• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1#! /usr/bin/perl -w
2##--------------------------------------------------------------------##
3##--- The cache simulation framework: instrumentation, recording   ---##
4##--- and results printing.                                        ---##
5##---                                           callgrind_annotate ---##
6##--------------------------------------------------------------------##
7
8#  This file is part of Callgrind, a cache-simulator and call graph
9#  tracer built on Valgrind.
10#
11#  Copyright (C) 2003 Josef Weidendorfer
12#     Josef.Weidendorfer@gmx.de
13#
14#  This file is based heavily on cg_annotate, part of Valgrind.
15#  Copyright (C) 2002 Nicholas Nethercote
16#     njn@valgrind.org
17#
18#  This program is free software; you can redistribute it and/or
19#  modify it under the terms of the GNU General Public License as
20#  published by the Free Software Foundation; either version 2 of the
21#  License, or (at your option) any later version.
22#
23#  This program is distributed in the hope that it will be useful, but
24#  WITHOUT ANY WARRANTY; without even the implied warranty of
25#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
26#  General Public License for more details.
27#
28#  You should have received a copy of the GNU General Public License
29#  along with this program; if not, write to the Free Software
30#  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
31#  02111-1307, USA.
32#
33#  The GNU General Public License is contained in the file COPYING.
34
35#----------------------------------------------------------------------------
36# Annotator for cachegrind/callgrind.
37#
38# File format is described in /docs/techdocs.html.
39#
40# Performance improvements record, using cachegrind.out for cacheprof, doing no
41# source annotation (irrelevant ones removed):
42#                                                               user time
43# 1. turned off warnings in add_hash_a_to_b()                   3.81 --> 3.48s
44#    [now add_array_a_to_b()]
45# 6. make line_to_CC() return a ref instead of a hash           3.01 --> 2.77s
46#
47#10. changed file format to avoid file/fn name repetition       2.40s
48#    (not sure why higher;  maybe due to new '.' entries?)
49#11. changed file format to drop unnecessary end-line "."s      2.36s
50#    (shrunk file by about 37%)
51#12. switched from hash CCs to array CCs                        1.61s
52#13. only adding b[i] to a[i] if b[i] defined (was doing it if
53#    either a[i] or b[i] was defined, but if b[i] was undefined
54#    it just added 0)                                           1.48s
55#14. Stopped converting "." entries to undef and then back      1.16s
56#15. Using foreach $i (x..y) instead of for ($i = 0...) in
57#    add_array_a_to_b()                                         1.11s
58#
59# Auto-annotating primes:
60#16. Finding count lengths by int((length-1)/3), not by
61#    commifying (halves the number of commify calls)            1.68s --> 1.47s
62
63use strict;
64
65#----------------------------------------------------------------------------
66# Overview: the running example in the comments is for:
67#   - events = A,B,C,D
68#   - --show=C,A,D
69#   - --sort=D,C
70#----------------------------------------------------------------------------
71
72#----------------------------------------------------------------------------
73# Global variables, main data structures
74#----------------------------------------------------------------------------
75# CCs are arrays, the counts corresponding to @events, with 'undef'
76# representing '.'.  This makes things fast (faster than using hashes for CCs)
77# but we have to use @sort_order and @show_order below to handle the --sort and
78# --show options, which is a bit tricky.
79#----------------------------------------------------------------------------
80
81# Total counts for summary (an array reference).
82my $summary_CC;
83my $totals_CC;
84
85# Totals for each function, for overall summary.
86# hash(filename:fn_name => CC array)
87my %fn_totals;
88
89# Individual CCs, organised by filename and line_num for easy annotation.
90# hash(filename => hash(line_num => CC array))
91my %all_ind_CCs;
92
93# Files chosen for annotation on the command line.
94# key = basename (trimmed of any directory), value = full filename
95my %user_ann_files;
96
97# Generic description string.
98my $desc = "";
99
100# Command line of profiled program.
101my $cmd = "";
102
103# Info on the profiled process.
104my $creator = "";
105my $pid = "";
106my $part = "";
107my $thread = "";
108
109# Positions used for cost lines; default: line numbers
110my $has_line = 1;
111my $has_addr = 0;
112
113# Events in input file, eg. (A,B,C,D)
114my @events;
115my $events;
116
117# Events to show, from command line, eg. (C,A,D)
118my @show_events;
119
120# Map from @show_events indices to @events indices, eg. (2,0,3).  Gives the
121# order in which we must traverse @events in order to show the @show_events,
122# eg. (@events[$show_order[1]], @events[$show_order[2]]...) = @show_events.
123# (Might help to think of it like a hash (0 => 2, 1 => 0, 2 => 3).)
124my @show_order;
125
126# Print out the function totals sorted by these events, eg. (D,C).
127my @sort_events;
128
129# Map from @sort_events indices to @events indices, eg. (3,2).  Same idea as
130# for @show_order.
131my @sort_order;
132
133# Thresholds, one for each sort event (or default to 1 if no sort events
134# specified).  We print out functions and do auto-annotations until we've
135# handled this proportion of all the events thresholded.
136my @thresholds;
137
138my $default_threshold = 99;
139
140my $single_threshold  = $default_threshold;
141
142# If on, automatically annotates all files that are involved in getting over
143# all the threshold counts.
144my $auto_annotate = 0;
145
146# Number of lines to show around each annotated line.
147my $context = 8;
148
149# Directories in which to look for annotation files.
150my @include_dirs = ("");
151
152# Verbose mode
153my $verbose = "1";
154
155# Inclusive statistics (with subroutine events)
156my $inclusive = 0;
157
158# Inclusive totals for each function, for overall summary.
