cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

firmware-assisted-dump.rst (17254B)


      1======================
      2Firmware-Assisted Dump
      3======================
      4
      5July 2011
      6
      7The goal of firmware-assisted dump is to enable the dump of
      8a crashed system, and to do so from a fully-reset system, and
      9to minimize the total elapsed time until the system is back
     10in production use.
     11
     12- Firmware-Assisted Dump (FADump) infrastructure is intended to replace
     13  the existing phyp assisted dump.
     14- Fadump uses the same firmware interfaces and memory reservation model
     15  as phyp assisted dump.
     16- Unlike phyp dump, FADump exports the memory dump through /proc/vmcore
     17  in the ELF format in the same way as kdump. This helps us reuse the
     18  kdump infrastructure for dump capture and filtering.
     19- Unlike phyp dump, userspace tool does not need to refer any sysfs
     20  interface while reading /proc/vmcore.
     21- Unlike phyp dump, FADump allows user to release all the memory reserved
     22  for dump, with a single operation of echo 1 > /sys/kernel/fadump_release_mem.
     23- Once enabled through kernel boot parameter, FADump can be
     24  started/stopped through /sys/kernel/fadump_registered interface (see
     25  sysfs files section below) and can be easily integrated with kdump
     26  service start/stop init scripts.
     27
     28Comparing with kdump or other strategies, firmware-assisted
     29dump offers several strong, practical advantages:
     30
     31-  Unlike kdump, the system has been reset, and loaded
     32   with a fresh copy of the kernel.  In particular,
     33   PCI and I/O devices have been reinitialized and are
     34   in a clean, consistent state.
     35-  Once the dump is copied out, the memory that held the dump
     36   is immediately available to the running kernel. And therefore,
     37   unlike kdump, FADump doesn't need a 2nd reboot to get back
     38   the system to the production configuration.
     39
     40The above can only be accomplished by coordination with,
     41and assistance from the Power firmware. The procedure is
     42as follows:
     43
     44-  The first kernel registers the sections of memory with the
     45   Power firmware for dump preservation during OS initialization.
     46   These registered sections of memory are reserved by the first
     47   kernel during early boot.
     48
     49-  When system crashes, the Power firmware will copy the registered
     50   low memory regions (boot memory) from source to destination area.
     51   It will also save hardware PTE's.
     52
     53   NOTE:
     54         The term 'boot memory' means size of the low memory chunk
     55         that is required for a kernel to boot successfully when
     56         booted with restricted memory. By default, the boot memory
     57         size will be the larger of 5% of system RAM or 256MB.
     58         Alternatively, user can also specify boot memory size
     59         through boot parameter 'crashkernel=' which will override
     60         the default calculated size. Use this option if default
     61         boot memory size is not sufficient for second kernel to
     62         boot successfully. For syntax of crashkernel= parameter,
     63         refer to Documentation/admin-guide/kdump/kdump.rst. If any
     64         offset is provided in crashkernel= parameter, it will be
     65         ignored as FADump uses a predefined offset to reserve memory
     66         for boot memory dump preservation in case of a crash.
     67
     68-  After the low memory (boot memory) area has been saved, the
     69   firmware will reset PCI and other hardware state.  It will
     70   *not* clear the RAM. It will then launch the bootloader, as
     71   normal.
     72
     73-  The freshly booted kernel will notice that there is a new node
     74   (rtas/ibm,kernel-dump on pSeries or ibm,opal/dump/mpipl-boot
     75   on OPAL platform) in the device tree, indicating that
     76   there is crash data available from a previous boot. During
     77   the early boot OS will reserve rest of the memory above
     78   boot memory size effectively booting with restricted memory
     79   size. This will make sure that this kernel (also, referred
     80   to as second kernel or capture kernel) will not touch any
     81   of the dump memory area.
     82
     83-  User-space tools will read /proc/vmcore to obtain the contents
     84   of memory, which holds the previous crashed kernel dump in ELF
     85   format. The userspace tools may copy this info to disk, or
     86   network, nas, san, iscsi, etc. as desired.
     87
     88-  Once the userspace tool is done saving dump, it will echo
     89   '1' to /sys/kernel/fadump_release_mem to release the reserved
     90   memory back to general use, except the memory required for
     91   next firmware-assisted dump registration.
