Browse Source

Remove remnants of blocking DTLS timeouts.

We only support non-blocking BIOs for DTLS as of
https://boringssl-review.googlesource.com/13945. This logic is a remnant
of that. It should not be necessary. All users of DTLSv1_get_timeout
call DTLSv1_handle_timeout. This gets it out of the way for
dtls_open_record calls which don't use dtls1_get_record.

We can restore it elsewhere if necessary, but I don't think we need it.

Change-Id: Idb737868358e4b59ad3cb2c994c7084ffcdb3709
Reviewed-on: https://boringssl-review.googlesource.com/21349
Commit-Queue: Steven Valdez <svaldez@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Steven Valdez <svaldez@google.com>
kris/onging/CECPQ3_patch15
David Benjamin 7 years ago
committed by CQ bot account: commit-bot@chromium.org
parent
commit
3b777adb61
1 changed files with 0 additions and 12 deletions
  1. +0
    -12
      ssl/d1_pkt.cc

+ 0
- 12
ssl/d1_pkt.cc View File

@@ -143,18 +143,6 @@ again:
// Read a new packet if there is no unconsumed one.
if (ssl_read_buffer(ssl).empty()) {
int read_ret = ssl_read_buffer_extend_to(ssl, 0 /* unused */);
if (read_ret < 0 && dtls1_is_timer_expired(ssl)) {
// Historically, timeouts were handled implicitly if the caller did not
// handle them.
//
// TODO(davidben): This was to support blocking sockets but affected
// non-blocking sockets. Can it be removed?
int timeout_ret = DTLSv1_handle_timeout(ssl);
if (timeout_ret <= 0) {
return timeout_ret;
}
goto again;
}
if (read_ret <= 0) {
return read_ret;
}


Loading…
Cancel
Save