159# hash(filename:fn_name => CC array)
160my %cfn_totals;
161
162# hash( file:func => [ called file:func ])
163my $called_funcs;
164
165# hash( file:func => [ calling file:func ])
166my $calling_funcs;
167
168# hash( file:func,line => [called file:func ])
169my $called_from_line;
170
171# hash( file:func,line => file:func
172my %func_of_line;
173
174# hash (file:func => object name)
175my %obj_name;
176
177# Print out the callers of a function
178my $tree_caller = 0;
179
180# Print out the called functions
181my $tree_calling = 0;
182
183# hash( file:func,cfile:cfunc => call CC[])
184my %call_CCs;
185
186# hash( file:func,cfile:cfunc => call counter)
187my %call_counter;
188
189# hash(context, index) => realname for compressed traces
190my %compressed;
191
192# Input file name, will be set in process_cmd_line
193my $input_file = "";
194
195# Version number
196my $version = "@VERSION@";
197
198# Usage message.
199my $usage = <<END
200usage: callgrind_annotate [options] [callgrind-out-file [source-files...]]
201
202  options for the user, with defaults in [ ], are:
203    -h --help             show this message
204    --version             show version
205    --show=A,B,C          only show figures for events A,B,C [all]
206    --sort=A,B,C          sort columns by events A,B,C [event column order]
207    --threshold=<0--100>  percentage of counts (of primary sort event) we
208                          are interested in [$default_threshold%]
209    --auto=yes|no         annotate all source files containing functions
210                          that helped reach the event count threshold [no]
211    --context=N           print N lines of context before and after
212                          annotated lines [8]
213    --inclusive=yes|no    add subroutine costs to functions calls [no]
214    --tree=none|caller|   print for each function their callers,
215           calling|both   the called functions or both [none]
216    -I --include=<dir>    add <dir> to list of directories to search for
217                          source files
218
219END
220;
221
222# Used in various places of output.
223my $fancy = '-' x 80 . "\n";
224
225#-----------------------------------------------------------------------------
226# Argument and option handling
227#-----------------------------------------------------------------------------
228sub process_cmd_line()
229{
230    for my $arg (@ARGV) {
231
232        # Option handling
233        if ($arg =~ /^-/) {
234
235            # --version
236            if ($arg =~ /^--version$/) {
237                die("callgrind_annotate-$version\n");
238
239            # --show=A,B,C
240            } elsif ($arg =~ /^--show=(.*)$/) {
241                @show_events = split(/,/, $1);
242
243            # --sort=A,B,C
244            } elsif ($arg =~ /^--sort=(.*)$/) {
245                @sort_events = split(/,/, $1);
246                my $th_specified = 0;
247                foreach my $i (0 .. scalar @sort_events - 1) {
248                    if ($sort_events[$i] =~ /.*:([\d\.]+)%?$/) {
249                        my $th = $1;
250                        ($th >= 0 && $th <= 100) or die($usage);
251                        $sort_events[$i] =~ s/:.*//;
252                        $thresholds[$i] = $th;
253                        $th_specified = 1;
254                    } else {
255                        $thresholds[$i] = 0;
256                    }
257                }
258                if (not $th_specified) {
259                    @thresholds = ();
260                }
261
262            # --threshold=X (tolerates a trailing '%')
263            } elsif ($arg =~ /^--threshold=([\d\.]+)%?$/) {
264                $single_threshold = $1;
265                ($1 >= 0 && $1 <= 100) or die($usage);
266
267            # --auto=yes|no
268            } elsif ($arg =~ /^--auto=(yes|no)$/) {
269                $auto_annotate = 1 if ($1 eq "yes");
270                $auto_annotate = 0 if ($1 eq "no");
271
272            # --context=N
273            } elsif ($arg =~ /^--context=([\d\.]+)$/) {
274                $context = $1;
275                if ($context < 0) {
276                    die($usage);
277                }
278
279            # --inclusive=yes|no
280            } elsif ($arg =~ /^--inclusive=(yes|no)$/) {
281                $inclusive = 1 if ($1 eq "yes");
282                $inclusive = 0 if ($1 eq "no");
283
284            # --tree=none|caller|calling|both
285            } elsif ($arg =~ /^--tree=(none|caller|calling|both)$/) {
286                $tree_caller  = 1 if ($1 eq "caller" || $1 eq "both");
287                $tree_calling = 1 if ($1 eq "calling" || $1 eq "both");
288
289            # --include=A,B,C
290            } elsif ($arg =~ /^(-I|--include)=(.*)$/) {
291                my $inc = $2;
292                $inc =~ s|/$||;         # trim trailing '/'
293                push(@include_dirs, "$inc/");
294
295            } else {            # -h and --help fall under this case
296                die($usage);
297            }
298
299        # Argument handling -- annotation file checking and selection.
300        # Stick filenames into a hash for quick 'n easy lookup throughout
301        } else {
302	  if ($input_file eq "") {
303	    $input_file = $arg;
304	  }
305	  else {
306            my $readable = 0;
307            foreach my $include_dir (@include_dirs) {
308                if (-r $include_dir . $arg) {
309                    $readable = 1;
310                }
311            }
312            $readable or die("File $arg not found in any of: @include_dirs\n");
313            $user_ann_files{$arg} = 1;
314        }
315    }
316    }
317
318    if ($input_file eq "") {
319      $input_file = (<callgrind.out*>)[0];
320      if (!defined $input_file) {
321	  $input_file = (<cachegrind.out*>)[0];
322      }
323
324      (defined $input_file) or die($usage);
325      print "Reading data from '$input_file'...\n";
326    }
327}
328
329#-----------------------------------------------------------------------------
330# Reading of input file
331#-----------------------------------------------------------------------------
332sub max ($$)
333{
334    my ($x, $y) = @_;
335    return ($x > $y ? $x : $y);
336}
337
338# Add the two arrays;  any '.' entries are ignored.  Two tricky things:
339# 1. If $a2->[$i] is undefined, it defaults to 0 which is what we want; we turn
340#    off warnings to allow this.  This makes things about 10% faster than
341#    checking for definedness ourselves.
