytes += this; + } + } + + /* One more for the empty terminator */ + return bytes + sizeof(struct suspend_module_header); +} + +/* + * suspend_memory_for_modules + * + * Returns the amount of memory requested by modules for + * doing their work during the cycle. + */ + +unsigned long suspend_memory_for_modules(void) +{ + unsigned long bytes = 0; + struct suspend_module_ops *this_module; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled) + continue; + if (this_module->memory_needed) + bytes += this_module->memory_needed(); + } + + return ((bytes + PAGE_SIZE - 1) >> PAGE_SHIFT); +} + +/* suspend_find_module_given_name + * Functionality : Return a module (if found), given a pointer + * to its name + */ + +struct suspend_module_ops *suspend_find_module_given_name(char *name) +{ + struct suspend_module_ops *this_module, *found_module = NULL; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!strcmp(name, this_module->name)) { + found_module = this_module; + break; + } + } + + return found_module; +} + +/* + * suspend_print_module_debug_info + * Functionality : Get debugging info from modules into a buffer. + */ +int suspend_print_module_debug_info(char *buffer, int buffer_size) +{ + struct suspend_module_ops *this_module; + int len = 0; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled) + continue; + if (this_module->print_debug_info) { + int result; + result = this_module->print_debug_info(buffer + len, + buffer_size - len); + len += result; + } + } + + /* Ensure null terminated */ + buffer[buffer_size] = 0; + + return len; +} + +/* + * suspend_register_module + * + * Register a module. + */ +int suspend_register_module(struct suspend_module_ops *module) +{ + module->enabled = 1; + + if (suspend_find_module_given_name(module->name)) { + printk(name_suspend "Trying to load module %s," + " which is already registered.\n", + module->name); + return -EBUSY; + } + + switch (module->type) { + case FILTER_MODULE: + list_add_tail(&module->type_list, + &suspend_filters); + suspend_num_filters++; + break; + + case WRITER_MODULE: + list_add_tail(&module->type_list, + &suspend_writers); + suspend_num_writers++; + break; + + case MISC_MODULE: + break; + + default: + printk("Hmmm. Module '%s' has an invalid type." + " It has been ignored.\n", module->name); + return -EINVAL; + } + list_add_tail(&module->module_list, &suspend_modules); + suspend_num_modules++; + + return 0; +} + +/* + * suspend_unregister_module + * + * Remove a module. + */ +void suspend_unregister_module(struct suspend_module_ops *module) +{ + switch (module->type) { + case FILTER_MODULE: + list_del(&module->type_list); + suspend_num_filters--; + break; + + case WRITER_MODULE: + list_del(&module->type_list); + suspend_num_writers--; + if (suspend_active_writer == module) { + suspend_active_writer = NULL; + clear_suspend_state(SUSPEND_CAN_RESUME); + clear_suspend_state(SUSPEND_CAN_SUSPEND); + } + break; + + case MISC_MODULE: + break; + + default: + printk("Hmmm. Module '%s' has an invalid type." + " It has been ignored.\n", module->name); + return; + } + list_del(&module->module_list); + suspend_num_modules--; +} + +/* + * suspend_move_module_tail + * + * Rearrange modules when reloading the config. + */ +void suspend_move_module_tail(struct suspend_module_ops *module) +{ + switch (module->type) { + case FILTER_MODULE: + if (suspend_num_filters > 1) + list_move_tail(&module->type_list, + &suspend_filters); + break; + + case WRITER_MODULE: + if (suspend_num_writers > 1) + list_move_tail(&module->type_list, + &suspend_writers); + break; + + case MISC_MODULE: + break; + default: + printk("Hmmm. Module '%s' has an invalid type." + " It has been ignored.\n", module->name); + return; + } + if ((suspend_num_filters + suspend_num_writers) > 1) + list_move_tail(&module->module_list, &suspend_modules); +} + +/* + * suspend_initialise_modules + * + * Get ready to do some work! + */ +int suspend_initialise_modules(int starting_cycle) +{ + struct suspend_module_ops *this_module; + int result; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + this_module->header_requested = 0; + this_module->header_used = 0; + if (!this_module->enabled) + continue; + if (this_module->initialise) { + suspend_message(SUSPEND_MEMORY, SUSPEND_MEDIUM, 1, + "Initialising module %s.\n", + this_module->name); + if ((result = this_module->initialise(starting_cycle))) { + printk("%s didn't initialise okay.\n", + this_module->name); + return result; + } + } + } + + return 0; +} + +/* + * suspend_cleanup_modules + * + * Tell modules the work is done. + */ +void suspend_cleanup_modules(int finishing_cycle) +{ + struct suspend_module_ops *this_module; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!this_module->enabled) + continue; + if (this_module->cleanup) { + suspend_message(SUSPEND_MEMORY, SUSPEND_MEDIUM, 1, + "Cleaning up module %s.\n", + this_module->name); + this_module->cleanup(finishing_cycle); + } + } +} + +/* + * suspend_get_next_filter + * + * Get the next filter in the pipeline. + */ +struct suspend_module_ops *suspend_get_next_filter(struct suspend_module_ops *filter_sought) +{ + struct suspend_module_ops *last_filter = NULL, *this_filter = NULL; + + list_for_each_entry(this_filter, &suspend_filters, type_list) { + if (!this_filter->enabled) + continue; + if ((last_filter == filter_sought) || (!filter_sought)) + return this_filter; + last_filter = this_filter; + } + + return suspend_active_writer; +} + +/* suspend_get_modules + * + * Take a reference to modules so they can't go away under us. + */ + +int suspend_get_modules(void) +{ + struct suspend_module_ops *this_module; + + list_for_each_entry(this_module, &suspend_modules, module_list) { + if (!try_module_get(this_module->module)) { + /* Failed! Reverse gets and return error */ + struct suspend_module_ops *this_module2; + list_for_each_entry(this_module2, &suspend_modules, module_list) { + if (this_module == this_module2) + return -EINVAL; + module_put(this_module2->module); + } + } + } + + return 0; +} + +/* suspend_put_modules + * + * Release our references to modules we used. + */ + +void suspend_put_modules(void) +{ + struct suspend_module_ops *this_module; + + list_for_each_entry(this_module, &suspend_modules, module_list) + module_put(this_module->module); +} + diff -urN oldtree/kernel/power/modules.h newtree/kernel/power/modules.h --- oldtree/kernel/power/modules.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/modules.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,160 @@ +/* + * kernel/power/modules.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It contains declarations for modules. Modules are additions to + * suspend2 that provide facilities such as image compression or + * encryption, backends for storage of the image and user interfaces. + * + */ + +#ifndef SUSPEND_MODULES_H +#define SUSPEND_MODULES_H + +/* This is the maximum size we store in the image header for a module name */ +#define SUSPEND_MAX_MODULE_NAME_LENGTH 30 + +/* Per-module metadata */ +struct suspend_module_header { + char name[SUSPEND_MAX_MODULE_NAME_LENGTH]; + int enabled; + int type; + int index; + int data_length; + unsigned long signature; +}; + +extern int suspend_num_modules, suspend_num_writers; + +enum { + FILTER_MODULE, + WRITER_MODULE, + MISC_MODULE /* Block writer, eg. */ +}; + +enum { + SUSPEND_ASYNC, + SUSPEND_SYNC +}; + +struct suspend_module_ops { + /* Functions common to all modules */ + int type; + char *name; + struct module *module; + int enabled; + struct list_head module_list; + + /* List of filters or writers */ + struct list_head list, type_list; + + /* + * Requirements for memory and storage in + * the image header.. + */ + unsigned long (*memory_needed) (void); + unsigned long (*storage_needed) (void); + + unsigned long header_requested, header_used; + + /* + * Debug info + */ + int (*print_debug_info) (char *buffer, int size); + int (*save_config_info) (char *buffer); + void (*load_config_info) (char *buffer, int len); + + /* + * Initialise & cleanup - general routines called + * at the start and end of a cycle. + */ + int (*initialise) (int starting_cycle); + void (*cleanup) (int finishing_cycle); + + /* + * Calls for allocating storage (writers only). + * + * Header space is allocated separately. Note that allocation + * of space for the header might result in allocated space + * being stolen from the main pool if there is no unallocated + * space. We have to be able to allocate enough space for + * the header. We can eat memory to ensure there is enough + * for the main pool. + */ + + int (*storage_available) (void); + int (*allocate_header_space) (int space_requested); + int (*allocate_storage) (int space_requested); + int (*storage_allocated) (void); + int (*release_storage) (void); + + /* + * Routines used in image I/O. + */ + int (*rw_init) (int rw, int stream_number); + int (*rw_cleanup) (int rw); + int (*write_chunk) (struct page *buffer_page); + int (*read_chunk) (struct page *buffer_page, int sync); + + /* Reset module if image exists but reading aborted */ + void (*noresume_reset) (void); + + /* Read and write the metadata */ + int (*write_header_init) (void); + int (*write_header_cleanup) (void); + + int (*read_header_init) (void); + int (*read_header_cleanup) (void); + + int (*rw_header_chunk) (int rw, struct suspend_module_ops *owner, + char *buffer_start, int buffer_size); + + /* Attempt to parse an image location */ + int (*parse_sig_location) (char *buffer, int only_writer); + + /* Determine whether image exists that we can restore */ + int (*image_exists) (void); + + /* Mark the image as having tried to resume */ + void (*mark_resume_attempted) (void); + + /* Destroy image if one exists */ + int (*invalidate_image) (void); + +}; + +extern struct suspend_module_ops *suspend_active_writer; +extern struct list_head suspend_filters, suspend_writers, suspend_modules; + +extern void suspend_prepare_console_modules(void); +extern void suspend_cleanup_console_modules(void); + +extern struct suspend_module_ops *suspend_find_module_given_name(char *name), + *suspend_get_next_filter(struct suspend_module_ops *); + +extern int suspend_register_module(struct suspend_module_ops *module); +extern void suspend_move_module_tail(struct suspend_module_ops *module); + +extern unsigned long suspend_header_storage_for_modules(void); +extern unsigned long suspend_memory_for_modules(void); + +extern int suspend_print_module_debug_info(char *buffer, int buffer_size); +extern int suspend_register_module(struct suspend_module_ops *module); +extern void suspend_unregister_module(struct suspend_module_ops *module); + +extern int suspend_initialise_modules(int starting_cycle); +extern void suspend_cleanup_modules(int finishing_cycle); + +int suspend_get_modules(void); +void suspend_put_modules(void); + +static inline void suspend_initialise_module_lists(void) { + INIT_LIST_HEAD(&suspend_filters); + INIT_LIST_HEAD(&suspend_writers); + INIT_LIST_HEAD(&suspend_modules); +} + +#endif diff -urN oldtree/kernel/power/netlink.c newtree/kernel/power/netlink.c --- oldtree/kernel/power/netlink.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/netlink.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,375 @@ +/* + * kernel/power/netlink.c + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Functions for communicating with a userspace helper via netlink. + */ + + +#include +#include "netlink.h" + +#ifdef CONFIG_NET +struct user_helper_data *uhd_list = NULL; + +/* + * Refill our pool of SKBs for use in emergencies (eg, when eating memory and none + * can be allocated). + */ +static void suspend_fill_skb_pool(struct user_helper_data *uhd) +{ + while (uhd->pool_level < uhd->pool_limit) { + struct sk_buff *new_skb = + alloc_skb(NLMSG_SPACE(uhd->skb_size), GFP_ATOMIC); + + if (!new_skb) + break; + + new_skb->next = uhd->emerg_skbs; + uhd->emerg_skbs = new_skb; + uhd->pool_level++; + } +} + +/* + * Try to allocate a single skb. If we can't get one, try to use one from + * our pool. + */ +static struct sk_buff *suspend_get_skb(struct user_helper_data *uhd) +{ + struct sk_buff *skb = + alloc_skb(NLMSG_SPACE(uhd->skb_size), GFP_ATOMIC); + + if (skb) + return skb; + + skb = uhd->emerg_skbs; + if (skb) { + uhd->pool_level--; + uhd->emerg_skbs = skb->next; + skb->next = NULL; + } + + return skb; +} + +static void put_skb(struct user_helper_data *uhd, struct sk_buff *skb) +{ + if (uhd->pool_level < uhd->pool_limit) { + skb->next = uhd->emerg_skbs; + uhd->emerg_skbs = skb; + } else + kfree_skb(skb); +} + + +static void suspend_notify_userspace(void* data) +{ + struct task_struct *t; + struct user_helper_data *uhd = (struct user_helper_data *) data; + + BUG_ON(!uhd); + + read_lock(&tasklist_lock); + if ((t = find_task_by_pid(uhd->pid))) + wake_up_process(t); + read_unlock(&tasklist_lock); +} + +DECLARE_WORK(suspend_notify_userspace_work, suspend_notify_userspace, NULL); + +void suspend_send_netlink_message(struct user_helper_data *uhd, + int type, void* params, size_t len) +{ + struct sk_buff *skb; + struct nlmsghdr *nlh; + void *dest; + + if (uhd->pid == -1) + return; + + skb = suspend_get_skb(uhd); + if (!skb) { + printk("suspend_netlink: Can't allocate skb!\n"); + return; + } + + /* NLMSG_PUT contains a hidden goto nlmsg_failure */ + nlh = NLMSG_PUT(skb, 0, uhd->sock_seq, type, len); + uhd->sock_seq++; + + dest = NLMSG_DATA(nlh); + if (params && len > 0) + memcpy(dest, params, len); + + netlink_unicast(uhd->nl, skb, uhd->pid, 0); + + /* We may be in an interrupt context so defer waking up userspace */ + suspend_notify_userspace_work.data = uhd; + schedule_work(&suspend_notify_userspace_work); + + return; + +nlmsg_failure: + if (skb) + put_skb(uhd, skb); +} + +#ifdef CONFIG_PM_DEBUG +static int is_debugging = 1; +#else +static int is_debugging = 0; +#endif + +static void send_whether_debugging(struct user_helper_data *uhd) +{ + suspend_send_netlink_message(uhd, NETLINK_MSG_IS_DEBUGGING, + &is_debugging, sizeof(int)); +} + +/* + * Set the PF_NOFREEZE flag on the given process to ensure it can run whilst we + * are suspending. + */ +static int nl_set_nofreeze(struct user_helper_data *uhd, int pid) +{ + struct task_struct *t; + + read_lock(&tasklist_lock); + if ((t = find_task_by_pid(pid)) == NULL) { + read_unlock(&tasklist_lock); + printk("Strange. Can't find the userspace task %d.\n", pid); + return -EINVAL; + } + + t->flags |= PF_NOFREEZE; + + read_unlock(&tasklist_lock); + uhd->pid = pid; + + suspend_send_netlink_message(uhd, NETLINK_MSG_NOFREEZE_ACK, NULL, 0); + + return 0; +} + +/* + * Called when the userspace process has informed us that it's ready to roll. + */ +static int nl_ready(struct user_helper_data *uhd, int version) +{ + if (version != uhd->interface_version) { + printk("%s userspace process using invalid interface version." + " Trying to continue without it.\n", + uhd->name); + if (uhd->not_ready) + uhd->not_ready(); + return 1; + } + + complete(&uhd->wait_for_process); + + return 0; +} + +static int suspend_nl_gen_rcv_msg(struct user_helper_data *uhd, + struct sk_buff *skb, struct nlmsghdr *nlh) +{ + int type; + int *data; + int err; + + /* Let the more specific handler go first. It returns + * 1 for valid messages that it doesn't know. */ + if ((err = uhd->rcv_msg(skb, nlh)) != 1) + return err; + + type = nlh->nlmsg_type; + + /* Only allow one task to receive NOFREEZE privileges */ + if (type == NETLINK_MSG_NOFREEZE_ME && uhd->pid != -1) { + printk("Received extra nofreeze me requests.\n"); + return -EBUSY; + } + + data = (int*)NLMSG_DATA(nlh); + + switch (type) { + case NETLINK_MSG_NOFREEZE_ME: + if ((err = nl_set_nofreeze(uhd, nlh->nlmsg_pid)) != 0) + return err; + break; + case NETLINK_MSG_GET_DEBUGGING: + send_whether_debugging(uhd); + break; + case NETLINK_MSG_READY: + if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(int))) { + printk("Invalid ready mesage.\n"); + return -EINVAL; + } + if ((err = nl_ready(uhd, *data)) != 0) + return err; + break; + } + + return 0; +} + +static void suspend_user_rcv_skb(struct user_helper_data *uhd, + struct sk_buff *skb) +{ + int err; + struct nlmsghdr *nlh; + + while (skb->len >= NLMSG_SPACE(0)) { + u32 rlen; + + nlh = (struct nlmsghdr *) skb->data; + if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) + return; + + rlen = NLMSG_ALIGN(nlh->nlmsg_len); + if (rlen > skb->len) + rlen = skb->len; + + if ((err = suspend_nl_gen_rcv_msg(uhd, skb, nlh)) != 0) + netlink_ack(skb, nlh, err); + else if (nlh->nlmsg_flags & NLM_F_ACK) + netlink_ack(skb, nlh, 0); + skb_pull(skb, rlen); + } +} + +static void suspend_netlink_input(struct sock *sk, int len) +{ + struct user_helper_data *uhd = uhd_list; + + while (uhd && uhd->netlink_id != sk->sk_protocol) + uhd= uhd->next; + + BUG_ON(!uhd); + + do { + struct sk_buff *skb; + while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + suspend_user_rcv_skb(uhd, skb); + put_skb(uhd, skb); + } + } while (uhd->nl && uhd->nl->sk_receive_queue.qlen); +} + +static int netlink_prepare(struct user_helper_data *uhd) +{ + uhd->next = uhd_list; + uhd_list = uhd; + + uhd->sock_seq = 0x42c0ffee; + uhd->nl = netlink_kernel_create(uhd->netlink_id, 0, + suspend_netlink_input, THIS_MODULE); + if (!uhd->nl) { + printk("Failed to allocate netlink socket for %s.\n", + uhd->name); + return -ENOMEM; + } + + suspend_fill_skb_pool(uhd); + + return 0; +} + +void suspend_netlink_close(struct user_helper_data *uhd) +{ + if (uhd->nl) { + sock_release(uhd->nl->sk_socket); + uhd->nl = NULL; + } + + while (uhd->emerg_skbs) { + struct sk_buff *next = uhd->emerg_skbs->next; + kfree_skb(uhd->emerg_skbs); + uhd->emerg_skbs = next; + } +} + +int suspend2_launch_userspace_program(char *command, int channel_no) +{ + int retval; + static char *envp[] = { + "HOME=/", + "TERM=linux", + "PATH=/sbin:/usr/sbin:/bin:/usr/bin", + NULL }; + static char *argv[] = { NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL }; + char *channel = kmalloc(6, GFP_KERNEL); + int arg = 0, size; + char test_read[255]; + char *orig_posn = command; + + if (!strlen(orig_posn)) + return 1; + + /* Up to 7 args supported */ + while (arg < 7) { + sscanf(orig_posn, "%s", test_read); + size = strlen(test_read); + if (!(size)) + break; + argv[arg] = kmalloc(size + 1, GFP_ATOMIC); + strcpy(argv[arg], test_read); + orig_posn += size + 1; + *test_read = 0; + arg++; + } + + if (channel_no) { + sprintf(channel, "-c%d", channel_no); + argv[arg] = channel; + } else + arg--; + + retval = call_usermodehelper(argv[0], argv, envp, 0); + + if (retval) + printk("Failed to launch userspace program '%s': Error %d\n", + command, retval); + + { + int i; + for (i = 0; i < arg; i++) + if (argv[i] && argv[i] != channel) + kfree(argv[i]); + } + + kfree(channel); + + return retval; +} + +int suspend_netlink_setup(struct user_helper_data *uhd) +{ + if (netlink_prepare(uhd) < 0) { + printk("Netlink prepare failed.\n"); + return 1; + } + + if (suspend2_launch_userspace_program(uhd->program, uhd->netlink_id) < 0) { + printk("Launch userspace program failed.\n"); + suspend_netlink_close(uhd); + return 1; + } + + /* Wait 2 seconds for the userspace process to make contact */ + wait_for_completion_timeout(&uhd->wait_for_process, 2*HZ); + + if (uhd->pid == -1) { + printk("%s: Failed to contact userspace process.\n", + uhd->name); + suspend_netlink_close(uhd); + return 1; + } + + return 0; +} +#endif diff -urN oldtree/kernel/power/netlink.h newtree/kernel/power/netlink.h --- oldtree/kernel/power/netlink.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/netlink.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,59 @@ +/* + * kernel/power/netlink.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Declarations for functions for communicating with a userspace helper + * via netlink. + */ + +#include +#include + +#define NETLINK_MSG_BASE 0x10 + +#define NETLINK_MSG_READY 0x10 +#define NETLINK_MSG_NOFREEZE_ME 0x16 +#define NETLINK_MSG_GET_DEBUGGING 0x19 +#define NETLINK_MSG_CLEANUP 0x24 +#define NETLINK_MSG_NOFREEZE_ACK 0x27 +#define NETLINK_MSG_IS_DEBUGGING 0x28 + +struct user_helper_data { + int (*rcv_msg) (struct sk_buff *skb, struct nlmsghdr *nlh); + void (* not_ready) (void); + struct sock *nl; + u32 sock_seq; + pid_t pid; + char *comm; + char program[256]; + int pool_level; + int pool_limit; + struct sk_buff *emerg_skbs; + int skb_size; + int netlink_id; + char *name; + struct user_helper_data *next; + struct completion wait_for_process; + int interface_version; + int must_init; +}; + + +#ifdef CONFIG_NET +int suspend_netlink_setup(struct user_helper_data *uhd); +void suspend_netlink_close(struct user_helper_data *uhd); +void suspend_send_netlink_message(struct user_helper_data *uhd, + int type, void* params, size_t len); +#else +static inline int suspend_netlink_setup(struct user_helper_data *uhd) +{ + return 0; +} + +static inline void suspend_netlink_close(struct user_helper_data *uhd) { }; +static inline void suspend_send_netlink_message(struct user_helper_data *uhd, + int type, void* params, size_t len) { }; +#endif diff -urN oldtree/kernel/power/pagedir.c newtree/kernel/power/pagedir.c --- oldtree/kernel/power/pagedir.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/pagedir.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,358 @@ +/* + * kernel/power/pagedir.c + * + * Copyright (C) 1998-2001 Gabor Kuti + * Copyright (C) 1998,2001,2002 Pavel Machek + * Copyright (C) 2002-2003 Florent Chabaud + * Copyright (C) 2002-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Routines for handling pagesets. + * Note that pbes aren't actually stored as such. They're stored as + * bitmaps and extents. + */ + +#include +#include +#include +#include +#include +#include + +#include "pageflags.h" +#include "ui.h" +#include "pagedir.h" + +int extra_pagedir_pages_allocated; + +struct extras { + struct page *page; + int order; + struct extras *next; +} *extras_list; + +/* suspend_free_extra_pagedir_memory + * + * Description: Free previously allocated extra pagedir memory. + */ +void suspend_free_extra_pagedir_memory(void) +{ + /* Free allocated pages */ + while (extras_list) { + struct extras *this = extras_list; + int i; + + extras_list = this->next; + + for (i = 0; i < (1 << this->order); i++) + ClearPageNosave(this->page + i); + + __free_pages(this->page, this->order); + kfree(this); + } + + extra_pagedir_pages_allocated = 0; +} + +/* suspend_allocate_extra_pagedir_memory + * + * Description: Allocate memory for making the atomic copy of pagedir1 in the + * case where it is bigger than pagedir2. + * Arguments: struct pagedir *: The pagedir for which we should + * allocate memory. + * int: Size of pageset 1. + * int: Size of pageset 2. + * Result: int. Zero on success. One if unable to allocate enough memory. + */ +int suspend_allocate_extra_pagedir_memory(struct pagedir *p, int pageset_size, + int alloc_from) +{ + int num_to_alloc = pageset_size - alloc_from - extra_pagedir_pages_allocated; + int j, order, num_added = 0; + + if (num_to_alloc < 1) + num_to_alloc = 0; + + if (num_to_alloc) { + order = fls(num_to_alloc); + if (order >= MAX_ORDER) + order = MAX_ORDER - 1; + + while (num_added < num_to_alloc) { + struct page *newpage; + unsigned long virt; + struct extras *extras_entry; + + while ((1 << order) > (num_to_alloc - num_added)) + order--; + + virt = __get_free_pages(GFP_ATOMIC | __GFP_NOWARN, order); + while ((!virt) && (order > 0)) { + order--; + virt = __get_free_pages(GFP_ATOMIC | __GFP_NOWARN, order); + } + + if (!virt) { + p->pageset_size += num_added; + extra_pagedir_pages_allocated += num_added; + return 1; + } + + newpage = virt_to_page(virt); + + extras_entry = (struct extras *) kmalloc(sizeof(struct extras), GFP_ATOMIC); + + if (!extras_entry) { + __free_pages(newpage, order); + extra_pagedir_pages_allocated += num_added; + return 1; + } + + extras_entry->page = newpage; + extras_entry->order = order; + extras_entry->next = NULL; + + if (extras_list) + extras_entry->next = extras_list; + + extras_list = extras_entry; + + for (j = 0; j < (1 << order); j++) { + SetPageNosave(newpage + j); + SetPagePageset1Copy(newpage + j); + } + num_added+= (1 << order); + } + } + + extra_pagedir_pages_allocated += num_added; + return 0; +} + +int suspend_protect_pageset2(void) +{ +#ifdef CONFIG_DEBUG_PAGEALLOC + int pfn; + + BITMAP_FOR_EACH_SET(pageset2_map, pfn) { + struct page *page = pfn_to_page(pfn); + if (PageHighMem(page)) + continue; + + change_page_attr(page, 1, __pgprot(0)); + __flush_tlb_all(); + } +#endif + + return 0; +} + +void suspend_restore_pageset2_permissions(void) +{ +#ifdef CONFIG_DEBUG_PAGEALLOC + int pfn; + + BITMAP_FOR_EACH_SET(pageset2_map, pfn) { + struct page *page = pfn_to_page(pfn); + if (PageHighMem(page)) + continue; + + change_page_attr(page, 1, PAGE_KERNEL); + __flush_tlb_all(); + } +#endif +} + +/* + * suspend_mark_task_as_pageset1 + * Functionality : Marks all the pages belonging to a given process as + * pageset 1 pages. + * Called From : pagedir.c - mark_pages_for_pageset2 + * + */ +extern struct page *suspend2_follow_page(struct mm_struct *mm, unsigned long address); + +static void suspend_mark_task_as_pageset1(struct task_struct *t) +{ + struct vm_area_struct *vma; + struct mm_struct *mm; + + mm = t->active_mm; + + if (!mm || !mm->mmap) return; + + /* Don't try to take the sem when processes are frozen, + * drivers are suspended and irqs are disabled. We're + * not racing with anything anyway. */ + BUG_ON(in_atomic() && !irqs_disabled()); + + if (!irqs_disabled()) + down_read(&mm->mmap_sem); + + for (vma = mm->mmap; vma; vma = vma->vm_next) { + if (vma->vm_flags & VM_PFNMAP) + continue; + if (vma->vm_start) { + unsigned long posn; + for (posn = vma->vm_start; posn < vma->vm_end; + posn += PAGE_SIZE) { + struct page *page = + suspend2_follow_page(mm, posn); + if (page) { + ClearPagePageset2(page); + SetPagePageset1(page); + } + } + } + } + + BUG_ON(in_atomic() && !irqs_disabled()); + + if (!irqs_disabled()) + up_read(&mm->mmap_sem); +} + +/* mark_pages_for_pageset2 + * + * Description: Mark unshared pages in processes not needed for suspend as + * being able to be written out in a separate pagedir. + * HighMem pages are simply marked as pageset2. They won't be + * needed during suspend. + */ + +struct attention_list { + struct task_struct *task; + struct attention_list *next; +}; + +void suspend_mark_pages_for_pageset2(void) +{ + struct zone *zone; + struct task_struct *p; + struct attention_list *attention_list = NULL, *last = NULL; + unsigned long flags; + + BUG_ON(in_atomic() && !irqs_disabled()); + + if (test_action_state(SUSPEND_NO_PAGESET2)) + return; + + clear_dyn_pageflags(pageset2_map); + + for_each_zone(zone) { + spin_lock_irqsave(&zone->lru_lock, flags); + if (zone->nr_inactive) { + struct page *page; + list_for_each_entry(page, &zone->inactive_list, lru) + SetPagePageset2(page); + } + if (zone->nr_active) { + struct page *page; + list_for_each_entry(page, &zone->active_list, lru) + SetPagePageset2(page); + } + spin_unlock_irqrestore(&zone->lru_lock, flags); + } + + BUG_ON(in_atomic() && !irqs_disabled()); + + /* Now we find all userspace process (with task->mm) marked PF_NOFREEZE + * and move them into pageset1. + */ + read_lock(&tasklist_lock); + for_each_process(p) + if ((p->flags & PF_NOFREEZE) || p == current) { + struct attention_list *this = kmalloc(sizeof(struct attention_list), GFP_ATOMIC); + BUG_ON(!this); + this->task = p; + this->next = NULL; + if (attention_list) { + last->next = this; + last = this; + } else + attention_list = last = this; + } + read_unlock(&tasklist_lock); + + BUG_ON(in_atomic() && !irqs_disabled()); + + /* Because the tasks in attention_list are ones related to suspending, + * we know that they won't go away under us. + */ + + while (attention_list) { + suspend_mark_task_as_pageset1(attention_list->task); + last = attention_list; + attention_list = attention_list->next; + kfree(last); + } + + BUG_ON(in_atomic() && !irqs_disabled()); + +} + +/* suspend_get_nonconflicting_page + * + * Description: Gets order zero pages that won't be overwritten + * while copying the original pages. + */ + +unsigned long suspend_get_nonconflicting_page(void) +{ + struct page *page; + + do { + page = alloc_pages(GFP_ATOMIC | __GFP_NOWARN | __GFP_ZERO, 0); + BUG_ON(!page); + } while(PagePageset1(page)); + + return (unsigned long) page_address(page); +} + +/* relocate_page_if_required + * + * Description: Given the address of a pointer to a page, we check if the page + * needs relocating and do so if needs be, adjusting the pointer + * too. + */ + +void suspend_relocate_if_required(unsigned long *current_value, unsigned int size) +{ + if (PagePageset1(virt_to_page(*current_value))) { + unsigned long new_page = suspend_get_nonconflicting_page(); + memcpy((char *) new_page, (char *) *current_value, size); + if (PageSlab(virt_to_page(*current_value))) + kfree((void *) *current_value); + else + free_page((unsigned long) *current_value); + *current_value = new_page; + } +} + +/* get_pageset1_load_addresses + * + * Description: We check here that pagedir & pages it points to won't collide + * with pages where we're going to restore from the loaded pages + * later. + * Returns: Zero on success, one if couldn't find enough pages (shouldn't + * happen). + */ + +int suspend_get_pageset1_load_addresses(void) +{ + int i, result = 0; + void *this; + + for(i=0; i < pagedir1.pageset_size; i++) { + this = (void *) suspend_get_nonconflicting_page(); + if (!this) { + abort_suspend("Error: Ran out of memory seeking locations for reloading data."); + result = 1; + break; + } + SetPagePageset1Copy(virt_to_page(this)); + } + + return result; +} diff -urN oldtree/kernel/power/pagedir.h newtree/kernel/power/pagedir.h --- oldtree/kernel/power/pagedir.