LIBPFM(3) Linux Programmer's Manual LIBPFM(3)NAME
pfm_regmask_set, pfm_regmask_isset, pfm_regmask_clr, pfm_reg‐
mask_weight, pfm_regmask_eq, pfm_regmask_and, pfm_regmask_or, pfm_reg‐
mask_copy - operations on pfmlib_regmask_t bitmasks
SYNOPSIS
#include <perfmon/pfmlib.h>
int pfm_regmask_isset(pfmlib_regmask_t *mask, unsigned int b);
int pfm_regmask_set(pfmlib_regmask_t *mask, unsigned int b);
int pfm_regmask_clr(pfmlib_regmask_t *mask, unsigned int b);
int pfm_regmask_weight(pfmlib_regmask_t *mask, unsigned int *w);
int pfm_regmask_eq(pfmlib_regmask_t *mask1, pfmlib_regmask_t *mask2);
int pfm_regmask_and(pfmlib_regmask_t *dest, pfmlib_regmask_t *m1, pmlib_regmask_t *m2);
int pfm_regmask_or(pfmlib_regmask_t *dest, pfmlib_regmask_t *m1, pmlib_regmask_t *m2);
int pfm_regmask_copy(pfmlib_regmask_t *dest, pfmlib_regmask_t *src);
DESCRIPTION
This set of function is used to operate of the pfmlib_regmask_t bit‐
masks that are returned by certain functions or passed to the pfm_dis‐
patch_events() function.
To ensure portability, it is important that applications use only the
functions specified here to access the bitmasks. It is strongly dis‐
couraged to access the internal fields of the pfm_regmask_t structure.
The pfm_regmask_set() function is used to set bit b in the bitmask
mask.
The pfm_regmask_clr() function is used to clear bit b in the bitmask
mask.
The pfm_regmask_isset() function returns a non-zero value if b is set
in the bitmask mask.
The pfm_regmask_weight() function returns in w the number of bits set
in the bitmask mask.
The pfm_regmask_eq() function returns a non-zero value if the bitmasks
mask1 and mask2 are identical.
The pfm_regmask_and() function returns in bitmask dest the result of
the logical AND operation between bitmask m1 and bitmask m2.
The pfm_regmask_or() function returns in bitmask dest the result of the
logical OR operation between bitmask m1 and bitmask m2.
The pfm_regmask_copy() function copies bitmask src into bitmask dest.
RETURN
The function returns whether or not it was successful. A return value
of PFMLIB_SUCCESS indicates success, otherwise the value is the error
code.
ERRORS
PFMLIB_ERR_INVAL the bit b exceeds the limit supported by the library
AUTHOR
Stephane Eranian <eranian@hpl.hp.com>
Apr, 2006 LIBPFM(3)
