binutils-gdb/gprof/gprof.c
Sean Eric Fagan 2ea5f3252a Use BFD to determine what a symbol prefix is, be better about determining
which symbols to study, and divide by the correct amount.
1993-08-13 21:38:41 +00:00

762 lines
18 KiB
C

/*
* Copyright (c) 1983 Regents of the University of California.
* All rights reserved.
*
* Redistribution and use in source and binary forms are permitted
* provided that: (1) source distributions retain this entire copyright
* notice and comment, and (2) distributions including binaries display
* the following acknowledgement: ``This product includes software
* developed by the University of California, Berkeley and its contributors''
* in the documentation or other materials provided with the distribution
* and in all advertising materials mentioning features or use of this
* software. Neither the name of the University nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
*/
#ifndef lint
char copyright[] =
"@(#) Copyright (c) 1983 Regents of the University of California.\n\
All rights reserved.\n";
#endif /* not lint */
#ifndef lint
static char sccsid[] = "@(#)gprof.c 5.6 (Berkeley) 6/1/90";
#endif /* not lint */
#include "gprof.h"
bfd *abfd;
char *whoami;
/*
* things which get -E excluded by default.
*/
char *defaultEs[] = { "mcount" , "__mcleanup" , 0 };
int discard_underscores = 1; /* Should we discard initial underscores? */
int bsd_style_output = 0; /* As opposed to FSF style output */
main(argc, argv)
int argc;
char **argv;
{
char **sp;
nltype **timesortnlp;
whoami = argv[0];
--argc;
argv++;
debug = 0;
bflag = TRUE;
while ( *argv != 0 && **argv == '-' ) {
(*argv)++;
switch ( **argv ) {
case 'a':
aflag = TRUE;
break;
case 'b':
bflag = FALSE;
break;
case 'c':
cflag = TRUE;
break;
case 'd':
dflag = TRUE;
(*argv)++;
debug |= atoi( *argv );
debug |= ANYDEBUG;
# ifdef DEBUG
printf("[main] debug = %d\n", debug);
# else not DEBUG
printf("%s: -d ignored\n", whoami);
# endif DEBUG
break;
case 'E':
++argv;
addlist( Elist , *argv );
Eflag = TRUE;
addlist( elist , *argv );
eflag = TRUE;
break;
case 'e':
addlist( elist , *++argv );
eflag = TRUE;
break;
case 'F':
++argv;
addlist( Flist , *argv );
Fflag = TRUE;
addlist( flist , *argv );
fflag = TRUE;
break;
case 'f':
addlist( flist , *++argv );
fflag = TRUE;
break;
case 'k':
addlist( kfromlist , *++argv );
addlist( ktolist , *++argv );
kflag = TRUE;
break;
case 's':
sflag = TRUE;
break;
case 'T': /* "Traditional" output format */
bsd_style_output = 1;
break;
case 'z':
zflag = TRUE;
break;
default:
fprintf (stderr, "\
Usage: %s [-a] [-b] [-c] [-d[num]] [-E function-name] [-e function-name]\n\
[-F function-name] [-f function-name] [-k from to] [-s] [-T] [-z]\n\
[image-file] [profile-file...]\n", whoami);
exit (1);
}
argv++;
}
if ( *argv != 0 ) {
a_outname = *argv;
argv++;
} else {
a_outname = A_OUTNAME;
}
if ( *argv != 0 ) {
gmonname = *argv;
argv++;
} else {
gmonname = GMONNAME;
}
/*
* turn off default functions
*/
for ( sp = &defaultEs[0] ; *sp ; sp++ ) {
Eflag = TRUE;
addlist( Elist , *sp );
eflag = TRUE;
addlist( elist , *sp );
}
/*
* how many ticks per second?
* if we can't tell, report time in ticks.
*/
hz = hertz();
if (hz == 0) {
hz = 1;
fprintf(stderr, "time is in ticks, not seconds\n");
}
/*
* get information about a.out file.
*/
getnfile();
/*
* get information about mon.out file(s).
