Date: Thu, 11 Jul 2002 17:40:27 +0200 From: Bernd Walter <ticso@cicely5.cicely.de> To: bruno schwander <bruno@tinkerbox.org> Cc: hackers@FreeBSD.ORG Subject: Re: termios guru ? Message-ID: <20020711154027.GC33078@cicely5.cicely.de> In-Reply-To: <Pine.BSF.4.21.0207102056320.287-100000@duron.bschwand.net> References: <Pine.BSF.4.21.0207102056320.287-100000@duron.bschwand.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jul 10, 2002 at 09:13:18PM -0700, bruno schwander wrote: > I making a port (not much really) of Irit > (http://www.cs.technion.ac.il/~irit/) a modelling environment. > > I am having some problems with terminal handling, so all termios guru out > there, please help ! :-) That's what I did for a terminal programm to setup the controlling tty: void settty(void) { struct termios buf; if (tcgetattr(STDIN_FILENO, &save_termios) < 0) { printf("tcgetattr failed: %s\n", strerror(errno)); exit (1); } memcpy(&buf, &save_termios, sizeof(buf)); buf.c_lflag &= ~(ECHO | ICANON | IEXTEN | ISIG); buf.c_iflag &= ~(ICRNL | INPCK | ISTRIP | IXON); buf.c_cflag &= ~(CSIZE | PARENB); buf.c_cflag |= CS8; buf.c_oflag &= ~OPOST; buf.c_cc[VMIN] = 1; buf.c_cc[VTIME] = 0; if (tcsetattr(STDIN_FILENO, TCSAFLUSH, &buf) < 0) { printf("tcsetattr failed: %s\n", strerror(errno)); exit (1); } ttyset = 1; return; } > At stratup, irit does the following > > Termio.c_cc[VEOF] = 0; /* MIN = 0, no minimal length to wait for. */ > Termio.c_cc[VEOL] = 1; /* TIME - 1 tenth of a second as time o Setting VEOL or VEOF with disabled ICANON is senseless. > which seems wrong, I think it should be > > Termio.c_cc[VMIN] = 0; /* MIN = 0, no minimal length to wait for. */ > Termio.c_cc[VTIME] = 1; /* TIME - 1 tenth of a second as time o > > then later: > > Termio.c_lflag &= ~ICANON; Sparse initialisation may require specific defaults to work. > basically, irit wants to manage line editing itself, to manage the irit > command prompt. There is some code doing the ^A, ^H, etc handling and line > printing, and reading periodically stdin. > > What I see happening, is that usually at the very beginning, input seems > locked. Running in the debugger, I see that characters are fgetc'ed > periodically, but fgetc always returns -1 even when there should be > characters available. > > I then tried using fcntl(0, F_SETFL, O_NONBLOCK) instead of the above 2 > lines. which I thought would do the right thing, ie: non blocking IO, but > anything available from stdin is buffered and provided on the next read. > This works, however I am seeing something strange on stdout now: when > outputting lots of lines, outputs stalls after a few dozen lines. Adding a > usleep between each fwrite() solves the problem but slows it all > down... (and is inherently wrong) I'm not shure if nonblocking or character mode is compatible with stdio. I always used direct io in such cases. > What is going on here ? I do not understand very well all the terminal/IO > discipline here. > > I agree that this is all bad design, and should probably multithread or > use select() but I am not Irit's author... -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020711154027.GC33078>