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/pagedir.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,40 @@ +/* + * kernel/power/pagedir.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Declarations for routines for handling pagesets. + */ + +/* Pagedir + * + * Contains the metadata for a set of pages saved in the image. + */ + +struct pagedir { + long pageset_size; + long lastpageset_size; +}; + +extern struct pagedir pagedir1, pagedir2; + +extern void suspend_copy_pageset1(void); + +extern void suspend_free_extra_pagedir_memory(void); + +extern int suspend_allocate_extra_pagedir_memory(struct pagedir *p, int pageset_size, int alloc_from); + +//extern void suspend_mark_task_as_pageset1 (struct task_struct *t); +extern void suspend_mark_pages_for_pageset2(void); + +extern void suspend_relocate_if_required(unsigned long *current_value, unsigned int size); +extern int suspend_get_pageset1_load_addresses(void); + +extern int extra_pagedir_pages_allocated; + +extern unsigned long suspend_get_nonconflicting_page(void); + +int suspend_protect_pageset2(void); +void suspend_restore_pageset2_permissions(void); diff -urN oldtree/kernel/power/pageflags.c newtree/kernel/power/pageflags.c --- oldtree/kernel/power/pageflags.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/pageflags.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,169 @@ +/* + * kernel/power/pageflags.c + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Routines for serialising and relocating pageflags in which we + * store our image metadata. + * + */ + +#include +#include +#include +#include +#include +#include +#include "pageflags.h" +#include "modules.h" +#include "pagedir.h" + +/* Maps used in copying the image back are in builtin.c */ +dyn_pageflags_t pageset1_map; +dyn_pageflags_t pageset1_copy_map; +dyn_pageflags_t pageset2_map; +dyn_pageflags_t pageset2_rw_map; +dyn_pageflags_t in_use_map; + +static int pages_for_zone(struct zone *zone) +{ + return (zone->spanned_pages + (PAGE_SIZE << 3) - 1) / + (PAGE_SIZE << 3); +} + +int suspend_pageflags_space_needed(void) +{ + int total = 0; + struct zone *zone; + + for_each_zone(zone) + if (populated_zone(zone)) + total += sizeof(int) * 2 + pages_for_zone(zone) * PAGE_SIZE; + + total += sizeof(int); + + return total; +} + +/* save_dyn_pageflags + * + * Description: Save a set of pageflags. + * Arguments: dyn_pageflags_t *: Pointer to the bitmap being saved. + */ + +void save_dyn_pageflags(dyn_pageflags_t pagemap) +{ + int i, zone_num, size; + struct zone *zone; + + if (!*pagemap) + return; + + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + size = pages_for_zone(zone); + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + suspend_active_writer->rw_header_chunk(WRITE, NULL, + (char *) &zone_num, sizeof(int)); + suspend_active_writer->rw_header_chunk(WRITE, NULL, + (char *) &size, sizeof(int)); + + for (i = 0; i < size; i++) + suspend_active_writer->rw_header_chunk(WRITE, NULL, + (char *) pagemap[zone_num][i], PAGE_SIZE); + } + zone_num = -1; + suspend_active_writer->rw_header_chunk(WRITE, NULL, + (char *) &zone_num, sizeof(int)); +} + +/* load_dyn_pageflags + * + * Description: Load a set of pageflags. + * Arguments: dyn_pageflags_t *: Pointer to the bitmap being loaded. + * (It must be allocated before calling this routine). + */ + +void load_dyn_pageflags(dyn_pageflags_t pagemap) +{ + int i, zone_num, zone_check = 0, size; + struct zone *zone; + + if (!pagemap) + return; + + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) &zone_check, sizeof(int)); + if (zone_check != zone_num) { + printk("Zone check (%d) != zone_num (%d).\n", + zone_check, zone_num); + BUG(); + } + suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) &size, sizeof(int)); + + for (i = 0; i < size; i++) + suspend_active_writer->rw_header_chunk(READ, NULL, + (char *) pagemap[zone_num][i], + PAGE_SIZE); + } + suspend_active_writer->rw_header_chunk(READ, NULL, (char *) &zone_check, + sizeof(int)); + if (zone_check != -1) { + printk("Didn't read end of dyn pageflag data marker.(%x)\n", + zone_check); + BUG(); + } +} + +/* relocate_dyn_pageflags + * + * Description: Relocate a set of pageflags to ensure they don't collide with + * pageset 1 data which will get overwritten on copyback. + * Arguments: dyn_pageflags_t *: Pointer to the bitmap being relocated. + */ + +void relocate_dyn_pageflags(dyn_pageflags_t *pagemap) +{ + int i, zone_num; + struct zone *zone; + + if (!*pagemap) + return; + + suspend_relocate_if_required((unsigned long *) pagemap, + (unsigned int) (sizeof(void *) * (1 << ZONEID_SHIFT))); + BUG_ON(PagePageset1(virt_to_page(*pagemap))); + + for_each_zone(zone) { + int pages = pages_for_zone(zone); + + if (!populated_zone(zone)) + continue; + + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + suspend_relocate_if_required((void *) &((*pagemap)[zone_num]), + sizeof(void *) * pages); + BUG_ON(PagePageset1(virt_to_page((*pagemap)[zone_num]))); + + for (i = 0; i < pages; i++) { + suspend_relocate_if_required( + (void *) &((*pagemap)[zone_num][i]), + PAGE_SIZE); + BUG_ON(PagePageset1(virt_to_page( + (*pagemap)[zone_num][i]))); + } + } +} diff -urN oldtree/kernel/power/pageflags.h newtree/kernel/power/pageflags.h --- oldtree/kernel/power/pageflags.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/pageflags.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,56 @@ +/* + * kernel/power/pageflags.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Suspend2 needs a few pageflags while working that aren't otherwise + * used. To save the struct page pageflags, we dynamically allocate + * a bitmap and use that. These are the only non order-0 allocations + * we do. + * + * NOTE!!! + * We assume that PAGE_SIZE - sizeof(void *) is a multiple of + * sizeof(unsigned long). Is this ever false? + */ + +#include +#include + +extern dyn_pageflags_t in_use_map; +extern dyn_pageflags_t pageset2_map; +extern dyn_pageflags_t pageset2_rw_map; + +/* + * inusemap is used in two ways: + * - During suspend, to tag pages which are not used (to speed up + * count_data_pages); + * - During resume, to tag pages which are in pagedir1. This does not tag + * pagedir2 pages, so !== first use. + */ + +#define PageInUse(page) (test_dynpageflag(&in_use_map, page)) +#define SetPageInUse(page) (set_dynpageflag(&in_use_map, page)) +#define ClearPageInUse(page) (clear_dynpageflag(&in_use_map, page)) + +#define PagePageset1(page) (test_dynpageflag(&pageset1_map, page)) +#define SetPagePageset1(page) (set_dynpageflag(&pageset1_map, page)) +#define ClearPagePageset1(page) (clear_dynpageflag(&pageset1_map, page)) + +#define PagePageset1Copy(page) (test_dynpageflag(&pageset1_copy_map, page)) +#define SetPagePageset1Copy(page) (set_dynpageflag(&pageset1_copy_map, page)) +#define ClearPagePageset1Copy(page) (clear_dynpageflag(&pageset1_copy_map, page)) + +#define PagePageset2(page) (test_dynpageflag(&pageset2_map, page)) +#define SetPagePageset2(page) (set_dynpageflag(&pageset2_map, page)) +#define ClearPagePageset2(page) (clear_dynpageflag(&pageset2_map, page)) + +#define PageWasRW(page) (test_dynpageflag(&pageset2_map, page)) +#define SetPageWasRW(page) (set_dynpageflag(&pageset2_map, page)) +#define ClearPageWasRW(page) (clear_dynpageflag(&pageset2_map, page)) + +extern void save_dyn_pageflags(dyn_pageflags_t pagemap); +extern void load_dyn_pageflags(dyn_pageflags_t pagemap); +extern void relocate_dyn_pageflags(dyn_pageflags_t *pagemap); +extern int suspend_pageflags_space_needed(void); diff -urN oldtree/kernel/power/power.h newtree/kernel/power/power.h --- oldtree/kernel/power/power.h 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/power.h 2006-10-06 17:28:20.000000000 -0400 @@ -1,5 +1,6 @@ #include #include +#include "suspend.h" struct swsusp_info { struct new_utsname uts; @@ -19,7 +20,15 @@ #else static inline int pm_suspend_disk(void) { +#ifdef CONFIG_SUSPEND2 + if (suspend_start_anything(1)) + return -EBUSY; + suspend_main(1); + suspend_finish_anything(1); + return 0; +#else return -EPERM; +#endif } #endif extern struct semaphore pm_sem; @@ -35,9 +44,6 @@ extern struct subsystem power_subsys; -/* References to section boundaries */ -extern const void __nosave_begin, __nosave_end; - /* Preferred image size in bytes (default 500 MB) */ extern unsigned long image_size; extern int in_suspend; diff -urN oldtree/kernel/power/power_off.c newtree/kernel/power/power_off.c --- oldtree/kernel/power/power_off.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/power_off.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,86 @@ +/* + * kernel/power/power_off.c + * + * Copyright (C) 2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Support for powering down. + */ + +#include +#include +#include +#include +#include +#include "suspend.h" +#include "ui.h" + +unsigned long suspend_powerdown_method = 0; /* 0 - Kernel power off */ + +extern struct pm_ops *pm_ops; + +/* Use suspend_enter from main.c */ +extern int suspend_enter(suspend_state_t state); + +int try_pm_state_powerdown(void) +{ + if (pm_ops && pm_ops->prepare && suspend_powerdown_method && + pm_ops->prepare(suspend_powerdown_method)) + return 0; + + if (suspend_powerdown_method > 3) + kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); + else { + if (device_suspend(PMSG_SUSPEND)) { + printk(KERN_ERR "Some devices failed to suspend\n"); + return 0; + } + } + + mdelay(1000); /* Give time for devices to power down */ + + if (suspend_enter(suspend_powerdown_method)) + return 0; + + device_resume(); + + if (pm_ops && pm_ops->finish && suspend_powerdown_method) + pm_ops->finish(suspend_powerdown_method); + + return 1; +} + +/* + * suspend_power_down + * Functionality : Powers down or reboots the computer once the image + * has been written to disk. + * Key Assumptions : Able to reboot/power down via code called or that + * the warning emitted if the calls fail will be visible + * to the user (ie printk resumes devices). + * Called From : do_suspend2_suspend_2 + */ + +void suspend_power_down(void) +{ + if (test_action_state(SUSPEND_REBOOT)) { + suspend_prepare_status(DONT_CLEAR_BAR, "Ready to reboot."); + kernel_restart(NULL); + } + + suspend_prepare_status(DONT_CLEAR_BAR, "Powering down."); + + if (pm_ops && pm_ops->enter && suspend_powerdown_method && try_pm_state_powerdown()) + return; + + kernel_shutdown_prepare(SYSTEM_POWER_OFF); + + mdelay(1000); /* Give time for devices to power down */ + + machine_power_off(); + machine_halt(); + suspend_prepare_status(DONT_CLEAR_BAR, "Powerdown failed"); + while (1) + cpu_relax(); +} + diff -urN oldtree/kernel/power/power_off.h newtree/kernel/power/power_off.h --- oldtree/kernel/power/power_off.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/power_off.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,13 @@ +/* + * kernel/power/power_off.h + * + * Copyright (C) 2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Support for the powering down. + */ + +int suspend_pm_state_finish(void); +void suspend_power_down(void); +extern unsigned long suspend_powerdown_method; diff -urN oldtree/kernel/power/prepare_image.c newtree/kernel/power/prepare_image.c --- oldtree/kernel/power/prepare_image.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/prepare_image.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,689 @@ +/* + * kernel/power/prepare_image.c + * + * Copyright (C) 2003-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * We need to eat memory until we can: + * 1. Perform the save without changing anything (RAM_NEEDED < max_pfn) + * 2. Fit it all in available space (suspend_active_writer->available_space() >= + * storage_needed()) + * 3. Reload the pagedir and pageset1 to places that don't collide with their + * final destinations, not knowing to what extent the resumed kernel will + * overlap with the one loaded at boot time. I think the resumed kernel + * should overlap completely, but I don't want to rely on this as it is + * an unproven assumption. We therefore assume there will be no overlap at + * all (worse case). + * 4. Meet the user's requested limit (if any) on the size of the image. + * The limit is in MB, so pages/256 (assuming 4K pages). + * + */ + +#include +#include +#include + +#include "pageflags.h" +#include "modules.h" +#include "io.h" +#include "ui.h" +#include "extent.h" +#include "prepare_image.h" +#include "block_io.h" +#include "suspend.h" + +static int are_frozen = 0, num_nosave = 0; +static long header_space_allocated = 0; +static long storage_allocated = 0; +static long storage_available = 0; +long extra_pd1_pages_allowance = MIN_EXTRA_PAGES_ALLOWANCE; + +/* + * num_pcp_pages: Count pcp pages. + */ +static long num_pcp_pages(void) +{ + struct zone *zone; + long result = 0, i = 0; + + /* PCP lists */ + for_each_zone(zone) { + struct per_cpu_pageset *pset; + int cpu; + + if (!zone->present_pages) + continue; + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + if (!cpu_possible(cpu)) + continue; + + pset = zone_pcp(zone, cpu); + + for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { + struct per_cpu_pages *pcp; + + pcp = &(pset->pcp[i]); + result += pcp->count; + } + } + } + return result; +} + +/* + * Number of free pages, including pcp pages. + */ +long real_nr_free_pages(void) +{ + return nr_free_pages() + num_pcp_pages(); +} + +/* + * Discover how much extra memory will be required by the drivers + * when they're asked to suspend. We can then ensure that amount + * of memory is available when we really want it. + */ +static void get_extra_pd1_allowance(void) +{ + int orig_num_free = real_nr_free_pages(), final; + + suspend_prepare_status(CLEAR_BAR, "Finding allowance for drivers."); + device_suspend(PMSG_FREEZE); + local_irq_disable(); /* irqs might have been re-enabled on us */ + device_power_down(PMSG_FREEZE); + + final = real_nr_free_pages(); + + device_power_up(); + local_irq_enable(); + + device_resume(); + + extra_pd1_pages_allowance = max( + orig_num_free - final + MIN_EXTRA_PAGES_ALLOWANCE, + MIN_EXTRA_PAGES_ALLOWANCE); +} + +/* + * Amount of storage needed, possibly taking into account the + * expected compression ratio and possibly also ignoring our + * allowance for extra pages. + */ +static long main_storage_needed(int use_ecr, + int ignore_extra_pd1_allow) +{ + return ((pagedir1.pageset_size + pagedir2.pageset_size + + (ignore_extra_pd1_allow ? 0 : extra_pd1_pages_allowance)) * + (use_ecr ? suspend_expected_compression_ratio() : 100) / 100); +} + +/* + * Storage needed for the image header, in bytes until the return. + */ +static int header_storage_needed(void) +{ + unsigned long bytes = + (int) sizeof(struct suspend_header) + + (int) suspend_header_storage_for_modules() + + suspend_pageflags_space_needed(); + + return ((int) ((bytes + (int) PAGE_SIZE - 1) >> PAGE_SHIFT)); +} + +static void display_stats(int always, int sub_extra_pd1_allow) +{ + char buffer[255]; + snprintf(buffer, 254, + "Free:%d(%d). Sets:%ld(%ld),%ld(%ld). Header:%d. Nosave:%d-%d=%d. Storage:%lu/%lu(%lu). Needed:%ld|%ld|%ld.\n", + + /* Free */ + nr_free_pages(), + nr_free_pages() - nr_free_highpages(), + + /* Sets */ + pagedir1.pageset_size, pageset1_sizelow, + pagedir2.pageset_size, pageset2_sizelow, + + /* Header */ + header_storage_needed(), + + /* Nosave */ + num_nosave, extra_pagedir_pages_allocated, + num_nosave - extra_pagedir_pages_allocated, + + /* Storage - converted to pages for comparison */ + storage_allocated, + storage_needed(1, sub_extra_pd1_allow), + storage_available, + + /* Needed */ + ram_to_suspend() - nr_free_pages() - nr_free_highpages(), + storage_needed(1, sub_extra_pd1_allow) - storage_available, + (image_size_limit > 0) ? (storage_needed(1, sub_extra_pd1_allow) - (image_size_limit << 8)) : 0); + if (always) + printk(buffer); + else + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_MEDIUM, 1, buffer); +} + +/* generate_free_page_map + * + * Description: This routine generates a bitmap of free pages from the + * lists used by the memory manager. We then use the bitmap + * to quickly calculate which pages to save and in which + * pagesets. + */ +static void generate_free_page_map(void) +{ + int order, loop, cpu; + struct page *page; + unsigned long flags, i; + struct zone *zone; + struct per_cpu_pageset *pset; + + for_each_zone(zone) { + if (!zone->present_pages) + continue; + for(i=0; i < zone->spanned_pages; i++) + SetPageInUse(pfn_to_page(zone->zone_start_pfn + i)); + } + + for_each_zone(zone) { + if (!zone->present_pages) + continue; + spin_lock_irqsave(&zone->lock, flags); + for (order = MAX_ORDER - 1; order >= 0; --order) { + list_for_each_entry(page, &zone->free_area[order].free_list, lru) + for(loop=0; loop < (1 << order); loop++) + ClearPageInUse(page+loop); + } + + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + if (!cpu_possible(cpu)) + continue; + + pset = zone_pcp(zone, cpu); + + for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) { + struct per_cpu_pages *pcp; + struct page *page; + + pcp = &pset->pcp[i]; + list_for_each_entry(page, &pcp->list, lru) + ClearPageInUse(page); + } + } + + spin_unlock_irqrestore(&zone->lock, flags); + } +} + +/* size_of_free_region + * + * Description: Return the number of pages that are free, beginning with and + * including this one. + */ +static int size_of_free_region(struct page *page) +{ + struct zone *zone = page_zone(page); + struct page *posn = page, *last_in_zone = + pfn_to_page(zone->zone_start_pfn) + zone->spanned_pages - 1; + + while (posn < last_in_zone && !PageInUse(posn)) { + BUG_ON(PagePageset2(posn)); + posn++; + } + return (posn - page); +} + +static struct page *rotext_start, *rotext_end; +static struct page *rodata_start, *rodata_end; +static struct page *nosave_start, *nosave_end; + +static __init int page_nosave_init(void) +{ + rodata_start = rodata_start_page(); + rodata_end = rodata_end_page(); + + rotext_start = rotext_start_page(); + rotext_end = rotext_end_page(); + + nosave_start = nosave_start_page(); + nosave_end = nosave_end_page(); + + return 0; +} + +subsys_initcall(page_nosave_init); + +/* count_data_pages + * + * This routine generates our lists of pages to be stored in each + * pageset. Since we store the data using extents, and adding new + * extents might allocate a new extent page, this routine may well + * be called more than once. + */ +static struct pageset_sizes_result count_data_pages(void) +{ + int num_free = 0; + unsigned long loop; + int use_pagedir2; + struct pageset_sizes_result result; + struct zone *zone; + + result.size1 = 0; + result.size1low = 0; + result.size2 = 0; + result.size2low = 0; + + num_nosave = 0; + + clear_dyn_pageflags(pageset1_map); + + generate_free_page_map(); + + if (test_result_state(SUSPEND_ABORTED)) + return result; + + /* + * Pages not to be saved are marked Nosave irrespective of being reserved + */ + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + for (loop = 0; loop < zone->spanned_pages; loop++) { + unsigned long pfn = zone->zone_start_pfn + loop; + struct page *page; + int chunk_size; + + if (!pfn_valid(pfn)) + continue; + + page = pfn_to_page(pfn); + chunk_size = size_of_free_region(page); + + if (PageNosave(page) || + (page >= rodata_start && page < rodata_end) || + (PageReserved(page) && + ((page >= nosave_start && page < nosave_end) || + is_highmem(zone)))) { + num_nosave++; + continue; + } + + if (chunk_size) { + num_free += chunk_size; + loop += chunk_size - 1; + continue; + } + + use_pagedir2 = PagePageset2(page); + + if (use_pagedir2) { + result.size2++; + if (!PageHighMem(page)) { + result.size2low++; + SetPagePageset1Copy(page); + } + } else { + result.size1++; + SetPagePageset1(page); + if (!PageHighMem(page)) + result.size1low++; + } + } + } + + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_MEDIUM, 0, + "Count data pages: Set1 (%d) + Set2 (%d) + Nosave (%d) + NumFree (%d) = %d.\n", + result.size1, result.size2, num_nosave, num_free, + result.size1 + result.size2 + num_nosave + num_free); + return result; +} + +/* amount_needed + * + * Calculates the amount by which the image size needs to be reduced to meet + * our constraints. + */ +static int amount_needed(int use_image_size_limit) +{ + + int max1 = max( (int) (ram_to_suspend() - real_nr_free_pages() - + nr_free_highpages()), + ((int) (storage_needed(1, 0) - + storage_available))); + if (use_image_size_limit) + return max( max1, + (image_size_limit > 0) ? + ((int) (storage_needed(1, 0) - (image_size_limit << 8))) : 0); + return max1; +} + +/* suspend_recalculate_image_contents + * + * Eaten is the number of pages which have been eaten. + * Pagedirincluded is the number of pages which have been allocated for the pagedir. + */ +void suspend_recalculate_image_contents(int atomic_copy) +{ + struct pageset_sizes_result result; + + clear_dyn_pageflags(pageset1_map); + if (!atomic_copy) { + int pfn; + BITMAP_FOR_EACH_SET(pageset2_map, pfn) + ClearPagePageset1Copy(pfn_to_page(pfn)); + /* Need to call this before getting pageset1_size! */ + suspend_mark_pages_for_pageset2(); + } + BUG_ON(in_atomic() && !irqs_disabled()); + result = count_data_pages(); + pageset1_sizelow = result.size1low; + pageset2_sizelow = result.size2low; + pagedir1.lastpageset_size = pagedir1.pageset_size = result.size1; + pagedir2.lastpageset_size = pagedir2.pageset_size = result.size2; + + if (!atomic_copy) { + storage_available = suspend_active_writer->storage_available(); + display_stats(1, 0); + } + BUG_ON(in_atomic() && !irqs_disabled()); + return; +} + +static void try_freeze_processes(void) +{ + if (freeze_processes()) { + set_result_state(SUSPEND_FREEZING_FAILED); + set_result_state(SUSPEND_ABORTED); + } +} + +/* update_image + * + * Allocate [more] memory and storage for the image. + */ +static int update_image(void) +{ + int result2, param_used; + + suspend_recalculate_image_contents(0); + + /* Include allowance for growth in pagedir1 while writing pagedir 2 */ + if (suspend_allocate_extra_pagedir_memory(&pagedir1, + pagedir1.pageset_size + extra_pd1_pages_allowance, + pageset2_sizelow)) { + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_LOW, 1, + "Still need to get more pages for pagedir 1.\n"); + return 1; + } + + thaw_processes(FREEZER_KERNEL_THREADS); + + param_used = main_storage_needed(1, 0); + if ((result2 = suspend_active_writer->allocate_storage(param_used))) { + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_LOW, 1, + "Allocate storage returned %d. Still need to get more" + " storage space for the image proper.\n", + result2); + storage_allocated = suspend_active_writer->storage_allocated(); + try_freeze_processes(); + return 1; + } + + /* + * Allocate remaining storage space, if possible, up to the + * maximum we know we'll need. It's okay to allocate the + * maximum if the writer is the swapwriter, but + * we don't want to grab all available space on an NFS share. + * We therefore ignore the expected compression ratio here, + * thereby trying to allocate the maximum image size we could + * need (assuming compression doesn't expand the image), but + * don't complain if we can't get the full amount we're after. + */ + + suspend_active_writer->allocate_storage( + min(storage_available, main_storage_needed(0, 1))); + + storage_allocated = suspend_active_writer->storage_allocated(); + + /* Allocate the header storage after allocating main storage + * so that the overhead for metadata doesn't change the amount + * of storage needed for the header itself. + */ + + param_used = header_storage_needed(); + + result2 = suspend_active_writer->allocate_header_space(param_used); + + try_freeze_processes(); + + if (result2) { + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_LOW, 1, + "Still need to get more storage space for header.\n"); + return 1; + } + + header_space_allocated = param_used; + + suspend_recalculate_image_contents(0); + + suspend_message(SUSPEND_EAT_MEMORY, SUSPEND_LOW, 1, + "Amount still needed (%d) > 0:%d. Header: %d < %d: %d," + " Storage allocd: %d < %d + %d: %d.\n", + amount_needed(0), + (amount_needed(0) > 0), + header_space_allocated, header_storage_needed(), + header_space_allocated < header_storage_needed(), + storage_allocated, + header_storage_needed(), main_storage_needed(1, 1), + storage_allocated < + (header_storage_needed() + main_storage_needed(1, 1))); + + suspend_cond_pause(0, NULL); + + return ((amount_needed(0) > 0) || + header_space_allocated < header_storage_needed() || + storage_allocated < + (header_storage_needed() + main_storage_needed(1, 1))); +} + +/* attempt_to_freeze + * + * Try to freeze processes. + */ + +static int attempt_to_freeze(void) +{ + int result; + + /* Stop processes before checking again */ + thaw_processes(FREEZER_ALL_THREADS); + suspend_prepare_status(CLEAR_BAR, "Freezing processes"); + result = freeze_processes(); + + if (result) { + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_FREEZING_FAILED); + } else + are_frozen = 1; + + return result; +} + +long storage_needed(int use_ecr, int ignore_extra_pd1_allow) +{ + return (main_storage_needed(use_ecr, ignore_extra_pd1_allow) + + header_storage_needed()); +} + +long ram_to_suspend(void) +{ + return (1 + + max_t(long, (pagedir1.pageset_size + extra_pd1_pages_allowance - + pageset2_sizelow - extra_pagedir_pages_allocated) / 2, 0) + + MIN_FREE_RAM + suspend_memory_for_modules()); +} + +/* eat_memory + * + * Try to free some memory, either to meet hard or soft constraints on the image + * characteristics. + * + * Hard constraints: + * - Pageset1 must be < half of memory; + * - We must have enough memory free at resume time to have pageset1 + * be able to be loaded in pages that don't conflict with where it has to + * be restored. + * Soft constraints + * - User specificied image size limit. + */ +static int eat_memory(void) +{ + int amount_wanted = 0; + int free_flags = 0, did_eat_memory = 0; + + /* + * Note that if we have enough storage space and enough free memory, we may + * exit without eating anything. We give up when the last 10 iterations ate + * no extra pages because we're not going to get much more anyway, but + * the few pages we get will take a lot of time. + * + * We freeze processes before beginning, and then unfreeze them if we + * need to eat memory until we think we have enough. If our attempts + * to freeze fail, we give up and abort. + */ + + /* -- Stage 1: Freeze Processes -- */ + + + suspend_recalculate_image_contents(0); + amount_wanted = amount_needed(1); + + switch (image_size_limit) { + case -1: /* Don't eat any memory */ + if (amount_wanted > 0) { + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_WOULD_EAT_MEMORY); + } + break; + case -2: /* Free caches only */ + free_flags = GFP_NOIO | __GFP_HIGHMEM; + amount_wanted = 1 << 31; /* As much cache as we can get */ + break; + default: + free_flags = GFP_ATOMIC | __GFP_HIGHMEM; + } + + thaw_processes(FREEZER_KERNEL_THREADS); + + /* -- Stage 2: Eat memory -- */ + + if (amount_wanted > 0 && !test_result_state(SUSPEND_ABORTED) && + image_size_limit != -1) { + + suspend_prepare_status(CLEAR_BAR, "Seeking to free %dMB of memory.", MB(amount_wanted)); + + shrink_all_memory(amount_wanted); + suspend_recalculate_image_contents(0); + + did_eat_memory = 1; + + suspend_cond_pause(0, NULL); + } + + if (freeze_processes()) { + set_result_state(SUSPEND_FREEZING_FAILED); + set_result_state(SUSPEND_ABORTED); + } + + if (did_eat_memory) { + unsigned long orig_state = get_suspend_state(); + /* Freeze_processes will call sys_sync too */ + restore_suspend_state(orig_state); + suspend_recalculate_image_contents(0); + } + + /* Blank out image size display */ + suspend_update_status(100, 100, NULL); + + if (!test_result_state(SUSPEND_ABORTED) && + (amount_needed(0) - extra_pd1_pages_allowance > 0)) { + printk("Unable to free sufficient memory to suspend. Still need %d pages.\n", + amount_needed(1)); + display_stats(1, 1); + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_UNABLE_TO_FREE_ENOUGH_MEMORY); + } + + return 0; +} + +/* suspend_prepare_image + * + * Entry point to the whole image preparation section. + * + * We do four things: + * - Freeze processes; + * - Ensure image size constraints are met; + * - Complete all the preparation for saving the image, + * including allocation of storage. The only memory + * that should be needed when we're finished is that + * for actually storing the image (and we know how + * much is needed for that because the modules tell + * us). + * - Make sure that all dirty buffers are written out. + */ +#define MAX_TRIES 4 +int suspend_prepare_image(void) +{ + int result = 1, tries = 0; + + are_frozen = 0; + + header_space_allocated = 0; + + if (attempt_to_freeze()) + return 1; + + if (!extra_pd1_pages_allowance) + get_extra_pd1_allowance(); + + storage_available = suspend_active_writer->storage_available(); + + if (!storage_available) { + printk(KERN_ERR "You need some storage available to be able to suspend.