*/
do {
getpfile( gmonname );
if ( *argv != 0 ) {
gmonname = *argv;
}
} while ( *argv++ != 0 );
/*
* dump out a gmon.sum file if requested
*/
if ( sflag ) {
dumpsum( GMONSUM );
}
/*
* assign samples to procedures
*/
asgnsamples();
/*
* assemble the dynamic profile
*/
timesortnlp = doarcs();
if (bsd_style_output) {
printgprof( timesortnlp ); /* print the dynamic profile */
printprof(); /* print the flat profile */
} else {
printprof(); /* print the flat profile */
printgprof( timesortnlp ); /* print the dynamic profile */
}
/*
* print the index
*/
printindex();
done();
}
/*
* Set up string and symbol tables from a.out.
* and optionally the text space.
* On return symbol table is sorted by value.
*/
getnfile()
{
int valcmp();
abfd = bfd_openr (a_outname, NULL);
if (abfd == NULL) {
perror (a_outname);
done();
}
if (!bfd_check_format (abfd, bfd_object)) {
fprintf (stderr, "%s: %s: bad format\n", whoami, a_outname);
done();
}
/* getstrtab(nfile); */
getsymtab(abfd);
gettextspace( abfd );
qsort(nl, nname, sizeof(nltype), valcmp);
# ifdef DEBUG
if ( debug & AOUTDEBUG ) {
register int j;
for (j = 0; j < nname; j++){
printf("[getnfile] 0X%08x\t%s\n", nl[j].value, nl[j].name);
}
}
# endif DEBUG
}
/*
* Read in symbol table
*/
getsymtab(abfd)
bfd *abfd;
{
register long i;
int askfor;
int nosyms;
asymbol **syms;
i = get_symtab_upper_bound (abfd); /* This will probably give us more
* than we need, but that's ok.
*/
syms = (asymbol**)xmalloc (i);
nosyms = bfd_canonicalize_symtab (abfd, syms);
nname = 0;
for (i = 0; i < nosyms; i++) {
if (!funcsymbol (syms[i]))
continue;
nname++;
}
if (nname == 0) {
fprintf(stderr, "%s: %s: no symbols\n", whoami , a_outname );
done();
}
askfor = nname + 1;
nl = (nltype *) calloc( askfor , sizeof(nltype) );
if (nl == 0) {
fprintf(stderr, "%s: No room for %d bytes of symbol table\n",
whoami, askfor * sizeof(nltype) );
done();
}
/* pass2 - read symbols */
npe = nl;
nname = 0;
for (i = 0; i < nosyms; i++) {
if (!funcsymbol (syms[i])) {
# ifdef DEBUG
if ( debug & AOUTDEBUG ) {
printf( "[getsymtab] rejecting: 0x%x %s\n" ,
syms[i]->value, syms[i]->name);
}
# endif DEBUG
continue;
}
/* Symbol offsets are always section-relative. */
npe->value = syms[i]->value + syms[i]->section->vma;
npe->name = syms[i]->name;
/* If we see "main" without an initial '_', we assume
names are *not* prefixed by '_'. */
if (npe->name[0] == 'm' && discard_underscores
&& strcmp(npe->name, "main") == 0)
discard_underscores = 0;
# ifdef DEBUG
if ( debug & AOUTDEBUG ) {
printf( "[getsymtab] %d %s 0x%08x\n" ,
nname , npe -> name , npe -> value );
}
# endif DEBUG
npe++;
nname++;
}
npe->value = -1;
}
/*
* read in the text space of an a.out file
*/
gettextspace( abfd )
bfd *abfd;
{
asection *texsec;
if ( cflag == 0 ) {
return;
}
texsec = bfd_get_section_by_name (abfd, ".text");
if (texsec == NULL) {
return;
}
textspace = (u_char *) malloc( texsec->_cooked_size );
if ( textspace == 0 ) {
fprintf( stderr , "%s: ran out room for %d bytes of text space: " ,
whoami , texsec->_cooked_size);
fprintf( stderr , "can't do -c\n" );
return;
}
bfd_get_section_contents (abfd, texsec, textspace, texsec->filepos,
texsec->_cooked_size);
}
/*
* information from a gmon.out file is in two parts:
* an array of sampling hits within pc ranges,
* and the arcs.