342# 2. We don't add an undefined count or a ".", even though it's value is 0,
343#    because we don't want to make an $a2->[$i] that is undef become 0
344#    unnecessarily.
345sub add_array_a_to_b ($$)
346{
347    my ($a1, $a2) = @_;
348
349    my $n = max(scalar @$a1, scalar @$a2);
350    $^W = 0;
351    foreach my $i (0 .. $n-1) {
352        $a2->[$i] += $a1->[$i] if (defined $a1->[$i] && "." ne $a1->[$i]);
353    }
354    $^W = 1;
355}
356
357# Is this a line with all events zero?
358sub is_zero ($)
359{
360    my ($CC) = @_;
361    my $isZero = 1;
362    foreach my $i (0 .. (scalar @$CC)-1) {
363	$isZero = 0 if ($CC->[$i] >0);
364    }
365    return $isZero;
366}
367
368# Add each event count to the CC array.  '.' counts become undef, as do
369# missing entries (implicitly).
370sub line_to_CC ($)
371{
372    my @CC = (split /\s+/, $_[0]);
373    (@CC <= @events) or die("Line $.: too many event counts\n");
374    return \@CC;
375}
376
377sub uncompressed_name($$)
378{
379   my ($context, $name) = @_;
380
381   if ($name =~ /^\((\d+)\)\s*(.*)$/) {
382     my $index = $1;
383     my $realname = $2;
384
385     if ($realname eq "") {
386       $realname = $compressed{$context,$index};
387     }
388     else {
389       $compressed{$context,$index} = $realname;
390     }
391     return $realname;
392   }
393   return $name;
394}
395
396sub read_input_file()
397{
398    open(INPUTFILE, "< $input_file") || die "File $input_file not opened\n";
399
400    my $line;
401
402    # Read header
403    while(<INPUTFILE>) {
404
405      # remove comments
406      s/#.*$//;
407
408      if (/^$/) { ; }
409
410      elsif (/^version:\s*(\d+)/) {
411	# Can't read format with major version > 1
412	($1<2) or die("Can't read format with major version $1.\n");
413      }
414
415      elsif (/^pid:\s+(.*)$/) { $pid = $1;  }
416      elsif (/^thread:\s+(.*)$/) { $thread = $1;  }
417      elsif (/^part:\s+(.*)$/) { $part = $1;  }
418      elsif (/^desc:\s+(.*)$/) {
419	my $dline = $1;
420	# suppress profile options in description output
421	if ($dline =~ /^Option:/) {;}
422	else { $desc .= "$dline\n"; }
423      }
424      elsif (/^cmd:\s+(.*)$/)  { $cmd = $1; }
425      elsif (/^creator:\s+(.*)$/)  { $creator = $1; }
426      elsif (/^positions:\s+(.*)$/) {
427	my $positions = $1;
428	$has_line = ($positions =~ /line/);
429	$has_addr = ($positions =~ /(addr|instr)/);
430      }
431      elsif (/^events:\s+(.*)$/) {
432	$events = $1;
433
434	# events line is last in header
435	last;
436      }
437      else {
438	warn("WARNING: header line $. malformed, ignoring\n");
439	if ($verbose) { chomp; warn("    line: '$_'\n"); }
440      }
441    }
442
443    # Check for needed header entries
444    ($cmd ne "") or die("Line $.: missing command line\n");
445
446    # Read "events:" line.  We make a temporary hash in which the Nth event's
447    # value is N, which is useful for handling --show/--sort options below.
448    ($events ne "") or die("Line $.: missing events line\n");
449    @events = split(/\s+/, $events);
450    my %events;
451    my $n = 0;
452    foreach my $event (@events) {
453        $events{$event} = $n;
454        $n++
455    }
456
457    # If no --show arg give, default to showing all events in the file.
458    # If --show option is used, check all specified events appeared in the
459    # "events:" line.  Then initialise @show_order.
460    if (@show_events) {
461        foreach my $show_event (@show_events) {
462            (defined $events{$show_event}) or
463                die("--show event `$show_event' did not appear in input\n");
464        }
465    } else {
466        @show_events = @events;
467    }
468    foreach my $show_event (@show_events) {
469        push(@show_order, $events{$show_event});
470    }
471
472    # Do as for --show, but if no --sort arg given, default to sorting by
473    # column order (ie. first column event is primary sort key, 2nd column is
474    # 2ndary key, etc).
475    if (@sort_events) {
476        foreach my $sort_event (@sort_events) {
477            (defined $events{$sort_event}) or
478                die("--sort event `$sort_event' did not appear in input\n");
479        }
480    } else {
481        @sort_events = @events;
482    }
483    foreach my $sort_event (@sort_events) {
484        push(@sort_order, $events{$sort_event});
485    }
486
487    # If multiple threshold args weren't given via --sort, stick in the single
488    # threshold (either from --threshold if used, or the default otherwise) for
489    # the primary sort event, and 0% for the rest.
