.\" Copyright (c) 1990 Regents of the University of California. .\" All rights reserved. .\" .\" %sccs.include.redist.man% .\" .\" %W% (Berkeley) %G% .\" .Dd %Q% .Dt mprof 1 .Sh NAME .Nm mprof .Nd display dynamic memory allocation data .Sh SYNTAX .Nm mprof .Op options .Op Ar a.out Op Ar mprof.data .Pp .Ft void .Fn set_mprof_autosave "int count" .Pp .Ft void .Fn mprof_stop .Pp .Ft void .Fn mprof_restart "char *filename" .Sh DESCRIPTION The .Nm mprof command produces four tables that summarize the memory allocation behavior of C programs, similar in style to the .Xr gprof 1 command. The arguments to .Nm mprof are the executable image .Pf ( Xr a.out 5 default) and the profile data file .Pf ( Ar mprof.data default). The .Ar mprof.data file is generated by linking a special version .Xr malloc 3 into the executing image. This new version, found in the library .Pa libc_mp.a must be linked in at the end of the command that creates the executable image. For example: .Pp .Dl I cc -g -o test main.o sub1.o sub2.o libc_mp.a .De .Pp Users' programs can contain additional calls to customize the user interface to .Nm mprof . The function .Fn set_mprof_autosave allows users to save the profile data periodically. The .Fa count parameter specifies to save after that number of allocations. A value of 10,000 or 100,000 is typical for the .Fa count parameter for long running programs. A value of 0 (the default) causes the the profile data to be written only when the program exits. The function .Fn mprof_stop causes memory profiling to be discontinued and the profile data to be written to the output file. The function .Fn mprof_restart restarts profiling. The .Ar filename parameter to .Fn mprof_restart specifies the name of the file to write the profile data to. .Pp The output of .Nm mprof consists of four tables, the fields of which are described in detail below. The first table breaks down the memory allocation of the program by the number of bytes requested. For each byte size the number of allocations and frees is listed along with the program structure types that correspond to that byte size. .Pp The second table lists partial call chains over which memory was allocated and never freed (call chains resulting in memory leaks). The table shows how much memory was allocated by each chain and how much each chain contributed to the total memory leakage. .Pp The third table lists the functions in which allocation occurred directly (i.e., called .Xr malloc ) , indicates how much memory was allocated, shows how much of that was not later freed, breaks down allocation roughly by size, and shows how many times each function was called. .Pp The fourth table contains the subgraph of the program's dynamic call graph in which allocation occurred. This table allows programmers to identify what functions were indirectly responsible for memory allocation. .Pp The following options are available: .Tw Ds .Tl Fl verbose Every bin in which memory was allocated is printed; the call chain for every memory leak is shown. .Tl Fl normal Only bins that contributed a reasonable fraction to the total allocation are printed; call chains for leaks contributing more than 0.5% to the total are shown. This is the default verbosity setting. .Tl Fl terse Only bins that contributed a significant fraction to the total allocation are printed. Call chains contributing more than 1% to the total leakage are shown. .Tl Fl leaktable Print out the memory leak table without printing out call site offsets. This is the default. .Tl Fl noleaktable Do not print out the memory leak table. .Tl Fl offsets Print out the memory leak table and distinguish different call sites within a function by indicating the offset in the function as part of the path. This is useful to identify a particular call site in a function with many call sites that allocate memory. .Tl .Sh FIELDS IN THE OUTPUT Often in the tables, percentages are presented in two column fields. In such a field, a blank indicates 0%, a dot indicates less than 1%, and two stars indicate 100%. .Pp When data is broken down by size categories, the categories mean the following: .Df I .Cw x_=_extra_largexxx .Cs s = small x <= 32 bytes m = medium 32 < x <= 256 bytes l = large 256 < x <= 2048 bytes x = extra large x > 2048 bytes .Cw .De .Pp where x is the exact size of the object being allocated by a call to malloc. When data is broken into categories, percentages are always given in a two column format. Throughout this document, we refer to such a listing as a .Dq breakdown . .Sh TABLE 1: ALLOCATION BINS .Pp The memory allocation is broken down by the sizes of objects requested and freed. .Tw kept_bytes_(%) .Tl size The size in bytes of the object allocated or freed. .Tl allocs The number of calls to malloc requesting allocation of this size. .Tl bytes (%) The total number of bytes allocated to objects of this size. The percent indicates the percent of the total bytes allocated. .Tl frees The number of times objects of this size were freed. .Tl kept (%) The number of bytes of objects of this size that were never freed. The percent indicates what fraction of unfreed bytes were allocated to objects of this size. .Tl types A list of the program names of structure types or typedefs that define objects of this size. .Sh TABLE 2: MEMORY LEAKS .Pp The memory leak table lists the partial call chains which allocated memory that was never freed. At most five functions in the call chain are listed. .Tw kept_bytes_(%) .Tl kept bytes (%) The number of bytes allocated on this partial call chain and not subsequently freed. The table is sorted by decreasing values in this field. The percent indicates the percent of total bytes not freed. .Tl allocs The number of allocations that occurred on this partial call chain. .Tl bytes (%) The number of bytes allocated on this partial call chain. The percent indicates the percent of the total bytes allocated and never freed. .Tl frees The number of frees that occurred on this partial call chain. If no objects were freed this and the following field are ommitted. .Tl bytes (%) The number of bytes freed on this partial call chain. This field is omitted if no bytes were freed. .Tl path The partial call chain. Call chains starting with "..." indicate that more callers were present, but were ommitted from the listing. Call chains consist of function names (and possible call site offsets) separated by ">". Call site offsets are indicated by a +n following the function name, where n is the distance in bytes of the call site from