Provided by: erlang-manpages_25.3.2.8+dfsg-1ubuntu4.4_all bug

NAME

       instrument - Analysis and Utility Functions for Instrumentation

DESCRIPTION

       The  module  instrument  contains  support  for  studying the resource usage in an Erlang runtime system.
       Currently, only the allocation of memory can be studied.

   Note:
       Since this module inspects internal details of the runtime system it may differ greatly from one  version
       to another. We make no compatibility guarantees in this module.

DATA TYPES

       block_histogram() = tuple()

              A  histogram  of  block sizes where each interval's upper bound is twice as high as the one before
              it.

              The upper bound of the first interval is provided by the function that returned the histogram, and
              the last interval has no upper bound.

              For example, the histogram below has 40 (message) blocks between 256-512 bytes in size, 78  blocks
              between 512-1024 bytes,2 blocks between 1-2KB, and 2 blocks between 2-4KB.

              > instrument:allocations(#{ histogram_start => 128, histogram_width => 15 }).
              {ok, {128, 0, #{ message => {0,40,78,2,2,0,0,0,0,0,0,0,0,0,0}, ... } }}

       allocation_summary() =
           {HistogramStart :: integer() >= 0,
            UnscannedSize :: integer() >= 0,
            Allocations ::
                #{Origin :: atom() =>
                      #{Type :: atom() => block_histogram()}}}

              A summary of allocated block sizes (including their headers) grouped by their Origin and Type.

              Origin  is generally which NIF or driver that allocated the blocks, or 'system' if it could not be
              determined.

              Type is the allocation category that the blocks belong to, e.g. db_term, message  or  binary.  The
              categories correspond to those in  erl_alloc.types.

              If  one  or  more  carriers could not be scanned in full without harming the responsiveness of the
              system, UnscannedSize is the number of bytes that had to be skipped.

       carrier_info_list() =
           {HistogramStart :: integer() >= 0,
            Carriers ::
                [{AllocatorType :: atom(),
                  InPool :: boolean(),
                  TotalSize :: integer() >= 0,
                  UnscannedSize :: integer() >= 0,
                  Allocations ::
                      {Type :: atom(),
                       Count :: integer() >= 0,
                       Size :: integer() >= 0},
                  FreeBlocks :: block_histogram()}]}

              AllocatorType is the type of the allocator that employs this carrier.

              InPool is whether the carrier is in the migration pool.

              TotalSize is the total size of the carrier, including its header.

              Allocations is a summary of the allocated blocks in the carrier. Note that  carriers  may  contain
              multiple  different  block  types  when carrier pools are shared between different allocator types
              (see the erts_alloc documentation for more details).

              FreeBlocks is a histogram of the free block sizes in the carrier.

              If the carrier could not be scanned in full without harming  the  responsiveness  of  the  system,
              UnscannedSize is the number of bytes that had to be skipped.

EXPORTS

       allocations() -> {ok, Result} | {error, Reason}

              Types:

                 Result = allocation_summary()
                 Reason = not_enabled

              Shorthand for allocations(#{}).

       allocations(Options) -> {ok, Result} | {error, Reason}

              Types:

                 Result = allocation_summary()
                 Reason = not_enabled
                 Options =
                     #{scheduler_ids => [integer() >= 0],
                       allocator_types => [atom()],
                       histogram_start => integer() >= 1,
                       histogram_width => integer() >= 1}

              Returns  a  summary of all tagged allocations in the system, optionally filtered by allocator type
              and scheduler id.

              Only binaries and allocations made by NIFs and drivers are tagged by  default,  but  this  can  be
              configured an a per-allocator basis with the +M<S>atags emulator option.

              If the specified allocator types are not enabled, the call will fail with {error, not_enabled}.

              The following options can be used:

                allocator_types:
                  The allocator types that will be searched.

                  Specifying  a  specific  allocator  type  may  lead  to strange results when carrier migration
                  between different allocator types has been enabled: you may see unexpected types (e.g. process
                  heaps when searching binary_alloc), or fewer blocks than expected if the carriers  the  blocks
                  are on have been migrated out to an allocator of a different type.