     92
     93   e.g.::
     94
     95     # echo 1 > /sys/kernel/fadump_release_mem
     96
     97Please note that the firmware-assisted dump feature
     98is only available on POWER6 and above systems on pSeries
     99(PowerVM) platform and POWER9 and above systems with OP940
    100or later firmware versions on PowerNV (OPAL) platform.
    101Note that, OPAL firmware exports ibm,opal/dump node when
    102FADump is supported on PowerNV platform.
    103
    104On OPAL based machines, system first boots into an intermittent
    105kernel (referred to as petitboot kernel) before booting into the
    106capture kernel. This kernel would have minimal kernel and/or
    107userspace support to process crash data. Such kernel needs to
    108preserve previously crash'ed kernel's memory for the subsequent
    109capture kernel boot to process this crash data. Kernel config
    110option CONFIG_PRESERVE_FA_DUMP has to be enabled on such kernel
    111to ensure that crash data is preserved to process later.
    112
    113-- On OPAL based machines (PowerNV), if the kernel is build with
    114   CONFIG_OPAL_CORE=y, OPAL memory at the time of crash is also
    115   exported as /sys/firmware/opal/mpipl/core file. This procfs file is
    116   helpful in debugging OPAL crashes with GDB. The kernel memory
    117   used for exporting this procfs file can be released by echo'ing
    118   '1' to /sys/firmware/opal/mpipl/release_core node.
    119
    120   e.g.
    121     # echo 1 > /sys/firmware/opal/mpipl/release_core
    122
    123Implementation details:
    124-----------------------
    125
    126During boot, a check is made to see if firmware supports
    127this feature on that particular machine. If it does, then
    128we check to see if an active dump is waiting for us. If yes
    129then everything but boot memory size of RAM is reserved during
    130early boot (See Fig. 2). This area is released once we finish
    131collecting the dump from user land scripts (e.g. kdump scripts)
    132that are run. If there is dump data, then the
    133/sys/kernel/fadump_release_mem file is created, and the reserved
    134memory is held.
    135
    136If there is no waiting dump data, then only the memory required to
    137hold CPU state, HPTE region, boot memory dump, FADump header and
    138elfcore header, is usually reserved at an offset greater than boot
    139memory size (see Fig. 1). This area is *not* released: this region
    140will be kept permanently reserved, so that it can act as a receptacle
    141for a copy of the boot memory content in addition to CPU state and
    142HPTE region, in the case a crash does occur.
    143
    144Since this reserved memory area is used only after the system crash,
    145there is no point in blocking this significant chunk of memory from
    146production kernel. Hence, the implementation uses the Linux kernel's
    147Contiguous Memory Allocator (CMA) for memory reservation if CMA is
    148configured for kernel. With CMA reservation this memory will be
    149available for applications to use it, while kernel is prevented from
    150using it. With this FADump will still be able to capture all of the
    151kernel memory and most of the user space memory except the user pages
    152that were present in CMA region::
    153
    154  o Memory Reservation during first kernel
    155
    156  Low memory                                                 Top of memory
    157  0    boot memory size   |<--- Reserved dump area --->|       |
    158  |           |           |    Permanent Reservation   |       |
    159  V           V           |                            |       V
    160  +-----------+-----/ /---+---+----+-------+-----+-----+----+--+
    161  |           |           |///|////|  DUMP | HDR | ELF |////|  |
    162  +-----------+-----/ /---+---+----+-------+-----+-----+----+--+
    163        |                   ^    ^     ^      ^           ^
    164        |                   |    |     |      |           |
    165        \                  CPU  HPTE   /      |           |
    166         ------------------------------       |           |
    167      Boot memory content gets transferred    |           |
    168      to reserved area by firmware at the     |           |
    169      time of crash.                          |           |
    170                                          FADump Header   |
    171                                           (meta area)    |
    172                                                          |
    173                                                          |
    174                      Metadata: This area holds a metadata structure whose
    175                      address is registered with f/w and retrieved in the
    176                      second kernel after crash, on platforms that support
    177                      tags (OPAL). Having such structure with info needed
    178                      to process the crashdump eases dump capture process.