490    if (not @thresholds) {
491        foreach my $e (@sort_order) {
492            push(@thresholds, 0);
493        }
494        $thresholds[0] = $single_threshold;
495    }
496
497    # Current directory, used to strip from file names if absolute
498    my $pwd = `pwd`;
499    chomp $pwd;
500    $pwd .= '/';
501
502    my $curr_obj = "";
503    my $curr_file;
504    my $curr_fn;
505    my $curr_name;
506    my $curr_line_num = 0;
507    my $prev_line_num = 0;
508
509    my $curr_cobj = "";
510    my $curr_cfile = "";
511    my $curr_cfunc = "";
512    my $curr_cname;
513    my $curr_call_counter = 0;
514    my $curr_cfn_CC = [];
515
516    my $curr_fn_CC = [];
517    my $curr_file_ind_CCs = {};     # hash(line_num => CC)
518
519    # Read body of input file.
520    while (<INPUTFILE>) {
521	$prev_line_num = $curr_line_num;
522
523        s/#.*$//;   # remove comments
524        s/^\+(\d+)/$prev_line_num+$1/e;
525        s/^\-(\d+)/$prev_line_num-$1/e;
526        s/^\*/$prev_line_num/e;
527        if (s/^(-?\d+|0x\w+)\s+//) {
528            $curr_line_num = $1;
529	    if ($has_addr) {
530	      if ($has_line) {
531                s/^\+(\d+)/$prev_line_num+$1/e;
532	        s/^\-(\d+)/$prev_line_num-$1/e;
533                s/^\*/$prev_line_num/e;
534
535	        if (s/^(\d+)\s+//) { $curr_line_num = $1; }
536	      }
537	      else { $curr_line_num = 0; }
538	    }
539            my $CC = line_to_CC($_);
540
541	    if ($curr_call_counter>0) {
542#	      print "Read ($curr_name => $curr_cname) $curr_call_counter\n";
543
544	      if (!defined $call_CCs{$curr_name,$curr_cname}) {
545		$call_CCs{$curr_name,$curr_cname} = [];
546		$call_counter{$curr_name,$curr_cname} = 0;
547	      }
548	      add_array_a_to_b($CC, $call_CCs{$curr_name,$curr_cname});
549	      $call_counter{$curr_name,$curr_cname} += $curr_call_counter;
550
551	      my $tmp = $called_from_line->{$curr_file,$curr_line_num};
552	      if (!defined $tmp) {
553		$func_of_line{$curr_file,$curr_line_num} = $curr_name;
554	      }
555	      $tmp = {} unless defined $tmp;
556	      $$tmp{$curr_cname} = 1;
557	      $called_from_line->{$curr_file,$curr_line_num} = $tmp;
558	      if (!defined $call_CCs{$curr_name,$curr_cname,$curr_line_num}) {
559		$call_CCs{$curr_name,$curr_cname,$curr_line_num} = [];
560		$call_counter{$curr_name,$curr_cname,$curr_line_num} = 0;
561	      }
562	      add_array_a_to_b($CC, $call_CCs{$curr_name,$curr_cname,$curr_line_num});
563	      $call_counter{$curr_name,$curr_cname,$curr_line_num} += $curr_call_counter;
564
565	      $curr_call_counter = 0;
566
567	      # inclusive costs
568	      $curr_cfn_CC = $cfn_totals{$curr_cname};
569	      $curr_cfn_CC = [] unless (defined $curr_cfn_CC);
570	      add_array_a_to_b($CC, $curr_cfn_CC);
571	      $cfn_totals{$curr_cname} = $curr_cfn_CC;
572
573	      if ($inclusive) {
574		add_array_a_to_b($CC, $curr_fn_CC);
575	      }
576	      next;
577	    }
578
579            add_array_a_to_b($CC, $curr_fn_CC);
580
581            # If curr_file is selected, add CC to curr_file list.  We look for
582            # full filename matches;  or, if auto-annotating, we have to
583            # remember everything -- we won't know until the end what's needed.
584            if ($auto_annotate || defined $user_ann_files{$curr_file}) {
585                my $tmp = $curr_file_ind_CCs->{$curr_line_num};
586                $tmp = [] unless defined $tmp;
587                add_array_a_to_b($CC, $tmp);
588                $curr_file_ind_CCs->{$curr_line_num} = $tmp;
589            }
590
591        } elsif (s/^fn=(.*)$//) {
592            # Commit result from previous function
593            $fn_totals{$curr_name} = $curr_fn_CC if (defined $curr_name);
594
595            # Setup new one
596            $curr_fn = uncompressed_name("fn",$1);
597            $curr_name = "$curr_file:$curr_fn";
598	    $obj_name{$curr_name} = $curr_obj;
599            $curr_fn_CC = $fn_totals{$curr_name};
600            $curr_fn_CC = [] unless (defined $curr_fn_CC);
601
602        } elsif (s/^ob=(.*)$//) {
603            $curr_obj = uncompressed_name("ob",$1);
604
605        } elsif (s/^fl=(.*)$//) {
606            $all_ind_CCs{$curr_file} = $curr_file_ind_CCs
607                if (defined $curr_file);
608
609            $curr_file = uncompressed_name("fl",$1);
610            $curr_file =~ s/^\Q$pwd\E//;
611            $curr_file_ind_CCs = $all_ind_CCs{$curr_file};
612            $curr_file_ind_CCs = {} unless (defined $curr_file_ind_CCs);
613
614        } elsif (s/^(fi|fe)=(.*)$//) {
615            (defined $curr_name) or die("Line $.: Unexpected fi/fe line\n");
616            $fn_totals{$curr_name} = $curr_fn_CC;
617            $all_ind_CCs{$curr_file} = $curr_file_ind_CCs;
618
619            $curr_file = uncompressed_name("fl",$2);
620            $curr_file =~ s/^\Q$pwd\E//;
621            $curr_name = "$curr_file:$curr_fn";
622            $curr_file_ind_CCs = $all_ind_CCs{$curr_file};
623            $curr_file_ind_CCs = {} unless (defined $curr_file_ind_CCs);
624            $curr_fn_CC = $fn_totals{$curr_name};
625            $curr_fn_CC = [] unless (defined $curr_fn_CC);
626
627        } elsif (s/^\s*$//) {
628            # blank, do nothing