*/
getpfile(filename)
char *filename;
{
FILE *pfile;
FILE *openpfile();
struct rawarc arc;
struct veryrawarc rawarc;
pfile = openpfile(filename);
readsamples(pfile);
/*
* the rest of the file consists of
* a bunch of <from,self,count> tuples.
*/
while ( fread( &rawarc , sizeof rawarc , 1 , pfile ) == 1 ) {
arc.raw_frompc = bfd_get_32 (abfd, (bfd_byte *) rawarc.raw_frompc);
arc.raw_selfpc = bfd_get_32 (abfd, (bfd_byte *) rawarc.raw_selfpc);
arc.raw_count = bfd_get_32 (abfd, (bfd_byte *) rawarc.raw_count);
# ifdef DEBUG
if ( debug & SAMPLEDEBUG ) {
printf( "[getpfile] frompc 0x%x selfpc 0x%x count %d\n" ,
arc.raw_frompc , arc.raw_selfpc , arc.raw_count );
}
# endif DEBUG
/*
* add this arc
*/
tally( &arc );
}
fclose(pfile);
}
FILE *
openpfile(filename)
char *filename;
{
struct hdr tmp;
struct rawhdr raw;
FILE *pfile;
if((pfile = fopen(filename, "r")) == NULL) {
perror(filename);
done();
}
if (sizeof(struct rawhdr) != fread(&raw, 1, sizeof(struct rawhdr), pfile))
{
fprintf(stderr, "%s: file too short to be a gmon file\n", filename);
done();
}
tmp.lowpc = (UNIT *)bfd_get_32 (abfd, (bfd_byte *) &raw.lowpc[0]);
tmp.highpc = (UNIT *)bfd_get_32 (abfd, (bfd_byte *) &raw.highpc[0]);
tmp.ncnt = bfd_get_32 (abfd, (bfd_byte *) &raw.ncnt[0]);
if ( s_highpc != 0 && ( tmp.lowpc != h.lowpc ||
tmp.highpc != h.highpc || tmp.ncnt != h.ncnt ) ) {
fprintf(stderr, "%s: incompatible with first gmon file\n", filename);
done();
}
h = tmp;
s_lowpc = (unsigned long) h.lowpc;
s_highpc = (unsigned long) h.highpc;
lowpc = (unsigned long)h.lowpc / sizeof(UNIT);
highpc = (unsigned long)h.highpc / sizeof(UNIT);
sampbytes = h.ncnt - sizeof(struct rawhdr);
nsamples = sampbytes / sizeof (UNIT);
# ifdef DEBUG
if ( debug & SAMPLEDEBUG ) {
printf( "[openpfile] hdr.lowpc 0x%x hdr.highpc 0x%x hdr.ncnt %d\n",
h.lowpc , h.highpc , h.ncnt );
printf( "[openpfile] s_lowpc 0x%x s_highpc 0x%x\n" ,
s_lowpc , s_highpc );
printf( "[openpfile] lowpc 0x%x highpc 0x%x\n" ,
lowpc , highpc );
printf( "[openpfile] sampbytes %d nsamples %d\n" ,
sampbytes , nsamples );
}
# endif DEBUG
return(pfile);
}
tally( rawp )
struct rawarc *rawp;
{
nltype *parentp;
nltype *childp;
parentp = nllookup( rawp -> raw_frompc );
childp = nllookup( rawp -> raw_selfpc );
if ( kflag
&& onlist( kfromlist , parentp -> name )
&& onlist( ktolist , childp -> name ) ) {
return;
}
childp -> ncall += rawp -> raw_count;
# ifdef DEBUG
if ( debug & TALLYDEBUG ) {
printf( "[tally] arc from %s to %s traversed %d times\n" ,
parentp -> name , childp -> name , rawp -> raw_count );
}
# endif DEBUG
addarc( parentp , childp , rawp -> raw_count );
}
/*
* dump out the gmon.sum file
*/
dumpsum( sumfile )
char *sumfile;
{
register nltype *nlp;
register arctype *arcp;
struct rawarc arc;
FILE *sfile;
if ( ( sfile = fopen ( sumfile , "w" ) ) == NULL ) {
perror( sumfile );
done();
}
/*
* dump the header; use the last header read in
*/
if ( fwrite( &h , sizeof h , 1 , sfile ) != 1 ) {
perror( sumfile );
done();
}
/*
* dump the samples
*/
if (fwrite(samples, sizeof (UNIT), nsamples, sfile) != nsamples) {
perror( sumfile );
