pkarashchenko commented on a change in pull request #5526:
URL: https://github.com/apache/incubator-nuttx/pull/5526#discussion_r809052145
##########
File path: net/tcp/tcp_send_buffered.c
##########
@@ -1172,7 +1204,12 @@ ssize_t psock_tcp_send(FAR struct socket *psock, FAR
const void *buf,
goto errout_with_lock;
}
- net_lockedwait_uninterruptible(&conn->snd_sem);
+ remain = tcp_send_gettimeout(anchor, remain);
+ ret = net_timedwait_uninterruptible(&conn->snd_sem, remain);
Review comment:
> 1. anchor time should not to be changed on sending process,
>
I do not propose to change anchor time on sending process. It will be done
once. I just used a different name `anchor` -> `start`.
> 2. remain, anchor, _SO_TIMEOUT(conn->sconn.s_sndtimeo), these are in
milliseconds, which will reduce unnecessary type cast
>
I do not fully understand this. In my proposal I don't think that there are
additional type casts.
> 3. what different with current tcp_send_gettimeout ?
In general I propose to track running time in `clock_t` (the same type that
is returned by `clock_systime_ticks`) and perform conversion to `ms` only
inside the `tcp_send_gettimeout` once. This will give better granularity
compared to current code. Also it will eliminate having `uint32_t remain;`
inside the `psock_tcp_send`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@nuttx.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