629
630        } elsif (s/^cob=(.*)$//) {
631	  $curr_cobj = uncompressed_name("ob",$1);
632
633	} elsif (s/^cfi=(.*)$//) {
634	  $curr_cfile = uncompressed_name("fl",$1);
635
636	} elsif (s/^cfn=(.*)$//) {
637	  $curr_cfunc = uncompressed_name("fn",$1);
638	  if ($curr_cfile eq "") {
639	    $curr_cname = "$curr_file:$curr_cfunc";
640	  }
641	  else {
642	    $curr_cname = "$curr_cfile:$curr_cfunc";
643	    $curr_cfile = "";
644	  }
645
646	  my $tmp = $calling_funcs->{$curr_cname};
647	  $tmp = {} unless defined $tmp;
648	  $$tmp{$curr_name} = 1;
649	  $calling_funcs->{$curr_cname} = $tmp;
650
651	  my $tmp2 = $called_funcs->{$curr_name};
652	  $tmp2 = {} unless defined $tmp2;
653	  $$tmp2{$curr_cname} = 1;
654	  $called_funcs->{$curr_name} = $tmp2;
655
656	} elsif (s/^calls=(\d+)//) {
657	  $curr_call_counter = $1;
658
659        } elsif (s/^(jump|jcnd)=//) {
660	  #ignore jump information
661
662        } elsif (s/^jfi=(.*)$//) {
663          # side effect needed: possibly add compression mapping
664          uncompressed_name("fl",$1);
665          # ignore jump information
666
667        } elsif (s/^jfn=(.*)$//) {
668          # side effect needed: possibly add compression mapping
669          uncompressed_name("fn",$1);
670          # ignore jump information
671
672        } elsif (s/^totals:\s+//) {
673	    $totals_CC = line_to_CC($_);
674
675        } elsif (s/^summary:\s+//) {
676            $summary_CC = line_to_CC($_);
677
678        } else {
679            warn("WARNING: line $. malformed, ignoring\n");
680	    if ($verbose) { chomp; warn("    line: '$_'\n"); }
681        }
682    }
683
684    if ((not defined $summary_CC) || is_zero($summary_CC)) {
685	$summary_CC = $totals_CC;
686    }
687
688    # Check if summary line was present
689    if (not defined $summary_CC) {
690        warn("WARNING: missing final summary line, no summary will be printed\n");
691    }
692
693    # Finish up handling final filename/fn_name counts
694    $fn_totals{"$curr_file:$curr_fn"} = $curr_fn_CC
695	if (defined $curr_file && defined $curr_fn);
696    $all_ind_CCs{$curr_file} =
697	$curr_file_ind_CCs if (defined $curr_file);
698
699    # Correct inclusive totals
700    if ($inclusive) {
701      foreach my $name (keys %cfn_totals) {
702	$fn_totals{$name} = $cfn_totals{$name};
703      }
704    }
705
706    close(INPUTFILE);
707}
708
709#-----------------------------------------------------------------------------
710# Print options used
711#-----------------------------------------------------------------------------
712sub print_options ()
713{
714    print($fancy);
715    print "Profile data file '$input_file'";
716    if ($creator ne "") { print " (creator: $creator)"; }
717    print "\n";
718
719    print($fancy);
720    print($desc);
721    my $target = $cmd;
722    if ($pid ne "") {
723      $target .= " (PID $pid";
724      if ($part ne "") { $target .= ", part $part"; }
725      if ($thread ne "") { $target .= ", thread $thread"; }
726      $target .= ")";
727    }
728    print("Profiled target:  $target\n");
729    print("Events recorded:  @events\n");
730    print("Events shown:     @show_events\n");
731    print("Event sort order: @sort_events\n");
732    print("Thresholds:       @thresholds\n");
733
734    my @include_dirs2 = @include_dirs;  # copy @include_dirs
735    shift(@include_dirs2);       # remove "" entry, which is always the first
736    unshift(@include_dirs2, "") if (0 == @include_dirs2);
737    my $include_dir = shift(@include_dirs2);
738    print("Include dirs:     $include_dir\n");
739    foreach my $include_dir (@include_dirs2) {
740        print("                  $include_dir\n");
741    }
742
743    my @user_ann_files = keys %user_ann_files;
744    unshift(@user_ann_files, "") if (0 == @user_ann_files);
745    my $user_ann_file = shift(@user_ann_files);
746    print("User annotated:   $user_ann_file\n");
747    foreach $user_ann_file (@user_ann_files) {
748        print("                  $user_ann_file\n");
749    }
750
751    my $is_on = ($auto_annotate ? "on" : "off");
752    print("Auto-annotation:  $is_on\n");
753    print("\n");
754}
755
756#-----------------------------------------------------------------------------
757# Print summary and sorted function totals
758#-----------------------------------------------------------------------------
759sub mycmp ($$)
760{
761    my ($c, $d) = @_;
762
763    # Iterate through sort events (eg. 3,2); return result if two are different
764    foreach my $i (@sort_order) {
765        my ($x, $y);
766        $x = $c->[$i];
767        $y = $d->[$i];
768        $x = -1 unless defined $x;
769        $y = -1 unless defined $y;
770
771        my $cmp = $y <=> $x;        # reverse sort
772        if (0 != $cmp) {
773            return $cmp;
774        }
775    }
776    # Exhausted events, equal
777    return 0;
778}
779
780sub commify ($) {
781    my ($val) = @_;
782    1 while ($val =~ s/^(\d+)(\d{3})/$1,$2/);
783    return $val;
784}
785
786# Because the counts can get very big, and we don't want to waste screen space
787# and make lines too long, we compute exactly how wide each column needs to be
788# by finding the widest entry for each one.