\n"); + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_NOSTORAGE_AVAILABLE); + return 1; + } + + do { + suspend_prepare_status(CLEAR_BAR, "Preparing Image."); + + if (eat_memory() || test_result_state(SUSPEND_ABORTED)) + break; + + result = update_image(); + + suspend_cond_pause(0, NULL); + + tries++; + + } while ((result) && (tries < MAX_TRIES) && (!test_result_state(SUSPEND_ABORTED)) && + (!test_result_state(SUSPEND_UNABLE_TO_FREE_ENOUGH_MEMORY))); + + if (tries == MAX_TRIES) { + abort_suspend("Unable to successfully prepare the image.\n"); + display_stats(1, 0); + } + + suspend_cond_pause(1, "Image preparation complete."); + + return result; +} diff -urN oldtree/kernel/power/prepare_image.h newtree/kernel/power/prepare_image.h --- oldtree/kernel/power/prepare_image.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/prepare_image.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,89 @@ +/* + * kernel/power/prepare_image.h + * + * Copyright (C) 2003-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + */ + +#include + +extern int suspend_prepare_image(void); +extern void suspend_recalculate_image_contents(int storage_available); +extern long real_nr_free_pages(void); +extern long image_size_limit; +extern long pageset1_sizelow, pageset2_sizelow; + +struct pageset_sizes_result { + long size1; /* Can't be unsigned - breaks MAX function */ + long size1low; + long size2; + long size2low; +}; + +#ifdef CONFIG_CRYPTO +extern int suspend_expected_compression_ratio(void); +#else +static inline int suspend_expected_compression_ratio(void) +{ + return 0; +}; +#endif + +#define MIN_FREE_RAM 2000 +#define MIN_EXTRA_PAGES_ALLOWANCE 500 + +extern long extra_pd1_pages_allowance; +extern long storage_needed(int use_ecr, int ignore_extra_p1_allowance); +extern long ram_to_suspend(void); + +#ifdef CONFIG_DEBUG_RODATA +static inline struct page* rodata_start_page(void) +{ + return virt_to_page(&__start_rodata); +} + +static inline struct page* rodata_end_page(void) +{ + return virt_to_page(&__end_rodata); +} + +#else +static inline struct page* rodata_start_page(void) +{ + return NULL; +} + +static inline struct page* rodata_end_page(void) +{ + return NULL; +} +#endif + +#ifdef CONFIG_PPC +static inline struct page* rotext_start_page(void) +{ + return virt_to_page(PAGE_OFFSET); +} +#else +static inline struct page* rotext_start_page(void) +{ + return virt_to_page(_text); +} +#endif + +static inline struct page* rotext_end_page(void) +{ + return virt_to_page(_etext); +} + +static inline struct page* nosave_start_page(void) +{ + return virt_to_page(&__nosave_begin); +} + +static inline struct page* nosave_end_page(void) +{ + return virt_to_page(&__nosave_end); +} diff -urN oldtree/kernel/power/process.c newtree/kernel/power/process.c --- oldtree/kernel/power/process.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/process.c 2006-10-06 17:28:20.000000000 -0400 @@ -5,7 +5,6 @@ * Originally from swsusp. */ - #undef DEBUG #include @@ -13,12 +12,74 @@ #include #include #include +#include +#include + +unsigned long freezer_state = 0; + +#ifdef CONFIG_PM_DEBUG +#define freezer_message(msg, a...) do { printk(msg, ##a); } while(0) +#else +#define freezer_message(msg, a...) do { } while(0) +#endif /* * Timeout for stopping processes */ #define TIMEOUT (20 * HZ) +struct frozen_fs +{ + struct list_head fsb_list; + struct super_block *sb; +}; + +LIST_HEAD(frozen_fs_list); + +void freezer_make_fses_rw(void) +{ + struct frozen_fs *fs, *next_fs; + + list_for_each_entry_safe(fs, next_fs, &frozen_fs_list, fsb_list) { + thaw_bdev(fs->sb->s_bdev, fs->sb); + + list_del(&fs->fsb_list); + kfree(fs); + } +} + +/* + * Done after userspace is frozen, so there should be no danger of + * fses being unmounted while we're in here. + */ +int freezer_make_fses_ro(void) +{ + struct frozen_fs *fs; + struct super_block *sb; + + /* Generate the list */ + list_for_each_entry(sb, &super_blocks, s_list) { + if (!sb->s_root || !sb->s_bdev || + (sb->s_frozen == SB_FREEZE_TRANS) || + (sb->s_flags & MS_RDONLY)) + continue; + + fs = kmalloc(sizeof(struct frozen_fs), GFP_ATOMIC); + if (!fs) + return 1; + fs->sb = sb; + list_add_tail(&fs->fsb_list, &frozen_fs_list); + }; + + /* Do the freezing in reverse order so filesystems dependant + * upon others are frozen in the right order. (Eg loopback + * on ext3). */ + list_for_each_entry_reverse(fs, &frozen_fs_list, fsb_list) + freeze_bdev(fs->sb->s_bdev); + + return 0; +} + static inline int freezeable(struct task_struct * p) { @@ -39,7 +100,7 @@ long save; save = current->state; pr_debug("%s entered refrigerator\n", current->comm); - printk("="); + freezer_message("="); frozen_process(current); spin_lock_irq(¤t->sighand->siglock); @@ -86,9 +147,13 @@ unsigned long start_time; struct task_struct *g, *p; - printk( "Stopping tasks: " ); + user_frozen = test_freezer_state(FREEZER_ON); + + if (!user_frozen) + set_freezer_state(FREEZER_ON); + + freezer_message( "Stopping tasks: " ); start_time = jiffies; - user_frozen = 0; do { nr_user = todo = 0; read_lock(&tasklist_lock); @@ -119,8 +184,10 @@ read_unlock(&tasklist_lock); todo += nr_user; if (!user_frozen && !nr_user) { - sys_sync(); + freezer_message("Freezing bdevs... "); + freezer_make_fses_ro(); start_time = jiffies; + freezer_message("Freezing kernel threads... "); } user_frozen = !nr_user; yield(); /* Yield is okay here */ @@ -134,41 +201,70 @@ * but it cleans up leftover PF_FREEZE requests. */ if (todo) { - printk( "\n" ); - printk(KERN_ERR " stopping tasks timed out " + freezer_message( "\n" ); + freezer_message(KERN_ERR " stopping tasks timed out " "after %d seconds (%d tasks remaining):\n", TIMEOUT / HZ, todo); read_lock(&tasklist_lock); do_each_thread(g, p) { if (freezeable(p) && !frozen(p)) - printk(KERN_ERR " %s\n", p->comm); + freezer_message(KERN_ERR " %s\n", p->comm); cancel_freezing(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); return todo; } - printk( "|\n" ); + freezer_message( "|\n" ); BUG_ON(in_atomic()); + set_freezer_state(FREEZING_COMPLETE); return 0; } -void thaw_processes(void) +void thaw_processes(int all) { struct task_struct *g, *p; + int pass = 0; /* Start on kernel space */ - printk( "Restarting tasks..." ); + if (!test_freezer_state(FREEZER_ON)) + return; + + if (!test_freezer_state(FREEZING_COMPLETE)) + pass++; + + clear_freezer_state(FREEZING_COMPLETE); + + freezer_message( "Restarting tasks... " ); read_lock(&tasklist_lock); - do_each_thread(g, p) { - if (!freezeable(p)) - continue; - if (!thaw_process(p)) - printk(KERN_INFO " Strange, %s not stopped\n", p->comm ); - } while_each_thread(g, p); + do { + do_each_thread(g, p) { + /* + * is_user = 0 if kernel thread or borrowed mm, + * 1 otherwise. + */ + int is_user = !!(p->mm && !(p->flags & PF_BORROWED_MM)); + if (!freezeable(p) || (is_user != pass)) + continue; + if (!thaw_process(p)) + freezer_message(KERN_INFO " Strange, %s not stopped\n", p->comm ); + } while_each_thread(g, p); + + if (!pass) { + read_unlock(&tasklist_lock); + freezer_message("Thawing bdevs... "); + freezer_make_fses_rw(); + read_lock(&tasklist_lock); + } + + pass++; + } while(pass < 2 && all); read_unlock(&tasklist_lock); schedule(); - printk( " done\n" ); + freezer_message( "Done\n" ); + + if (all) + clear_freezer_state(FREEZER_ON); } EXPORT_SYMBOL(refrigerator); diff -urN oldtree/kernel/power/snapshot.c newtree/kernel/power/snapshot.c --- oldtree/kernel/power/snapshot.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/snapshot.c 2006-10-06 17:28:20.000000000 -0400 @@ -34,6 +34,13 @@ #include "power.h" +#ifdef CONFIG_SUSPEND2 +#include "pagedir.h" + +extern int suspend2_running; +int suspend_post_context_save(void); +#endif + /* List of PBEs used for creating and restoring the suspend image */ struct pbe *restore_pblist; @@ -195,6 +202,10 @@ unsigned long get_safe_page(gfp_t gfp_mask) { +#ifdef CONFIG_SUSPEND2 + if (suspend2_running) + return suspend_get_nonconflicting_page(); +#endif return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE); } @@ -868,6 +879,11 @@ { unsigned int nr_pages; +#ifdef CONFIG_SUSPEND2 + if (suspend2_running) + return suspend_post_context_save(); +#endif + pr_debug("swsusp: critical section: \n"); drain_local_pages(); diff -urN oldtree/kernel/power/storage.c newtree/kernel/power/storage.c --- oldtree/kernel/power/storage.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/storage.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,304 @@ +/* + * kernel/power/storage.c + * + * Copyright (C) 2005-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Routines for talking to a userspace program that manages storage. + * + * The kernel side: + * - starts the userspace program; + * - sends messages telling it when to open and close the connection; + * - tells it when to quit; + * + * The user space side: + * - passes messages regarding status; + * + */ + +#include +#include + +#include "sysfs.h" +#include "modules.h" +#include "netlink.h" +#include "storage.h" +#include "ui.h" + +static struct user_helper_data usm_helper_data; +static struct suspend_module_ops usm_ops; +static int message_received = 0; +static int activations = 0; +static int usm_prepare_count = 0; +static int storage_manager_last_action = 0; +static int storage_manager_action = 0; + +static int usm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + int type; + int *data; + + type = nlh->nlmsg_type; + + /* A control message: ignore them */ + if (type < NETLINK_MSG_BASE) + return 0; + + /* Unknown message: reply with EINVAL */ + if (type >= USM_MSG_MAX) + return -EINVAL; + + /* All operations require privileges, even GET */ + if (security_netlink_recv(skb, CAP_NET_ADMIN)) + return -EPERM; + + /* Only allow one task to receive NOFREEZE privileges */ + if (type == NETLINK_MSG_NOFREEZE_ME && usm_helper_data.pid != -1) + return -EBUSY; + + data = (int*)NLMSG_DATA(nlh); + + switch (type) { + case USM_MSG_SUCCESS: + case USM_MSG_FAILED: + message_received = type; + complete(&usm_helper_data.wait_for_process); + break; + default: + printk("Storage manager doesn't recognise message %d.\n", type); + } + + return 1; +} + +int suspend_activate_storage(int force) +{ + int tries = 1; + + if (usm_helper_data.pid == -1 || !usm_ops.enabled) + return 0; + + message_received = 0; + activations++; + + if (activations > 1 && !force) + return 0; + + while ((!message_received || message_received == USM_MSG_FAILED) && tries < 2) { + suspend_prepare_status(DONT_CLEAR_BAR, "Activate storage attempt %d.\n", tries); + + init_completion(&usm_helper_data.wait_for_process); + + suspend_send_netlink_message(&usm_helper_data, + USM_MSG_CONNECT, + NULL, 0); + + /* Wait 2 seconds for the userspace process to make contact */ + wait_for_completion_timeout(&usm_helper_data.wait_for_process, 2*HZ); + + tries++; + } + + return 0; +} + +int suspend_deactivate_storage(int force) +{ + if (usm_helper_data.pid == -1 || !usm_ops.enabled) + return 0; + + message_received = 0; + activations--; + + if (activations && !force) + return 0; + + init_completion(&usm_helper_data.wait_for_process); + + suspend_send_netlink_message(&usm_helper_data, + USM_MSG_DISCONNECT, + NULL, 0); + + wait_for_completion_timeout(&usm_helper_data.wait_for_process, 2*HZ); + + if (!message_received || message_received == USM_MSG_FAILED) { + printk("Returning failure disconnecting storage.\n"); + return 1; + } + + return 0; +} + +#ifdef CONFIG_PM_DEBUG +static void storage_manager_simulate(void) +{ + printk("--- Storage manager simulate ---\n"); + suspend_prepare_usm(); + schedule(); + printk("--- Activate storage 1 ---\n"); + suspend_activate_storage(1); + schedule(); + printk("--- Deactivate storage 1 ---\n"); + suspend_deactivate_storage(1); + schedule(); + printk("--- Cleanup usm ---\n"); + suspend_cleanup_usm(); + schedule(); + printk("--- Storage manager simulate ends ---\n"); +} +#endif + +static unsigned long usm_storage_needed(void) +{ + return strlen(usm_helper_data.program); +} + +static int usm_save_config_info(char *buf) +{ + int len = strlen(usm_helper_data.program); + memcpy(buf, usm_helper_data.program, len); + return len; +} + +static void usm_load_config_info(char *buf, int size) +{ + /* Don't load the saved path if one has already been set */ + if (usm_helper_data.program[0]) + return; + + memcpy(usm_helper_data.program, buf, size); +} + +static unsigned long usm_memory_needed(void) +{ + /* ball park figure of 32 pages */ + return (32 * PAGE_SIZE); +} + +/* suspend_prepare_usm + */ +int suspend_prepare_usm(void) +{ + usm_prepare_count++; + + if (usm_prepare_count > 1 || !usm_ops.enabled) + return 0; + + usm_helper_data.pid = -1; + + if (!*usm_helper_data.program) + return 0; + + suspend_netlink_setup(&usm_helper_data); + + if (usm_helper_data.pid == -1) + printk("Suspend2 Storage Manager wanted, but couldn't start it.\n"); + + suspend_activate_storage(0); + + return (usm_helper_data.pid != -1); +} + +void suspend_cleanup_usm(void) +{ + usm_prepare_count--; + + if (usm_helper_data.pid > -1 && !usm_prepare_count) { + struct task_struct *t; + + suspend_deactivate_storage(0); + + suspend_send_netlink_message(&usm_helper_data, + NETLINK_MSG_CLEANUP, NULL, 0); + + read_lock(&tasklist_lock); + if ((t = find_task_by_pid(usm_helper_data.pid))) + t->flags &= ~PF_NOFREEZE; + read_unlock(&tasklist_lock); + + suspend_netlink_close(&usm_helper_data); + + usm_helper_data.pid = -1; + } +} + +static void storage_manager_activate(void) +{ + if (storage_manager_action == storage_manager_last_action) + return; + + if (storage_manager_action) + suspend_prepare_usm(); + else + suspend_cleanup_usm(); + + storage_manager_last_action = storage_manager_action; +} + +/* + * User interface specific /sys/power/suspend2 entries. + */ + +static struct suspend_sysfs_data sysfs_params[] = { + { SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&usm_ops.enabled, 0, 1) + }, + + { SUSPEND2_ATTR("program", SYSFS_RW), + SYSFS_STRING(usm_helper_data.program, 254, SYSFS_SM_NOT_NEEDED) + }, + + { SUSPEND2_ATTR("activate_storage", SYSFS_RW), + SYSFS_INT(&storage_manager_action, 0, 1), + .write_side_effect = storage_manager_activate, + }, + +#ifdef CONFIG_PM_DEBUG + { SUSPEND2_ATTR("simulate_atomic_copy", SYSFS_RW), + .type = SUSPEND_SYSFS_DATA_NONE, + .write_side_effect = storage_manager_simulate, + } +#endif +}; + +static struct suspend_module_ops usm_ops = { + .type = MISC_MODULE, + .name = "Userspace Storage Manager", + .module = THIS_MODULE, + .storage_needed = usm_storage_needed, + .save_config_info = usm_save_config_info, + .load_config_info = usm_load_config_info, + .memory_needed = usm_memory_needed, +}; + +/* suspend_usm_sysfs_init + * Description: Boot time initialisation for user interface. + */ +static __init int suspend_usm_sysfs_init(void) +{ + int result, i, + numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + if (!(result = suspend_register_module(&usm_ops))) { + struct kobject *kobj = make_suspend2_sysdir("storage_manager"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, &sysfs_params[i]); + } + + usm_helper_data.nl = NULL; + usm_helper_data.program[0] = '\0'; + usm_helper_data.pid = -1; + usm_helper_data.skb_size = 0; + usm_helper_data.pool_limit = 6; + usm_helper_data.netlink_id = NETLINK_SUSPEND2_USM; + usm_helper_data.name = "userspace storage manager"; + usm_helper_data.rcv_msg = usm_user_rcv_msg; + usm_helper_data.interface_version = 1; + usm_helper_data.must_init = 0; + init_completion(&usm_helper_data.wait_for_process); + + return result; +} + +late_initcall(suspend_usm_sysfs_init); diff -urN oldtree/kernel/power/storage.h newtree/kernel/power/storage.h --- oldtree/kernel/power/storage.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/storage.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,37 @@ +/* + * kernel/power/storage.h + * + * Copyright (C) 2005-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + */ + +int suspend_prepare_usm(void); +void suspend_cleanup_usm(void); + +#ifdef CONFIG_NET +int suspend_activate_storage(int force); +int suspend_deactivate_storage(int force); +#else +static inline int suspend_activate_storage(int force) +{ + return 0; +} + +static inline int suspend_deactivate_storage(int force) +{ + return 0; +} +#endif + +enum { + USM_MSG_BASE = 0x10, + + /* Kernel -> Userspace */ + USM_MSG_CONNECT = 0x30, + USM_MSG_DISCONNECT = 0x31, + USM_MSG_SUCCESS = 0x40, + USM_MSG_FAILED = 0x41, + + USM_MSG_MAX, +}; diff -urN oldtree/kernel/power/suspend.c newtree/kernel/power/suspend.c --- oldtree/kernel/power/suspend.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/suspend.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,954 @@ +/* + * kernel/power/suspend.c + */ +/** \mainpage Suspend2. + * + * Suspend2 provides support for saving and restoring an image of + * system memory to an arbitrary storage device, either on the local computer, + * or across some network. The support is entirely OS based, so Suspend2 + * works without requiring BIOS, APM or ACPI support. The vast majority of the + * code is also architecture independant, so it should be very easy to port + * the code to new architectures. Suspend includes support for SMP, 4G HighMem + * and preemption. Initramfses and initrds are also supported. + * + * Suspend2 uses a modular design, in which the method of storing the image is + * completely abstracted from the core code, as are transformations on the data + * such as compression and/or encryption (multiple 'modules' can be used to + * provide arbitrary combinations of functionality). The user interface is also + * modular, so that arbitrarily simple or complex interfaces can be used to + * provide anything from debugging information through to eye candy. + * + * \section Copyright + * + * Suspend2 is released under the GPLv2. + * + * Copyright (C) 1998-2001 Gabor Kuti
+ * Copyright (C) 1998,2001,2002 Pavel Machek
+ * Copyright (C) 2002-2003 Florent Chabaud
+ * Copyright (C) 2002-2006 Nigel Cunningham
+ * + * \section Credits + * + * Nigel would like to thank the following people for their work: + * + * Pavel Machek
+ * Modifications, defectiveness pointing, being with Gabor 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 + * + * Jeff Snyder
+ * ACPI patch + * + * Nathan Friess
+ * Some patches. + * + * Michael Frank
+ * Extensive testing and help with improving stability. Nigel was constantly + * amazed by the quality and quantity of Michael's help. + * + * Bernard Blackham
+ * Web page & Wiki administration, some coding. Another person without whom + * Suspend would not be where it is. + * + * ..and of course the myriads of Suspend2 users who have helped diagnose + * and fix bugs, made suggestions on how to improve the code, proofread + * documentation, and donated time and money. + * + * Thanks also to corporate sponsors: + * + * Cyclades.com. Nigel's employers from Dec 2004, who allow him to work on + * Suspend and PM related issues on company time. + * + * LinuxFund.org. Sponsored Nigel's work on Suspend for four months Oct 2003 + * to Jan 2004. + * + * LAC Linux. Donated P4 hardware that enabled development and ongoing + * maintenance of SMP and Highmem support. + * + * OSDL. Provided access to various hardware configurations, make occasional + * small donations to the project. + */ + +#define SUSPEND_MAIN_C + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* For LRU unlinking prototypes */ +#include +#include +#include + +#include "suspend.h" +#include "modules.h" +#include "sysfs.h" +#include "pageflags.h" +#include "prepare_image.h" +#include "io.h" +#include "ui.h" +#include "extent.h" +#include "power_off.h" +#include "storage.h" + +#ifdef CONFIG_X86 +#include /* for kernel_fpu_end */ +#endif + +int suspend2_running; + +/* Variables to be preserved over suspend */ +long pageset1_sizelow = 0, pageset2_sizelow = 0, image_size_limit = 0; +unsigned long suspend_orig_mem_free = 0; + +static char suspend_core_version[] = SUSPEND_CORE_VERSION; + +extern __nosavedata char suspend_resume_commandline[COMMAND_LINE_SIZE]; + +#ifdef CONFIG_SUSPEND2_REPLACE_SWSUSP +unsigned long suspend_action = 1 << SUSPEND_REPLACE_SWSUSP; +#else +unsigned long suspend_action = 0; +#endif +unsigned long suspend_result = 0; +unsigned long suspend_debug_state = 0; + +int suspend2_in_suspend __nosavedata; +extern void copyback_post(void); +extern int suspend2_suspend(void); +extern int extra_pd1_pages_used; + +static int orig_system_state; +extern int driver_model_beeping; + +/* + * --- Variables ----- + * + * The following are used by the arch specific low level routines + * and only needed if suspend2 is compiled in. Other variables, + * used by the freezer even if suspend2 is not compiled in, are + * found in process.c + */ + +/*! How long I/O took. */ +int suspend_io_time[2][2]; + +/* Compression ratio */ +__nosavedata unsigned long bytes_in = 0, bytes_out = 0; + +/*! Pageset metadata. */ +struct pagedir pagedir1 = { 0, 0}, pagedir2 = { 0, 0}; + +/* Suspend2 variables used by built-in routines. */ + +/*! The number of suspends we have started (some may have been cancelled) */ +unsigned int nr_suspends = 0; + +/* + * For resume2= kernel option. It's pointless to compile + * suspend2 without any writers, but compilation shouldn't + * fail if you do. + */ + +unsigned long suspend_state = ((1 << SUSPEND_BOOT_TIME) | + (1 << SUSPEND_RESUME_NOT_DONE) | (1 << SUSPEND_IGNORE_LOGLEVEL)); + +mm_segment_t oldfs; + +char resume2_file[256] = CONFIG_SUSPEND2_DEFAULT_RESUME2; +#ifdef CONFIG_SOFTWARE_SUSPEND +extern char resume_file[256]; +#endif + +static atomic_t actions_running; + +extern int block_dump; + +int block_dump_save; + +/* + * Basic clean-up routine. + */ +void suspend_finish_anything(int finishing_cycle) +{ + if (atomic_dec_and_test(&actions_running)) { + suspend_cleanup_modules(finishing_cycle); + suspend_put_modules(); + clear_suspend_state(SUSPEND_RUNNING); + } + + set_fs(oldfs); + + if (finishing_cycle) + block_dump = block_dump_save; +} + +/* + * Basic set-up routine. + */ +int suspend_start_anything(int starting_cycle) +{ + oldfs = get_fs(); + + if (atomic_add_return(1, &actions_running) == 1) { + set_fs(KERNEL_DS); + + set_suspend_state(SUSPEND_RUNNING); + + if (suspend_get_modules()) { + printk(name_suspend "Get modules failed!\n"); + clear_suspend_state(SUSPEND_RUNNING); + set_fs(oldfs); + return -EBUSY; + } + + if (suspend_initialise_modules(starting_cycle)) { + printk(name_suspend "Initialise modules failed!\n"); + suspend_finish_anything(starting_cycle); + return -EBUSY; + } + + if (starting_cycle) { + block_dump_save = block_dump; + block_dump = 0; + } + } + + return 0; +} + +/* + * save_image + * Functionality : High level routine which performs the steps necessary + * to prepare and save the image after preparatory steps + * have been taken. + * Key Assumptions : Processes frozen, sufficient memory available, drivers + * suspended. + * Called from : suspend_suspend_2 + */ +static void save_image(void) +{ + int temp_result; + + suspend_message(SUSPEND_ANY_SECTION, SUSPEND_LOW, 1, + " - Final values: %d and %d.\n", + pagedir1.pageset_size, + pagedir2.pageset_size); + + suspend_cond_pause(1, "About to write pagedir2."); + + temp_result = write_pageset(&pagedir2, 2); + + if (temp_result == -1 || test_result_state(SUSPEND_ABORTED)) + goto backout; + + suspend_cond_pause(1, "About to copy pageset 1."); + + if (test_result_state(SUSPEND_ABORTED)) + goto backout; + + suspend_deactivate_storage(1); + + suspend_prepare_status(DONT_CLEAR_BAR, "Doing atomic copy."); + + suspend2_running = 1; /* For the swsusp code we use :< */ + + suspend2_in_suspend = 1; + + if (device_suspend(PMSG_FREEZE)) { + set_result_state(SUSPEND_DEVICE_REFUSED); + set_result_state(SUSPEND_ABORTED); + goto backout; + } + + if (suspend2_suspend()) { + device_resume(); + return; + } + + suspend2_running = 0; + + device_resume(); + + /* Resume time? */ + if (!suspend2_in_suspend) { + copyback_post(); + return; + } + + /* Nope. Suspending. So, see if we can save the image... */ + if (!save_image_part1()) { + suspend_power_down(); + + if (read_pageset2(1)) + panic("Attempt to reload pagedir 2 failed. Try rebooting."); + + if (!test_result_state(SUSPEND_ABORT_REQUESTED) && + !test_action_state(SUSPEND_TEST_FILTER_SPEED) && + !test_action_state(SUSPEND_TEST_BIO) && + suspend_powerdown_method != PM_SUSPEND_MEM) + printk(KERN_EMERG name_suspend + "Suspend failed, trying to recover...\n"); + barrier(); + mb(); + } + + return; + +backout: + return; +} + +/* + * Save the second part of the image. + */ +int save_image_part1(void) +{ + int temp_result; + + if (suspend_activate_storage(1)) + panic("Failed to reactivate our storage."); + + suspend_update_status(pagedir2.pageset_size, + pagedir1.pageset_size + pagedir2.pageset_size, + NULL); + + if (test_result_state(SUSPEND_ABORTED)) + goto abort_reloading_pagedir_two; + + suspend_cond_pause(1, "About to write pageset1."); + + /* + * End of critical section. + */ + + suspend_message(SUSPEND_ANY_SECTION, SUSPEND_LOW, 1, + "-- Writing pageset1\n"); + + temp_result = write_pageset(&pagedir1, 1); + + /* We didn't overwrite any memory, so no reread needs to be done. */ + if (test_action_state(SUSPEND_TEST_FILTER_SPEED)) + return -1; + + if (temp_result == -1 || test_result_state(SUSPEND_ABORTED)) + goto abort_reloading_pagedir_two; + + suspend_cond_pause(1, "About to write header."); + + if (test_result_state(SUSPEND_ABORTED)) + goto abort_reloading_pagedir_two; + + temp_result = write_image_header(); + + if (test_action_state(SUSPEND_TEST_BIO)) + return -1; + + if (temp_result || (test_result_state(SUSPEND_ABORTED))) + goto abort_reloading_pagedir_two; + + suspend_cond_pause(1, "About to power down or reboot."); + + return 0; + +abort_reloading_pagedir_two: + temp_result = read_pageset2(1); + + /* If that failed, we're sunk. Panic! */ + if (temp_result) + panic("Attempt to reload pagedir 2 while aborting " + "a suspend failed."); + + return -1; + +} + +static int io_MB_per_second(int read_write) +{ + if (!suspend_io_time[read_write][1]) + return 0; + + return MB((unsigned long) suspend_io_time[read_write][0]) * HZ / + suspend_io_time[read_write][1]; +} + +/* get_debug_info + * Functionality: Store debug info in a buffer. + * Called from: suspend2_try_suspend. + */ +#define SNPRINTF(a...) len += snprintf_used(((char *)buffer) + len, \ + count - len - 1, ## a) +static int get_suspend_debug_info(const char *buffer, int count) +{ + int len = 0; + + SNPRINTF("Suspend2 debugging info:\n"); + SNPRINTF("- SUSPEND core : %s\n", SUSPEND_CORE_VERSION); + SNPRINTF("- Kernel Version : %s\n", UTS_RELEASE); + SNPRINTF("- Compiler vers. : %d.%d\n", __GNUC__, __GNUC_MINOR__); + SNPRINTF("- Attempt number : %d\n", nr_suspends); + SNPRINTF("- Parameters : %ld %ld %ld %d %d %ld\n", + suspend_result, + suspend_action, + suspend_debug_state, + suspend_default_console_level, + image_size_limit, + suspend_powerdown_method); + SNPRINTF("- Overall expected compression percentage: %d.\n", + 100 - suspend_expected_compression_ratio()); + len+= suspend_print_module_debug_info(((char *) buffer) + len, + PAGE_SIZE - len - 1); + if (suspend_io_time[0][1]) { + if ((io_MB_per_second(0) < 5) || (io_MB_per_second(1) < 5)) { + SNPRINTF("- I/O speed: Write %d KB/s", + (KB((unsigned long) suspend_io_time[0][0]) * HZ / + suspend_io_time[0][1])); + if (suspend_io_time[1][1]) + SNPRINTF(", Read %d KB/s", + (KB((unsigned long) suspend_io_time[1][0]) * HZ / + suspend_io_time[1][1])); + } else { + SNPRINTF("- I/O speed: Write %d MB/s", + (MB((unsigned long) suspend_io_time[0][0]) * HZ / + suspend_io_time[0][1])); + if (suspend_io_time[1][1]) + SNPRINTF(", Read %d MB/s", + (MB((unsigned long) suspend_io_time[1][0]) * HZ / + suspend_io_time[1][1])); + } + SNPRINTF(".\n"); + } + else + SNPRINTF("- No I/O speed stats available.\n"); + SNPRINTF("- Extra pages : %d used/%d.\n", + extra_pd1_pages_used, extra_pd1_pages_allowance); + + return len; +} + +static int allocate_bitmaps(void) +{ + if (allocate_dyn_pageflags(&in_use_map) || + allocate_dyn_pageflags(&pageset1_map) || + allocate_dyn_pageflags(&pageset1_copy_map) || + allocate_dyn_pageflags(&pageset2_map) || + allocate_dyn_pageflags(&pageset2_rw_map)) + return 1; + + return 0; +} + +static void free_metadata(void) +{ + free_dyn_pageflags(&pageset1_map); + free_dyn_pageflags(&pageset1_copy_map); + free_dyn_pageflags(&pageset2_map); + free_dyn_pageflags(&pageset2_rw_map); + free_dyn_pageflags(&in_use_map); +} + +static int check_still_keeping_image(void) +{ + if (test_action_state(SUSPEND_KEEP_IMAGE)) { + printk("Image already stored: powering down immediately."); + suspend_power_down(); + return 1; /* Just in case we're using S3 */ + } + + printk("Invalidating previous image.\n"); + suspend_active_writer->invalidate_image(); + + return 0; +} + +static int suspend_init(void) +{ + suspend_result = 0; + + printk(name_suspend "Initiating a software suspend cycle.