DWS/chrony
1
0
Fork 0
mirror of https://gitlab.com/chrony/chrony.git synced 2025-12-05 02:05:07 -05:00
Code Issues Packages Projects Releases Wiki Activity

ntp: don't require previous HW TX timestamp to wait for another

Browse Source 
Clients sockets are closed immediately after receiving valid response.
Don't wait for the first early HW TX timestamp to enable waiting for
late timestamps. It may take a long time or never come if the HW/driver
is consistently slow. It's a chicken and egg problem.

Instead, simply check if HW timestamping is enabled on at least one
interface. Responses from NTP sources on other interfaces will always be
saved (for 1 millisecond by default).
This commit is contained in:
Miroslav Lichvar
2023-08-08 15:44:25 +02:00
parent 7bc7d00297
commit 3dc9f1ff92
6 changed files with 28 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
test/unit/ntp_core.c
View File

@@ -35,6 +35,7 @@ static struct timespec current_time;
static NTP_Packet req_buffer, res_buffer;
static int req_length, res_length;
#define NIO_IsHwTsEnabled() 1
#define NIO_OpenServerSocket(addr) ((addr)->ip_addr.family != IPADDR_UNSPEC ? 100 : 0)
#define NIO_CloseServerSocket(fd) assert(fd == 100)
#define NIO_OpenClientSocket(addr) ((addr)->ip_addr.family != IPADDR_UNSPEC ? 101 : 0)
@@ -106,13 +107,9 @@ send_request(NCR_Instance inst, int late_hwts)
}
if (late_hwts) {
inst->had_hw_tx_timestamp = 1;
inst->report.total_good_count++;
} else {
if (random() % 2)
inst->had_hw_tx_timestamp = 0;
else
inst->report.total_good_count = 0;
inst->report.total_good_count = 0;
}
}