789sub compute_CC_col_widths (@)
790{
791    my @CCs = @_;
792    my $CC_col_widths = [];
793
794    # Initialise with minimum widths (from event names)
795    foreach my $event (@events) {
796        push(@$CC_col_widths, length($event));
797    }
798
799    # Find maximum width count for each column.  @CC_col_width positions
800    # correspond to @CC positions.
801    foreach my $CC (@CCs) {
802        foreach my $i (0 .. scalar(@$CC)-1) {
803            if (defined $CC->[$i]) {
804                # Find length, accounting for commas that will be added
805                my $length = length $CC->[$i];
806                my $clength = $length + int(($length - 1) / 3);
807                $CC_col_widths->[$i] = max($CC_col_widths->[$i], $clength);
808            }
809        }
810    }
811    return $CC_col_widths;
812}
813
814# Print the CC with each column's size dictated by $CC_col_widths.
815sub print_CC ($$)
816{
817    my ($CC, $CC_col_widths) = @_;
818
819    foreach my $i (@show_order) {
820        my $count = (defined $CC->[$i] ? commify($CC->[$i]) : ".");
821        my $space = ' ' x ($CC_col_widths->[$i] - length($count));
822        print("$space$count ");
823    }
824}
825
826sub print_events ($)
827{
828    my ($CC_col_widths) = @_;
829
830    foreach my $i (@show_order) {
831        my $event       = $events[$i];
832        my $event_width = length($event);
833        my $col_width   = $CC_col_widths->[$i];
834        my $space       = ' ' x ($col_width - $event_width);
835        print("$space$event ");
836    }
837}
838
839# Prints summary and function totals (with separate column widths, so that
840# function names aren't pushed over unnecessarily by huge summary figures).
841# Also returns a hash containing all the files that are involved in getting the
842# events count above the thresholds (ie. all the interesting ones).
843sub print_summary_and_fn_totals ()
844{
845    my @fn_fullnames = keys   %fn_totals;
846
847    # Work out the size of each column for printing (summary and functions
848    # separately).
849    my $summary_CC_col_widths = compute_CC_col_widths($summary_CC);
850    my      $fn_CC_col_widths = compute_CC_col_widths(values %fn_totals);
851
852    # Header and counts for summary
853    print($fancy);
854    print_events($summary_CC_col_widths);
855    print("\n");
856    print($fancy);
857    print_CC($summary_CC, $summary_CC_col_widths);
858    print(" PROGRAM TOTALS\n");
859    print("\n");
860
861    # Header for functions
862    print($fancy);
863    print_events($fn_CC_col_widths);
864    print(" file:function\n");
865    print($fancy);
866
867    # Sort function names into order dictated by --sort option.
868    @fn_fullnames = sort {
869        mycmp($fn_totals{$a}, $fn_totals{$b})
870    } @fn_fullnames;
871
872
873    # Assertion
874    (scalar @sort_order == scalar @thresholds) or
875        die("sort_order length != thresholds length:\n",
876            "  @sort_order\n  @thresholds\n");
877
878    my $threshold_files       = {};
879    # @curr_totals has the same shape as @sort_order and @thresholds
880    my @curr_totals = ();
881    foreach my $e (@thresholds) {
882        push(@curr_totals, 0);
883    }
884
885    # Print functions, stopping when the threshold has been reached.
886    foreach my $fn_name (@fn_fullnames) {
887
888        # Stop when we've reached all the thresholds
889        my $reached_all_thresholds = 1;
890        foreach my $i (0 .. scalar @thresholds - 1) {
891            my $prop = $curr_totals[$i] * 100;
892	    if ($summary_CC->[$sort_order[$i]] >0) {
893	      $prop = $prop / $summary_CC->[$sort_order[$i]];
894	    }
895            $reached_all_thresholds &&= ($prop >= $thresholds[$i]);
896        }
897        last if $reached_all_thresholds;
898
899	if ($tree_caller || $tree_calling) { print "\n"; }
900
901	if ($tree_caller && ($fn_name ne "???:???")) {
902	  # Print function callers
903	  my $tmp1 = $calling_funcs->{$fn_name};
904	  if (defined $tmp1) {
905	    foreach my $calling (keys %$tmp1) {
906	      if (defined $call_counter{$calling,$fn_name}) {
907		print_CC($call_CCs{$calling,$fn_name}, $fn_CC_col_widths);
908		print" < $calling (";
909		print $call_counter{$calling,$fn_name} . "x)";
910		if (defined $obj_name{$calling}) {
911		  print " [$obj_name{$calling}]";
912		}
913		print "\n";
914	      }
915	    }
916	  }
917	}
918
919        # Print function results
920        my $fn_CC = $fn_totals{$fn_name};
921        print_CC($fn_CC, $fn_CC_col_widths);
922	if ($tree_caller || $tree_calling) { print " * "; }
923        print(" $fn_name");
924	if (defined $obj_name{$fn_name}) {
925	  print " [$obj_name{$fn_name}]";
926	}
927	print "\n";
928
929	if ($tree_calling && ($fn_name ne "???:???")) {
930	  # Print called functions
931	  my $tmp2 = $called_funcs->{$fn_name};
932	  if (defined $tmp2) {
933	    foreach my $called (keys %$tmp2) {
934	      if (defined $call_counter{$fn_name,$called}) {
935		print_CC($call_CCs{$fn_name,$called}, $fn_CC_col_widths);
936		print" >   $called (";
937		print $call_counter{$fn_name,$called} . "x)";
938		if (defined $obj_name{$called}) {
939		  print " [$obj_name{$called}]";
940		}
941		print "\n";
942	      }
943	    }
944	  }
945	}
946
947        # Update the threshold counts
948        my $filename = $fn_name;
949        $filename =~ s/:.+$//;    # remove function name
950        $threshold_files->{$filename} = 1;
951        foreach my $i (0 .. scalar @sort_order - 1) {
952	  if ($inclusive) {
953	    $curr_totals[$i] = $summary_CC->[$sort_order[$i]] -
954                               $fn_CC->[$sort_order[$i]]
955	      if (defined $fn_CC->[$sort_order[$i]]);
956	  } else {
957            $curr_totals[$i] += $fn_CC->[$sort_order[$i]]
958                if (defined $fn_CC->[$sort_order[$i]]);
959        }
960    }
961    }
962    print("\n");
963
964    return $threshold_files;
965}
966
967#-----------------------------------------------------------------------------
968# Annotate selected files
969#-----------------------------------------------------------------------------
970
971# Issue a warning that the source file is more recent than the input file.