\n"); + + nr_suspends++; + clear_suspend_state(SUSPEND_NOW_RESUMING); + + orig_system_state = system_state; + + suspend_io_time[0][0] = suspend_io_time[0][1] = + suspend_io_time[1][0] = + suspend_io_time[1][1] = 0; + + free_metadata(); /* We might have kept it */ + + if (!test_suspend_state(SUSPEND_CAN_SUSPEND)) + return 0; + + if (allocate_bitmaps()) + return 0; + + suspend_prepare_console(); + disable_nonboot_cpus(); + + return 1; +} + +void suspend_cleanup(int had_pmsem) +{ + int i = 0; + char *buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + if (buffer) + i = get_suspend_debug_info(buffer, PAGE_SIZE); + + suspend_free_extra_pagedir_memory(); + + pagedir1.pageset_size = pagedir2.pageset_size = 0; + + system_state = orig_system_state; + + thaw_processes(FREEZER_KERNEL_THREADS); + +#ifdef CONFIG_SUSPEND2_KEEP_IMAGE + if (test_action_state(SUSPEND_KEEP_IMAGE) && + !test_result_state(SUSPEND_ABORTED)) { + suspend_message(SUSPEND_ANY_SECTION, SUSPEND_LOW, 1, + name_suspend "Not invalidating the image due " + "to Keep Image being enabled.\n"); + set_result_state(SUSPEND_KEPT_IMAGE); + } else +#endif + if (suspend_active_writer) + suspend_active_writer->invalidate_image(); + + free_metadata(); + + if (buffer && i) { + /* Printk can only handle 1023 bytes, including + * its level mangling. */ + for (i = 0; i < 3; i++) + printk("%s", buffer + (1023 * i)); + free_page((unsigned long) buffer); + buffer = NULL; + } + + thaw_processes(FREEZER_ALL_THREADS); + enable_nonboot_cpus(); + suspend_cleanup_console(); + suspend2_running = 0; + if (!had_pmsem) + up(&pm_sem); +} + +static int can_suspend(int had_pmsem) +{ + if (!had_pmsem && down_trylock(&pm_sem)) { + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_PM_SEM); + return 0; + } + + if (!test_suspend_state(SUSPEND_CAN_SUSPEND)) + suspend_attempt_to_parse_resume_device(); + + if (!test_suspend_state(SUSPEND_CAN_SUSPEND)) { + printk(name_suspend "Software suspend is disabled.\n" + "This may be because you haven't put something along " + "the lines of\n\nresume2=swap:/dev/hda1\n\n" + "in lilo.conf or equivalent. (Where /dev/hda1 is your " + "swap partition).\n"); + set_result_state(SUSPEND_ABORTED); + if (!had_pmsem) + up(&pm_sem); + return 0; + } + + return 1; +} + +/* + * suspend_main + * Functionality : First level of code for software suspend invocations. + * Stores and restores load averages (to avoid a spike), + * allocates bitmaps, freezes processes and eats memory + * as required before suspending drivers and invoking + * the 'low level' code to save the state to disk. + * By the time we return from do_suspend2_suspend, we + * have either failed to save the image or successfully + * suspended and reloaded the image. The difference can + * be discerned by checking SUSPEND_ABORTED. + * Called From : + */ +void suspend_main(int had_pmsem) +{ + if (suspend_activate_storage(0)) + return; + + if (!can_suspend(had_pmsem)) + goto cleanup_deactivate_storage; + + /* + * If kept image and still keeping image and suspending to RAM, we will + * return 1 after suspending and resuming (provided the power doesn't + * run out. + */ + if (test_result_state(SUSPEND_KEPT_IMAGE) && check_still_keeping_image()) + goto cleanup; + + + if (suspend_init() && !suspend_prepare_image() && !test_result_state(SUSPEND_ABORTED) && + !test_action_state(SUSPEND_FREEZER_TEST)) { + suspend_prepare_status(DONT_CLEAR_BAR, "Starting to save the image.."); + unlink_lru_lists(); + save_image(); + relink_lru_lists(); + } + +cleanup: + suspend_cleanup(had_pmsem); +cleanup_deactivate_storage: + suspend_deactivate_storage(0); +} + +/* image_exists_read + * + * Return 0 or 1, depending on whether an image is found. + * Incoming buffer is PAGE_SIZE and result is guaranteed + * to be far less than that, so we don't worry about + * overflow. + */ +static int image_exists_read(const char *page, int count) +{ + int len = 0; + char *result; + + if (suspend_activate_storage(0)) + return count; + + if (!test_suspend_state(SUSPEND_RESUME_DEVICE_OK)) + suspend_attempt_to_parse_resume_device(); + + if (!suspend_active_writer) { + len = sprintf((char *) page, "-1\n"); + } else { + result = get_have_image_data(); + if (result) { + len = sprintf((char *) page, "%s", result); + free_page((unsigned long) result); + } + } + + suspend_deactivate_storage(0); + + return len; +} + +/* image_exists_write + * + * Invalidate an image if one exists. + */ +static int image_exists_write(const char *buffer, int count) +{ + if (suspend_activate_storage(0)) + return count; + + if (suspend_active_writer && suspend_active_writer->image_exists()) + suspend_active_writer->invalidate_image(); + + suspend_deactivate_storage(0); + + return count; +} + +/* + * Core sysfs entries that aren't built in. + * + * This array contains entries that are automatically registered at + * boot. Modules and the console code register their own entries separately. + */ +static struct suspend_sysfs_data sysfs_params[] = { + { SUSPEND2_ATTR("driver_model_beeping", SYSFS_RW), + SYSFS_INT(&driver_model_beeping, 0, 1) + }, + + { SUSPEND2_ATTR("debug_info", SYSFS_READONLY), + SYSFS_CUSTOM(get_suspend_debug_info, NULL, SYSFS_SM_NOT_NEEDED) + }, + + { SUSPEND2_ATTR("extra_pages_allowance", SYSFS_RW), + SYSFS_UL(&extra_pd1_pages_allowance, 0, INT_MAX) + }, + + { SUSPEND2_ATTR("ignore_rootfs", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_IGNORE_ROOTFS) + }, + + { SUSPEND2_ATTR("image_exists", SYSFS_RW), + SYSFS_CUSTOM(image_exists_read, image_exists_write, + SYSFS_NEEDS_FOR_BOTH) + }, + + { SUSPEND2_ATTR("image_size_limit", SYSFS_RW), + SYSFS_LONG(&image_size_limit, -2, INT_MAX) + }, + + { SUSPEND2_ATTR("last_result", SYSFS_READONLY), + SYSFS_UL(&suspend_result, 0, 0) + }, + + { SUSPEND2_ATTR("reboot", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_REBOOT) + }, + +#ifdef CONFIG_SOFTWARE_SUSPEND + { SUSPEND2_ATTR("replace_swsusp", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_REPLACE_SWSUSP) + }, +#endif + + { SUSPEND2_ATTR("resume2", SYSFS_RW), + SYSFS_STRING(resume2_file, 255, SYSFS_NEEDS_FOR_WRITE), + .write_side_effect = attempt_to_parse_resume_device2, + }, + + { SUSPEND2_ATTR("resume_commandline", SYSFS_RW), + SYSFS_STRING(suspend_resume_commandline, COMMAND_LINE_SIZE, + SYSFS_SM_NOT_NEEDED) + }, + + { SUSPEND2_ATTR("version", SYSFS_READONLY), + SYSFS_STRING(suspend_core_version, 0, SYSFS_SM_NOT_NEEDED) + }, + +#ifdef CONFIG_PM_DEBUG + { SUSPEND2_ATTR("freezer_test", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_FREEZER_TEST) + }, + + { SUSPEND2_ATTR("test_bio", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_TEST_BIO) + }, + + { SUSPEND2_ATTR("test_filter_speed", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_TEST_FILTER_SPEED) + }, + + { SUSPEND2_ATTR("slow", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_SLOW) + }, + + { SUSPEND2_ATTR("no_pageset2", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_NO_PAGESET2) + }, +#endif + +#if defined(CONFIG_ACPI) + { SUSPEND2_ATTR("powerdown_method", SYSFS_RW), + SYSFS_UL(&suspend_powerdown_method, 0, 5) + }, +#endif + +#ifdef CONFIG_SUSPEND2_KEEP_IMAGE + { SUSPEND2_ATTR("keep_image", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_KEEP_IMAGE) + }, +#endif +}; + +static __init int core_load(void) +{ + int i, + numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + printk("Suspend2 Core.\n"); + + suspend_initialise_module_lists(); + + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(&suspend2_subsys.kset.kobj, + &sysfs_params[i]); + +#ifdef CONFIG_SOFTWARE_SUSPEND + /* Overriding resume2= with resume=? */ + if (test_action_state(SUSPEND_REPLACE_SWSUSP) && resume_file[0]) + strncpy(resume2_file, resume_file, 256); +#endif + + return 0; +} + +/* + * Called from init kernel_thread. + * We check if we have an image and if so we try to resume. + * We also start ksuspendd if configuration looks right. + */ +int suspend_resume(void) +{ + int read_image_result = 0; + + if (sizeof(swp_entry_t) != sizeof(long)) { + printk(KERN_WARNING name_suspend + "The size of swp_entry_t != size of long. " + "Please report this!\n"); + return 1; + } + + if (!resume2_file[0]) + printk(KERN_WARNING name_suspend + "You need to use a resume2= command line parameter to " + "tell Suspend2 where to look for an image.\n"); + + suspend_activate_storage(0); + + if (!(test_suspend_state(SUSPEND_RESUME_DEVICE_OK)) && + !suspend_attempt_to_parse_resume_device()) { + /* + * Without a usable storage device we can do nothing - + * even if noresume is given + */ + + if (!suspend_num_writers) + printk(KERN_ALERT name_suspend + "No writers have been registered.\n"); + else + printk(KERN_ALERT name_suspend + "Missing or invalid storage location " + "(resume2= parameter). Please correct and " + "rerun lilo (or equivalent) before " + "suspending.\n"); + suspend_deactivate_storage(0); + return 1; + } + + suspend_orig_mem_free = real_nr_free_pages(); + + read_image_result = read_pageset1(); /* non fatal error ignored */ + + if (test_suspend_state(SUSPEND_NORESUME_SPECIFIED)) + printk(KERN_WARNING name_suspend "Resuming disabled as requested.\n"); + + suspend_deactivate_storage(0); + + if (read_image_result) { + printk(KERN_WARNING name_suspend "Read image returned %d.\n", read_image_result); + return 1; + } + + suspend2_running = 1; + + suspend_atomic_restore(); + + BUG(); + + return 0; +} + +/* -- Functions for kickstarting a suspend or resume --- */ + +/* + * Check if we have an image and if so try to resume. + */ +void __suspend_try_resume(void) +{ + set_suspend_state(SUSPEND_TRYING_TO_RESUME); + clear_suspend_state(SUSPEND_RESUME_NOT_DONE); + + suspend_resume(); + + clear_suspend_state(SUSPEND_IGNORE_LOGLEVEL); + clear_suspend_state(SUSPEND_TRYING_TO_RESUME); +} + +/* Wrapper for when called from init/do_mounts.c */ +void suspend2_try_resume(void) +{ + if (suspend_start_anything(0)) + return; + + down(&pm_sem); + __suspend_try_resume(); + + /* + * For initramfs, we have to clear the boot time + * flag after trying to resume + */ + clear_suspend_state(SUSPEND_BOOT_TIME); + + up(&pm_sem); + + suspend_finish_anything(0); +} + +/* + * suspend2_try_suspend + * Functionality : Wrapper around suspend_main. + * Called From : drivers/acpi/sleep/main.c + * kernel/reboot.c + */ +void suspend2_try_suspend(int have_pmsem) +{ + if (suspend_start_anything(0)) + return; + + suspend_main(0); + + suspend_finish_anything(0); +} + +/* -- Commandline Parameter Handling --- + * + * Resume setup: obtain the storage device. + */ +static int __init resume2_setup(char *str) +{ + if (!*str) + return 0; + + strncpy(resume2_file, str, 255); + return 0; +} + +/* + * Allow the user to set the debug parameter from lilo, prior to resuming. + */ +/* + * Allow the user to specify that we should ignore any image found and + * invalidate the image if necesssary. This is equivalent to running + * the task queue and a sync and then turning off the power. The same + * precautions should be taken: fsck if you're not journalled. + */ +static int __init noresume2_setup(char *str) +{ + set_suspend_state(SUSPEND_NORESUME_SPECIFIED); + return 0; +} + +static int __init suspend_retry_resume_setup(char *str) +{ + set_suspend_state(SUSPEND_RETRY_RESUME); + return 0; +} + +__setup("noresume2", noresume2_setup); +__setup("resume2=", resume2_setup); +__setup("suspend_retry_resume", suspend_retry_resume_setup); + +late_initcall(core_load); +EXPORT_SYMBOL(suspend_state); diff -urN oldtree/kernel/power/suspend.h newtree/kernel/power/suspend.h --- oldtree/kernel/power/suspend.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/suspend.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,62 @@ +/* + * kernel/power/suspend.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It contains declarations used throughout swsusp. + * + */ + +#ifndef KERNEL_POWER_SUSPEND_H +#define KERNEL_POWER_SUSPEND_H + +#include +#include + +#define SUSPEND_CORE_VERSION "2.2.8" +#define name_suspend "Suspend2 " SUSPEND_CORE_VERSION ": " + +extern __nosavedata struct pbe *restore_pblist; +extern int save_image_part1(void); +extern int suspend_atomic_restore(void); +extern void suspend_main(int had_pmsem); + +extern unsigned long suspend_orig_mem_free; + +#define KB(x) ((x) << (PAGE_SHIFT - 10)) +#define MB(x) ((x) >> (20 - PAGE_SHIFT)) + +extern int suspend_start_anything(int starting_cycle); +extern void suspend_finish_anything(int finishing_cycle); + +#ifdef CONFIG_PM_DEBUG +#define set_debug_state(bit) (test_and_set_bit(bit, &suspend_debug_state)) +#define clear_debug_state(bit) (test_and_clear_bit(bit, &suspend_debug_state)) +#else +#define set_debug_state(bit) (0) +#define clear_debug_state(bit) (0) +#endif + +#define set_result_state(bit) (test_and_set_bit(bit, &suspend_result)) +#define clear_result_state(bit) (test_and_clear_bit(bit, &suspend_result)) + +enum { + SUSPEND_ABORTED, + SUSPEND_ABORT_REQUESTED, + SUSPEND_NOSTORAGE_AVAILABLE, + SUSPEND_INSUFFICIENT_STORAGE, + SUSPEND_FREEZING_FAILED, + SUSPEND_UNEXPECTED_ALLOC, + SUSPEND_KEPT_IMAGE, + SUSPEND_WOULD_EAT_MEMORY, + SUSPEND_UNABLE_TO_FREE_ENOUGH_MEMORY, + SUSPEND_ENCRYPTION_SETUP_FAILED, + SUSPEND_PM_SEM, + SUSPEND_DEVICE_REFUSED, +}; + +extern unsigned int nr_suspends; +extern char resume2_file[256]; +#endif diff -urN oldtree/kernel/power/suspend_block_io.c newtree/kernel/power/suspend_block_io.c --- oldtree/kernel/power/suspend_block_io.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/suspend_block_io.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,1085 @@ +/* + * kernel/power/suspend_block_io.c + * + * Copyright 2004-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * This file contains block io functions for suspend2. These are + * used by the swapwriter and it is planned that they will also + * be used by the NFSwriter. + * + */ + +#include +#include +#include + +#include "suspend.h" +#include "sysfs.h" +#include "modules.h" +#include "prepare_image.h" +#include "block_io.h" +#include "ui.h" + +/* Bits in struct io_info->flags */ +enum { + IO_AWAITING_READ, + IO_AWAITING_SUBMIT, + IO_AWAITING_CLEANUP, +}; + +#define MAX_OUTSTANDING_IO 2048 + +/* + * + * IO in progress information storage and helpers + * + */ + +struct io_info { + struct bio *sys_struct; + sector_t block[MAX_BUF_PER_PAGE]; + struct page *buffer_page; + struct page *data_page; + unsigned long flags; + struct block_device *dev; + struct list_head list; + int readahead_index; +}; + +/* + * submit_params + */ +struct submit_params { + swp_entry_t swap_address; + struct page *page; + struct block_device *dev; + sector_t block[MAX_BUF_PER_PAGE]; + int readahead_index; + struct submit_params *next; + int printme; +}; + +/* Locks separated to allow better SMP support. + * An io_struct moves through the lists as follows. + * free -> submit_batch -> busy -> ready_for_cleanup -> free + */ +static LIST_HEAD(ioinfo_free); +static DEFINE_SPINLOCK(ioinfo_free_lock); + +static LIST_HEAD(ioinfo_ready_for_cleanup); +static DEFINE_SPINLOCK(ioinfo_ready_lock); + +static LIST_HEAD(ioinfo_submit_batch); +static DEFINE_SPINLOCK(ioinfo_submit_lock); + +static LIST_HEAD(ioinfo_busy); +static DEFINE_SPINLOCK(ioinfo_busy_lock); + +static atomic_t submit_batch; +static int submit_batch_size = 64; +static int submit_batched(void); + +/* [Max] number of I/O operations pending */ +static atomic_t outstanding_io; +static int max_outstanding_io = 0; +static atomic_t buffer_allocs, buffer_frees; + +/* [Max] number of pages used for above struct */ +static int infopages = 0; +static int maxinfopages = 0; + +static int extra_page_forward = 0; + +static volatile unsigned long suspend_readahead_flags[ + (MAX_OUTSTANDING_IO + BITS_PER_LONG - 1) / BITS_PER_LONG]; +static spinlock_t suspend_readahead_flags_lock = SPIN_LOCK_UNLOCKED; +static struct page *suspend_readahead_pages[MAX_OUTSTANDING_IO]; +static int readahead_index, readahead_submit_index; + +static int current_stream; +struct extent_iterate_saved_state suspend_writer_posn_save[3]; + +/* Pointer to current entry being loaded/saved. */ +struct extent_iterate_state suspend_writer_posn; + +/* Not static, so that the allocators can setup and complete + * writing the header */ +char *suspend_writer_buffer; +int suspend_writer_buffer_posn; + +int suspend_read_fd; + +static unsigned long nr_schedule_calls[8]; + +static char *sch_caller[] = { + "get_io_info_struct #1 ", + "get_io_info_struct #2 ", + "get_io_info_struct #3 ", + "suspend_finish_all_io ", + "wait_on_one_page ", + "submit ", + "start_one ", + "suspend_wait_on_readahead", +}; + +static struct suspend_bdev_info *suspend_devinfo; + +int suspend_header_bytes_used = 0; + +/* + * suspend_reset_io_stats + * + * Description: Reset all our sanity-checking statistics. + */ +static void suspend_reset_io_stats(void) +{ + int i; + + max_outstanding_io = 0; + maxinfopages = 0; + + for (i = 0; i < 8; i++) + nr_schedule_calls[i] = 0; +} + +/* + * suspend_check_io_stats + * + * Description: Check that our statistics look right and print + * any debugging info wanted. + */ +static void suspend_check_io_stats(void) +{ + int i; + + BUG_ON(atomic_read(&outstanding_io)); + BUG_ON(infopages); + BUG_ON(!list_empty(&ioinfo_submit_batch)); + BUG_ON(!list_empty(&ioinfo_busy)); + BUG_ON(!list_empty(&ioinfo_ready_for_cleanup)); + BUG_ON(!list_empty(&ioinfo_free)); + BUG_ON(atomic_read(&buffer_allocs) != atomic_read(&buffer_frees)); + + suspend_message(SUSPEND_WRITER, SUSPEND_LOW, 0, + "Maximum outstanding_io was %d.\n", + max_outstanding_io); + suspend_message(SUSPEND_WRITER, SUSPEND_LOW, 0, + "Max info pages was %d.\n", + maxinfopages); + if (atomic_read(&buffer_allocs) != atomic_read(&buffer_frees)) + suspend_message(SUSPEND_WRITER, SUSPEND_MEDIUM, 0, + "Buffer allocs (%d) != buffer frees (%d)", + atomic_read(&buffer_allocs), + atomic_read(&buffer_frees)); + for(i = 0; i < 8; i++) + suspend_message(SUSPEND_WRITER, SUSPEND_MEDIUM, 0, + "Nr schedule calls %s: %lu.\n", sch_caller[i], + nr_schedule_calls[i]); +} + +/* + * __suspend_bio_cleanup_one + * + * Description: Clean up after completing I/O on a page. + * Arguments: struct io_info: Data for I/O to be completed. + */ +static void __suspend_bio_cleanup_one(struct io_info *io_info) +{ + struct page *buffer_page; + struct page *data_page; + char *buffer_address, *data_address; + int reading; + + buffer_page = io_info->buffer_page; + data_page = io_info->data_page; + + reading = test_bit(IO_AWAITING_READ, &io_info->flags); + suspend_message(SUSPEND_WRITER, SUSPEND_HIGH, 0, + "Cleanup IO: [%p]\n", + io_info); + + if (reading && io_info->readahead_index == -1) { + /* + * Copy the page we read into the buffer our caller provided. + */ + data_address = (char *) kmap(data_page); + buffer_address = (char *) kmap(buffer_page); + memcpy(data_address, buffer_address, PAGE_SIZE); + kunmap(data_page); + kunmap(buffer_page); + + } + + if (!reading || io_info->readahead_index == -1) { + /* Sanity check */ + if (page_count(buffer_page) != 2) + printk(KERN_EMERG "Cleanup IO: Page count on page %p" + " is %d. Not good!\n", + buffer_page, page_count(buffer_page)); + put_page(buffer_page); + __free_page(buffer_page); + atomic_inc(&buffer_frees); + } else + put_page(buffer_page); + + bio_put(io_info->sys_struct); + io_info->sys_struct = NULL; + io_info->flags = 0; +} + +/* __suspend_io_cleanup + */ + +static int suspend_bio_cleanup_one(void *data) +{ + struct io_info *io_info = (struct io_info *) data; + int readahead_index; + unsigned long flags; + + /* + * If this I/O was a readahead, remember its index. + */ + readahead_index = io_info->readahead_index; + + /* + * Add it to the free list. + */ + list_del_init(&io_info->list); + + /* + * Do the cleanup. + */ + __suspend_bio_cleanup_one(io_info); + + /* + * Record the readahead as done. + */ + if (readahead_index > -1) { + int index = readahead_index/BITS_PER_LONG; + int bit = readahead_index - (index * BITS_PER_LONG); + spin_lock_irqsave(&suspend_readahead_flags_lock, flags); + set_bit(bit, &suspend_readahead_flags[index]); + spin_unlock_irqrestore(&suspend_readahead_flags_lock, flags); + } + + spin_lock_irqsave(&ioinfo_free_lock, flags); + list_add_tail(&io_info->list, &ioinfo_free); + spin_unlock_irqrestore(&ioinfo_free_lock, flags); + + /* Important: Must be last thing we do to avoid a race with + * finish_all_io when using keventd to do the cleanup */ + atomic_dec(&outstanding_io); + + return 0; +} + +/* suspend_cleanup_some_completed_io + * + * NB: This is designed so that multiple callers can be in here simultaneously. + */ + +static void suspend_cleanup_some_completed_io(void) +{ + int num_cleaned = 0; + struct io_info *first; + unsigned long flags; + + spin_lock_irqsave(&ioinfo_ready_lock, flags); + while(!list_empty(&ioinfo_ready_for_cleanup)) { + int result; + first = list_entry(ioinfo_ready_for_cleanup.next, + struct io_info, list); + + BUG_ON(!test_and_clear_bit(IO_AWAITING_CLEANUP, &first->flags)); + + list_del_init(&first->list); + + spin_unlock_irqrestore(&ioinfo_ready_lock, flags); + + result = suspend_bio_cleanup_one((void *) first); + + spin_lock_irqsave(&ioinfo_ready_lock, flags); + if (result) + continue; + num_cleaned++; + if (num_cleaned == submit_batch_size) + break; + } + spin_unlock_irqrestore(&ioinfo_ready_lock, flags); +} + +/* do_bio_wait + * + * Actions taken when we want some I/O to get run. + * + * Submit any I/O that's batched up (if we're not already doing + * that, unplug queues, schedule and clean up whatever we can. + */ +static void do_bio_wait(int caller) +{ + int num_submitted = 0; + + nr_schedule_calls[caller]++; + + /* Don't want to wait on I/O we haven't submitted! */ + num_submitted = submit_batched(); + + kblockd_flush(); + + io_schedule(); + + suspend_cleanup_some_completed_io(); +} + +/* + * suspend_finish_all_io + * + * Description: Finishes all IO and frees all IO info struct pages. + */ +static void suspend_finish_all_io(void) +{ + struct io_info *this, *next = NULL; + unsigned long flags; + + /* Wait for all I/O to complete. */ + while (atomic_read(&outstanding_io)) + do_bio_wait(2); + + spin_lock_irqsave(&ioinfo_free_lock, flags); + + /* + * Two stages, to avoid using freed pages. + * + * First free all io_info structs on a page except the first. + */ + list_for_each_entry_safe(this, next, &ioinfo_free, list) { + if (((unsigned long) this) & ~PAGE_MASK) + list_del(&this->list); + } + + /* + * Now we have only one reference to each page, and can safely + * free pages, knowing we're not going to be trying to access the + * same page after freeing it. + */ + list_for_each_entry_safe(this, next, &ioinfo_free, list) { + list_del(&this->list); + free_page((unsigned long) this); + infopages--; + suspend_message(SUSPEND_MEMORY, SUSPEND_VERBOSE, 0, + "[FreedIOPage %lx]", this); + } + + spin_unlock_irqrestore(&ioinfo_free_lock, flags); +} + +/* + * wait_on_one_page + * + * Description: Wait for a particular I/O to complete. + */ +static void wait_on_one_page(struct io_info *io_info) +{ + do { do_bio_wait(3); } while (io_info->flags); +} + +/* + * wait_on_readahead + * + * Wait until a particular readahead is ready. + */ +static void suspend_wait_on_readahead(int readahead_index) +{ + int index = readahead_index / BITS_PER_LONG; + int bit = readahead_index - index * BITS_PER_LONG; + + /* read_ahead_index is the one we want to return */ + while (!test_bit(bit, &suspend_readahead_flags[index])) + do_bio_wait(6); +} + +/* + * readahead_done + * + * Returns whether the readahead requested is ready. + */ + +static int suspend_readahead_ready(int readahead_index) +{ + int index = readahead_index / BITS_PER_LONG; + int bit = readahead_index - (index * BITS_PER_LONG); + + return test_bit(bit, &suspend_readahead_flags[index]); +} + +/* suspend_readahead_prepare + * Set up for doing readahead on an image */ +static int suspend_prepare_readahead(int index) +{ + unsigned long new_page = get_zeroed_page(GFP_ATOMIC); + + if(!new_page) + return -ENOMEM; + + suspend_readahead_pages[index] = virt_to_page(new_page); + return 0; +} + +/* suspend_readahead_cleanup + * Clean up structures used for readahead */ +static void suspend_cleanup_readahead(int page) +{ + __free_page(suspend_readahead_pages[page]); + suspend_readahead_pages[page] = 0; + return; +} + +/* + * suspend_end_bio + * + * Description: Function called by block driver from interrupt context when I/O + * is completed. This is the reason we use spinlocks in + * manipulating the io_info lists. + * Nearly the fs/buffer.c version, but we want to mark the page as + * done in our own structures too. + */ + +static int suspend_end_bio(struct bio *bio, unsigned int num, int err) +{ + struct io_info *io_info = bio->bi_private; + unsigned long flags; + + spin_lock_irqsave(&ioinfo_busy_lock, flags); + list_del_init(&io_info->list); + spin_unlock_irqrestore(&ioinfo_busy_lock, flags); + + set_bit(IO_AWAITING_CLEANUP, &io_info->flags); + + spin_lock_irqsave(&ioinfo_ready_lock, flags); + list_add_tail(&io_info->list, &ioinfo_ready_for_cleanup); + spin_unlock_irqrestore(&ioinfo_ready_lock, flags); + return 0; +} + +/** + * submit - submit BIO request. + * @rw: READ or WRITE. + * @io_info: IO info structure. + * + * Based on Patrick's pmdisk code from long ago: + * "Straight from the textbook - allocate and initialize the bio. + * If we're writing, make sure the page is marked as dirty. + * Then submit it and carry on." + * + * With a twist, though - we handle block_size != PAGE_SIZE. + * Caller has already checked that our page is not fragmented. + */ + +static int submit(int rw, struct io_info *io_info) +{ + int error = 0; + struct bio *bio = NULL; + unsigned long flags; + + while (!bio) { + bio = bio_alloc(GFP_ATOMIC,1); + if (!bio) + do_bio_wait(4); + } + + bio->bi_bdev = io_info->dev; + bio->bi_sector = io_info->block[0]; + bio->bi_private = io_info; + bio->bi_end_io = suspend_end_bio; + bio->bi_flags |= (1 << BIO_SUSPEND_DATA); + io_info->sys_struct = bio; + + if (bio_add_page(bio, io_info->buffer_page, PAGE_SIZE, 0) < PAGE_SIZE) { + printk("ERROR: adding page to bio at %lld\n", + (unsigned long long) io_info->block[0]); + bio_put(bio); + return -EFAULT; + } + + if (rw == WRITE) + bio_set_pages_dirty(bio); + + spin_lock_irqsave(&ioinfo_busy_lock, flags); + list_add_tail(&io_info->list, &ioinfo_busy); + spin_unlock_irqrestore(&ioinfo_busy_lock, flags); + + submit_bio(rw,bio); + + return error; +} + +/* + * submit a batch. The submit function can wait on I/O, so we have + * simple locking to avoid infinite recursion. + */ +static int submit_batched(void) +{ + static int running_already = 0; + struct io_info *first; + unsigned long flags; + int num_submitted = 0; + + running_already = 1; + spin_lock_irqsave(&ioinfo_submit_lock, flags); + while(!list_empty(&ioinfo_submit_batch)) { + first = list_entry(ioinfo_submit_batch.next, struct io_info, + list); + + BUG_ON(!test_and_clear_bit(IO_AWAITING_SUBMIT, &first->flags)); + + list_del_init(&first->list); + + atomic_dec(&submit_batch); + + spin_unlock_irqrestore(&ioinfo_submit_lock, flags); + + if (test_bit(IO_AWAITING_READ, &first->flags)) + submit(READ, first); + else + submit(WRITE, first); + + spin_lock_irqsave(&ioinfo_submit_lock, flags); + + num_submitted++; + if (num_submitted == submit_batch_size) + break; + } + spin_unlock_irqrestore(&ioinfo_submit_lock, flags); + running_already = 0; + + return num_submitted; +} + +static void add_to_batch(struct io_info *io_info) +{ + unsigned long flags; + + set_bit(IO_AWAITING_SUBMIT, &io_info->flags); + + /* Put our prepared I/O struct on the batch list. */ + spin_lock_irqsave(&ioinfo_submit_lock, flags); + list_add_tail(&io_info->list, &ioinfo_submit_batch); + spin_unlock_irqrestore(&ioinfo_submit_lock, flags); + + atomic_inc(&submit_batch); + + if (atomic_read(&submit_batch) >= submit_batch_size) + submit_batched(); +} + +/* + * get_io_info_struct + * + * Description: Get an I/O struct. + * Returns: Pointer to the struct prepared for use. + */ +static struct io_info *get_io_info_struct(void) +{ + unsigned long newpage = 0, flags; + struct io_info *this = NULL; + int remaining = 0; + + do { + while (atomic_read(&outstanding_io) >= MAX_OUTSTANDING_IO) + do_bio_wait(0); + + /* Can start a new I/O. Is there a free one? */ + if (!list_empty(&ioinfo_free)) { + /* Yes. Grab it. */ + spin_lock_irqsave(&ioinfo_free_lock, flags); + break; + } + + /* No. Need to allocate a new page for I/O info structs. */ + newpage = get_zeroed_page(GFP_ATOMIC); + if (!newpage) { + do_bio_wait(1); + continue; + } + + suspend_message(SUSPEND_MEMORY, SUSPEND_VERBOSE, 0, + "[NewIOPage %lx]", newpage); + infopages++; + if (infopages > maxinfopages) + maxinfopages++; + + /* Prepare the new page for use. */ + this = (struct io_info *) newpage; + remaining = PAGE_SIZE; + spin_lock_irqsave(&ioinfo_free_lock, flags); + while (remaining >= (sizeof(struct io_info))) { + list_add_tail(&this->list, &ioinfo_free); + this = (struct io_info *) (((char *) this) + + sizeof(struct io_info)); + remaining -= sizeof(struct io_info); + } + break; + } while (1); + + /* + * We have an I/O info struct. Remove it from the free list. + * It will be added to the submit or busy list later. + */ + this = list_entry(ioinfo_free.next, struct io_info, list); + list_del_init(&this->list); + spin_unlock_irqrestore(&ioinfo_free_lock, flags); + return this; +} + +/* + * start_one + * + * Description: Prepare and start a read or write operation. + * Note that we use our own buffer for reading or writing. + * This simplifies doing readahead and asynchronous writing. + * We can begin a read without knowing the location into which + * the data will eventually be placed, and the buffer passed + * for a write can be reused immediately (essential for the + * modules system). + * Failure? What's that? + * Returns: The io_info struct created. + */ +static struct io_info *start_one(int rw, struct submit_params *submit_info) +{ + struct io_info *io_info = get_io_info_struct(); + unsigned long buffer_virt = 0; + char *to, *from; + struct page *buffer_page; + + if (!io_info) + return NULL; + + /* Get our local buffer */ + suspend_message(SUSPEND_WRITER, SUSPEND_HIGH, 1, + "Start_IO: [%p]", io_info); + + /* Copy settings to the io_info struct */ + io_info->data_page = submit_info->page; + io_info->readahead_index = submit_info->readahead_index; + + if (io_info->readahead_index == -1) { + while (!