diff options
author | Dominik Brodowski <linux@dominikbrodowski.net> | 2011-03-30 16:30:11 +0200 |
---|---|---|
committer | Dominik Brodowski <linux@dominikbrodowski.net> | 2011-07-29 18:35:36 +0200 |
commit | 7fe2f6399a84760a9af8896ac152728250f82adb (patch) | |
tree | fa4bf236359b8d6d9f8d6ff823ddd3e839da5768 /tools/power/cpupower/utils/helpers/bitmask.c | |
parent | 02f8c6aee8df3cdc935e9bdd4f2d020306035dbe (diff) |
cpupowerutils - cpufrequtils extended with quite some features
CPU power consumption vs performance tuning is no longer
limited to CPU frequency switching anymore: deep sleep states,
traditional dynamic frequency scaling and hidden turbo/boost
frequencies are tied close together and depend on each other.
The first two exist on different architectures like PPC, Itanium and
ARM, the latter (so far) only on X86. On X86 the APU (CPU+GPU) will
only run most efficiently if CPU and GPU has proper power management
in place.
Users and Developers want to have *one* tool to get an overview what
their system supports and to monitor and debug CPU power management
in detail. The tool should compile and work on as many architectures
as possible.
Once this tool stabilizes a bit, it is intended to replace the
Intel-specific tools in tools/power/x86
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Diffstat (limited to 'tools/power/cpupower/utils/helpers/bitmask.c')
-rw-r--r-- | tools/power/cpupower/utils/helpers/bitmask.c | 290 |
1 files changed, 290 insertions, 0 deletions
diff --git a/tools/power/cpupower/utils/helpers/bitmask.c b/tools/power/cpupower/utils/helpers/bitmask.c new file mode 100644 index 000000000000..60f4d69bb20d --- /dev/null +++ b/tools/power/cpupower/utils/helpers/bitmask.c @@ -0,0 +1,290 @@ +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include <helpers/bitmask.h> + +/* How many bits in an unsigned long */ +#define bitsperlong (8 * sizeof(unsigned long)) + +/* howmany(a,b) : how many elements of size b needed to hold all of a */ +#define howmany(x,y) (((x)+((y)-1))/(y)) + +/* How many longs in mask of n bits */ +#define longsperbits(n) howmany(n, bitsperlong) + +#define max(a,b) ((a) > (b) ? (a) : (b)) + +/* + * Allocate and free `struct bitmask *` + */ + +/* Allocate a new `struct bitmask` with a size of n bits */ +struct bitmask *bitmask_alloc(unsigned int n) +{ + struct bitmask *bmp; + + bmp = malloc(sizeof(*bmp)); + if (bmp == 0) + return 0; + bmp->size = n; + bmp->maskp = calloc(longsperbits(n), sizeof(unsigned long)); + if (bmp->maskp == 0) { + free(bmp); + return 0; + } + return bmp; +} + +/* Free `struct bitmask` */ +void bitmask_free(struct bitmask *bmp) +{ + if (bmp == 0) + return; + free(bmp->maskp); + bmp->maskp = (unsigned long *)0xdeadcdef; /* double free tripwire */ + free(bmp); +} + +/* + * The routines _getbit() and _setbit() are the only + * routines that actually understand the layout of bmp->maskp[]. + * + * On little endian architectures, this could simply be an array of + * bytes. But the kernel layout of bitmasks _is_ visible to userspace + * via the sched_(set/get)affinity calls in Linux 2.6, and on big + * endian architectures, it is painfully obvious that this is an + * array of unsigned longs. + */ + +/* Return the value (0 or 1) of bit n in bitmask bmp */ +static unsigned int _getbit(const struct bitmask *bmp, unsigned int n) +{ + if (n < bmp->size) + return (bmp->maskp[n/bitsperlong] >> (n % bitsperlong)) & 1; + else + return 0; +} + +/* Set bit n in bitmask bmp to value v (0 or 1) */ +static void _setbit(struct bitmask *bmp, unsigned int n, unsigned int v) +{ + if (n < bmp->size) { + if (v) + bmp->maskp[n/bitsperlong] |= 1UL << (n % bitsperlong); + else + bmp->maskp[n/bitsperlong] &= ~(1UL << (n % bitsperlong)); + } +} + +/* + * When parsing bitmask lists, only allow numbers, separated by one + * of the allowed next characters. + * + * The parameter 'sret' is the return from a sscanf "%u%c". It is + * -1 if the sscanf input string was empty. It is 0 if the first + * character in the sscanf input string was not a decimal number. + * It is 1 if the unsigned number matching the "%u" was the end of the + * input string. It is 2 if one or more additional characters followed + * the matched unsigned number. If it is 2, then 'nextc' is the first + * character following the number. The parameter 'ok_next_chars' + * is the nul-terminated list of allowed next characters. + * + * The mask term just scanned was ok if and only if either the numbers + * matching the %u were all of the input or if the next character in + * the input past the numbers was one of the allowed next characters. + */ +static int scan_was_ok(int sret, char nextc, const char *ok_next_chars) +{ + return sret == 1 || + (sret == 2 && strchr(ok_next_chars, nextc) != NULL); +} + +static const char *nexttoken(const char *q, int sep) +{ + if (q) + q = strchr(q, sep); + if (q) + q++; + return q; +} + +/* Set a single bit i in bitmask */ +struct bitmask *bitmask_setbit(struct bitmask *bmp, unsigned int i) +{ + _setbit(bmp, i, 1); + return bmp; +} + +/* Set all bits in bitmask: bmp = ~0 */ +struct bitmask *bitmask_setall(struct bitmask *bmp) +{ + unsigned int i; + for (i = 0; i < bmp->size; i++) + _setbit(bmp, i, 1); + return bmp; +} + +/* Clear all bits in bitmask: bmp = 0 */ +struct bitmask *bitmask_clearall(struct bitmask *bmp) +{ + unsigned int i; + for (i = 0; i < bmp->size; i++) + _setbit(bmp, i, 0); + return bmp; +} + +/* True if all bits are clear */ +int bitmask_isallclear(const struct bitmask *bmp) +{ + unsigned int i; + for (i = 0; i < bmp->size; i++) + if (_getbit(bmp, i)) + return 0; + return 1; +} + +/* True if specified bit i is set */ +int bitmask_isbitset(const struct bitmask *bmp, unsigned int i) +{ + return _getbit(bmp, i); +} + +/* Number of lowest set bit (min) */ +unsigned int bitmask_first(const struct bitmask *bmp) +{ + return bitmask_next(bmp, 0); +} + +/* Number of highest set bit (max) */ +unsigned int bitmask_last(const struct bitmask *bmp) +{ + unsigned int i; + unsigned int m = bmp->size; + for (i = 0; i < bmp->size; i++) + if (_getbit(bmp, i)) + m = i; + return m; +} + +/* Number of next set bit at or above given bit i */ +unsigned int bitmask_next(const struct bitmask *bmp, unsigned int i) +{ + unsigned int n; + for (n = i; n < bmp->size; n++) + if (_getbit(bmp, n)) + break; + return n; +} + +/* + * Parses a comma-separated list of numbers and ranges of numbers, + * with optional ':%u' strides modifying ranges, into provided bitmask. + * Some examples of input lists and their equivalent simple list: + * Input Equivalent to + * 0-3 0,1,2,3 + * 0-7:2 0,2,4,6 + * 1,3,5-7 1,3,5,6,7 + * 0-3:2,8-15:4 0,2,8,12 + */ +int bitmask_parselist(const char *buf, struct bitmask *bmp) +{ + const char *p, *q; + + bitmask_clearall(bmp); + + q = buf; + while (p = q, q = nexttoken(q, ','), p) { + unsigned int a; /* begin of range */ + unsigned int b; /* end of range */ + unsigned int s; /* stride */ + const char *c1, *c2; /* next tokens after '-' or ',' */ + char nextc; /* char after sscanf %u match */ + int sret; /* sscanf return (number of matches) */ + + sret = sscanf(p, "%u%c", &a, &nextc); + if (!scan_was_ok(sret, nextc, ",-")) + goto err; + b = a; + s = 1; + c1 = nexttoken(p, '-'); + c2 = nexttoken(p, ','); + if (c1 != NULL && (c2 == NULL || c1 < c2)) { + sret = sscanf(c1, "%u%c", &b, &nextc); + if (!scan_was_ok(sret, nextc, ",:")) + goto err; + c1 = nexttoken(c1, ':'); + if (c1 != NULL && (c2 == NULL || c1 < c2)) { + sret = sscanf(c1, "%u%c", &s, &nextc); + if (!scan_was_ok(sret, nextc, ",")) + goto err; + } + } + if (!(a <= b)) + goto err; + if (b >= bmp->size) + goto err; + while (a <= b) { + _setbit(bmp, a, 1); + a += s; + } + } + return 0; +err: + bitmask_clearall(bmp); + return -1; +} + +/* + * emit(buf, buflen, rbot, rtop, len) + * + * Helper routine for bitmask_displaylist(). Write decimal number + * or range to buf+len, suppressing output past buf+buflen, with optional + * comma-prefix. Return len of what would be written to buf, if it + * all fit. + */ + +static inline int emit(char *buf, int buflen, int rbot, int rtop, int len) +{ + if (len > 0) + len += snprintf(buf + len, max(buflen - len, 0), ","); + if (rbot == rtop) + len += snprintf(buf + len, max(buflen - len, 0), "%d", rbot); + else + len += snprintf(buf + len, max(buflen - len, 0), "%d-%d", rbot, rtop); + return len; +} + +/* + * Write decimal list representation of bmp to buf. + * + * Output format is a comma-separated list of decimal numbers and + * ranges. Consecutively set bits are shown as two hyphen-separated + * decimal numbers, the smallest and largest bit numbers set in + * the range. Output format is compatible with the format + * accepted as input by bitmap_parselist(). + * + * The return value is the number of characters which would be + * generated for the given input, excluding the trailing '\0', as + * per ISO C99. + */ + +int bitmask_displaylist(char *buf, int buflen, const struct bitmask *bmp) +{ + int len = 0; + /* current bit is 'cur', most recently seen range is [rbot, rtop] */ + unsigned int cur, rbot, rtop; + + if (buflen > 0) + *buf = 0; + rbot = cur = bitmask_first(bmp); + while (cur < bmp->size) { + rtop = cur; + cur = bitmask_next(bmp, cur+1); + if (cur >= bmp->size || cur > rtop + 1) { + len = emit(buf, buflen, rbot, rtop, len); + rbot = cur; + } + } + return len; +} |