Provided by: erlang-manpages_25.2.3+dfsg-1_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)