(buffer_virt = get_zeroed_page(GFP_ATOMIC))) + do_bio_wait(5); + + atomic_inc(&buffer_allocs); + suspend_message(SUSPEND_WRITER, SUSPEND_HIGH, 0, + "[ALLOC BUFFER]->%d", + real_nr_free_pages()); + buffer_page = virt_to_page(buffer_virt); + + io_info->buffer_page = buffer_page; + } else { + unsigned long flags; + int index = io_info->readahead_index / BITS_PER_LONG; + int bit = io_info->readahead_index - index * BITS_PER_LONG; + + spin_lock_irqsave(&suspend_readahead_flags_lock, flags); + clear_bit(bit, &suspend_readahead_flags[index]); + spin_unlock_irqrestore(&suspend_readahead_flags_lock, flags); + + io_info->buffer_page = buffer_page = submit_info->page; + } + + /* If writing, copy our data. The data is probably in + * lowmem, but we cannot be certain. If there is no + * compression/encryption, we might be passed the + * actual source page's address. */ + if (rw == WRITE) { + to = (char *) buffer_virt; + from = kmap_atomic(io_info->data_page, KM_USER1); + memcpy(to, from, PAGE_SIZE); + kunmap_atomic(from, KM_USER1); + } + + /* Submit the page */ + get_page(buffer_page); + + io_info->dev = submit_info->dev; + io_info->block[0] = submit_info->block[0]; + + if (rw == READ) + set_bit(IO_AWAITING_READ, &io_info->flags); + + suspend_message(SUSPEND_WRITER, SUSPEND_HIGH, 1, + "-> (PRE BRW) %d\n", + real_nr_free_pages()); + + if (submit_batch_size > 1) + add_to_batch(io_info); + else + submit(rw, io_info); + + atomic_inc(&outstanding_io); + if (atomic_read(&outstanding_io) > max_outstanding_io) + max_outstanding_io++; + + return io_info; +} + +static int suspend_do_io(int rw, + struct submit_params *submit_info, int syncio) +{ + struct io_info *io_info; + + if(!submit_info->dev) { + printk("Suspend_do_io: submit_info->dev is NULL!\n"); + return 1; + } + + io_info = start_one(rw, submit_info); + + if (!io_info) { + printk("Unable to allocate an io_info struct.\n"); + return 1; + } else if (syncio) + wait_on_one_page(io_info); + + /* If we were the only one, clean everything up */ + if (!atomic_read(&outstanding_io)) + suspend_finish_all_io(); + return 0; +} + +/* We used to use bread here, but it doesn't correctly handle + * blocksize != PAGE_SIZE. Now we create a submit_info to get the data we + * want and use our normal routines (synchronously). + */ + +static int suspend_bdev_page_io(int rw, struct block_device *bdev, long pos, + struct page *page) +{ + struct submit_params submit_info; + + if (!bdev) + return 0; + + submit_info.page = page; + submit_info.dev = bdev; + submit_info.block[0] = pos; + submit_info.readahead_index = -1; + return suspend_do_io(rw, &submit_info, 1); +} + +static unsigned long suspend_bio_memory_needed(void) +{ + /* We want to have at least enough memory so as to have + * MAX_OUTSTANDING_IO transactions on the fly at once. If we + * can to more, fine. */ + return (MAX_OUTSTANDING_IO * (PAGE_SIZE + sizeof(struct request) + + sizeof(struct bio) + sizeof(struct io_info))); +} + +static void suspend_set_devinfo(struct suspend_bdev_info *info) +{ + suspend_devinfo = info; +} + +static int forward_extra_blocks(void) +{ + int i; + + for (i = 1; i < suspend_devinfo[suspend_writer_posn.current_chain]. + blocks_per_page; i++) + suspend_extent_state_next(&suspend_writer_posn); + + if (suspend_extent_state_eof(&suspend_writer_posn)) { + printk("Extent state eof.\n"); + return -ENODATA; + } + + return 0; +} + +static int forward_one_page(void) +{ + int at_start = (suspend_writer_posn.current_chain == -1); + + /* Have to go forward one to ensure we're on the right chain, + * before we can know how many more blocks to skip.*/ + suspend_extent_state_next(&suspend_writer_posn); + + if (!at_start) + if (forward_extra_blocks()) + return -ENODATA; + + if (extra_page_forward) { + extra_page_forward = 0; + return forward_one_page(); + } + + return 0; +} + +static void set_extra_page_forward(void) +{ + extra_page_forward = 1; +} + +static int suspend_rw_page(int rw, struct page *page, + int readahead_index, int sync, int debug) +{ + int i, current_chain; + struct submit_params submit_params; + + if (test_action_state(SUSPEND_TEST_FILTER_SPEED)) + return 0; + + submit_params.readahead_index = readahead_index; + submit_params.page = page; + + if (forward_one_page()) { + printk("Failed to advance a page in the extent data.\n"); + return -ENODATA; + } + + current_chain = suspend_writer_posn.current_chain; + submit_params.dev = suspend_devinfo[current_chain].bdev; + submit_params.block[0] = suspend_writer_posn.current_offset << + suspend_devinfo[current_chain].bmap_shift; + + if (debug) + printk("%s: %lx:%lx.\n", rw ? "Write" : "Read", + (long) submit_params.dev->bd_dev, + (long) submit_params.block[0]); + + i = suspend_do_io(rw, &submit_params, sync); + + if (i) + return -EIO; + + return 0; +} + +static int suspend_bio_read_chunk(struct page *buffer_page, int sync) +{ + static int last_result; + unsigned long *virt; + + if (sync == SUSPEND_ASYNC) + return suspend_rw_page(READ, buffer_page, -1, sync, 0); + + /* Start new readahead while we wait for our page */ + if (readahead_index == -1) { + last_result = 0; + readahead_index = readahead_submit_index = 0; + } + + /* Start a new readahead? */ + if (last_result) { + /* We failed to submit a read, and have cleaned up + * all the readahead previously submitted */ + if (readahead_submit_index == readahead_index) + return -EPERM; + goto wait; + } + + do { + if (suspend_prepare_readahead(readahead_submit_index)) + break; + + last_result = suspend_rw_page( + READ, + suspend_readahead_pages[readahead_submit_index], + readahead_submit_index, SUSPEND_ASYNC, 0); + if (last_result) { + printk("Begin read chunk for page %d returned %d.\n", + readahead_submit_index, last_result); + suspend_cleanup_readahead(readahead_submit_index); + break; + } + + readahead_submit_index++; + + if (readahead_submit_index == MAX_OUTSTANDING_IO) + readahead_submit_index = 0; + + } while((!last_result) && (readahead_submit_index != readahead_index) && + (!suspend_readahead_ready(readahead_index))); + +wait: + suspend_wait_on_readahead(readahead_index); + + virt = kmap_atomic(buffer_page, KM_USER1); + memcpy(virt, page_address(suspend_readahead_pages[readahead_index]), + PAGE_SIZE); + kunmap_atomic(virt, KM_USER1); + + suspend_cleanup_readahead(readahead_index); + + readahead_index++; + if (readahead_index == MAX_OUTSTANDING_IO) + readahead_index = 0; + + return 0; +} + +static int suspend_rw_init(int rw, int stream_number) +{ + suspend_extent_state_restore(&suspend_writer_posn, + &suspend_writer_posn_save[stream_number]); + current_stream = stream_number; + + BUG_ON(!suspend_writer_posn.current_extent); + + suspend_reset_io_stats(); + + readahead_index = readahead_submit_index = -1; + + return 0; +} + +static int suspend_rw_cleanup(int rw) +{ + if (rw == WRITE && current_stream == 2) + suspend_extent_state_save(&suspend_writer_posn, + &suspend_writer_posn_save[1]); + + suspend_finish_all_io(); + + if (rw == READ) { + while (readahead_index != readahead_submit_index) { + suspend_cleanup_readahead(readahead_index); + readahead_index++; + if (readahead_index == MAX_OUTSTANDING_IO) + readahead_index = 0; + } + } + + suspend_check_io_stats(); + + return 0; +} + +static int suspend_write_chunk(struct page *buffer_page) +{ + return suspend_rw_page(WRITE, buffer_page, -1, 0, 0); +} + +static int suspend_rw_header_chunk(int rw, struct suspend_module_ops *owner, + char *buffer, int buffer_size) +{ + int bytes_left = buffer_size; + + if (owner) { + owner->header_used += buffer_size; + if (owner->header_used > owner->header_requested) { + printk(KERN_EMERG "Suspend2 module %s is using more" + "header space (%lu) than it requested (%lu).\n", + owner->name, + owner->header_used, + owner->header_requested); + BUG(); + } + } + + /* Read a chunk of the header */ + while (bytes_left) { + char *source_start = buffer + buffer_size - bytes_left; + char *dest_start = suspend_writer_buffer + suspend_writer_buffer_posn; + int capacity = PAGE_SIZE - suspend_writer_buffer_posn; + char *to = rw ? dest_start : source_start; + char *from = rw ? source_start : dest_start; + + if (bytes_left <= capacity) { + if (test_debug_state(SUSPEND_HEADER)) + printk("Copy %d bytes %d-%d from %p to %p.\n", + bytes_left, + suspend_header_bytes_used, + suspend_header_bytes_used + bytes_left, + from, to); + memcpy(to, from, bytes_left); + suspend_writer_buffer_posn += bytes_left; + suspend_header_bytes_used += bytes_left; + return rw ? 0 : buffer_size; + } + + /* Next to read the next page */ + if (test_debug_state(SUSPEND_HEADER)) + printk("Copy %d bytes (%d-%d) from %p to %p.\n", + capacity, + suspend_header_bytes_used, + suspend_header_bytes_used + capacity, + from, to); + memcpy(to, from, capacity); + bytes_left -= capacity; + suspend_header_bytes_used += capacity; + + if (rw == READ && test_suspend_state(SUSPEND_TRY_RESUME_RD)) + sys_read(suspend_read_fd, + suspend_writer_buffer, BLOCK_SIZE); + else { + if (suspend_rw_page(rw, + virt_to_page(suspend_writer_buffer), + -1, !rw, + test_debug_state(SUSPEND_HEADER))) + return -EIO; + } + + suspend_writer_buffer_posn = 0; + suspend_cond_pause(0, NULL); + } + + return rw ? 0 : buffer_size; +} + +static int write_header_chunk_finish(void) +{ + return suspend_rw_page(WRITE, + virt_to_page(suspend_writer_buffer), + -1, 0, test_debug_state(SUSPEND_HEADER)) ? -EIO : 0; +} + +struct suspend_bio_ops suspend_bio_ops = { + .bdev_page_io = suspend_bdev_page_io, + .check_io_stats = suspend_check_io_stats, + .reset_io_stats = suspend_reset_io_stats, + .finish_all_io = suspend_finish_all_io, + .prepare_readahead = suspend_prepare_readahead, + .cleanup_readahead = suspend_cleanup_readahead, + .readahead_pages = suspend_readahead_pages, + .readahead_ready = suspend_readahead_ready, + .forward_one_page = forward_one_page, + .set_extra_page_forward = set_extra_page_forward, + .set_devinfo = suspend_set_devinfo, + .read_chunk = suspend_bio_read_chunk, + .write_chunk = suspend_write_chunk, + .rw_init = suspend_rw_init, + .rw_cleanup = suspend_rw_cleanup, + .rw_header_chunk = suspend_rw_header_chunk, + .write_header_chunk_finish = write_header_chunk_finish, +}; + +static struct suspend_module_ops suspend_blockwriter_ops = +{ + .name = "Block I/O", + .type = MISC_MODULE, + .module = THIS_MODULE, + .memory_needed = suspend_bio_memory_needed, +}; + +static __init int suspend_block_io_load(void) +{ + return suspend_register_module(&suspend_blockwriter_ops); +} + +#ifdef MODULE +static __exit void suspend_block_io_unload(void) +{ + suspend_unregister_module(&suspend_blockwriter_ops); +} + +module_init(suspend_block_io_load); +module_exit(suspend_block_io_unload); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nigel Cunningham"); +MODULE_DESCRIPTION("Suspend2 block io functions"); +#else +late_initcall(suspend_block_io_load); +#endif diff -urN oldtree/kernel/power/suspend_file.c newtree/kernel/power/suspend_file.c --- oldtree/kernel/power/suspend_file.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/suspend_file.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,1125 @@ +/* + * kernel/power/suspend_file.c + * + * Copyright 2005-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * This file encapsulates functions for usage of a simple file as a + * backing store. It is based upon the swapwriter, and shares the + * same basic working. Here, though, we have nothing to do with + * swapspace, and only one device to worry about. + * + * The user can just + * + * echo Suspend2 > /path/to/my_file + * + * and + * + * echo /path/to/my_file > /sys/power/suspend2/filewriter/target + * + * then put what they find in /sys/power/suspend2/resume2 + * as their resume2= parameter in lilo.conf (and rerun lilo if using it). + * + * Having done this, they're ready to suspend and resume. + * + * TODO: + * - File resizing. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "suspend.h" +#include "sysfs.h" +#include "modules.h" +#include "ui.h" +#include "extent.h" +#include "io.h" +#include "storage.h" +#include "block_io.h" + +static struct suspend_module_ops filewriterops; + +/* Details of our target. */ + +char filewriter_target[256]; +static struct inode *target_inode; +static struct file *target_file; +static struct block_device *filewriter_target_bdev; +static dev_t resume_dev_t; +static int used_devt = 0; +static int setting_filewriter_target = 0; +static sector_t target_firstblock = 0, target_header_start = 0; +static int target_storage_available = 0; +static int target_claim = 0; + +static char HaveImage[] = "HaveImage\n"; +static char NoImage[] = "Suspend2\n"; +#define sig_size (sizeof(HaveImage) + 1) + +struct filewriter_header { + char sig[sig_size]; + int resumed_before; + unsigned long first_header_block; +}; + +extern dev_t ROOT_DEV; +extern char *__initdata root_device_name; + +/* Header_pages must be big enough for signature */ +static int header_pages, main_pages; + +#define target_is_normal_file() (S_ISREG(target_inode->i_mode)) + +static struct suspend_bdev_info devinfo; + +/* Extent chain for blocks */ +static struct extent_chain block_chain; + +/* Signature operations */ +enum { + GET_IMAGE_EXISTS, + INVALIDATE, + MARK_RESUME_ATTEMPTED, +}; + +static void set_devinfo(struct block_device *bdev, int target_blkbits) +{ + devinfo.bdev = bdev; + if (!target_blkbits) { + devinfo.bmap_shift = devinfo.blocks_per_page = 0; + } else { + devinfo.bmap_shift = target_blkbits - 9; + devinfo.blocks_per_page = (1 << (PAGE_SHIFT - target_blkbits)); + } +} + +static int filewriter_storage_available(void) +{ + int result = 0; + struct block_device *bdev=filewriter_target_bdev; + + if (!target_inode) + return 0; + + switch (target_inode->i_mode & S_IFMT) { + case S_IFSOCK: + case S_IFCHR: + case S_IFIFO: /* Socket, Char, Fifo */ + return -1; + case S_IFREG: /* Regular file: current size - holes + free + space on part */ + result = target_storage_available; + break; + case S_IFBLK: /* Block device */ + if (!bdev->bd_disk) { + printk("bdev->bd_disk null.\n"); + return 0; + } + + result = (bdev->bd_part ? + bdev->bd_part->nr_sects : + bdev->bd_disk->capacity) >> (PAGE_SHIFT - 9); + } + + return result; +} + +static int has_contiguous_blocks(int page_num) +{ + int j; + sector_t last = 0; + + for (j = 0; j < devinfo.blocks_per_page; j++) { + sector_t this = bmap(target_inode, + page_num * devinfo.blocks_per_page + j); + + if (!this || (last && (last + 1) != this)) + break; + + last = this; + } + + return (j == devinfo.blocks_per_page); +} + +static int size_ignoring_ignored_pages(void) +{ + int mappable = 0, i; + + if (target_is_normal_file()) { + for (i = 0; i < (target_inode->i_size >> PAGE_SHIFT) ; i++) + if (has_contiguous_blocks(i)) + mappable++; + + return mappable; + } else + return filewriter_storage_available(); +} + +static void __populate_block_list(int min, int max) +{ + if (test_action_state(SUSPEND_TEST_BIO)) + printk("Adding extent %d-%d.\n", min << devinfo.bmap_shift, + ((max + 1) << devinfo.bmap_shift) - 1); + + suspend_add_to_extent_chain(&block_chain, min, max); +} + +static void populate_block_list(void) +{ + int i; + + if (block_chain.first) + suspend_put_extent_chain(&block_chain); + + if (target_is_normal_file()) { + int extent_min = -1, extent_max = -1, got_header = 0; + + for (i = 0; + i < (target_inode->i_size >> PAGE_SHIFT); + i++) { + sector_t new_sector; + + if (!has_contiguous_blocks(i)) + continue; + + new_sector = bmap(target_inode, + (i * devinfo.blocks_per_page)); + + /* + * Ignore the first block in the file. + * It gets the header. + */ + if (new_sector == target_firstblock >> devinfo.bmap_shift) { + got_header = 1; + continue; + } + + /* + * I'd love to be able to fill in holes and resize + * files, but not yet... + */ + + if (new_sector == extent_max + 1) + extent_max+= devinfo.blocks_per_page; + else { + if (extent_min > -1) + __populate_block_list(extent_min, + extent_max); + + extent_min = new_sector; + extent_max = extent_min + + devinfo.blocks_per_page - 1; + } + } + if (extent_min > -1) + __populate_block_list(extent_min, extent_max); + + BUG_ON(!got_header); + } else + if (target_storage_available > 0) + __populate_block_list(devinfo.blocks_per_page, + (min(main_pages, target_storage_available) + 1) * + devinfo.blocks_per_page - 1); +} + +static void filewriter_cleanup(int finishing_cycle) +{ + if (filewriter_target_bdev) { + if (target_claim) { + bd_release(filewriter_target_bdev); + target_claim = 0; + } + + if (used_devt) { + blkdev_put(filewriter_target_bdev); + used_devt = 0; + } + filewriter_target_bdev = NULL; + target_inode = NULL; + set_devinfo(NULL, 0); + target_storage_available = 0; + } + + if (target_file > 0) { + filp_close(target_file, NULL); + target_file = NULL; + } +} + +/* + * reopen_resume_devt + * + * Having opened resume2= once, we remember the major and + * minor nodes and use them to reopen the bdev for checking + * whether an image exists (possibly when starting a resume). + */ +static void reopen_resume_devt(void) +{ + filewriter_target_bdev = open_by_devnum(resume_dev_t, FMODE_READ); + if (IS_ERR(filewriter_target_bdev)) { + printk("Got a dev_num (%lx) but failed to open it.\n", + (unsigned long) resume_dev_t); + return; + } + target_inode = filewriter_target_bdev->bd_inode; + set_devinfo(filewriter_target_bdev, target_inode->i_blkbits); +} + +static void filewriter_get_target_info(char *target, int get_size, + int resume2) +{ + if (target_file) + filewriter_cleanup(0); + + if (!target || !strlen(target)) + return; + + target_file = filp_open(target, O_RDWR, 0); + + if (IS_ERR(target_file) || !target_file) { + + if (!resume2) { + printk("Open file %s returned %p.\n", + target, target_file); + target_file = NULL; + return; + } + + target_file = NULL; + resume_dev_t = name_to_dev_t(target); + if (!resume_dev_t) { + printk("Open file %s returned %p and name_to_devt " + "failed.\n", + target, target_file); + if (!resume_dev_t) { + struct kstat stat; + int error = vfs_stat(target, &stat); + if (error) { + printk("Stating the file also failed." + " Nothing more we can do.\n"); + return; + } + resume_dev_t = stat.rdev; + } + return; + } + filewriter_target_bdev = open_by_devnum(resume_dev_t, FMODE_READ); + if (IS_ERR(filewriter_target_bdev)) { + printk("Got a dev_num (%lx) but failed to open it.\n", + (unsigned long) resume_dev_t); + return; + } + used_devt = 1; + target_inode = filewriter_target_bdev->bd_inode; + } else + target_inode = target_file->f_mapping->host; + + if (S_ISLNK(target_inode->i_mode) || + S_ISDIR(target_inode->i_mode) || + S_ISSOCK(target_inode->i_mode) || + S_ISFIFO(target_inode->i_mode)) { + printk("The filewriter works with regular files, character " + "files and block devices.\n"); + goto cleanup; + } + + if (!used_devt) { + if (S_ISBLK(target_inode->i_mode)) { + filewriter_target_bdev = I_BDEV(target_inode); + if (!bd_claim(filewriter_target_bdev, &filewriterops)) + target_claim = 1; + } else + filewriter_target_bdev = target_inode->i_sb->s_bdev; + resume_dev_t = filewriter_target_bdev->bd_dev; + } + + set_devinfo(filewriter_target_bdev, target_inode->i_blkbits); + + if (get_size) + target_storage_available = size_ignoring_ignored_pages(); + + if (!resume2) + target_firstblock = bmap(target_inode, 0) << devinfo.bmap_shift; + + return; +cleanup: + target_inode = NULL; + if (target_file) { + filp_close(target_file, NULL); + target_file = NULL; + } + set_devinfo(NULL, 0); + target_storage_available = 0; +} + +static int parse_signature(struct filewriter_header *header) +{ + int have_image = !memcmp(HaveImage, header->sig, sizeof(HaveImage) - 1); + int no_image_header = !memcmp(NoImage, header->sig, sizeof(NoImage) - 1); + + if (no_image_header) + return 0; + + if (!have_image) + return -1; + + if (header->resumed_before) + set_suspend_state(SUSPEND_RESUMED_BEFORE); + else + clear_suspend_state(SUSPEND_RESUMED_BEFORE); + + target_header_start = header->first_header_block; + return 1; +} + +/* prepare_signature */ + +static int prepare_signature(struct filewriter_header *current_header, + unsigned long first_header_block) +{ + strncpy(current_header->sig, HaveImage, sizeof(HaveImage)); + current_header->resumed_before = 0; + current_header->first_header_block = first_header_block; + return 0; +} + +static int filewriter_storage_allocated(void) +{ + int result; + + if (!target_inode) + return 0; + + if (target_is_normal_file()) { + result = (int) target_storage_available; + } else + result = header_pages + main_pages; + + return result; +} + +static int filewriter_release_storage(void) +{ + if ((test_action_state(SUSPEND_KEEP_IMAGE)) && + test_suspend_state(SUSPEND_NOW_RESUMING)) + return 0; + + suspend_put_extent_chain(&block_chain); + + header_pages = main_pages = 0; + return 0; +} + +static int filewriter_allocate_header_space(int space_requested) +{ + int i; + + if (!block_chain.first) + return 0; + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.forward_one_page(); /* To first page */ + + for (i = 0; i < space_requested; i++) + if (suspend_bio_ops.forward_one_page()) + return -ENOSPC; + + /* The end of header pages will be the start of pageset 2 */ + suspend_extent_state_save(&suspend_writer_posn, + &suspend_writer_posn_save[2]); + header_pages = space_requested; + return 0; +} + +static int filewriter_allocate_storage(int space_requested) +{ + int result = 0, prev_header_pages; + int blocks_to_get = (space_requested << devinfo.bmap_shift) - + block_chain.size; + + /* Only release_storage reduces the size */ + if (blocks_to_get < 1) + return 0; + + main_pages = space_requested; + + populate_block_list(); + + suspend_message(SUSPEND_WRITER, SUSPEND_MEDIUM, 0, + "Finished with block_chain.size == %d.\n", + block_chain.size); + + if (block_chain.size < (header_pages + main_pages)) { + printk("Block chain size (%d) < header pages (%d) + main pages (%d) (=%d).\n", + block_chain.size, + header_pages, main_pages, + header_pages + main_pages); + result = -ENOSPC; + } + + prev_header_pages = header_pages; + header_pages = 0; + filewriter_allocate_header_space(prev_header_pages); + return result; +} + +static int filewriter_write_header_init(void) +{ + suspend_extent_state_goto_start(&suspend_writer_posn); + + suspend_writer_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + suspend_writer_buffer_posn = suspend_header_bytes_used = 0; + + /* Info needed to bootstrap goes at the start of the header. + * First we save the basic info needed for reading, including the number + * of header pages. Then we save the structs containing data needed + * for reading the header pages back. + * Note that even if header pages take more than one page, when we + * read back the info, we will have restored the location of the + * next header page by the time we go to use it. + */ + + suspend_bio_ops.rw_header_chunk(WRITE, &filewriterops, + (char *) &suspend_writer_posn_save, + sizeof(suspend_writer_posn_save)); + + suspend_bio_ops.rw_header_chunk(WRITE, &filewriterops, + (char *) &devinfo, sizeof(devinfo)); + + suspend_serialise_extent_chain(&filewriterops, &block_chain); + + return 0; +} + +static int filewriter_write_header_cleanup(void) +{ + struct filewriter_header *header; + + /* Write any unsaved data */ + if (suspend_writer_buffer_posn) + suspend_bio_ops.write_header_chunk_finish(); + + suspend_bio_ops.finish_all_io(); + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.forward_one_page(); + + /* Adjust image header */ + suspend_bio_ops.bdev_page_io(READ, filewriter_target_bdev, + target_firstblock, + virt_to_page(suspend_writer_buffer)); + + header = (struct filewriter_header *) suspend_writer_buffer; + + prepare_signature(header, + suspend_writer_posn.current_offset << + devinfo.bmap_shift); + + suspend_bio_ops.bdev_page_io(WRITE, filewriter_target_bdev, + target_firstblock, + virt_to_page(suspend_writer_buffer)); + + free_page((unsigned long) suspend_writer_buffer); + suspend_writer_buffer = NULL; + + suspend_bio_ops.finish_all_io(); + + return 0; +} + +/* HEADER READING */ + +#ifdef CONFIG_DEVFS_FS +int create_dev(char *name, dev_t dev, char *devfs_name); +#else +static int create_dev(char *name, dev_t dev, char *devfs_name) +{ + sys_unlink(name); + return sys_mknod(name, S_IFBLK|0600, new_encode_dev(dev)); +} +#endif + +static int rd_init(void) +{ + suspend_writer_buffer_posn = 0; + + create_dev("/dev/root", ROOT_DEV, root_device_name); + create_dev("/dev/ram", MKDEV(RAMDISK_MAJOR, 0), NULL); + + suspend_read_fd = sys_open("/dev/root", O_RDONLY, 0); + if (suspend_read_fd < 0) + goto out; + + sys_read(suspend_read_fd, suspend_writer_buffer, BLOCK_SIZE); + + memcpy(&suspend_writer_posn_save, + suspend_writer_buffer + suspend_writer_buffer_posn, + sizeof(suspend_writer_posn_save)); + + suspend_writer_buffer_posn += sizeof(suspend_writer_posn_save); + + return 0; +out: + sys_unlink("/dev/ram"); + sys_unlink("/dev/root"); + return -EIO; +} + +static int file_init(void) +{ + suspend_writer_buffer_posn = 0; + + /* Read filewriter configuration */ + suspend_bio_ops.bdev_page_io(READ, filewriter_target_bdev, + target_header_start, + virt_to_page((unsigned long) suspend_writer_buffer)); + + return 0; +} + +/* + * read_header_init() + * + * Ramdisk support based heavily on init/do_mounts_rd.c + * + * Description: + * 1. Attempt to read the device specified with resume2=. + * 2. Check the contents of the header for our signature. + * 3. Warn, ignore, reset and/or continue as appropriate. + * 4. If continuing, read the filewriter configuration section + * of the header and set up block device info so we can read + * the rest of the header & image. + * + * Returns: + * May not return if user choose to reboot at a warning. + * -EINVAL if cannot resume at this time. Booting should continue + * normally. + */ + +static int filewriter_read_header_init(void) +{ + int result; + struct block_device *tmp; + + suspend_writer_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + if (test_suspend_state(SUSPEND_TRY_RESUME_RD)) + result = rd_init(); + else + result = file_init(); + + if (result) { + printk("Filewriter read header init: Failed to initialise " + "reading the first page of data.\n"); + return result; + } + + memcpy(&suspend_writer_posn_save, + suspend_writer_buffer + suspend_writer_buffer_posn, + sizeof(suspend_writer_posn_save)); + + suspend_writer_buffer_posn += sizeof(suspend_writer_posn_save); + + tmp = devinfo.bdev; + + memcpy(&devinfo, + suspend_writer_buffer + suspend_writer_buffer_posn, + sizeof(devinfo)); + + devinfo.bdev = tmp; + suspend_writer_buffer_posn += sizeof(devinfo); + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.set_extra_page_forward(); + + suspend_header_bytes_used = suspend_writer_buffer_posn; + + suspend_load_extent_chain(&block_chain); + + return 0; +} + +static int filewriter_read_header_cleanup(void) +{ + free_page((unsigned long) suspend_writer_buffer); + suspend_writer_buffer = NULL; + return 0; +} + +static int filewriter_signature_op(int op) +{ + char *cur; + int result = 0, changed = 0; + struct filewriter_header *header; + + if(filewriter_target_bdev <= 0) + return -1; + + cur = (char *) get_zeroed_page(GFP_ATOMIC); + if (!cur) { + printk("Unable to allocate a page for reading the image " + "signature.\n"); + return -ENOMEM; + } + + suspend_bio_ops.bdev_page_io(READ, filewriter_target_bdev, + target_firstblock, + virt_to_page(cur)); + + header = (struct filewriter_header *) cur; + result = parse_signature(header); + + switch (op) { + case INVALIDATE: + if (result == -1) + goto out; + + strcpy(header->sig, NoImage); + header->resumed_before = 0; + result = changed = 1; + break; + case MARK_RESUME_ATTEMPTED: + if (result == 1) { + header->resumed_before = 1; + changed = 1; + } + break; + } + + if (changed) + suspend_bio_ops.bdev_page_io(WRITE, filewriter_target_bdev, + target_firstblock, + virt_to_page(cur)); + +out: + suspend_bio_ops.finish_all_io(); + free_page((unsigned long) cur); + return result; +} + +/* Print debug info + * + * Description: + */ + +static int filewriter_print_debug_stats(char *buffer, int size) +{ + int len = 0; + + if (suspend_active_writer != &filewriterops) { + len = snprintf_used(buffer, size, "- Filewriter inactive.\n"); + return len; + } + + len = snprintf_used(buffer, size, "- Filewriter active.\n"); + + len+= snprintf_used(buffer+len, size-len, " Storage available for image: " + "%ld pages.