mirror of
https://github.com/tuxbox-neutrino/neutrino.git
synced 2025-08-27 23:42:58 +02:00
rcinput: remove unused code, comment FIXME
I think I have mostly deciphered the meaning of this :-)
This commit is contained in:
committed by
svenhoefer
parent
364dd41655
commit
89f6da98f1
@@ -884,39 +884,11 @@ void CRCInput::getMsg_us(neutrino_msg_t * msg, neutrino_msg_data_t * data, uint6
|
||||
{
|
||||
case CSectionsdClient::EVT_TIMESET:
|
||||
{
|
||||
#if 0
|
||||
struct timeval ltv;
|
||||
gettimeofday(<v, NULL);
|
||||
int64_t timeOld = ltv.tv_usec + ltv.tv_sec * (int64_t)1000000;
|
||||
time_t dvbtime = *((time_t*)p);
|
||||
if (dvbtime) {
|
||||
printf("[neutrino] timeset event. ");
|
||||
time_t difftime = dvbtime - ltv.tv_sec;
|
||||
if (abs(difftime) > 120)
|
||||
{
|
||||
printf("difference is %ld s, stepping...\n", difftime);
|
||||
if (stime(&dvbtime))
|
||||
perror("stime");
|
||||
} else if (difftime != 0) {
|
||||
struct timeval oldd;
|
||||
ltv.tv_sec = difftime;
|
||||
ltv.tv_usec = 0;
|
||||
if (adjtime(<v, &oldd))
|
||||
perror("adjtime");
|
||||
int64_t t = oldd.tv_sec * 1000000LL + oldd.tv_usec;
|
||||
printf("difference is %ld s, using adjtime(). oldd: %lld us\n", difftime, t);
|
||||
} else
|
||||
printf("difference is 0 s, nothing to do...\n");
|
||||
}
|
||||
gettimeofday( <v, NULL );
|
||||
int64_t timeNew = ltv.tv_usec + ltv.tv_sec * (int64_t)1000000;
|
||||
|
||||
delete[] p;//new [] delete []
|
||||
p = new unsigned char[sizeof(int64_t)];
|
||||
*(int64_t*) p = timeNew - timeOld;
|
||||
#endif
|
||||
printf("[neutrino] CSectionsdClient::EVT_TIMESET: timediff %" PRId64 "\n", *(int64_t*) p);
|
||||
/* FIXME what this code really do ? */
|
||||
/* compensate last_keypress for autorepeat / long press detection
|
||||
* I doubt this works correcty, if we had kernel 3.4+, using
|
||||
* EVIOCSCLOCKID ioctl would be better.
|
||||
* Still guessing the logic behind the condition... */
|
||||
if ((int64_t)last_keypress > *(int64_t*)p)
|
||||
last_keypress += *(int64_t *)p;
|
||||
|
||||
|
Reference in New Issue
Block a user