                  Defaults to all alloc_util allocators.

                scheduler_ids:
                  The  scheduler  ids whose allocator instances will be searched. A scheduler id of 0 will refer
                  to the global instance that  is  not  tied  to  any  particular  scheduler.  Defaults  to  all
                  schedulers and the global instance.

                histogram_start:
                  The upper bound of the first interval in the allocated block size histograms. Defaults to 128.

                histogram_width:
                  The number of intervals in the allocated block size histograms. Defaults to 18.

              Example:

              > instrument:allocations(#{ histogram_start => 128, histogram_width => 15 }).
              {ok,{128,0,
                   #{udp_inet =>
                         #{driver_event_state => {0,0,0,0,0,0,0,0,0,1,0,0,0,0,0}},
                     system =>
                         #{heap => {0,0,0,0,20,4,2,2,2,3,0,1,0,0,1},
                           db_term => {271,3,1,52,80,1,0,0,0,0,0,0,0,0,0},
                           code => {0,0,0,5,3,6,11,22,19,20,10,2,1,0,0},
                           binary => {18,0,0,0,7,0,0,1,0,0,0,0,0,0,0},
                           message => {0,40,78,2,2,0,0,0,0,0,0,0,0,0,0},
                           ... }
                     spawn_forker =>
                         #{driver_select_data_state =>
                               {1,0,0,0,0,0,0,0,0,0,0,0,0,0,0}},
                     ram_file_drv => #{drv_binary => {0,0,0,0,0,0,1,0,0,0,0,0,0,0,0}},
                     prim_file =>
                         #{process_specific_data => {2,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
                           nif_trap_export_entry => {0,4,0,0,0,0,0,0,0,0,0,0,0,0,0},
                           monitor_extended => {0,1,0,0,0,0,0,0,0,0,0,0,0,0,0},
                           drv_binary => {0,0,0,0,0,0,1,0,3,5,0,0,0,1,0},
                           binary => {0,4,0,0,0,0,0,0,0,0,0,0,0,0,0}},
                     prim_buffer =>
                         #{nif_internal => {0,4,0,0,0,0,0,0,0,0,0,0,0,0,0},
                           binary => {0,4,0,0,0,0,0,0,0,0,0,0,0,0,0}}}}}

       carriers() -> {ok, Result} | {error, Reason}

              Types:

                 Result = carrier_info_list()
                 Reason = not_enabled

              Shorthand for carriers(#{}).

       carriers(Options) -> {ok, Result} | {error, Reason}

              Types:

                 Result = carrier_info_list()
                 Reason = not_enabled
                 Options =
                     #{scheduler_ids => [integer() >= 0],
                       allocator_types => [atom()],
                       histogram_start => integer() >= 1,
                       histogram_width => integer() >= 1}

              Returns  a  summary  of  all  carriers  in  the  system, optionally filtered by allocator type and
              scheduler id.

              If the specified allocator types are not enabled, the call will fail with {error, not_enabled}.

              The following options can be used:

                allocator_types:
                  The allocator types that will be searched. Defaults to all alloc_util allocators.

                scheduler_ids:
                  The scheduler ids whose allocator instances will be searched. A scheduler id of 0  will  refer
                  to  the  global  instance  that  is  not  tied  to  any  particular scheduler. Defaults to all
                  schedulers and the global instance.

                histogram_start:
                  The upper bound of the first interval in the free block size histograms. Defaults to 512.

                histogram_width:
                  The number of intervals in the free block size histograms. Defaults to 14.

              Example:

              > instrument:carriers(#{ histogram_start => 512, histogram_width => 8 }).
              {ok,{512,
                   [{driver_alloc,false,262144,0,
                                  [{driver_alloc,1,32784}],
                                  {0,0,0,0,0,0,0,1}},
                    {binary_alloc,false,32768,0,
                                  [{binary_alloc,15,4304}],
                                  {3,0,0,0,1,0,0,0}},
                    {...}|...]}}

SEE ALSO

       erts_alloc(3erl), erl(1)

Ericsson AB                                        tools 3.5.3                                  instrument(3erl)