\n", + filewriter_storage_allocated()); + + return len; +} + +/* + * Storage needed + * + * Returns amount of space in the image header required + * for the filewriter's data. + * + * We ensure the space is allocated, but actually save the + * data from write_header_init and therefore don't also define a + * save_config_info routine. + */ +static unsigned long filewriter_storage_needed(void) +{ + return sig_size + strlen(filewriter_target) + 1 + + 3 * sizeof(struct extent_iterate_saved_state) + + sizeof(devinfo) + + sizeof(struct extent_chain) - 2 * sizeof(void *) + + (2 * sizeof(unsigned long) * + (block_chain.allocs - block_chain.frees)); +} + +/* + * filewriter_invalidate_image + * + */ +static int filewriter_invalidate_image(void) +{ + int result; + + if (nr_suspends > 0) + filewriter_release_storage(); + + result = filewriter_signature_op(INVALIDATE); + if (result == 1 && !nr_suspends) + printk(KERN_WARNING name_suspend "Image invalidated.\n"); + + return result; +} + +/* + * Image_exists + * + */ + +static int filewriter_image_exists(void) +{ + if (!filewriter_target_bdev) + reopen_resume_devt(); + + return filewriter_signature_op(GET_IMAGE_EXISTS); +} + +/* + * Mark resume attempted. + * + * Record that we tried to resume from this image. + */ + +static void filewriter_mark_resume_attempted(void) +{ + filewriter_signature_op(MARK_RESUME_ATTEMPTED); +} + +static void filewriter_set_resume2(void) +{ + char *buffer = (char *) get_zeroed_page(GFP_ATOMIC); + char *buffer2 = (char *) get_zeroed_page(GFP_ATOMIC); + unsigned long sector = bmap(target_inode, 0); + int offset = 0; + + if (filewriter_target_bdev) { + set_devinfo(filewriter_target_bdev, target_inode->i_blkbits); + + bdevname(filewriter_target_bdev, buffer2); + offset += snprintf(buffer + offset, PAGE_SIZE - offset, + "/dev/%s", buffer2); + + if (sector) + offset += snprintf(buffer + offset, PAGE_SIZE - offset, + ":0x%lx", sector << devinfo.bmap_shift); + } else + offset += snprintf(buffer + offset, PAGE_SIZE - offset, + "%s is not a valid target.", filewriter_target); + + sprintf(resume2_file, "file:%s", buffer); + + free_page((unsigned long) buffer); + free_page((unsigned long) buffer2); + + suspend_attempt_to_parse_resume_device(); +} + +static int __test_filewriter_target(char *target, int resume_time) +{ + filewriter_get_target_info(target, 0, resume_time); + if (filewriter_signature_op(GET_IMAGE_EXISTS) > -1) { + printk(name_suspend "Filewriter: File signature found.\n"); + if (!resume_time) + filewriter_set_resume2(); + + suspend_bio_ops.set_devinfo(&devinfo); + suspend_writer_posn.chains = &block_chain; + suspend_writer_posn.num_chains = 1; + + set_suspend_state(SUSPEND_CAN_SUSPEND); + return 0; + } + + clear_suspend_state(SUSPEND_CAN_SUSPEND); + + if (*target) + printk(name_suspend + "Filewriter: Sorry. No signature found at %s.\n", + target); + else + if (!resume_time) + printk(name_suspend + "Filewriter: Sorry. Target is not set for suspending.\n"); + + return 1; +} + +static void test_filewriter_target(void) +{ + int cant_suspend; + + setting_filewriter_target = 1; + + printk(name_suspend "Filewriter: Testing whether you can suspend:\n"); + cant_suspend =__test_filewriter_target(filewriter_target, 0); + + printk(name_suspend "Suspending %sabled.\n", cant_suspend ? "dis" : "en"); + + setting_filewriter_target = 0; +} + +/* + * Parse Image Location + * + * Attempt to parse a resume2= parameter. + * Swap Writer accepts: + * resume2=file:DEVNAME[:FIRSTBLOCK] + * + * Where: + * DEVNAME is convertable to a dev_t by name_to_dev_t + * FIRSTBLOCK is the location of the first block in the file. + * BLOCKSIZE is the logical blocksize >= SECTOR_SIZE & <= PAGE_SIZE, + * mod SECTOR_SIZE == 0 of the device. + * Data is validated by attempting to read a header from the + * location given. Failure will result in filewriter refusing to + * save an image, and a reboot with correct parameters will be + * necessary. + */ + +static int filewriter_parse_sig_location(char *commandline, int only_writer) +{ + char *thischar, *devstart = NULL, *colon = NULL, *at_symbol = NULL; + int result = -EINVAL, target_blocksize = 0; + + if (strncmp(commandline, "file:", 5)) { + if (!only_writer) + return 1; + } else + commandline += 5; + + /* + * Don't check signature again if we're beginning a cycle. If we already + * did the initialisation successfully, assume we'll be okay when it comes + * to resuming. + */ + if (filewriter_target_bdev) + return 0; + + devstart = thischar = commandline; + while ((*thischar != ':') && (*thischar != '@') && + ((thischar - commandline) < 250) && (*thischar)) + thischar++; + + if (*thischar == ':') { + colon = thischar; + *colon = 0; + thischar++; + } + + while ((*thischar != '@') && ((thischar - commandline) < 250) && (*thischar)) + thischar++; + + if (*thischar == '@') { + at_symbol = thischar; + *at_symbol = 0; + } + + /* + * For the filewriter, you can be able to resume, but not suspend, + * because the resume2= is set correctly, but the filewriter_target + * isn't. + * + * We may have come here as a result of setting resume2 or + * filewriter_target. We only test the filewriter target in the + * former case (it's already done in the later), and we do it before + * setting the block number ourselves. It will overwrite the values + * given on the command line if we don't. + */ + + if (!setting_filewriter_target) + __test_filewriter_target(filewriter_target, 1); + + if (colon) + target_firstblock = (int) simple_strtoul(colon + 1, NULL, 0); + else + target_firstblock = 0; + + if (at_symbol) { + target_blocksize = (int) simple_strtoul(at_symbol + 1, NULL, 0); + if (target_blocksize & (SECTOR_SIZE - 1)) { + printk("Filewriter: Blocksizes are multiples of %d.\n", SECTOR_SIZE); + result = -EINVAL; + goto out; + } + } + + printk("Suspend2 Filewriter: Testing whether you can resume:\n"); + + filewriter_get_target_info(commandline, 0, 1); + + if (!filewriter_target_bdev || IS_ERR(filewriter_target_bdev)) { + filewriter_target_bdev = NULL; + result = -1; + goto out; + } + + if (target_blocksize) + set_devinfo(filewriter_target_bdev, ffs(target_blocksize)); + + result = __test_filewriter_target(commandline, 1); + +out: + if (result) + clear_suspend_state(SUSPEND_CAN_SUSPEND); + else + set_suspend_state(SUSPEND_CAN_SUSPEND); + + printk("Resuming %sabled.\n", result ? "dis" : "en"); + + if (colon) + *colon = ':'; + if (at_symbol) + *at_symbol = '@'; + + return result; +} + +/* filewriter_save_config_info + * + * Description: Save the target's name, not for resume time, but for all_settings. + * Arguments: Buffer: Pointer to a buffer of size PAGE_SIZE. + * Returns: Number of bytes used for saving our data. + */ + +static int filewriter_save_config_info(char *buffer) +{ + strcpy(buffer, filewriter_target); + return strlen(filewriter_target) + 1; +} + +/* filewriter_load_config_info + * + * Description: Reload target's name. + * Arguments: Buffer: Pointer to the start of the data. + * Size: Number of bytes that were saved. + */ + +static void filewriter_load_config_info(char *buffer, int size) +{ + strcpy(filewriter_target, buffer); +} + +static int filewriter_initialise(int starting_cycle) +{ + int result = 0; + + if (starting_cycle) { + if (suspend_active_writer != &filewriterops) + return 0; + + if (!*filewriter_target) { + printk("Filewriter is the active writer, but no filename has been set.\n"); + return 1; + } + } + + if (filewriter_target) + filewriter_get_target_info(filewriter_target, starting_cycle, 0); + + if (starting_cycle && (filewriter_image_exists() == -1)) { + printk("%s is does not have a valid signature for suspending.\n", + filewriter_target); + result = 1; + } + + return result; +} + +static struct suspend_sysfs_data filewriter_sysfs_data[] = { + + { + SUSPEND2_ATTR("target", SYSFS_RW), + SYSFS_STRING(filewriter_target, 256, SYSFS_NEEDS_FOR_WRITE), + .write_side_effect = test_filewriter_target, + }, + + { + SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&filewriterops.enabled, 0, 1), + .write_side_effect = attempt_to_parse_resume_device2, + } +}; + +static struct suspend_module_ops filewriterops = { + .type = WRITER_MODULE, + .name = "File Writer", + .module = THIS_MODULE, + .print_debug_info = filewriter_print_debug_stats, + .save_config_info = filewriter_save_config_info, + .load_config_info = filewriter_load_config_info, + .storage_needed = filewriter_storage_needed, + .initialise = filewriter_initialise, + .cleanup = filewriter_cleanup, + + .storage_available = filewriter_storage_available, + .storage_allocated = filewriter_storage_allocated, + .release_storage = filewriter_release_storage, + .allocate_header_space = filewriter_allocate_header_space, + .allocate_storage = filewriter_allocate_storage, + .image_exists = filewriter_image_exists, + .mark_resume_attempted = filewriter_mark_resume_attempted, + .write_header_init = filewriter_write_header_init, + .write_header_cleanup = filewriter_write_header_cleanup, + .read_header_init = filewriter_read_header_init, + .read_header_cleanup = filewriter_read_header_cleanup, + .invalidate_image = filewriter_invalidate_image, + .parse_sig_location = filewriter_parse_sig_location, +}; + +/* ---- Registration ---- */ +static __init int filewriter_load(void) +{ + int result; + int i, + numfiles = sizeof(filewriter_sysfs_data) / + sizeof(struct suspend_sysfs_data); + + printk("Suspend2 FileWriter loading.\n"); + + filewriterops.read_chunk = suspend_bio_ops.read_chunk; + filewriterops.write_chunk = suspend_bio_ops.write_chunk; + filewriterops.rw_init = suspend_bio_ops.rw_init; + filewriterops.rw_cleanup = suspend_bio_ops.rw_cleanup; + filewriterops.rw_header_chunk = + suspend_bio_ops.rw_header_chunk; + + if (!(result = suspend_register_module(&filewriterops))) { + struct kobject *kobj = make_suspend2_sysdir("filewriter"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, + &filewriter_sysfs_data[i]); + } else + printk("Suspend2 FileWriter unable to register!\n"); + + return result; +} + +#ifdef MODULE +static __exit void filewriter_unload(void) +{ + int i, numfiles = sizeof(filewriter_sysfs_data) / + sizeof(struct suspend_sysfs_data); + + printk("Suspend2 FileWriter unloading.\n"); + + for (i=0; i< numfiles; i++) + suspend_unregister_sysfs_file(&filewriter_sysfs_data[i]); + suspend_unregister_module(&filewriterops); +} + +module_init(filewriter_load); +module_exit(filewriter_unload); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nigel Cunningham"); +MODULE_DESCRIPTION("Suspend2 filewriter"); +#else +late_initcall(filewriter_load); +#endif diff -urN oldtree/kernel/power/suspend_swap.c newtree/kernel/power/suspend_swap.c --- oldtree/kernel/power/suspend_swap.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/suspend_swap.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,1229 @@ +/* + * kernel/power/suspend_swap.c + * + * Copyright 2004-2006 Nigel Cunningham + * + * Distributed under GPLv2. + * + * This file encapsulates functions for usage of swap space as a + * backing store. + */ + +#include +#include +#include +#include +#include + +#include "suspend.h" +#include "sysfs.h" +#include "modules.h" +#include "io.h" +#include "ui.h" +#include "extent.h" +#include "block_io.h" + +static struct suspend_module_ops swapwriterops; + +#define SIGNATURE_VER 6 + +/* --- Struct of pages stored on disk */ + +union diskpage { + union swap_header swh; /* swh.magic is the only member used */ +}; + +union p_diskpage { + union diskpage *pointer; + char *ptr; + unsigned long address; +}; + +/* Devices used for swap */ +static struct suspend_bdev_info devinfo[MAX_SWAPFILES]; + +/* Extent chains for swap & blocks */ +struct extent_chain swapextents; +struct extent_chain block_chain[MAX_SWAPFILES]; + +static dev_t header_dev_t; +static struct block_device *header_block_device; +static unsigned long headerblock; + +/* For swapfile automatically swapon/off'd. */ +static char swapfilename[32] = ""; +extern asmlinkage long sys_swapon(const char *specialfile, int swap_flags); +extern asmlinkage long sys_swapoff(const char *specialfile); +static int suspend_swapon_status; + +/* Header Page Information */ +static int header_pages_allocated; + +/* User Specified Parameters. */ + +static unsigned long resume_firstblock; +static int resume_blocksize; +static dev_t resume_dev_t; +static struct block_device *resume_block_device; + +struct sysinfo swapinfo; +static int swapwriter_invalidate_image(void); + +/* Block devices open. */ +struct bdev_opened +{ + dev_t device; + struct block_device *bdev; + int claimed; +}; + +/* + * Entry MAX_SWAPFILES is the resume block device, which may + * not be a swap device enabled when we suspend. + * Entry MAX_SWAPFILES + 1 is the header block device, which + * is needed before we find out which slot it occupies. + */ +static struct bdev_opened *bdev_info_list[MAX_SWAPFILES + 2]; + +static void close_bdev(int i) +{ + struct bdev_opened *this = bdev_info_list[i]; + + if (this->claimed) + bd_release(this->bdev); + + /* Release our reference. */ + blkdev_put(this->bdev); + + /* Free our info. */ + kfree(this); + + bdev_info_list[i] = NULL; +} + +static void close_bdevs(void) +{ + int i; + + for (i = 0; i < MAX_SWAPFILES; i++) + if (bdev_info_list[i]) + close_bdev(i); + + resume_block_device = header_block_device = NULL; +} + +static struct block_device *open_bdev(int index, dev_t device) +{ + struct bdev_opened *this; + struct block_device *bdev; + + if (bdev_info_list[index] && (bdev_info_list[index]->device == device)){ + bdev = bdev_info_list[index]->bdev; + return bdev; + } + + if (bdev_info_list[index] && bdev_info_list[index]->device != device) + close_bdev(index); + + bdev = open_by_devnum(device, FMODE_READ); + + if (IS_ERR(bdev) || !bdev) { + suspend_early_boot_message(1,SUSPEND_CONTINUE_REQ, + "Failed to get access to block device " + "\"%s\" (error %d).\n Maybe you need " + "to run mknod and/or lvmsetup in an " + "initrd/ramfs?", device, bdev); + return ERR_PTR(-EINVAL); + } + + this = kmalloc(sizeof(struct bdev_opened), GFP_KERNEL); + BUG_ON(!this); + + bdev_info_list[index] = this; + this->device = device; + this->bdev = bdev; + + if (index < MAX_SWAPFILES) + devinfo[index].bdev = bdev; + + return bdev; +} + +/* Must be silent - might be called from cat /sys/power/suspend2/debug_info + * Returns 0 if was off, -EBUSY if was on, error value otherwise. + */ +static int enable_swapfile(void) +{ + int activateswapresult = -EINVAL; + + if (suspend_swapon_status) + return 0; + + if (swapfilename[0]) { + /* Attempt to swap on with maximum priority */ + activateswapresult = sys_swapon(swapfilename, 0xFFFF); + if ((activateswapresult) && (activateswapresult != -EBUSY)) + printk(name_suspend + "The swapfile/partition specified by " + "/sys/power/suspend2/swapwriter/swapfile (%s) could not" + " be turned on (error %d). Attempting " + "to continue.\n", + swapfilename, activateswapresult); + if (!activateswapresult) + suspend_swapon_status = 1; + } + return activateswapresult; +} + +/* Returns 0 if was on, -EINVAL if was off, error value otherwise */ +static int disable_swapfile(void) +{ + int result = -EINVAL; + + if (!suspend_swapon_status) + return 0; + + if (swapfilename[0]) { + result = sys_swapoff(swapfilename); + if (result == -EINVAL) + return 0; /* Wasn't on */ + if (!result) + suspend_swapon_status = 0; + } + + return result; +} + +static int try_to_parse_resume_device(char *commandline) +{ + struct kstat stat; + int error; + + resume_dev_t = name_to_dev_t(commandline); + + if (!resume_dev_t) { + error = vfs_stat(commandline, &stat); + if (!error) + resume_dev_t = stat.rdev; + } + + if (!resume_dev_t) { + if (test_suspend_state(SUSPEND_TRYING_TO_RESUME)) + suspend_early_boot_message(1, SUSPEND_CONTINUE_REQ, + "Failed to translate \"%s\" into a device id.\n", + commandline); + else + printk(name_suspend + "Can't translate \"%s\" into a device id yet.\n", + commandline); + return 1; + } + + if (IS_ERR(resume_block_device = + open_bdev(MAX_SWAPFILES, resume_dev_t))) { + suspend_early_boot_message(1, SUSPEND_CONTINUE_REQ, + "Failed to get access to \"%s\", where" + " the swap header should be found.", + commandline); + return 1; + } + + return 0; +} + +/* + * If we have read part of the image, we might have filled memory with + * data that should be zeroed out. + */ +static void swapwriter_noresume_reset(void) +{ + memset((char *) &devinfo, 0, sizeof(devinfo)); + close_bdevs(); +} + +static int parse_signature(char *header, int restore) +{ + int type = -1; + + if (!memcmp("SWAP-SPACE",header,10)) + return 0; + else if (!memcmp("SWAPSPACE2",header,10)) + return 1; + + else if (!memcmp("S1SUSP",header,6)) + type = 2; + else if (!memcmp("S2SUSP",header,6)) + type = 3; + else if (!memcmp("S1SUSPEND",header,9)) + type = 4; + + else if (!memcmp("z",header,1)) + type = 12; + else if (!memcmp("Z",header,1)) + type = 13; + + /* + * Put bdev of suspend header in last byte of swap header + * (unsigned short) + */ + if (type > 11) { + dev_t *header_ptr = (dev_t *) &header[1]; + unsigned char *headerblocksize_ptr = + (unsigned char *) &header[5]; + u32 *headerblock_ptr = (u32 *) &header[6]; + header_dev_t = *header_ptr; + /* + * We are now using the highest bit of the char to indicate + * whether we have attempted to resume from this image before. + */ + clear_suspend_state(SUSPEND_RESUMED_BEFORE); + if (((int) *headerblocksize_ptr) & 0x80) + set_suspend_state(SUSPEND_RESUMED_BEFORE); + headerblock = (unsigned long) *headerblock_ptr; + } + + if ((restore) && (type > 5)) { + /* We only reset our own signatures */ + if (type & 1) + memcpy(header,"SWAPSPACE2",10); + else + memcpy(header,"SWAP-SPACE",10); + } + + return type; +} + +/* + * prepare_signature + */ +static int prepare_signature(dev_t bdev, unsigned long block, + char *current_header) +{ + int current_type = parse_signature(current_header, 0); + dev_t *header_ptr = (dev_t *) (¤t_header[1]); + unsigned long *headerblock_ptr = + (unsigned long *) (¤t_header[6]); + + if ((current_type > 1) && (current_type < 6)) + return 1; + + /* At the moment, I don't have a way to handle the block being + * > 32 bits. Not enough room in the signature and no way to + * safely put the data elsewhere. */ + + if (BITS_PER_LONG == 64 && ffs(block) > 31) { + suspend_prepare_status(DONT_CLEAR_BAR, + "Header sector requires 33+ bits. " + "Would not be able to resume."); + return 1; + } + + if (current_type & 1) + current_header[0] = 'Z'; + else + current_header[0] = 'z'; + *header_ptr = bdev; + /* prev is the first/last swap page of the resume area */ + *headerblock_ptr = (unsigned long) block; + return 0; +} + +static int swapwriter_allocate_storage(int space_requested); + +static int swapwriter_allocate_header_space(int space_requested) +{ + int i; + + if (!swapextents.size) + swapwriter_allocate_storage(space_requested); + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.forward_one_page(); /* To first page */ + + for (i = 0; i < space_requested; i++) { + if (suspend_bio_ops.forward_one_page()) { + printk("Out of space while seeking to allocate " + "header pages,\n"); + header_pages_allocated = i; + return -ENOSPC; + } + + } + + header_pages_allocated = space_requested; + + /* The end of header pages will be the start of pageset 2; + * we are now sitting on the first pageset2 page. */ + suspend_extent_state_save(&suspend_writer_posn, + &suspend_writer_posn_save[2]); + return 0; +} + +static void get_main_pool_phys_params(void) +{ + struct extent *extentpointer = NULL; + unsigned long address; + int i, extent_min = -1, extent_max = -1, last_chain = -1; + + for (i = 0; i < MAX_SWAPFILES; i++) + if (block_chain[i].first) + suspend_put_extent_chain(&block_chain[i]); + + suspend_extent_for_each(&swapextents, extentpointer, address) { + swp_entry_t swap_address = extent_val_to_swap_entry(address); + pgoff_t offset = swp_offset(swap_address); + unsigned swapfilenum = swp_type(swap_address); + struct swap_info_struct *sis = get_swap_info_struct(swapfilenum); + sector_t new_sector = map_swap_page(sis, offset); + + if ((new_sector == extent_max + 1) && + (last_chain == swapfilenum)) + extent_max++; + else { + if (extent_min > -1) { + if (test_action_state(SUSPEND_TEST_BIO)) + printk("Adding extent chain %d %d-%d.\n", + swapfilenum, + extent_min << + devinfo[last_chain].bmap_shift, + extent_max << + devinfo[last_chain].bmap_shift); + + suspend_add_to_extent_chain( + &block_chain[last_chain], + extent_min, extent_max); + } + extent_min = extent_max = new_sector; + last_chain = swapfilenum; + } + } + + if (extent_min > -1) { + if (test_action_state(SUSPEND_TEST_BIO)) + printk("Adding extent chain %d %d-%d.\n", + last_chain, + extent_min << + devinfo[last_chain].bmap_shift, + extent_max << + devinfo[last_chain].bmap_shift); + suspend_add_to_extent_chain( + &block_chain[last_chain], + extent_min, extent_max); + } + + swapwriter_allocate_header_space(header_pages_allocated); +} + +static int swapwriter_storage_allocated(void) +{ + return swapextents.size; +} + +static int swapwriter_storage_available(void) +{ + si_swapinfo(&swapinfo); + return swapinfo.freeswap + swapwriter_storage_allocated(); +} + +static int swapwriter_initialise(int starting_cycle) +{ + if (!starting_cycle) + return 0; + + enable_swapfile(); + + if (resume_dev_t && !resume_block_device && + IS_ERR(resume_block_device = + open_bdev(MAX_SWAPFILES, resume_dev_t))) + return 1; + + return 0; +} + +static void swapwriter_cleanup(int ending_cycle) +{ + if (ending_cycle) + disable_swapfile(); + + close_bdevs(); +} + +static int swapwriter_release_storage(void) +{ + int i = 0; + + if (test_action_state(SUSPEND_KEEP_IMAGE) && + test_suspend_state(SUSPEND_NOW_RESUMING)) + return 0; + + header_pages_allocated = 0; + + if (swapextents.first) { + /* Free swap entries */ + struct extent *extentpointer; + unsigned long extentvalue; + swp_entry_t entry; + suspend_extent_for_each(&swapextents, extentpointer, + extentvalue) { + entry = extent_val_to_swap_entry(extentvalue); + swap_free(entry); + } + + suspend_put_extent_chain(&swapextents); + + for (i = 0; i < MAX_SWAPFILES; i++) + if (block_chain[i].first) + suspend_put_extent_chain(&block_chain[i]); + } + + return 0; +} + +/* + * Round robin allocation (where swap storage has the same priority). + * could make this very inefficient, so we track extents allocated on + * a per-swapfiles basis. + */ +static int swapwriter_allocate_storage(int space_requested) +{ + int i, result = 0, first[MAX_SWAPFILES]; + int pages_to_get = space_requested - swapextents.size; + unsigned long extent_min[MAX_SWAPFILES], extent_max[MAX_SWAPFILES]; + + if (pages_to_get < 1) + return 0; + + for (i=0; i < MAX_SWAPFILES; i++) { + struct swap_info_struct *si = get_swap_info_struct(i); + if ((devinfo[i].bdev = si->bdev)) + devinfo[i].dev_t = si->bdev->bd_dev; + devinfo[i].bmap_shift = 3; + devinfo[i].blocks_per_page = 1; + first[i] = 1; + } + + for(i=0; i < pages_to_get; i++) { + swp_entry_t entry; + unsigned long new_value; + unsigned swapfilenum; + + entry = get_swap_page(); + if (!entry.val) { + printk("Failed to get a swap page.\n"); + result = -ENOSPC; + break; + } + + swapfilenum = swp_type(entry); + new_value = swap_entry_to_extent_val(entry); + if (first[swapfilenum]) { + first[swapfilenum] = 0; + extent_min[swapfilenum] = extent_max[swapfilenum] = + new_value; + } else { + if (new_value == extent_max[swapfilenum] + 1) + extent_max[swapfilenum]++; + else { + suspend_add_to_extent_chain( + &swapextents, + extent_min[swapfilenum], + extent_max[swapfilenum]); + extent_min[swapfilenum] = + extent_max[swapfilenum] = new_value; + } + } + } + + for (i = 0; i < MAX_SWAPFILES; i++) + if (!first[i]) + suspend_add_to_extent_chain( + &swapextents, + extent_min[i], extent_max[i]); + + get_main_pool_phys_params(); + return result; +} + +static int swapwriter_write_header_init(void) +{ + int i, result; + struct swap_info_struct *si; + + suspend_header_bytes_used = 0; + + suspend_extent_state_goto_start(&suspend_writer_posn); + /* Forward one page will be done prior to the read */ + + for (i = 0; i < MAX_SWAPFILES; i++) { + si = get_swap_info_struct(i); + if (si->swap_file) + devinfo[i].dev_t = si->bdev->bd_dev; + else + devinfo[i].dev_t = (dev_t) 0; + } + + suspend_writer_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + if (!suspend_writer_buffer) { + printk("Failed to get swapwriter buffer.\n"); + return -ENOMEM; + } + + suspend_writer_buffer_posn = 0; + + /* Info needed to bootstrap goes at the start of the header. + * First we save the positions and devinfo, including the number + * of header pages. Then we save the structs containing data needed + * for reading the header pages back. + * Note that even if header pages take more than one page, when we + * read back the info, we will have restored the location of the + * next header page by the time we go to use it. + */ + + if ((result = suspend_bio_ops.rw_header_chunk(WRITE, + &swapwriterops, + (char *) &suspend_writer_posn_save, + sizeof(suspend_writer_posn_save)))) + return result; + + if ((result = suspend_bio_ops.rw_header_chunk(WRITE, + &swapwriterops, + (char *) &devinfo, sizeof(devinfo)))) + return result; + + for (i=0; i < MAX_SWAPFILES; i++) + suspend_serialise_extent_chain(&swapwriterops, &block_chain[i]); + + return 0; +} + +static int swapwriter_write_header_cleanup(void) +{ + int result; + struct swap_info_struct *si; + + /* Write any unsaved data */ + if (suspend_writer_buffer_posn) + suspend_bio_ops.write_header_chunk_finish(); + + suspend_bio_ops.finish_all_io(); + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.forward_one_page(); + + /* Adjust swap header */ + suspend_bio_ops.bdev_page_io(READ, resume_block_device, + resume_firstblock, + virt_to_page(suspend_writer_buffer)); + + si = get_swap_info_struct(suspend_writer_posn.current_chain); + result = prepare_signature(si->bdev->bd_dev, + suspend_writer_posn.current_offset, + ((union swap_header *) suspend_writer_buffer)->magic.magic); + + if (!result) + suspend_bio_ops.bdev_page_io(WRITE, resume_block_device, + resume_firstblock, + virt_to_page(suspend_writer_buffer)); + + free_page((unsigned long) suspend_writer_buffer); + suspend_writer_buffer = NULL; + + suspend_bio_ops.finish_all_io(); + + return result; +} + +/* ------------------------- HEADER READING ------------------------- */ + +/* + * read_header_init() + * + * Description: + * 1. Attempt to read the device specified with resume2=. + * 2. Check the contents of the swap header for our signature. + * 3. Warn, ignore, reset and/or continue as appropriate. + * 4. If continuing, read the swapwriter configuration section + * of the header and set up block device info so we can read + * the rest of the header & image. + * + * Returns: + * May not return if user choose to reboot at a warning. + * -EINVAL if cannot resume at this time. Booting should continue + * normally. + */ + +static int swapwriter_read_header_init(void) +{ + int i; + + BUG_ON(!resume_block_device); + BUG_ON(!resume_dev_t); + + suspend_header_bytes_used = 0; + + suspend_writer_buffer = (char *) get_zeroed_page(GFP_ATOMIC); + + BUG_ON(!suspend_writer_buffer); + + if (!header_dev_t) { + printk("read_header_init called when we haven't " + "verified there is an image!\n"); + return -EINVAL; + } + + /* + * If the header is not on the resume_dev_t, get the resume device first. + */ + if (header_dev_t != resume_dev_t) { + header_block_device = open_bdev(MAX_SWAPFILES + 1, + header_dev_t); + + if (IS_ERR(header_block_device)) + return PTR_ERR(header_block_device); + } else + header_block_device = resume_block_device; + + /* + * Read swapwriter configuration. + * Headerblock size taken into account already. + */ + suspend_bio_ops.bdev_page_io(READ, header_block_device, + headerblock << 3, + virt_to_page((unsigned long) suspend_writer_buffer)); + + memcpy(&suspend_writer_posn_save, suspend_writer_buffer, 3 * sizeof(struct extent_iterate_saved_state)); + + suspend_writer_buffer_posn = 3 * sizeof(struct extent_iterate_saved_state); + suspend_header_bytes_used += 3 * sizeof(struct extent_iterate_saved_state); + + memcpy(&devinfo, suspend_writer_buffer + suspend_writer_buffer_posn, sizeof(devinfo)); + + suspend_writer_buffer_posn += sizeof(devinfo); + suspend_header_bytes_used += sizeof(devinfo); + + /* Restore device info */ + for (i = 0; i < MAX_SWAPFILES; i++) { + dev_t thisdevice = devinfo[i].dev_t; + struct block_device *result; + + devinfo[i].bdev = NULL; + + if (!thisdevice) + continue; + + if (thisdevice == resume_dev_t) { + devinfo[i].bdev = resume_block_device; + bdev_info_list[i] = bdev_info_list[MAX_SWAPFILES]; + BUG_ON(!bdev_info_list[i]); + bdev_info_list[MAX_SWAPFILES] = NULL; + continue; + } + + if (thisdevice == header_dev_t) { + devinfo[i].bdev = header_block_device; + bdev_info_list[i] = bdev_info_list[MAX_SWAPFILES + 1]; + BUG_ON(!bdev_info_list[i]); + bdev_info_list[MAX_SWAPFILES + 1] = NULL; + continue; + } + + result = open_bdev(i, thisdevice); + if (IS_ERR(result)) { + close_bdevs(); + return PTR_ERR(result); + } + } + + suspend_extent_state_goto_start(&suspend_writer_posn); + suspend_bio_ops.set_extra_page_forward(); + + for (i = 0; i < MAX_SWAPFILES; i++) + suspend_load_extent_chain(&block_chain[i]); + + return 0; +} + +static int swapwriter_read_header_cleanup(void) +{ + free_page((unsigned long) suspend_writer_buffer); + return 0; +} + +/* swapwriter_invalidate_image + * + */ +static int swapwriter_invalidate_image(void) +{ + union p_diskpage cur; + int result = 0; + char newsig[11]; + + cur.address = get_zeroed_page(GFP_ATOMIC); + if (!cur.address) { + printk("Unable to allocate a page for restoring the swap signature.