On Fri, 9 Sep 2011, liuzl wrote:
> The EAGAIN returned from adjusting the receive window is not really EAGAIN,
> actually it is a BAD_USE.
I don't get this explanation.
In the function _libssh2_channel_receive_window_adjust() there's a call to
_libssh2_transport_send(). If this transport function returns EAGAIN, then
that will be returned from the receive window function. All other errors will
cause a LIBSSH2_ERROR_SOCKET_SEND to get returned.
I don't see how the receive window function would cause a BAD_USE return code.
Can you elaborate?
> If we don't receiving data until the window-adjust packet sent. Actually we
> will wait until all request packets sent(4 * buffer_size / 2000).
Why? There's a check of the window_size in every call to
_libssh2_channel_read() ?
-- / daniel.haxx.se _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-develReceived on 2011-09-08