    179
    180                   Fig. 1
    181
    182
    183  o Memory Reservation during second kernel after crash
    184
    185  Low memory                                              Top of memory
    186  0      boot memory size                                      |
    187  |           |<------------ Crash preserved area ------------>|
    188  V           V           |<--- Reserved dump area --->|       |
    189  +-----------+-----/ /---+---+----+-------+-----+-----+----+--+
    190  |           |           |///|////|  DUMP | HDR | ELF |////|  |
    191  +-----------+-----/ /---+---+----+-------+-----+-----+----+--+
    192        |                                           |
    193        V                                           V
    194   Used by second                             /proc/vmcore
    195   kernel to boot
    196
    197        +---+
    198        |///| -> Regions (CPU, HPTE & Metadata) marked like this in the above
    199        +---+    figures are not always present. For example, OPAL platform
    200                 does not have CPU & HPTE regions while Metadata region is
    201                 not supported on pSeries currently.
    202
    203                   Fig. 2
    204
    205
    206Currently the dump will be copied from /proc/vmcore to a new file upon
    207user intervention. The dump data available through /proc/vmcore will be
    208in ELF format. Hence the existing kdump infrastructure (kdump scripts)
    209to save the dump works fine with minor modifications. KDump scripts on
    210major Distro releases have already been modified to work seamlessly (no
    211user intervention in saving the dump) when FADump is used, instead of
    212KDump, as dump mechanism.
    213
    214The tools to examine the dump will be same as the ones
    215used for kdump.
    216
    217How to enable firmware-assisted dump (FADump):
    218----------------------------------------------
    219
    2201. Set config option CONFIG_FA_DUMP=y and build kernel.
    2212. Boot into linux kernel with 'fadump=on' kernel cmdline option.
    222   By default, FADump reserved memory will be initialized as CMA area.
    223   Alternatively, user can boot linux kernel with 'fadump=nocma' to
    224   prevent FADump to use CMA.
    2253. Optionally, user can also set 'crashkernel=' kernel cmdline
    226   to specify size of the memory to reserve for boot memory dump
    227   preservation.
    228
    229NOTE:
    230     1. 'fadump_reserve_mem=' parameter has been deprecated. Instead
    231        use 'crashkernel=' to specify size of the memory to reserve
    232        for boot memory dump preservation.
    233     2. If firmware-assisted dump fails to reserve memory then it
    234        will fallback to existing kdump mechanism if 'crashkernel='
    235        option is set at kernel cmdline.
    236     3. if user wants to capture all of user space memory and ok with
    237        reserved memory not available to production system, then
    238        'fadump=nocma' kernel parameter can be used to fallback to
    239        old behaviour.
    240
    241Sysfs/debugfs files:
    242--------------------
    243
    244Firmware-assisted dump feature uses sysfs file system to hold
    245the control files and debugfs file to display memory reserved region.
    246
    247Here is the list of files under kernel sysfs:
    248
    249 /sys/kernel/fadump_enabled
    250    This is used to display the FADump status.
    251
    252    - 0 = FADump is disabled
    253    - 1 = FADump is enabled
    254
    255    This interface can be used by kdump init scripts to identify if
    256    FADump is enabled in the kernel and act accordingly.
    257
    258 /sys/kernel/fadump_registered
    259    This is used to display the FADump registration status as well
    260    as to control (start/stop) the FADump registration.
    261
    262    - 0 = FADump is not registered.
    263    - 1 = FADump is registered and ready to handle system crash.
    264
    265    To register FADump echo 1 > /sys/kernel/fadump_registered and
    266    echo 0 > /sys/kernel/fadump_registered for un-register and stop the
    267    FADump. Once the FADump is un-registered, the system crash will not
    268    be handled and vmcore will not be captured. This interface can be
    269    easily integrated with kdump service start/stop.
    270
    271 /sys/kernel/fadump/mem_reserved
    272
    273   This is used to display the memory reserved by FADump for saving the
    274   crash dump.
    275
    276 /sys/kernel/fadump_release_mem
    277    This file is available only when FADump is active during
    278    second kernel. This is used to release the reserved memory
    279    region that are held for saving crash dump. To release the
    280    reserved memory echo 1 to it::
    281
    282	echo 1  > /sys/kernel/fadump_release_mem
    283
    284    After echo 1, the content of the /sys/kernel/debug/powerpc/fadump_region
    285    file will change to reflect the new memory reservations.
    286
    287    The existing userspace tools (kdump infrastructure) can be easily
    288    enhanced to use this interface to release the memory reserved for
    289    dump and continue without 2nd reboot.