\n"); + return -ENOMEM; + } + + /* + * If nr_suspends == 0, we must be booting, so no swap pages + * will be recorded as used yet. + */ + + if (nr_suspends > 0) + swapwriter_release_storage(); + + /* + * We don't do a sanity check here: we want to restore the swap + * whatever version of kernel made the suspend image. + * + * We need to write swap, but swap may not be enabled so + * we write the device directly + */ + + suspend_bio_ops.bdev_page_io(READ, resume_block_device, + resume_firstblock, + virt_to_page(cur.pointer)); + + result = parse_signature(cur.pointer->swh.magic.magic, 1); + + if (result < 5) + goto out; + + strncpy(newsig, cur.pointer->swh.magic.magic, 10); + newsig[10] = 0; + + suspend_bio_ops.bdev_page_io(WRITE, resume_block_device, + resume_firstblock, + virt_to_page(cur.pointer)); + + if (!nr_suspends) + printk(KERN_WARNING name_suspend "Image invalidated.\n"); +out: + suspend_bio_ops.finish_all_io(); + free_page(cur.address); + return 0; +} + +/* + * workspace_size + * + * Description: + * Returns the number of bytes of RAM needed for this + * code to do its work. (Used when calculating whether + * we have enough memory to be able to suspend & resume). + * + */ +static unsigned long swapwriter_memory_needed(void) +{ + return 1; +} + +/* + * Print debug info + * + * Description: + */ +static int swapwriter_print_debug_stats(char *buffer, int size) +{ + int len = 0; + struct sysinfo sysinfo; + + if (suspend_active_writer != &swapwriterops) { + len = snprintf_used(buffer, size, "- Swapwriter inactive.\n"); + return len; + } + + len = snprintf_used(buffer, size, "- Swapwriter active.\n"); + if (swapfilename[0]) + len+= snprintf_used(buffer+len, size-len, + " Attempting to automatically swapon: %s.\n", swapfilename); + + si_swapinfo(&sysinfo); + + len+= snprintf_used(buffer+len, size-len, " Swap available for image: %ld pages.\n", + sysinfo.freeswap + swapwriter_storage_allocated()); + + return len; +} + +/* + * Storage needed + * + * Returns amount of space in the swap header required + * for the swapwriter's data. This ignores the links between + * pages, which we factor in when allocating the space. + * + * We ensure the space is allocated, but actually save the + * data from write_header_init and therefore don't also define a + * save_config_info routine. + */ +static unsigned long swapwriter_storage_needed(void) +{ + int i, result; + result = sizeof(suspend_writer_posn_save) + sizeof(devinfo); + + for (i = 0; i < MAX_SWAPFILES; i++) { + result += 3 * sizeof(int); + result += (2 * sizeof(unsigned long) * + (block_chain[i].allocs - block_chain[i].frees)); + } + + return result; +} + +/* + * Image_exists + */ +static int swapwriter_image_exists(void) +{ + int signature_found; + union p_diskpage diskpage; + + if (!resume_dev_t) { + printk("Not even trying to read header " + "because resume_dev_t is not set.\n"); + return 0; + } + + if (!resume_block_device && + IS_ERR(resume_block_device = open_bdev(MAX_SWAPFILES, resume_dev_t))) { + printk("Failed to open resume dev_t (%x).\n", resume_dev_t); + return 0; + } + + diskpage.address = get_zeroed_page(GFP_ATOMIC); + + suspend_bio_ops.bdev_page_io(READ, resume_block_device, + resume_firstblock, + virt_to_page(diskpage.ptr)); + suspend_bio_ops.finish_all_io(); + + signature_found = parse_signature(diskpage.pointer->swh.magic.magic, 0); + free_page(diskpage.address); + + if (signature_found < 2) { + printk(name_suspend "Normal swapspace found.\n"); + return 0; /* Normal swap space */ + } else if (signature_found == -1) { + printk(KERN_ERR name_suspend + "Unable to find a signature. Could you have moved " + "a swap file?\n"); + return 0; + } else if (signature_found < 6) { + printk(name_suspend "Dectected another implementation's signature.\n"); + return 0; + } else if ((signature_found >> 1) != SIGNATURE_VER) { + if ((!(test_suspend_state(SUSPEND_NORESUME_SPECIFIED))) && + suspend_early_boot_message(1, SUSPEND_CONTINUE_REQ, + "Found a different style suspend image signature.")) { + set_suspend_state(SUSPEND_NORESUME_SPECIFIED); + printk(name_suspend "Dectected another implementation's signature.\n"); + } + } + + return 1; +} + +/* + * Mark resume attempted. + * + * Record that we tried to resume from this image. + */ +static void swapwriter_mark_resume_attempted(void) +{ + union p_diskpage diskpage; + int signature_found; + + if (!resume_dev_t) { + printk("Not even trying to record attempt at resuming" + " because resume_dev_t is not set.\n"); + return; + } + + diskpage.address = get_zeroed_page(GFP_ATOMIC); + + suspend_bio_ops.bdev_page_io(READ, resume_block_device, + resume_firstblock, + virt_to_page(diskpage.ptr)); + signature_found = parse_signature(diskpage.pointer->swh.magic.magic, 0); + + switch (signature_found) { + case 12: + case 13: + diskpage.pointer->swh.magic.magic[5] |= 0x80; + break; + } + + suspend_bio_ops.bdev_page_io(WRITE, resume_block_device, + resume_firstblock, + virt_to_page(diskpage.ptr)); + suspend_bio_ops.finish_all_io(); + free_page(diskpage.address); + + close_bdevs(); + return; +} + +/* + * Parse Image Location + * + * Attempt to parse a resume2= parameter. + * Swap Writer accepts: + * resume2=swap:DEVNAME[:FIRSTBLOCK][@BLOCKSIZE] + * + * Where: + * DEVNAME is convertable to a dev_t by name_to_dev_t + * FIRSTBLOCK is the location of the first block in the swap file + * (specifying for a swap partition is nonsensical but not prohibited). + * Data is validated by attempting to read a swap header from the + * location given. Failure will result in swapwriter refusing to + * save an image, and a reboot with correct parameters will be + * necessary. + */ +static int swapwriter_parse_sig_location(char *commandline, int only_writer) +{ + char *thischar, *devstart, *colon = NULL, *at_symbol = NULL; + union p_diskpage diskpage; + int signature_found, result = -EINVAL, temp_result; + + if (strncmp(commandline, "swap:", 5)) { + /* + * Failing swap:, we'll take a simple + * resume2=/dev/hda2, but fall through to + * other writers if /dev/ isn't matched. + */ + if (strncmp(commandline, "/dev/", 5)) + return 1; + } else + commandline += 5; + + devstart = thischar = commandline; + while ((*thischar != ':') && (*thischar != '@') && + ((thischar - commandline) < 250) && (*thischar)) + thischar++; + + if (*thischar == ':') { + colon = thischar; + *colon = 0; + thischar++; + } + + while ((*thischar != '@') && ((thischar - commandline) < 250) && (*thischar)) + thischar++; + + if (*thischar == '@') { + at_symbol = thischar; + *at_symbol = 0; + } + + if (colon) + resume_firstblock = (int) simple_strtoul(colon + 1, NULL, 0); + else + resume_firstblock = 0; + + clear_suspend_state(SUSPEND_CAN_SUSPEND); + clear_suspend_state(SUSPEND_CAN_RESUME); + + /* Legacy */ + if (at_symbol) { + resume_blocksize = (int) simple_strtoul(at_symbol + 1, NULL, 0); + if (resume_blocksize & (SECTOR_SIZE - 1)) { + printk("Swapwriter: Blocksizes are multiples of %d!\n", SECTOR_SIZE); + return -EINVAL; + } + resume_firstblock = resume_firstblock * (resume_blocksize / SECTOR_SIZE); + } + + temp_result = try_to_parse_resume_device(devstart); + + if (colon) + *colon = ':'; + if (at_symbol) + *at_symbol = '@'; + + if (temp_result) + return -EINVAL; + + diskpage.address = get_zeroed_page(GFP_ATOMIC); + if (!diskpage.address) { + printk(KERN_ERR name_suspend "Swapwriter: Failed to allocate a diskpage for I/O.\n"); + return -ENOMEM; + } + + temp_result = suspend_bio_ops.bdev_page_io(READ, + resume_block_device, + resume_firstblock, + virt_to_page(diskpage.ptr)); + + suspend_bio_ops.finish_all_io(); + + if (temp_result) { + printk(KERN_ERR name_suspend "Swapwriter: Failed to submit I/O.\n"); + goto invalid; + } + + signature_found = parse_signature(diskpage.pointer->swh.magic.magic, 0); + + if (signature_found != -1) { + printk(name_suspend "Swapwriter: Signature found.\n"); + result = 0; + + suspend_bio_ops.set_devinfo(devinfo); + suspend_writer_posn.chains = &block_chain[0]; + suspend_writer_posn.num_chains = MAX_SWAPFILES; + set_suspend_state(SUSPEND_CAN_SUSPEND); + set_suspend_state(SUSPEND_CAN_RESUME); + } else + printk(KERN_ERR name_suspend "Swapwriter: No swap signature found at specified location.\n"); +invalid: + free_page((unsigned long) diskpage.address); + return result; + +} + +static int header_locations_read_sysfs(const char *page, int count) +{ + int i, printedpartitionsmessage = 0, len = 0, haveswap = 0; + struct inode *swapf = 0; + int zone; + char *path_page = (char *) __get_free_page(GFP_KERNEL); + char *path, *output = (char *) page; + int path_len; + + if (!page) + return 0; + + for (i = 0; i < MAX_SWAPFILES; i++) { + struct swap_info_struct *si = get_swap_info_struct(i); + + if (!si->swap_file) + continue; + + if (S_ISBLK(si->swap_file->f_mapping->host->i_mode)) { + haveswap = 1; + if (!printedpartitionsmessage) { + len += sprintf(output + len, + "For swap partitions, simply use the format: resume2=swap:/dev/hda1.\n"); + printedpartitionsmessage = 1; + } + } else { + path_len = 0; + + path = d_path(si->swap_file->f_dentry, + si->swap_file->f_vfsmnt, + path_page, + PAGE_SIZE); + path_len = snprintf(path_page, 31, "%s", path); + + haveswap = 1; + swapf = si->swap_file->f_mapping->host; + if (!(zone = bmap(swapf,0))) { + len+= sprintf(output + len, + "Swapfile %s has been corrupted. Reuse mkswap on it and try again.\n", + path_page); + } else { + char name_buffer[255]; + len+= sprintf(output + len, "For swapfile `%s`, use resume2=swap:/dev/%s:0x%x.\n", + path_page, + bdevname(si->bdev, name_buffer), + zone << (swapf->i_blkbits - 9)); + } + + } + } + + if (!haveswap) + len = sprintf(output, "You need to turn on swap partitions before examining this file.\n"); + + free_page((unsigned long) path_page); + return len; +} + +static struct suspend_sysfs_data swapwriter_sysfs_data[] = { + { + SUSPEND2_ATTR("swapfilename", SYSFS_RW), + SYSFS_STRING(swapfilename, 255, SYSFS_SM_NOT_NEEDED) + }, + + { + SUSPEND2_ATTR("headerlocations", SYSFS_READONLY), + SYSFS_CUSTOM(header_locations_read_sysfs, NULL, SYSFS_SM_NOT_NEEDED) + }, + + { SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&swapwriterops.enabled, 0, 1), + .write_side_effect = attempt_to_parse_resume_device2, + } +}; + +static struct suspend_module_ops swapwriterops = { + .type = WRITER_MODULE, + .name = "Swap Writer", + .module = THIS_MODULE, + .memory_needed = swapwriter_memory_needed, + .print_debug_info = swapwriter_print_debug_stats, + .storage_needed = swapwriter_storage_needed, + .initialise = swapwriter_initialise, + .cleanup = swapwriter_cleanup, + + .noresume_reset = swapwriter_noresume_reset, + .storage_available = swapwriter_storage_available, + .storage_allocated = swapwriter_storage_allocated, + .release_storage = swapwriter_release_storage, + .allocate_header_space = swapwriter_allocate_header_space, + .allocate_storage = swapwriter_allocate_storage, + .image_exists = swapwriter_image_exists, + .mark_resume_attempted = swapwriter_mark_resume_attempted, + .write_header_init = swapwriter_write_header_init, + .write_header_cleanup = swapwriter_write_header_cleanup, + .read_header_init = swapwriter_read_header_init, + .read_header_cleanup = swapwriter_read_header_cleanup, + .invalidate_image = swapwriter_invalidate_image, + .parse_sig_location = swapwriter_parse_sig_location, +}; + +/* ---- Registration ---- */ +static __init int swapwriter_load(void) +{ + int result; + int i, numfiles = sizeof(swapwriter_sysfs_data) / sizeof(struct suspend_sysfs_data); + + printk("Suspend2 Swap Writer loading.\n"); + + swapwriterops.rw_init = suspend_bio_ops.rw_init; + swapwriterops.rw_cleanup = suspend_bio_ops.rw_cleanup; + swapwriterops.read_chunk = suspend_bio_ops.read_chunk; + swapwriterops.write_chunk = suspend_bio_ops.write_chunk; + swapwriterops.rw_header_chunk = suspend_bio_ops.rw_header_chunk; + + if (!(result = suspend_register_module(&swapwriterops))) { + struct kobject *kobj = make_suspend2_sysdir("swapwriter"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, + &swapwriter_sysfs_data[i]); + } else + printk("Suspend2 Swap Writer unable to register!\n"); + return result; +} + +#ifdef MODULE +static __exit void swapwriter_unload(void) +{ + int i, numfiles = sizeof(swapwriter_sysfs_data) / + sizeof(struct suspend_sysfs_data); + + printk("Suspend2 Swap Writer unloading.\n"); + + for (i=0; i< numfiles; i++) + suspend_unregister_sysfs_file(&swapwriter_sysfs_data[i]); + suspend_unregister_module(&swapwriterops); +} + +module_init(swapwriter_load); +module_exit(swapwriter_unload); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Nigel Cunningham"); +MODULE_DESCRIPTION("Suspend2 swap writer"); +#else +late_initcall(swapwriter_load); +#endif diff -urN oldtree/kernel/power/swsusp.c newtree/kernel/power/swsusp.c --- oldtree/kernel/power/swsusp.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/swsusp.c 2006-10-06 17:28:20.000000000 -0400 @@ -49,6 +49,7 @@ #include #include #include +#include #include "power.h" @@ -190,6 +191,8 @@ unsigned int i = 0; char *p = "-\\|/"; + thaw_processes(FREEZER_KERNEL_THREADS); + printk("Shrinking memory... "); do { size = 2 * count_highmem_pages(); @@ -214,6 +217,8 @@ } while (tmp > 0); printk("\bdone (%lu pages freed)\n", pages); + freeze_processes(); + return 0; } diff -urN oldtree/kernel/power/sysfs.c newtree/kernel/power/sysfs.c --- oldtree/kernel/power/sysfs.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/sysfs.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,298 @@ +/* + * kernel/power/sysfs.c + * + * Copyright (C) 2002-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * This file contains support for sysfs entries for tuning Suspend2. + * + * We have a generic handler that deals with the most common cases, and + * hooks for special handlers to use. + */ + +#include +#include +#include + +#include "sysfs.h" +#include "suspend.h" +#include "storage.h" + +static int suspend_sysfs_initialised = 0; + +static struct suspend_sysfs_data sysfs_params[]; + +extern void __suspend_try_resume(void); + +extern struct subsystem power_subsys; + +#define to_sysfs_data(_attr) container_of(_attr, struct suspend_sysfs_data, attr) + +static void suspend_main_wrapper(void) +{ + suspend_main(0); +} + +ssize_t suspend2_attr_show(struct kobject *kobj, struct attribute *attr, + char *page) +{ + struct suspend_sysfs_data *sysfs_data = to_sysfs_data(attr); + int len = 0; + + if (suspend_start_anything(0)) + return -EBUSY; + + if (sysfs_data->needs_storage_manager & SYSFS_NEEDS_FOR_READ) + suspend_prepare_usm(); + + switch (sysfs_data->type) { + case SUSPEND_SYSFS_DATA_CUSTOM: + len = (sysfs_data->data.special.read_sysfs) ? + (sysfs_data->data.special.read_sysfs)(page, PAGE_SIZE) + : 0; + break; + case SUSPEND_SYSFS_DATA_BIT: + len = sprintf(page, "%d\n", + -test_bit(sysfs_data->data.bit.bit, + sysfs_data->data.bit.bit_vector)); + break; + case SUSPEND_SYSFS_DATA_INTEGER: + len = sprintf(page, "%d\n", + *(sysfs_data->data.integer.variable)); + break; + case SUSPEND_SYSFS_DATA_LONG: + len = sprintf(page, "%ld\n", + *(sysfs_data->data.a_long.variable)); + break; + case SUSPEND_SYSFS_DATA_UL: + len = sprintf(page, "%lu\n", + *(sysfs_data->data.ul.variable)); + break; + case SUSPEND_SYSFS_DATA_STRING: + len = sprintf(page, "%s\n", + sysfs_data->data.string.variable); + break; + } + /* Side effect routine? */ + if (sysfs_data->read_side_effect) + sysfs_data->read_side_effect(); + + if (sysfs_data->needs_storage_manager & SYSFS_NEEDS_FOR_READ) + suspend_cleanup_usm(); + + suspend_finish_anything(0); + + return len; +} + +#define BOUND(_variable, _type) \ + if (sysfs_data->data._type.maximum) { \ + if (*_variable < sysfs_data->data._type.minimum) \ + *_variable = sysfs_data->data._type.minimum; \ + else if (*_variable > sysfs_data->data._type.maximum) \ + *_variable = sysfs_data->data._type.maximum; \ + } + +ssize_t suspend2_attr_store(struct kobject *kobj, struct attribute *attr, + const char *my_buf, size_t count) +{ + int assigned_temp_buffer = 0, result = count; + struct suspend_sysfs_data *sysfs_data = to_sysfs_data(attr); + + if (suspend_start_anything(sysfs_data == &sysfs_params[0])) + return -EBUSY; + + ((char *) my_buf)[count] = 0; + + if (sysfs_data->needs_storage_manager & SYSFS_NEEDS_FOR_WRITE) + suspend_prepare_usm(); + + switch (sysfs_data->type) { + case SUSPEND_SYSFS_DATA_CUSTOM: + if (sysfs_data->data.special.write_sysfs) + result = (sysfs_data->data.special.write_sysfs) + (my_buf, count); + break; + case SUSPEND_SYSFS_DATA_BIT: + { + int value = simple_strtoul(my_buf, NULL, 0); + if (value) + set_bit(sysfs_data->data.bit.bit, + (sysfs_data->data.bit.bit_vector)); + else + clear_bit(sysfs_data->data.bit.bit, + (sysfs_data->data.bit.bit_vector)); + } + break; + case SUSPEND_SYSFS_DATA_INTEGER: + { + int *variable = sysfs_data->data.integer.variable; + *variable = simple_strtol(my_buf, NULL, 0); + BOUND(variable, integer); + break; + } + case SUSPEND_SYSFS_DATA_LONG: + { + long *variable = sysfs_data->data.a_long.variable; + *variable = simple_strtol(my_buf, NULL, 0); + BOUND(variable, a_long); + break; + } + case SUSPEND_SYSFS_DATA_UL: + { + unsigned long *variable = sysfs_data->data.ul.variable; + *variable = simple_strtoul(my_buf, NULL, 0); + BOUND(variable, ul); + break; + } + break; + case SUSPEND_SYSFS_DATA_STRING: + { + int copy_len = count; + char *variable = + sysfs_data->data.string.variable; + + if (sysfs_data->data.string.max_length && + (copy_len > sysfs_data->data.string.max_length)) + copy_len = sysfs_data->data.string.max_length; + + if (!variable) { + sysfs_data->data.string.variable = + variable = (char *) get_zeroed_page(GFP_ATOMIC); + assigned_temp_buffer = 1; + } + strncpy(variable, my_buf, copy_len); + if ((copy_len) && + (my_buf[copy_len - 1] == '\n')) + variable[count - 1] = 0; + variable[count] = 0; + } + break; + } + + /* Side effect routine? */ + if (sysfs_data->write_side_effect) + sysfs_data->write_side_effect(); + + /* Free temporary buffers */ + if (assigned_temp_buffer) { + free_page((unsigned long) sysfs_data->data.string.variable); + sysfs_data->data.string.variable = NULL; + } + + if (sysfs_data->needs_storage_manager & SYSFS_NEEDS_FOR_WRITE) + suspend_cleanup_usm(); + + suspend_finish_anything(sysfs_data == &sysfs_params[0]); + + return result; +} + +static struct sysfs_ops suspend2_sysfs_ops = { + .show = &suspend2_attr_show, + .store = &suspend2_attr_store, +}; + +static struct kobj_type suspend2_ktype = { + .sysfs_ops = &suspend2_sysfs_ops, +}; + +decl_subsys(suspend2, &suspend2_ktype, NULL); + +/* Non-module sysfs entries. + * + * This array contains entries that are automatically registered at + * boot. Modules and the console code register their own entries separately. + * + * NB: If you move do_suspend, change suspend_write_sysfs's test so that + * suspend_start_anything still gets a 1 when the user echos > do_suspend! + */ + +static struct suspend_sysfs_data sysfs_params[] = { + { SUSPEND2_ATTR("do_suspend", SYSFS_WRITEONLY), + SYSFS_CUSTOM(NULL, NULL, SYSFS_NEEDS_FOR_WRITE), + .write_side_effect = suspend_main_wrapper + }, + + { SUSPEND2_ATTR("do_resume", SYSFS_WRITEONLY), + SYSFS_CUSTOM(NULL, NULL, SYSFS_NEEDS_FOR_WRITE), + .write_side_effect = __suspend_try_resume + }, +}; + +struct kobject *make_suspend2_sysdir(char *name) +{ + struct kobject *kobj = kzalloc(sizeof(struct kobject), GFP_KERNEL); + int err; + + BUG_ON(!kobj); + + err = kobject_set_name(kobj, "%s", name); + + if (err) { + kfree(kobj); + return NULL; + } + + kobj->kset = &suspend2_subsys.kset; + + err = kobject_register(kobj); + + return err ? NULL : kobj; +} + +/* suspend_initialise_sysfs + * + * Initialise the /sysfs/suspend2 directory. + */ + +static void suspend_initialise_sysfs(void) +{ + int i, error; + int numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + if (suspend_sysfs_initialised) + return; + + /* Make our suspend2 directory a child of /sys/power */ + kobj_set_kset_s(&suspend2_subsys.kset, power_subsys); + error = subsystem_register(&suspend2_subsys); + + if (error) + return; + + /* Make it use the .store and .show routines above */ + kobj_set_kset_s(&suspend2_subsys.kset, suspend2_subsys); + + suspend_sysfs_initialised = 1; + + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(&suspend2_subsys.kset.kobj, + &sysfs_params[i]); +} + +/* suspend_register_sysfs_file + * + * Helper for registering a new /sysfs/suspend2 entry. + */ + +void suspend_register_sysfs_file( + struct kobject *kobj, + struct suspend_sysfs_data *suspend_sysfs_data) +{ + if (!suspend_sysfs_initialised) + suspend_initialise_sysfs(); + + sysfs_create_file(kobj, &suspend_sysfs_data->attr); +} + +/* suspend_unregister_sysfs_file + * + * Helper for removing unwanted /sysfs/suspend2 entries. + * + */ +void suspend_unregister_sysfs_file(struct suspend_sysfs_data *suspend_sysfs_data) +{ + sysfs_remove_file(&suspend2_subsys.kset.kobj, &suspend_sysfs_data->attr); +} diff -urN oldtree/kernel/power/sysfs.h newtree/kernel/power/sysfs.h --- oldtree/kernel/power/sysfs.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/sysfs.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,117 @@ +/* + * kernel/power/sysfs.h + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * It provides declarations for suspend to use in managing + * /sysfs/suspend2. When we switch to kobjects, + * this will become redundant. + * + */ + +#include +#include "power.h" + +struct suspend_sysfs_data { + struct attribute attr; + int type; + int needs_storage_manager; + union { + struct { + unsigned long *bit_vector; + int bit; + } bit; + struct { + int *variable; + int minimum; + int maximum; + } integer; + struct { + long *variable; + long minimum; + long maximum; + } a_long; + struct { + unsigned long *variable; + unsigned long minimum; + unsigned long maximum; + } ul; + struct { + char *variable; + int max_length; + } string; + struct { + int (*read_sysfs) (const char *buffer, int count); + int (*write_sysfs) (const char *buffer, int count); + void *data; + } special; + } data; + + /* Side effects routines. Used, eg, for reparsing the + * resume2 entry when it changes */ + void (*read_side_effect) (void); + void (*write_side_effect) (void); + struct list_head sysfs_data_list; +}; + +enum { + SUSPEND_SYSFS_DATA_NONE, + SUSPEND_SYSFS_DATA_CUSTOM, + SUSPEND_SYSFS_DATA_BIT, + SUSPEND_SYSFS_DATA_INTEGER, + SUSPEND_SYSFS_DATA_UL, + SUSPEND_SYSFS_DATA_LONG, + SUSPEND_SYSFS_DATA_STRING +}; + +#define SUSPEND2_ATTR(_name, _mode) \ + .attr = {.name = _name , .mode = _mode } + +#define SYSFS_BIT(_ul, _bit) \ + .type = SUSPEND_SYSFS_DATA_BIT, \ + .data = { .bit = { .bit_vector = _ul, .bit = _bit } } + +#define SYSFS_INT(_int, _min, _max) \ + .type = SUSPEND_SYSFS_DATA_INTEGER, \ + .data = { .integer = { .variable = _int, .minimum = _min, \ + .maximum = _max } } + +#define SYSFS_UL(_ul, _min, _max) \ + .type = SUSPEND_SYSFS_DATA_UL, \ + .data = { .ul = { .variable = _ul, .minimum = _min, \ + .maximum = _max } } + +#define SYSFS_LONG(_long, _min, _max) \ + .type = SUSPEND_SYSFS_DATA_LONG, \ + .data = { .a_long = { .variable = _long, .minimum = _min, \ + .maximum = _max } } + +#define SYSFS_STRING(_string, _max_len, _needs_usm) \ + .type = SUSPEND_SYSFS_DATA_STRING, \ + .needs_storage_manager = _needs_usm, \ + .data = { .string = { .variable = _string, .max_length = _max_len } } + +#define SYSFS_CUSTOM(_read, _write, _needs_storage_manager) \ + .type = SUSPEND_SYSFS_DATA_CUSTOM, \ + .needs_storage_manager = _needs_storage_manager, \ + .data = { .special = { .read_sysfs = _read, .write_sysfs = _write } } + +#define SYSFS_WRITEONLY 0200 +#define SYSFS_READONLY 0444 +#define SYSFS_RW 0644 + +/* Storage manager needed? */ +#define SYSFS_SM_NOT_NEEDED 0 +#define SYSFS_NEEDS_FOR_READ 1 +#define SYSFS_NEEDS_FOR_WRITE 2 +#define SYSFS_NEEDS_FOR_BOTH 3 + +void suspend_register_sysfs_file(struct kobject *kobj, + struct suspend_sysfs_data *suspend_sysfs_data); +void suspend_unregister_sysfs_file(struct suspend_sysfs_data *suspend_sysfs_data); + +extern struct subsystem suspend2_subsys; + +struct kobject *make_suspend2_sysdir(char *name); diff -urN oldtree/kernel/power/ui.c newtree/kernel/power/ui.c --- oldtree/kernel/power/ui.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/ui.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,756 @@ +/* + * kernel/power/ui.c + * + * Copyright (C) 1998-2001 Gabor Kuti + * Copyright (C) 1998,2001,2002 Pavel Machek + * Copyright (C) 2002-2003 Florent Chabaud + * Copyright (C) 2002-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Routines for Suspend2's user interface. + * + * The user interface code talks to a userspace program via a + * netlink socket. + * + * The kernel side: + * - starts the userui program; + * - sends text messages and progress bar status; + * + * The user space side: + * - passes messages regarding user requests (abort, toggle reboot etc) + * + */ + +#define __KERNEL_SYSCALLS__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "sysfs.h" +#include "modules.h" +#include "suspend.h" +#include "ui.h" +#include "netlink.h" +#include "power_off.h" + +static char local_printf_buf[1024]; /* Same as printk - should be safe */ + +/*! The console log level we default to. */ +int suspend_default_console_level = 0; + +#ifdef CONFIG_NET +static struct user_helper_data ui_helper_data; +static struct suspend_module_ops userui_ops; +static int orig_loglevel; +static int orig_default_message_loglevel; +static int orig_kmsg; + +static char lastheader[512]; +static int lastheader_message_len = 0; +extern int console_suspended; + +/* Number of distinct progress amounts that userspace can display */ +static int progress_granularity = 30; + +DECLARE_WAIT_QUEUE_HEAD(userui_wait_for_key); + +static void ui_nl_set_state(int n) +{ + /* Only let them change certain settings */ + static const int suspend_action_mask = + (1 << SUSPEND_REBOOT) | (1 << SUSPEND_PAUSE) | (1 << SUSPEND_SLOW) | + (1 << SUSPEND_LOGALL) | (1 << SUSPEND_SINGLESTEP) | + (1 << SUSPEND_PAUSE_NEAR_PAGESET_END); + + suspend_action = (suspend_action & (~suspend_action_mask)) | + (n & suspend_action_mask); + + if (!test_action_state(SUSPEND_PAUSE) && + !test_action_state(SUSPEND_SINGLESTEP)) + wake_up_interruptible(&userui_wait_for_key); +} + +void userui_redraw(void) +{ + suspend_send_netlink_message(&ui_helper_data, + USERUI_MSG_REDRAW, NULL, 0); +} + +/* request_abort_suspend + * + * Description: Handle the user requesting the cancellation of a suspend by + * pressing escape. + * Callers: Invoked from a netlink packet from userspace when the user presses + * escape. + */ +void request_abort_suspend(void) +{ + if (test_result_state(SUSPEND_ABORT_REQUESTED)) + return; + + if (test_suspend_state(SUSPEND_NOW_RESUMING)) { + suspend_prepare_status(CLEAR_BAR, "Escape pressed. " + "Powering down again."); + suspend_power_down(); + } else { + suspend_prepare_status(CLEAR_BAR, "--- ESCAPE PRESSED :" + " ABORTING SUSPEND ---"); + set_result_state(SUSPEND_ABORTED); + set_result_state(SUSPEND_ABORT_REQUESTED); + + wake_up_interruptible(&userui_wait_for_key); + } +} + +static int userui_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + int type; + int *data; + + type = nlh->nlmsg_type; + + /* A control message: ignore them */ + if (type < NETLINK_MSG_BASE) + return 0; + + /* Unknown message: reply with EINVAL */ + if (type >= USERUI_MSG_MAX) + return -EINVAL; + + /* All operations require privileges, even GET */ + if (security_netlink_recv(skb, CAP_NET_ADMIN)) + return -EPERM; + + /* Only allow one task to receive NOFREEZE privileges */ + if (type == NETLINK_MSG_NOFREEZE_ME && ui_helper_data.pid != -1) + return -EBUSY; + + data = (int*)NLMSG_DATA(nlh); + + switch (type) { + case USERUI_MSG_ABORT: + request_abort_suspend(); + break; + case USERUI_MSG_GET_STATE: + suspend_send_netlink_message(&ui_helper_data, + USERUI_MSG_GET_STATE, &suspend_action, + sizeof(suspend_action)); + break; + case USERUI_MSG_GET_DEBUG_STATE: + suspend_send_netlink_message(&ui_helper_data, + USERUI_MSG_GET_DEBUG_STATE, + &suspend_debug_state, + sizeof(suspend_debug_state)); + break; + case USERUI_MSG_SET_STATE: + if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(int))) + return -EINVAL; + ui_nl_set_state(*data); + break; + case USERUI_MSG_SET_DEBUG_STATE: + if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(int))) + return -EINVAL; + suspend_debug_state = (*data); + break; + case USERUI_MSG_SPACE: + wake_up_interruptible(&userui_wait_for_key); + break; + } + + return 1; +} + +static unsigned long userui_storage_needed(void) +{ + return sizeof(ui_helper_data.program) + 1 + sizeof(int); +} + +static int userui_save_config_info(char *buf) +{ + *((int *) buf) = progress_granularity; + memcpy(buf + sizeof(int), ui_helper_data.program, sizeof(ui_helper_data.program)); + return sizeof(ui_helper_data.program) + sizeof(int) + 1; +} + +static void userui_load_config_info(char *buf, int size) +{ + progress_granularity = *((int *) buf); + size -= sizeof(int); + + /* Don't load the saved path if one has already been set */ + if (ui_helper_data.program[0]) + return; + + if (size > sizeof(ui_helper_data.program)) + size = sizeof(ui_helper_data.program); + + memcpy(ui_helper_data.program, buf + sizeof(int), size); + ui_helper_data.program[sizeof(ui_helper_data.program)-1] = '\0'; +} + +static unsigned long userui_memory_needed(void) +{ + /* ball park figure of 128 pages */ + return (128 * PAGE_SIZE); +} + +/* suspend_update_status + * + * Description: Update the progress bar and (if on) in-bar message. + * Arguments: UL value, maximum: Current progress percentage (value/max). + * const char *fmt, ...: Message to be displayed in the middle + * of the progress bar. + * Note that a NULL message does not mean that any previous + * message is erased! For that, you need suspend_prepare_status with + * clearbar on. + * Returns: Unsigned long: The next value where status needs to be updated. + * This is to reduce unnecessary calls to update_status. + */ +unsigned long suspend_update_status(unsigned long value, unsigned long maximum, + const char *fmt, ...) +{ + static int last_step = -1; + struct userui_msg_params msg; + int bitshift; + int this_step; + unsigned long next_update; + + if (ui_helper_data.pid == -1) + return 0; + + if ((!maximum) || (!progress_granularity)) + return maximum; + + if (value < 0) + value = 0; + + if (value > maximum) + value = maximum; + + /* Try to avoid math problems - we can't do 64 bit math here + * (and shouldn't need it - anyone got screen resolution + * of 65536 pixels or more?) */ + bitshift = fls(maximum) - 16; + if (bitshift > 0) { + unsigned long temp_maximum = maximum >> bitshift; + unsigned long temp_value = value >> bitshift; + this_step = (int) + (temp_value * progress_granularity / temp_maximum); + next_update = (((this_step + 1) * temp_maximum / + progress_granularity) + 1) << bitshift; + } else { + this_step = (int) (value * progress_granularity / maximum); + next_update = ((this_step + 1) * maximum / + progress_granularity) + 1; + } + + if (this_step == last_step) + return next_update; + + memset(&msg, 0, sizeof(msg)); + + msg.a = this_step; + msg.b = progress_granularity; + + if (fmt) { + va_list args; + va_start(args, fmt); + vsnprintf(msg.text, sizeof(msg.text), fmt, args); + va_end(args); + msg.text[sizeof(msg.text)-1] = '\0'; + } + + suspend_send_netlink_message(&ui_helper_data, USERUI_MSG_PROGRESS, + &msg, sizeof(msg)); + last_step = this_step; + + return next_update; +} + +/* __suspend_message. + * + * Description: This function is intended to do the same job as printk, but + * without normally logging what is printed. The point is to be + * able to get debugging info on screen without filling the logs + * with "1/534. ^M 2/534^M. 3/534^M" + * + * It may be called from an interrupt context - can't sleep! + * + * Arguments: int mask: The debugging section(s) this message belongs to. + * int level: The level of verbosity of this message. + * int restartline: Whether to output a \r or \n with this line + * (\n if we're logging all output). + * const char *fmt, ...: Message to be displayed a la printk. + */ +void __suspend_message(unsigned long section, unsigned long level, + int normally_logged, + const char *fmt, ...) +{ + struct userui_msg_params msg; + + if ((level) && (level > console_loglevel)) + return; + + memset(&msg, 0, sizeof(msg)); + + msg.a = section; + msg.b = level; + msg.c = normally_logged; + + if (fmt) { + va_list args; + va_start(args, fmt); + vsnprintf(msg.text, sizeof(msg.text), fmt, args); + va_end(args); + msg.text[sizeof(msg.text)-1] = '\0'; + } + + if (test_action_state(SUSPEND_LOGALL)) + printk("%s\n", msg.text); + + suspend_send_netlink_message(&ui_helper_data, USERUI_MSG_MESSAGE, + &msg, sizeof(msg)); +} + +static void wait_for_key_via_userui(void) +{ + DECLARE_WAITQUEUE(wait, current); + + add_wait_queue(&userui_wait_for_key, &wait); + set_current_state(TASK_INTERRUPTIBLE); + + interruptible_sleep_on(&userui_wait_for_key); + + set_current_state(TASK_RUNNING); + remove_wait_queue(&userui_wait_for_key, &wait); +} + +static char suspend_wait_for_keypress(int timeout) +{ + int fd; + char key = '\0'; + struct termios t, t_backup; + + if (ui_helper_data.pid != -1) { + wait_for_key_via_userui(); + key = ' '; + goto out; + } + + /* We should be guaranteed /dev/console exists after populate_rootfs() in + * init/main.c + */ + if ((fd = sys_open("/dev/console", O_RDONLY, 0)) < 0) { + printk("Couldn't open /dev/console.