apps/{s_client.c|s_socket}.c: omit usleep calls.

Even though removed calls were oiriginally added on Windows, problem
they tried to mitigate is not Windows-specific.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5887)
This commit is contained in:
Andy Polyakov 2018-04-05 19:19:35 +02:00
parent dcf3d83faf
commit f69d050ee3
2 changed files with 13 additions and 26 deletions

View File

@ -3050,19 +3050,6 @@ int s_client_main(int argc, char **argv)
print_stuff(bio_c_out, con, full_log);
do_ssl_shutdown(con);
/*
* Give the socket time to send its last data before we close it.
* No amount of setting SO_LINGER etc on the socket seems to persuade
* Windows to send the data before closing the socket...but sleeping
* for a short time seems to do it (units in ms)
* TODO: Find a better way to do this
*/
#if defined(OPENSSL_SYS_WINDOWS)
Sleep(50);
#elif defined(OPENSSL_SYS_CYGWIN)
usleep(50000);
#endif
/*
* If we ended with an alert being sent, but still with data in the
* network buffer to be read, then calling BIO_closesocket() will
@ -3074,6 +3061,19 @@ int s_client_main(int argc, char **argv)
* TCP-RST. This seems to allow the peer to read the alert data.
*/
shutdown(SSL_get_fd(con), 1); /* SHUT_WR */
/*
* We just said we have nothing else to say, but it doesn't mean that
* the other side has nothing. It's even recommended to consume incoming
* data. [In testing context this ensures that alerts are passed on...]
*/
timeout.tv_sec = 0;
timeout.tv_usec = 500000; /* some extreme round-trip */
do {
FD_ZERO(&readfds);
openssl_fdset(s, &readfds);
} while (select(s + 1, &readfds, NULL, NULL, &timeout) > 0
&& BIO_read(sbio, sbuf, BUFSIZZ) > 0);
BIO_closesocket(SSL_get_fd(con));
end:
if (con != NULL) {

View File

@ -333,19 +333,6 @@ int do_server(int *accept_sock, const char *host, const char *port,
BIO_set_tcp_ndelay(sock, 1);
i = (*cb)(sock, type, protocol, context);
/*
* Give the socket time to send its last data before we close it.
* No amount of setting SO_LINGER etc on the socket seems to
* persuade Windows to send the data before closing the socket...
* but sleeping for a short time seems to do it (units in ms)
* TODO: Find a better way to do this
*/
#if defined(OPENSSL_SYS_WINDOWS)
Sleep(50);
#elif defined(OPENSSL_SYS_CYGWIN)
usleep(50000);
#endif
/*
* If we ended with an alert being sent, but still with data in the
* network buffer to be read, then calling BIO_closesocket() will