nils/shairport-sync
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

rename absolute_time_in_ns to absolute_time_in_fp (more accurate name). Fix osx timing calculation.

Browse Source
This commit is contained in:
Mike Brady
2014-09-07 12:31:23 +01:00
parent 344cf45feb
commit 0733e962bb
1 changed files with 10 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common.c
+10 -9
View File
@@ -421,12 +421,12 @@ double vol2attn(double vol, long max_db, long min_db) {
return vol_setting;
}
uint64_t get_absolute_time_in_ns() {
uint64_t time_now_ns;
uint64_t get_absolute_time_in_fp() {
uint64_t time_now_fp;
#ifdef COMPILE_FOR_LINUX
struct timespec tn;
clock_gettime(CLOCK_MONOTONIC,&tn);
time_now_ns=((uint64_t)tn.tv_sec<<32)+((uint64_t)tn.tv_nsec<<32)/1000000000;
time_now_fp=((uint64_t)tn.tv_sec<<32)+((uint64_t)tn.tv_nsec<<32)/1000000000;
#endif
#ifdef COMPILE_FOR_OSX
uint64_t time_now_mach;
@@ -438,7 +438,7 @@ uint64_t get_absolute_time_in_ns() {
// If this is the first time we've run, get the timebase.
// We can use denom == 0 to indicate that sTimebaseInfo is
// uninitialised because it makes no sense to have a zero
// denominator is a fraction.
// denominator in a fraction.
if ( sTimebaseInfo.denom == 0 ) {
debug(1,"Mac initialise timebase info.");
@@ -448,12 +448,13 @@ uint64_t get_absolute_time_in_ns() {
// Do the maths. We hope that the multiplication doesn't
// overflow; the price you pay for working in fixed point.
// need to turn this into NTP 64-bit fixed point format
// and change the very misleading title of this function to
// get_absolute_fp_time or something
time_now_ns = time_now_mach * sTimebaseInfo.numer / sTimebaseInfo.denom;
// this gives us nanoseconds
uint64_t time_now_ns = time_now_mach * sTimebaseInfo.numer / sTimebaseInfo.denom;
// take the units and shift them to the upper half of the fp, and take the nanoseconds, shift them to the upper half and then divide the result to 1000000000
time_now_fp = ((time_now_ns/1000000000)<<32) + (((time_now_ns%1000000000)<<32)/1000000000);
#endif
return time_now_ns;
return time_now_fp;
}