\n"); + goto out; + } + + if (sys_ioctl(fd, TCGETS, (long)&t) < 0) + goto out_close; + + memcpy(&t_backup, &t, sizeof(t)); + + t.c_lflag &= ~(ISIG|ICANON|ECHO); + t.c_cc[VMIN] = 0; + if (timeout) + t.c_cc[VTIME] = timeout*10; + + if (sys_ioctl(fd, TCSETS, (long)&t) < 0) + goto out_restore; + + while (1) { + if (sys_read(fd, &key, 1) <= 0) { + key = '\0'; + break; + } + key = tolower(key); + if (test_suspend_state(SUSPEND_SANITY_CHECK_PROMPT)) { + if (key == 'c') { + set_suspend_state(SUSPEND_CONTINUE_REQ); + break; + } else if (key == ' ') + break; + } else + break; + } + +out_restore: + sys_ioctl(fd, TCSETS, (long)&t_backup); +out_close: + sys_close(fd); +out: + return key; +} + +/* abort_suspend + * + * Description: Begin to abort a cycle. If this wasn't at the user's request + * (and we're displaying output), tell the user why and wait for + * them to acknowledge the message. + * Arguments: A parameterised string (imagine this is printk) to display, + * telling the user why we're aborting. + */ + +void abort_suspend(const char *fmt, ...) +{ + va_list args; + int printed_len = 0; + + if (!test_result_state(SUSPEND_ABORTED)) { + if (!test_result_state(SUSPEND_ABORT_REQUESTED)) { + va_start(args, fmt); + printed_len = vsnprintf(local_printf_buf, + sizeof(local_printf_buf), fmt, args); + va_end(args); + if (ui_helper_data.pid != -1) + printed_len = sprintf(local_printf_buf + printed_len, + " (Press SPACE to continue)"); + suspend_prepare_status(CLEAR_BAR, local_printf_buf); + + /* + * Make sure message seen - wait for shift to be + * released if being pressed + */ + if (ui_helper_data.pid != -1) + suspend_wait_for_keypress(0); + } + /* Turn on aborting flag */ + set_result_state(SUSPEND_ABORTED); + } +} + +/* suspend_prepare_status + * Description: Prepare the 'nice display', drawing the header and version, + * along with the current action and perhaps also resetting the + * progress bar. + * Arguments: + * int clearbar: Whether to reset the progress bar. + * const char *fmt, ...: The action to be displayed. + */ +void suspend_prepare_status(int clearbar, const char *fmt, ...) +{ + va_list args; + + if (fmt) { + va_start(args, fmt); + lastheader_message_len = vsnprintf(lastheader, 512, fmt, args); + va_end(args); + } + + if (clearbar) + suspend_update_status(0, 1, NULL); + + suspend_message(0, SUSPEND_STATUS, 1, lastheader, NULL); + + if (ui_helper_data.pid == -1) + printk(KERN_EMERG "%s\n", lastheader); +} + +/* suspend_cond_pause + * + * Description: Potentially pause and wait for the user to tell us to continue. + * We normally only pause when @pause is set. + * Arguments: int pause: Whether we normally pause. + * char *message: The message to display. Not parameterised + * because it's normally a constant. + */ + +void suspend_cond_pause(int pause, char *message) +{ +#ifdef CONFIG_PM_DEBUG + int displayed_message = 0, last_key = 0; + + while (last_key != 32 && + ui_helper_data.pid != -1 && + (!test_result_state(SUSPEND_ABORTED)) && + ((test_action_state(SUSPEND_PAUSE) && pause) || + (test_action_state(SUSPEND_SINGLESTEP)))) { + if (!displayed_message) { + suspend_prepare_status(DONT_CLEAR_BAR, + "%s Press SPACE to continue.%s", + message ? message : "", + (test_action_state(SUSPEND_SINGLESTEP)) ? + " Single step on." : ""); + displayed_message = 1; + } + last_key = suspend_wait_for_keypress(0); + } +#endif + schedule(); +} + +extern asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd, + unsigned long arg); + +/* suspend_prepare_console + * + * Description: Prepare a console for use, save current settings. + * Returns: Boolean: Whether an error occured. Errors aren't + * treated as fatal, but a warning is printed. + */ +void suspend_prepare_console(void) +{ + orig_loglevel = console_loglevel; + orig_default_message_loglevel = default_message_loglevel; + orig_kmsg = kmsg_redirect; + kmsg_redirect = fg_console + 1; + default_message_loglevel = 1; + console_loglevel = suspend_default_console_level; + + ui_helper_data.pid = -1; + + if (!userui_ops.enabled) + return; + + if (!*ui_helper_data.program) { + printk("suspend_userui: program not configured. suspend_userui disabled.\n"); + return; + } + + suspend_netlink_setup(&ui_helper_data); + + return; +} + +/* suspend_cleanup_console + * + * Description: Restore the settings we saved above. + */ + +void suspend_cleanup_console(void) +{ + suspend_default_console_level = console_loglevel; + + if (ui_helper_data.pid > -1) { + struct task_struct *t; + + suspend_send_netlink_message(&ui_helper_data, + NETLINK_MSG_CLEANUP, NULL, 0); + + read_lock(&tasklist_lock); + if ((t = find_task_by_pid(ui_helper_data.pid))) + t->flags &= ~PF_NOFREEZE; + read_unlock(&tasklist_lock); + + suspend_netlink_close(&ui_helper_data); + + ui_helper_data.pid = -1; + } + + console_loglevel = orig_loglevel; + kmsg_redirect = orig_kmsg; + default_message_loglevel = orig_default_message_loglevel; +} +#endif + +/* suspend_early_boot_message() + * Description: Handle errors early in the process of booting. + * The user may press C to continue booting, perhaps + * invalidating the image, or space to reboot. + * This works from either the serial console or normally + * attached keyboard. + * + * Note that we come in here from init, while the kernel is + * locked. If we want to get events from the serial console, + * we need to temporarily unlock the kernel. + * + * suspend_early_boot_message may also be called post-boot. + * In this case, it simply printks the message and returns. + * + * Arguments: int Whether we are able to erase the image. + * int default_answer. What to do when we timeout. This + * will normally be continue, but the user might + * provide command line options (__setup) to override + * particular cases. + * Char *. Pointer to a string explaining why we're moaning. + */ + +#define say(message, a...) printk(KERN_EMERG message, ##a) +#define message_timeout 25 /* message_timeout * 10 must fit in 8 bits */ + +int suspend_early_boot_message(int message_detail, int default_answer, char *warning_reason, ...) +{ + unsigned long orig_state = get_suspend_state(), continue_req = 0; + va_list args; + int printed_len; + + if (warning_reason) { + va_start(args, warning_reason); + printed_len = vsnprintf(local_printf_buf, + sizeof(local_printf_buf), + warning_reason, + args); + va_end(args); + } + + if (!test_suspend_state(SUSPEND_BOOT_TIME)) { + printk(name_suspend "%s\n", local_printf_buf); + return default_answer; + } + + /* We might be called directly from do_mounts_initrd if the + * user fails to set up their initrd properly. We need to + * enable the keyboard handler by setting the running flag */ + set_suspend_state(SUSPEND_RUNNING); + +#if defined(CONFIG_VT) || defined(CONFIG_SERIAL_CONSOLE) + console_loglevel = 7; + + say("=== Suspend2 ===\n\n"); + if (warning_reason) { + say("BIG FAT WARNING!! %s\n\n", local_printf_buf); + switch (message_detail) { + case 0: + say("If you continue booting, note that any image WILL NOT BE REMOVED.\n"); + say("Suspend is unable to do so because the appropriate modules aren't\n"); + say("loaded. You should manually remove the image to avoid any\n"); + say("possibility of corrupting your filesystem(s) later.\n"); + break; + case 1: + say("If you want to use the current suspend image, reboot and try\n"); + say("again with the same kernel that you suspended from. If you want\n"); + say("to forget that image, continue and the image will be erased.\n"); + break; + } + say("Press SPACE to reboot or C to continue booting with this kernel\n\n"); + say("Default action if you don't select one in %d seconds is: %s.\n", + message_timeout, + default_answer == SUSPEND_CONTINUE_REQ ? + "continue booting" : "reboot"); + } else { + say("BIG FAT WARNING!!\n\n"); + say("You have tried to resume from this image before.\n"); + say("If it failed once, it may well fail again.\n"); + say("Would you like to remove the image and boot normally?\n"); + say("This will be equivalent to entering noresume2 on the\n"); + say("kernel command line.\n\n"); + say("Press SPACE to remove the image or C to continue resuming.\n\n"); + say("Default action if you don't select one in %d seconds is: %s.\n", + message_timeout, + !!default_answer ? + "continue resuming" : "remove the image"); + } + + set_suspend_state(SUSPEND_SANITY_CHECK_PROMPT); + clear_suspend_state(SUSPEND_CONTINUE_REQ); + + if (suspend_wait_for_keypress(message_timeout) == 0) /* We timed out */ + continue_req = !!default_answer; + else + continue_req = test_suspend_state(SUSPEND_CONTINUE_REQ); + + if ((warning_reason) && (!continue_req)) + machine_restart(NULL); + + restore_suspend_state(orig_state); + if (continue_req) + set_suspend_state(SUSPEND_CONTINUE_REQ); + +#endif /* CONFIG_VT or CONFIG_SERIAL_CONSOLE */ + return -EPERM; +} +#undef say + +/* + * User interface specific /sys/power/suspend2 entries. + */ + +static struct suspend_sysfs_data sysfs_params[] = { +#if defined(CONFIG_NET) && defined(CONFIG_SYSFS) + { SUSPEND2_ATTR("default_console_level", SYSFS_RW), +#ifdef CONFIG_PM_DEBUG + SYSFS_INT(&suspend_default_console_level, 0, 7) +#else + SYSFS_INT(&suspend_default_console_level, 0, 1) +#endif + }, + + { SUSPEND2_ATTR("enable_escape", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_CAN_CANCEL) + }, + +#ifdef CONFIG_PM_DEBUG + { SUSPEND2_ATTR("debug_sections", SYSFS_RW), + SYSFS_UL(&suspend_debug_state, 0, 2 << 30) + }, + + { SUSPEND2_ATTR("log_everything", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_LOGALL) + }, + + { SUSPEND2_ATTR("pause_between_steps", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_PAUSE) + }, +#endif + { SUSPEND2_ATTR("enabled", SYSFS_RW), + SYSFS_INT(&userui_ops.enabled, 0, 1) + }, + + { SUSPEND2_ATTR("progress_granularity", SYSFS_RW), + SYSFS_INT(&progress_granularity, 1, 2048) + }, + + { SUSPEND2_ATTR("program", SYSFS_RW), + SYSFS_STRING(ui_helper_data.program, 255, SYSFS_SM_NOT_NEEDED) + }, +#endif + { SUSPEND2_ATTR("pm_prepare_console", SYSFS_RW), + SYSFS_BIT(&suspend_action, SUSPEND_PM_PREPARE_CONSOLE) + } +}; + +static struct suspend_module_ops userui_ops = { + .type = MISC_MODULE, + .name = "Userspace UI Support", + .module = THIS_MODULE, +#ifdef CONFIG_NET + .storage_needed = userui_storage_needed, + .save_config_info = userui_save_config_info, + .load_config_info = userui_load_config_info, + .memory_needed = userui_memory_needed, +#endif +}; + +/* suspend_console_sysfs_init + * Description: Boot time initialisation for user interface. + */ +static __init int suspend_console_sysfs_init(void) +{ + int result, i, + numfiles = sizeof(sysfs_params) / sizeof(struct suspend_sysfs_data); + + if (!(result = suspend_register_module(&userui_ops))) { + struct kobject *kobj = make_suspend2_sysdir("user_interface"); + for (i=0; i< numfiles; i++) + suspend_register_sysfs_file(kobj, &sysfs_params[i]); + } + +#ifdef CONFIG_NET + ui_helper_data.nl = NULL; + ui_helper_data.program[0] = '\0'; + ui_helper_data.pid = -1; + ui_helper_data.skb_size = sizeof(struct userui_msg_params); + ui_helper_data.pool_limit = 6; + ui_helper_data.netlink_id = NETLINK_SUSPEND2_USERUI; + ui_helper_data.name = "userspace ui"; + ui_helper_data.rcv_msg = userui_user_rcv_msg; + ui_helper_data.interface_version = 6; + ui_helper_data.must_init = 0; + ui_helper_data.not_ready = suspend_cleanup_console; + init_completion(&ui_helper_data.wait_for_process); +#endif + + return result; +} + +late_initcall(suspend_console_sysfs_init); diff -urN oldtree/kernel/power/ui.h newtree/kernel/power/ui.h --- oldtree/kernel/power/ui.h 1969-12-31 19:00:00.000000000 -0500 +++ newtree/kernel/power/ui.h 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,71 @@ +/* + * kernel/power/ui.h + * + * Copyright 2004-2006 Nigel Cunningham + */ + +extern int suspend_default_console_level; + +enum { + DONT_CLEAR_BAR, + CLEAR_BAR +}; + +enum { + /* Userspace -> Kernel */ + USERUI_MSG_ABORT = 0x11, + USERUI_MSG_SET_STATE = 0x12, + USERUI_MSG_GET_STATE = 0x13, + USERUI_MSG_GET_DEBUG_STATE = 0x14, + USERUI_MSG_SET_DEBUG_STATE = 0x15, + USERUI_MSG_SET_PROGRESS_GRANULARITY = 0x17, + USERUI_MSG_SPACE = 0x18, + + /* Kernel -> Userspace */ + USERUI_MSG_MESSAGE = 0x21, + USERUI_MSG_PROGRESS = 0x22, + USERUI_MSG_REDRAW = 0x25, + USERUI_MSG_KEYPRESS = 0x26, + USERUI_MSG_DEBUG_STATE = 0x29, + + USERUI_MSG_MAX, +}; + +struct userui_msg_params { + unsigned long a, b, c, d; + char text[255]; +}; + +#if defined(CONFIG_NET) + +extern unsigned long suspend_update_status(unsigned long value, unsigned long maximum, + const char *fmt, ...); +extern void suspend_prepare_status (int clearbar, const char *fmt, ...); +extern void suspend_cond_pause(int pause, char *message); +extern void abort_suspend(const char *fmt, ...); +extern void suspend_prepare_console(void); +extern void suspend_cleanup_console(void); +extern void userui_redraw(void); + +#else +static inline char suspend_wait_for_keypress(int timeout) +{ + return 0; +} + +static inline unsigned long suspend_update_status(unsigned long value, unsigned long maximum, + const char *fmt, ...) +{ + return maximum; +} + +static inline void __suspend_message(unsigned long section, unsigned long level, + int normally_logged, + const char *fmt, ...) { } +static inline void suspend_prepare_status(int clearbar, const char *fmt, ...) { } +static inline void suspend_cond_pause(int pause, char *message) { } +static inline void abort_suspend(const char *fmt, ...) { } +static inline void suspend_prepare_console(void) { } +static inline void suspend_cleanup_console(void) { } +static inline void userui_redraw(void) { } +#endif diff -urN oldtree/kernel/power/user.c newtree/kernel/power/user.c --- oldtree/kernel/power/user.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/power/user.c 2006-10-06 17:28:20.000000000 -0400 @@ -22,6 +22,7 @@ #include #include #include +#include #include @@ -79,7 +80,7 @@ free_bitmap(data->bitmap); if (data->frozen) { down(&pm_sem); - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); enable_nonboot_cpus(); up(&pm_sem); } @@ -148,7 +149,7 @@ if (!error) { error = freeze_processes(); if (error) { - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); error = -EBUSY; } } @@ -162,7 +163,7 @@ if (!data->frozen) break; down(&pm_sem); - thaw_processes(); + thaw_processes(FREEZER_ALL_THREADS); enable_nonboot_cpus(); up(&pm_sem); data->frozen = 0; diff -urN oldtree/kernel/timer.c newtree/kernel/timer.c --- oldtree/kernel/timer.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/kernel/timer.c 2006-10-06 17:28:20.000000000 -0400 @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -1010,6 +1011,16 @@ active_tasks = count_active_tasks(); for (count -= ticks; count < 0; count += LOAD_FREQ) { + + /* Suspend2 does a lot of work (pagecache I/O) before + * and after the atomic copy. If we let the load average + * be updated while suspending, it will be very high post + * resume. Processes such as some MTAs that stop work + * while the average is high will be unnecessarily disrupted. + */ + if (freezer_is_on()) + continue; + CALC_LOAD(avenrun[0], EXP_1, active_tasks); CALC_LOAD(avenrun[1], EXP_5, active_tasks); CALC_LOAD(avenrun[2], EXP_15, active_tasks); diff -urN oldtree/lib/Kconfig newtree/lib/Kconfig --- oldtree/lib/Kconfig 2006-10-05 15:26:55.000000000 -0400 +++ newtree/lib/Kconfig 2006-10-06 17:28:20.000000000 -0400 @@ -43,6 +43,9 @@ depends on AUDIT && !AUDIT_ARCH default y +config DYN_PAGEFLAGS + bool + # # compression support is select'ed if needed # diff -urN oldtree/lib/Makefile newtree/lib/Makefile --- oldtree/lib/Makefile 2006-10-05 15:26:55.000000000 -0400 +++ newtree/lib/Makefile 2006-10-06 17:28:20.000000000 -0400 @@ -35,6 +35,8 @@ lib-y += dec_and_lock.o endif +obj-$(CONFIG_DYN_PAGEFLAGS) += dyn_pageflags.o + obj-$(CONFIG_CRC_CCITT) += crc-ccitt.o obj-$(CONFIG_CRC16) += crc16.o obj-$(CONFIG_CRC32) += crc32.o diff -urN oldtree/lib/dyn_pageflags.c newtree/lib/dyn_pageflags.c --- oldtree/lib/dyn_pageflags.c 1969-12-31 19:00:00.000000000 -0500 +++ newtree/lib/dyn_pageflags.c 2006-10-06 17:28:20.000000000 -0400 @@ -0,0 +1,257 @@ +/* + * lib/dyn_pageflags.c + * + * Copyright (C) 2004-2006 Nigel Cunningham + * + * This file is released under the GPLv2. + * + * Routines for dynamically allocating and releasing bitmaps + * used as pseudo-pageflags. + */ + +#include +#include +#include +#include + +#define page_to_zone_offset(pg) (page_to_pfn(pg) - page_zone(pg)->zone_start_pfn) + +/* + * pages_for_zone(struct zone *zone) + * + * How many pages do we need for a bitmap for this zone? + * + */ + +static int pages_for_zone(struct zone *zone) +{ + return (zone->spanned_pages + (PAGE_SIZE << 3) - 1) >> + (PAGE_SHIFT + 3); +} + +/* + * clear_dyn_pageflags(dyn_pageflags_t pagemap) + * + * Clear an array used to store local page flags. + * + */ + +void clear_dyn_pageflags(dyn_pageflags_t pagemap) +{ + int i = 0, zone_num = 0; + struct zone *zone; + + BUG_ON(!pagemap); + + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + for (i = 0; i < pages_for_zone(zone); i++) + memset((pagemap[zone_num][i]), 0, PAGE_SIZE); + } +} + +/* + * allocate_dyn_pageflags(dyn_pageflags_t *pagemap) + * + * Allocate a bitmap for dynamic page flags. + * + */ +int allocate_dyn_pageflags(dyn_pageflags_t *pagemap) +{ + int i, zone_num, zone_pages; + struct zone *zone; + + BUG_ON(*pagemap); + + *pagemap = kmalloc(sizeof(void *) * (1 << ZONEID_SHIFT), GFP_ATOMIC); + + if (!*pagemap) + return -ENOMEM; + + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + zone_pages = pages_for_zone(zone); + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + (*pagemap)[zone_num] = kmalloc(sizeof(void *) * zone_pages, + GFP_ATOMIC); + + if (!(*pagemap)[zone_num]) { + free_dyn_pageflags(pagemap); + return -ENOMEM; + } + + for (i = 0; i < zone_pages; i++) { + unsigned long address = get_zeroed_page(GFP_ATOMIC); + (*pagemap)[zone_num][i] = (unsigned long *) address; + if (!(*pagemap)[zone_num][i]) { + printk("Error. Unable to allocate memory for " + "dynamic pageflags."); + free_dyn_pageflags(pagemap); + return -ENOMEM; + } + } + } + + return 0; +} + +/* + * free_dyn_pageflags(dyn_pageflags_t *pagemap) + * + * Free a dynamically allocated pageflags bitmap. For Suspend2 usage, we + * support data being relocated from slab to pages that don't conflict + * with the image that will be copied back. This is the reason for the + * PageSlab tests below. + * + */ +void free_dyn_pageflags(dyn_pageflags_t *pagemap) +{ + int i = 0, zone_num, zone_pages; + struct zone *zone; + + if (!*pagemap) + return; + + for_each_zone(zone) { + if (!populated_zone(zone)) + continue; + + zone_pages = pages_for_zone(zone); + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + + /* + * May be called on an error path in allocating, so this + * isn't redundant. + */ + if (!((*pagemap)[zone_num])) + continue; + + for (i = 0; i < zone_pages; i++) + if ((*pagemap)[zone_num][i]) + free_page((unsigned long) (*pagemap)[zone_num][i]); + + if (PageSlab(virt_to_page((*pagemap)[zone_num]))) + kfree((*pagemap)[zone_num]); + else + free_page((unsigned long) (*pagemap)[zone_num]); + } + + if (PageSlab(virt_to_page((*pagemap)))) + kfree(*pagemap); + else + free_page((unsigned long) (*pagemap)); + + *pagemap = NULL; + return; +} + +#define GET_BIT_AND_UL(bitmap, page) \ + unsigned long zone_pfn = page_to_zone_offset(page); \ + int zone_num = page_zone_id(page); \ + int pagenum = PAGENUMBER(zone_pfn); \ + int page_offset = PAGEINDEX(zone_pfn); \ + unsigned long *ul = ((*bitmap)[zone_num][pagenum]) + page_offset; \ + int bit = PAGEBIT(zone_pfn); + +/* + * test_dynpageflag(dyn_pageflags_t *bitmap, struct page *page) + * + * Is the page flagged in the given bitmap? + * + */ + +int test_dynpageflag(dyn_pageflags_t *bitmap, struct page *page) +{ + GET_BIT_AND_UL(bitmap, page); + return test_bit(bit, ul); +} + +/* + * set_dynpageflag(dyn_pageflags_t *bitmap, struct page *page) + * + * Set the flag for the page in the given bitmap. + * + */ + +void set_dynpageflag(dyn_pageflags_t *bitmap, struct page *page) +{ + GET_BIT_AND_UL(bitmap, page); + set_bit(bit, ul); +} + +/* + * clear_dynpageflags(dyn_pageflags_t *bitmap, struct page *page) + * + * Clear the flag for the page in the given bitmap. + * + */ + +void clear_dynpageflag(dyn_pageflags_t *bitmap, struct page *page) +{ + GET_BIT_AND_UL(bitmap, page); + clear_bit(bit, ul); +} + +/* + * get_next_bit_on(dyn_pageflags_t bitmap, int counter) + * + * Given a pfn (possibly -1), find the next pfn in the bitmap that + * is set. If there are no more flags set, return -1. + * + */ + +unsigned long get_next_bit_on(dyn_pageflags_t bitmap, unsigned long counter) +{ + struct page *page; + struct zone *zone; + unsigned long *ul = NULL; + unsigned long zone_offset; + int pagebit, zone_num, first = (counter == max_pfn); + + if (first) + counter = first_online_pgdat()->node_zones->zone_start_pfn; + + page = pfn_to_page(counter); + zone = page_zone(page); + zone_num = page_zone_id(page); + zone_offset = counter - zone->zone_start_pfn; + + if (first) + goto test; + + do { + zone_offset++; + + if (zone_offset >= zone->spanned_pages) { + do { + zone = next_zone(zone); + if (!zone) + return max_pfn; + } while(!zone->spanned_pages); + + zone_num = page_zone_id(pfn_to_page(zone->zone_start_pfn)); + zone_offset = 0; + } +test: + pagebit = PAGEBIT(zone_offset); + + if (!pagebit || !ul) + ul = (bitmap[zone_num][PAGENUMBER(zone_offset)]) + PAGEINDEX(zone_offset); + + if (!(*ul & ~((1 << pagebit) - 1))) { + zone_offset += BITS_PER_LONG - pagebit - 1; + continue; + } + + } while(!test_bit(pagebit, ul)); + + return zone->zone_start_pfn + zone_offset; +} + diff -urN oldtree/lib/vsprintf.c newtree/lib/vsprintf.c --- oldtree/lib/vsprintf.c 2006-10-05 15:26:55.000000000 -0400 +++ newtree/lib/vsprintf.c 2006-10-06 17:28:20.000000000 -0400 @@ -236,6 +236,29 @@ return buf; } +/* + * vsnprintf_used + * + * Functionality : Print a string with parameters to a buffer of a + * limited size. Unlike vsnprintf, we return the number + * of bytes actually put in the buffer, not the number + * that would have been put in if it was big enough. + */ +int snprintf_used(char *buffer, int buffer_size, const char *fmt, ...) +{ + int result; + va_list args; + + if (!buffer_size) + return 0; + + va_start(args, fmt); + result = vsnprintf(buffer, buffer_size, fmt, args); + va_end(args); + + return result > buffer_size ? buffer_size : result; +} + /** * vsnprintf - Format a string and place it in a buffer * @buf: The buffer to place the result into diff -urN oldtree/mm/memory.c newtree/mm/memory.c --- oldtree/mm/memory.c 2006-10-06 16:28:30.000000000 -0400 +++ newtree/mm/memory.c 2006-10-06 17:28:20.000000000 -0400 @@ -976,6 +976,15 @@ return page; } +/* + * We want the address of the page for Suspend2 to mark as being in pageset1. + */ + +struct page *suspend2_follow_page(struct mm_struct *mm, unsigned long address) +{ + return follow_page(mm->mmap, address, 0); +} + int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, unsigned long start, int len, int write, int force, struct page **pages, struct vm_area_struct **vmas) diff -urN oldtree/mm/vmscan.c newtree/mm/vmscan.c --- oldtree/mm/vmscan.c 2006-10-06 16:52:04.000000000 -0400 +++ newtree/mm/vmscan.c 2006-10-06 17:28:20.000000000 -0400 @@ -1151,6 +1151,77 @@ return ret; } +struct lru_save { + struct zone *zone; + struct list_head active_list; + struct list_head inactive_list; + unsigned long nr_active; + unsigned long nr_inactive; + struct lru_save *next; +}; + +struct lru_save *lru_save_list; + +void unlink_lru_lists(void) +{ + struct zone *zone; + + for_each_zone(zone) { + struct lru_save *this; + + if (!zone->spanned_pages) + continue; + + this = (struct lru_save *) + kzalloc(sizeof(struct lru_save), GFP_ATOMIC); + + BUG_ON(!this); + + this->next = lru_save_list; + lru_save_list = this; + + this->zone = zone; + + spin_lock_irq(&zone->lru_lock); + this->nr_active = zone->nr_active; + this->nr_inactive = zone->nr_inactive; + memcpy((char *) &this->active_list, (char *) &zone->active_list, + sizeof(struct list_head)); + memcpy((char *) &this->inactive_list, + (char *) &zone->inactive_list, + sizeof(struct list_head)); + + zone->nr_active = zone->nr_inactive = 0; + INIT_LIST_HEAD(&zone->active_list); + INIT_LIST_HEAD(&zone->inactive_list); + + spin_unlock_irq(&zone->lru_lock); + } +} + +void relink_lru_lists(void) +{ + while(lru_save_list) { + struct lru_save *this = lru_save_list; + struct zone *zone = this->zone; + + lru_save_list = this->next; + + spin_lock_irq(&zone->lru_lock); + zone->nr_active = this->nr_active; + zone->nr_inactive = this->nr_inactive; + memcpy((char *) &zone->active_list, (char *) &this->active_list, + sizeof(struct list_head)); + memcpy((char *) &zone->inactive_list, + (char *) &this->inactive_list, + sizeof(struct list_head)); + + spin_unlock_irq(&zone->lru_lock); + + kfree(this); + } +} + /* * For kswapd, balance_pgdat() will work across all this node's zones until * they are all at pages_high.