972sub warning_on_src_more_recent_than_inputfile ($)
973{
974    my $src_file = $_[0];
975
976    my $warning = <<END
977@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
978@@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@
979@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
980@ Source file '$src_file' is more recent than input file '$input_file'.
981@ Annotations may not be correct.
982@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
983
984END
985;
986    print($warning);
987}
988
989# If there is information about lines not in the file, issue a warning
990# explaining possible causes.
991sub warning_on_nonexistent_lines ($$$)
992{
993    my ($src_more_recent_than_inputfile, $src_file, $excess_line_nums) = @_;
994    my $cause_and_solution;
995
996    if ($src_more_recent_than_inputfile) {
997        $cause_and_solution = <<END
998@@ cause:    '$src_file' has changed since information was gathered.
999@@           If so, a warning will have already been issued about this.
1000@@ solution: Recompile program and rerun under "valgrind --cachesim=yes" to
1001@@           gather new information.
1002END
1003    # We suppress warnings about .h files
1004    } elsif ($src_file =~ /\.h$/) {
1005        $cause_and_solution = <<END
1006@@ cause:    bug in the Valgrind's debug info reader that screws up with .h
1007@@           files sometimes
1008@@ solution: none, sorry
1009END
1010    } else {
1011        $cause_and_solution = <<END
1012@@ cause:    not sure, sorry
1013END
1014    }
1015
1016    my $warning = <<END
1017@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
1018@@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@ WARNING @@
1019@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
1020@@
1021@@ Information recorded about lines past the end of '$src_file'.
1022@@
1023@@ Probable cause and solution:
1024$cause_and_solution@@
1025@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
1026END
1027;
1028    print($warning);
1029}
1030
1031sub annotate_ann_files($)
1032{
1033    my ($threshold_files) = @_;
1034
1035    my %all_ann_files;
1036    my @unfound_auto_annotate_files;
1037    my $printed_totals_CC = [];
1038
1039    # If auto-annotating, add interesting files (but not "???")
1040    if ($auto_annotate) {
1041        delete $threshold_files->{"???"};
1042        %all_ann_files = (%user_ann_files, %$threshold_files)
1043    } else {
1044        %all_ann_files = %user_ann_files;
1045    }
1046
1047    # Track if we did any annotations.
1048    my $did_annotations = 0;
1049
1050    LOOP:
1051    foreach my $src_file (keys %all_ann_files) {
1052
1053        my $opened_file = "";
1054        my $full_file_name = "";
1055        foreach my $include_dir (@include_dirs) {
1056            my $try_name = $include_dir . $src_file;
1057            if (open(INPUTFILE, "< $try_name")) {
1058                $opened_file    = $try_name;
1059                $full_file_name = ($include_dir eq ""
1060                                  ? $src_file
1061                                  : "$include_dir + $src_file");
1062                last;
1063            }
1064        }
1065
1066        if (not $opened_file) {
1067            # Failed to open the file.  If chosen on the command line, die.
1068            # If arose from auto-annotation, print a little message.
1069            if (defined $user_ann_files{$src_file}) {
1070                die("File $src_file not opened in any of: @include_dirs\n");
1071
1072            } else {
1073                push(@unfound_auto_annotate_files, $src_file);
1074            }
1075
1076        } else {
1077            # File header (distinguish between user- and auto-selected files).
1078            print("$fancy");
1079            my $ann_type =
1080                (defined $user_ann_files{$src_file} ? "User" : "Auto");
1081            print("-- $ann_type-annotated source: $full_file_name\n");
1082            print("$fancy");
1083
1084            # Get file's CCs
1085            my $src_file_CCs = $all_ind_CCs{$src_file};
1086            if (!defined $src_file_CCs) {
1087                print("  No information has been collected for $src_file\n\n");
1088                next LOOP;
1089            }
1090
1091            $did_annotations = 1;
1092
1093            # Numeric, not lexicographic sort!
