Hi,
There's something in the examples that sounds strange to me.
In the write loop of some non blocking examples, we have:
do {
nread = fread(mem, 1, sizeof(mem), local);
if (nread <= 0) {
/* end of file */
break;
}
ptr = mem;
do {
/* write data in a loop until we block */
while ((rc = libssh2_sftp_write(sftp_handle, ptr, nread)) ==
LIBSSH2_ERROR_EAGAIN) {
;
}
ptr += rc;
nread -= nread;
} while (rc > 0);
} while (1);
In the inner do while loop, shouldn't it be:
ptr += rc;
nread -= rc
instead of this curious:
ptr += rc;
nread -= nread;
=> after this; nread is always 0 (right ?) and if rc returned by
libssh2_sftp_write was < nread, then the next loop will be in fact
rc = libssh2_sftp_write(sftp_handle, ptr, 0)...
Other question:
when libssh2_sftp_write returns LIBSSH2_ERROR_EAGAIN, why not do a
select() on the socket instead of recall libssh2_sftp_write immediately
which I presume is consuming CPU cycles?
JL
------------------------------------------------------------------------------
_______________________________________________
libssh2-devel mailing list
libssh2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libssh2-devel