done();
}
/*
* dump the normalized raw arc information
*/
for ( nlp = nl ; nlp < npe ; nlp++ ) {
for ( arcp = nlp -> children ; arcp ; arcp = arcp -> arc_childlist ) {
arc.raw_frompc = arcp -> arc_parentp -> value;
arc.raw_selfpc = arcp -> arc_childp -> value;
arc.raw_count = arcp -> arc_count;
if ( fwrite ( &arc , sizeof arc , 1 , sfile ) != 1 ) {
perror( sumfile );
done();
}
# ifdef DEBUG
if ( debug & SAMPLEDEBUG ) {
printf( "[dumpsum] frompc 0x%x selfpc 0x%x count %d\n" ,
arc.raw_frompc , arc.raw_selfpc , arc.raw_count );
}
# endif DEBUG
}
}
fclose( sfile );
}
valcmp(p1, p2)
nltype *p1, *p2;
{
if ( p1 -> value < p2 -> value ) {
return LESSTHAN;
}
if ( p1 -> value > p2 -> value ) {
return GREATERTHAN;
}
return EQUALTO;
}
readsamples(pfile)
FILE *pfile;
{
register i;
if (samples == 0) {
samples = (int *) calloc (nsamples, sizeof(int));
if (samples == 0) {
fprintf( stderr , "%s: No room for %d sample pc's\n",
whoami , nsamples);
done();
}
}
for (i = 0; i < nsamples; i++) {
UNIT raw;
int value;
fread(raw, sizeof (raw), 1, pfile);
value = bfd_get_16 (abfd, (bfd_byte *) raw);
if (feof(pfile))
break;
samples[i] += value;
}
if (i != nsamples) {
fprintf(stderr,
"%s: unexpected EOF after reading %d/%d samples\n",
whoami , --i , nsamples );
done();
}
}
/*
* Assign samples to the procedures to which they belong.
*
* There are three cases as to where pcl and pch can be
* with respect to the routine entry addresses svalue0 and svalue1
* as shown in the following diagram. overlap computes the
* distance between the arrows, the fraction of the sample
* that is to be credited to the routine which starts at svalue0.
*
* svalue0 svalue1
* | |
* v v
*
* +-----------------------------------------------+
* | |
* | ->| |<- ->| |<- ->| |<- |
* | | | | | |
* +---------+ +---------+ +---------+
*
* ^ ^ ^ ^ ^ ^
* | | | | | |
* pcl pch pcl pch pcl pch
*
* For the vax we assert that samples will never fall in the first
* two bytes of any routine, since that is the entry mask,
* thus we give call alignentries() to adjust the entry points if
* the entry mask falls in one bucket but the code for the routine
* doesn't start until the next bucket. In conjunction with the
* alignment of routine addresses, this should allow us to have
* only one sample for every four bytes of text space and never
* have any overlap (the two end cases, above).
*/
asgnsamples()
{
register int j;
int ccnt;
double time;
unsigned long pcl, pch;
register int i;
unsigned long overlap;
unsigned long svalue0, svalue1;
/* read samples and assign to namelist symbols */
scale = highpc - lowpc;
scale /= nsamples - 1;
alignentries();
for (i = 0, j = 1; i < nsamples; i++) {
ccnt = samples[i];
if (ccnt == 0)
continue;
pcl = lowpc + scale * i;
pch = lowpc + scale * (i + 1);
time = ccnt;
# ifdef DEBUG
if ( debug & SAMPLEDEBUG ) {
printf( "[asgnsamples] pcl 0x%x pch 0x%x ccnt %d\n" ,
pcl , pch , ccnt );
}
# endif DEBUG
totime += time;
for (j = j - 1; j < nname; j++) {
svalue0 = nl[j].svalue;
svalue1 = nl[j+1].svalue;
/*
* if high end of tick is below entry address,
* go for next tick.