1094            my @line_nums = sort {$a <=> $b} keys %$src_file_CCs;
1095
1096            # If $src_file more recent than cachegrind.out, issue warning
1097            my $src_more_recent_than_inputfile = 0;
1098            if ((stat $opened_file)[9] > (stat $input_file)[9]) {
1099                $src_more_recent_than_inputfile = 1;
1100                warning_on_src_more_recent_than_inputfile($src_file);
1101            }
1102
1103            # Work out the size of each column for printing
1104            my $CC_col_widths = compute_CC_col_widths(values %$src_file_CCs);
1105
1106            # Events header
1107            print_events($CC_col_widths);
1108            print("\n\n");
1109
1110            # Shift out 0 if it's in the line numbers (from unknown entries,
1111            # likely due to bugs in Valgrind's stabs debug info reader)
1112            shift(@line_nums) if (0 == $line_nums[0]);
1113
1114            # Finds interesting line ranges -- all lines with a CC, and all
1115            # lines within $context lines of a line with a CC.
1116            my $n = @line_nums;
1117            my @pairs;
1118            for (my $i = 0; $i < $n; $i++) {
1119                push(@pairs, $line_nums[$i] - $context);   # lower marker
1120                while ($i < $n-1 &&
1121                       $line_nums[$i] + 2*$context >= $line_nums[$i+1]) {
1122                    $i++;
1123                }
1124                push(@pairs, $line_nums[$i] + $context);   # upper marker
1125            }
1126
1127            # Annotate chosen lines, tracking total counts of lines printed
1128            $pairs[0] = 1 if ($pairs[0] < 1);
1129            while (@pairs) {
1130                my $low  = shift @pairs;
1131                my $high = shift @pairs;
1132                while ($. < $low-1) {
1133                    my $tmp = <INPUTFILE>;
1134                    last unless (defined $tmp);     # hack to detect EOF
1135                }
1136                my $src_line;
1137                # Print line number, unless start of file
1138                print("-- line $low " . '-' x 40 . "\n") if ($low != 1);
1139                while (($. < $high) && ($src_line = <INPUTFILE>)) {
1140                    if (defined $line_nums[0] && $. == $line_nums[0]) {
1141                        print_CC($src_file_CCs->{$.}, $CC_col_widths);
1142                        add_array_a_to_b($src_file_CCs->{$.},
1143                                         $printed_totals_CC);
1144                        shift(@line_nums);
1145
1146                    } else {
1147                        print_CC( [], $CC_col_widths);
1148                    }
1149
1150                    print(" $src_line");
1151
1152		    my $tmp  = $called_from_line->{$src_file,$.};
1153		    my $func = $func_of_line{$src_file,$.};
1154		    if (defined $tmp) {
1155		      foreach my $called (keys %$tmp) {
1156			if (defined $call_CCs{$func,$called,$.}) {
1157			  print_CC($call_CCs{$func,$called,$.}, $CC_col_widths);
1158			  print " => $called (";
1159			  print $call_counter{$func,$called,$.} . "x)\n";
1160			}
1161		      }
1162		    }
1163                }
1164                # Print line number, unless EOF
1165                if ($src_line) {
1166                    print("-- line $high " . '-' x 40 . "\n");
1167                } else {
1168                    last;
1169                }
1170            }
1171
1172            # If there was info on lines past the end of the file...
1173            if (@line_nums) {
1174                foreach my $line_num (@line_nums) {
1175                    print_CC($src_file_CCs->{$line_num}, $CC_col_widths);
1176                    print(" <bogus line $line_num>\n");
1177                }
1178                print("\n");
1179                warning_on_nonexistent_lines($src_more_recent_than_inputfile,
1180                                             $src_file, \@line_nums);
1181            }
1182            print("\n");
1183
1184            # Print summary of counts attributed to file but not to any
1185            # particular line (due to incomplete debug info).
1186            if ($src_file_CCs->{0}) {
1187                print_CC($src_file_CCs->{0}, $CC_col_widths);
1188                print(" <counts for unidentified lines in $src_file>\n\n");
1189            }
1190
1191            close(INPUTFILE);
1192        }
1193    }
1194
1195    # Print list of unfound auto-annotate selected files.
1196    if (@unfound_auto_annotate_files) {
1197        print("$fancy");
1198        print("The following files chosen for auto-annotation could not be found:\n");
1199        print($fancy);
1200        foreach my $f (@unfound_auto_annotate_files) {
1201            print("  $f\n");
1202        }
1203        print("\n");
1204    }
1205
1206    # If we did any annotating, print what proportion of events were covered by
1207    # annotated lines above.
1208    if ($did_annotations) {
1209        my $percent_printed_CC;
1210        foreach (my $i = 0; $i < @$summary_CC; $i++) {
1211            $percent_printed_CC->[$i] =
1212                sprintf("%.0f",
1213                        $printed_totals_CC->[$i] / $summary_CC->[$i] * 100);
1214        }
1215        my $pp_CC_col_widths = compute_CC_col_widths($percent_printed_CC);
1216        print($fancy);
1217        print_events($pp_CC_col_widths);
1218        print("\n");
1219        print($fancy);
1220        print_CC($percent_printed_CC, $pp_CC_col_widths);
1221        print(" percentage of events annotated\n\n");
1222    }
1223}
1224
1225#----------------------------------------------------------------------------
1226# "main()"
1227#----------------------------------------------------------------------------
1228process_cmd_line();
1229read_input_file();
1230print_options();
1231my $threshold_files = print_summary_and_fn_totals();
1232annotate_ann_files($threshold_files);
1233
1234##--------------------------------------------------------------------##
1235##--- end                                           vg_annotate.in ---##
1236##--------------------------------------------------------------------##
1237
1238
1239