    290
    291Note: /sys/kernel/fadump_release_opalcore sysfs has moved to
    292      /sys/firmware/opal/mpipl/release_core
    293
    294 /sys/firmware/opal/mpipl/release_core
    295
    296    This file is available only on OPAL based machines when FADump is
    297    active during capture kernel. This is used to release the memory
    298    used by the kernel to export /sys/firmware/opal/mpipl/core file. To
    299    release this memory, echo '1' to it:
    300
    301    echo 1  > /sys/firmware/opal/mpipl/release_core
    302
    303Note: The following FADump sysfs files are deprecated.
    304
    305+----------------------------------+--------------------------------+
    306| Deprecated                       | Alternative                    |
    307+----------------------------------+--------------------------------+
    308| /sys/kernel/fadump_enabled       | /sys/kernel/fadump/enabled     |
    309+----------------------------------+--------------------------------+
    310| /sys/kernel/fadump_registered    | /sys/kernel/fadump/registered  |
    311+----------------------------------+--------------------------------+
    312| /sys/kernel/fadump_release_mem   | /sys/kernel/fadump/release_mem |
    313+----------------------------------+--------------------------------+
    314
    315Here is the list of files under powerpc debugfs:
    316(Assuming debugfs is mounted on /sys/kernel/debug directory.)
    317
    318 /sys/kernel/debug/powerpc/fadump_region
    319    This file shows the reserved memory regions if FADump is
    320    enabled otherwise this file is empty. The output format
    321    is::
    322
    323      <region>: [<start>-<end>] <reserved-size> bytes, Dumped: <dump-size>
    324
    325    and for kernel DUMP region is:
    326
    327    DUMP: Src: <src-addr>, Dest: <dest-addr>, Size: <size>, Dumped: # bytes
    328
    329    e.g.
    330    Contents when FADump is registered during first kernel::
    331
    332      # cat /sys/kernel/debug/powerpc/fadump_region
    333      CPU : [0x0000006ffb0000-0x0000006fff001f] 0x40020 bytes, Dumped: 0x0
    334      HPTE: [0x0000006fff0020-0x0000006fff101f] 0x1000 bytes, Dumped: 0x0
    335      DUMP: [0x0000006fff1020-0x0000007fff101f] 0x10000000 bytes, Dumped: 0x0
    336
    337    Contents when FADump is active during second kernel::
    338
    339      # cat /sys/kernel/debug/powerpc/fadump_region
    340      CPU : [0x0000006ffb0000-0x0000006fff001f] 0x40020 bytes, Dumped: 0x40020
    341      HPTE: [0x0000006fff0020-0x0000006fff101f] 0x1000 bytes, Dumped: 0x1000
    342      DUMP: [0x0000006fff1020-0x0000007fff101f] 0x10000000 bytes, Dumped: 0x10000000
    343          : [0x00000010000000-0x0000006ffaffff] 0x5ffb0000 bytes, Dumped: 0x5ffb0000
    344
    345
    346NOTE:
    347      Please refer to Documentation/filesystems/debugfs.rst on
    348      how to mount the debugfs filesystem.
    349
    350
    351TODO:
    352-----
    353 - Need to come up with the better approach to find out more
    354   accurate boot memory size that is required for a kernel to
    355   boot successfully when booted with restricted memory.
    356 - The FADump implementation introduces a FADump crash info structure
    357   in the scratch area before the ELF core header. The idea of introducing
    358   this structure is to pass some important crash info data to the second
    359   kernel which will help second kernel to populate ELF core header with
    360   correct data before it gets exported through /proc/vmcore. The current
    361   design implementation does not address a possibility of introducing
    362   additional fields (in future) to this structure without affecting
    363   compatibility. Need to come up with the better approach to address this.
    364
    365   The possible approaches are:
    366
    367	1. Introduce version field for version tracking, bump up the version
    368	whenever a new field is added to the structure in future. The version
    369	field can be used to find out what fields are valid for the current
    370	version of the structure.
    371	2. Reserve the area of predefined size (say PAGE_SIZE) for this
    372	structure and have unused area as reserved (initialized to zero)
    373	for future field additions.
    374
    375   The advantage of approach 1 over 2 is we don't need to reserve extra space.
    376
    377Author: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
    378
    379This document is based on the original documentation written for phyp
    380
    381assisted dump by Linas Vepstas and Manish Ahuja.