the start of the function. Call site offsets are printed using the -offset option. .Sh TABLE 3: DIRECT ALLOCATION .Pp The row of the direct allocation listing contains a summary of all the functions where such a summary makes sense. .Tw kept_bytes_(%) .Tl % mem Percentage of the total memory allocated that was allocated by this function. .Tl bytes The total number of bytes allocated by this function. .Tl % mem(size) Size breakdown of the memory allocated by this function as a percentage of the total memory allocated by the program. For example, if the values for function MAIN are s=5, m=20, l=4, x=0, then direct calls to MALLOC from MAIN account for 5+20+4+0 = 29% of the total memory allocated by the program. Moreover, 20% of the total memory allocated by the program was of medium sized objects (between 33 and 256 bytes) by the function MAIN . The row represents the breakdown by size of all the memory allocated by the program. .Tl bytes kept The number of bytes allocated by this function that were never freed (by calls to FREE). .Tl % all kept The size breakdown of objects never freed by this function as a percentage of all objects never freed. For example, if <% all kept> values for function MAIN are s=2, m=10, l=, x=, then 10% of the total bytes not freed were allocated by MAIN and were allocated in medium-sized chunks. The row represents the size breakdown of all the memory allocated but never freed. .Tl calls The number of times this function was called to allocate an object. .Tl name The name of the function. .Sh TABLE 4: ALLOCATION CALL GRAPH A star (*) indicates that this field is omitted for ancestors or descendents in the same cycle as the function. .Pp Cycles are listed twice. The first appearance shows all the functions that are members of the cycle and the amount of memory allocated locally in each function, including the breakdown of the local allocation by size and the breakdown by size as a fraction of the total cycle. The second appearance shows what the call graph would look like if all the functions in the cycle were merged into a single function. .Tw kept_bytes_(%) .Tl index A unique index used to aid searching for functions in the call graph listing. .Tl self + desc The percent of the total allocated memory that was allocated by this function and its descendents. .Tl self (%) The number of bytes allocated by the function itself. The percentage indicates the fraction of the bytes allocated by the function and its descendents that were allocated in the function itself. .Tl size-func The size breakdown of objects allocated in the function itself (not including its descendents.) .Tl called The number of times this function was called while allocating memory. .Tl recur The number of recursive function calls while allocating memory. .Tl name The function name including possible cycle membership and index. .Sh ANCESTOR LISTINGS If the word ``all'' appears in the column, then this row represents a summary of all the ancestors and presents the total number of bytes requested by all ancestors in the column, and the breakdown of these bytes by size in the breakdown columns. If there is only one ancestor, then this summary is omitted. .Tw kept_bytes_(%) .Tl *self (%) The number of bytes allocated by the function and its descendents that were allocated on behalf of this parent. The percentage indicates what fraction of the total bytes allocated by the function and its descendents were allocated on behalf of this parent. .Tl *size-ances The size breakdown of the bytes allocated by the function and its descendents on behalf of this parent. .Tl *frac-ances The size breakdown of the objects allocated in the function and its descendents on behalf of this parent as a percentage of all objects allocated by the function and its descendents. For example if parent P1 of function F has values s=, m=, l=30, x=, then 30% of all objects allocated by F and its descendents are of large objects allocated on behalf of parent P1. .Tl called The number of times this parent called this function while requesting memory. .Tl *total The number of calls this parent made requesting memory from any function. .Tl ancestors The name of the parent including possible cycle membership and index. .Sh DESCENDENT LISTINGS If the word ``all'' appears in the column, then this row represents a summary of all the descendents and presents the total number of bytes allocated by all descendents in the column, and the breakdown of these bytes by size in the breakdown columns. If there is only one descendent, then this summary is omitted. .Tw kept_bytes_(%) .Tl *self (%) The number of bytes allocated in this descendent that were allocated at the request of the function. The percentage indicates what fraction of the total bytes allocated in descendents of the function were allocated in this descendent. .Tl *size-ances The size breakdown of the bytes allocated by this descendent on behalf of the function. .Tl *frac-desc The size breakdown of the objects allocated in this descendent on behalf of the function as a percentage of all objects allocated by all descendents on behalf of this function. For example if descendent C1 of function F has values s=35, m=, l=, x=, then 35% of all objects allocated by children of F on its behalf were allocated in child C1 and were small objects. .Tl called The number of times the function called this descendent while requesting memory. .Tl *total The number of times this descendent was called during a memory request. .Tl descendents The name of the child including possible cycle membership and index. .Sh FILES .Tw libc_mp.axx .Tl Pa a.out contains symbol table information. .Tl Pa mprof.data memory allocation call graph information. .Tl Pa libc_mp.a special version of malloc which profiles allocation. (eventually to be put in .Pa /lib/local/mprof/libc_mp.a ) .Tl .Sh SEE ALSO .Xr cc 1 , .Xr gprof 1 .Rf .Rt "A Memory Allocation Profiler for C and Lisp Programs" , Benjamin Zorn and Paul Hilfinger, Summer 1988 .Tn USENIX Conference. .Sh AUTHOR Written by Benjamin Zorn, zorn@ernie.berkeley.edu, as part of Ph.D. research sponsored by the .Tn SPUR research project. .Sh BUGS The code that determines the names and sizes of user types is poorly written and depends on the program being compiled with the -g option. In some cases (mostly very simple cases) the user type names are not correctly determined. .Pp If the user application calls .Xr valloc or .Xr memalign and later tries to free that memory, .Nm mprof will cause a segmentation fault.