*/
if (pch < svalue0)
break;
/*
* if low end of tick into next routine,
* go for next routine.
*/
if (pcl >= svalue1)
continue;
overlap = min(pch, svalue1) - max(pcl, svalue0);
if (overlap > 0) {
# ifdef DEBUG
if (debug & SAMPLEDEBUG) {
printf("[asgnsamples] (0x%x->0x%x-0x%x) %s gets %f ticks %d overlap\n",
nl[j].value/sizeof(UNIT), svalue0, svalue1,
nl[j].name,
overlap * time / scale, overlap);
}
# endif DEBUG
nl[j].time += overlap * time / scale;
}
}
}
# ifdef DEBUG
if (debug & SAMPLEDEBUG) {
printf("[asgnsamples] totime %f\n", totime);
}
# endif DEBUG
}
unsigned long
min(a, b)
unsigned long a,b;
{
if (a<b)
return(a);
return(b);
}
unsigned long
max(a, b)
unsigned long a,b;
{
if (a>b)
return(a);
return(b);
}
/*
* calculate scaled entry point addresses (to save time in asgnsamples),
* and possibly push the scaled entry points over the entry mask,
* if it turns out that the entry point is in one bucket and the code
* for a routine is in the next bucket.
*/
alignentries()
{
register struct nl *nlp;
unsigned long bucket_of_entry;
unsigned long bucket_of_code;
for (nlp = nl; nlp < npe; nlp++) {
nlp -> svalue = nlp -> value / sizeof(UNIT);
bucket_of_entry = (nlp->svalue - lowpc) / scale;
bucket_of_code = (nlp->svalue + UNITS_TO_CODE - lowpc) / scale;
if (bucket_of_entry < bucket_of_code) {
# ifdef DEBUG
if (debug & SAMPLEDEBUG) {
printf("[alignentries] pushing svalue 0x%x to 0x%x\n",
nlp->svalue, nlp->svalue + UNITS_TO_CODE);
}
# endif DEBUG
nlp->svalue += UNITS_TO_CODE;
}
}
}
bool
funcsymbol( symp )
asymbol *symp;
{
extern char *strtab; /* string table from a.out */
extern int aflag; /* if static functions aren't desired */
CONST char *name;
int i;
char symprefix;
/*
* must be a text symbol,
* and static text symbols don't qualify if aflag set.
*/
if (!symp->section)
return FALSE;
if (aflag && (symp->flags&BSF_LOCAL)) {
#if defined(DEBUG)
fprintf (stderr, "%s(%d): %s: not a function\n", __FILE__, __LINE__, symp->name);
#endif
return FALSE;
}
symprefix = bfd_get_symbol_leading_char (abfd);
i = bfd_decode_symclass (symp);
#if defined(DEBUG) && 0
if (i != 'T' && i != 't')
fprintf (stderr, "%s(%d): %s is of class %c\n", __FILE__, __LINE__, symp->name, i);
#endif
/*
* Any external text symbol should be okay. (Only problem would be
* variables in the text section.)
*/
if (i == 'T')
return TRUE;
/*
* 't' is static text; -a says to ignore it. So if it's not
* a static text symbol, *or* it is and the user gave -a, we
* ignore it.
*/
if (i != 't' || aflag)
return FALSE;
/*
* can't have any `funny' characters in name,
* where `funny' includes `.', .o file names
* and `$', pascal labels.
*/
if (!symp->name)
return FALSE;
for (name = symp->name; *name; name++) {
if ( *name == '.' || *name == '$' ) {
return FALSE;
}
}
/* On systems where the C compiler adds an underscore to all names,
* static names without underscores seem usually to be labels in
* hand written assembler in the library. We don't want these
* names. This is certainly necessary on a Sparc running SunOS 4.1
* (try profiling a program that does a lot of division). I don't
* know whether it has harmful side effects on other systems.
* Perhaps it should be made configurable.
*/
if (symprefix && symprefix != *symp->name)
return FALSE;
return TRUE;
}
done()
{
exit(0);
}