commit 926ee8364e8c83d7a22dcd5bceb65e235486703a
from: Omar Polo <op@omarpolo.com>
date: Thu May 19 14:43:06 2022 UTC

consume all enqueued messages before calling imsg_read

player_dispatch reads only one imsg from the ibuf.  Next time it's
called, the other messages on the ibuf (if any) are discarded and new
ones are read.  This can cause the player process to go out-of-sync with
the main process if multiple messages were "bundled" in the same chunk.

To avoid this, always try to imsg_read before.  If it succeedes, we read
process the equeued messages one by one, if it fails then we poll for
new data and call imsg_read to process them and retry.

This fixes a bug where amused could be "confused" by running

	$ amused pause ; amused stop ; amused play

in a loop a few times.  This bug and the repro were reported two months
ago by Dirk-Wilhelm Peters, thanks!  (and sorry it took so long to
understand and fix the issue)

commit - fc4a38afafeec3a23f264e76659b3405d2d7cc48
commit + 926ee8364e8c83d7a22dcd5bceb65e235486703a
blob - 5fdc9aaf6d8e883142898ac8286585135a723535
blob + e65ba1ff5f86266696f18deedc4069455e3eae51
--- player.c
+++ player.c
@@ -146,20 +146,20 @@ player_dispatch(void)
 	if (halted != 0)
 		return IMSG_STOP;
 
-	pfd.fd = ibuf->fd;
-	pfd.events = POLLIN;
-	if (poll(&pfd, 1, INFTIM) == -1)
-		fatal("poll");
-
-	if ((n = imsg_read(ibuf)) == -1 && errno != EAGAIN)
-		fatalx("imsg_read");
-	if (n == 0)
-		fatalx("pipe closed");
-
+again:
 	if ((n = imsg_get(ibuf, &imsg)) == -1)
 		fatal("imsg_get");
-	if (n == 0) /* no more messages */
-		fatalx("expected at least a message");
+	if (n == 0) {
+		pfd.fd = ibuf->fd;
+		pfd.events = POLLIN;
+		if (poll(&pfd, 1, INFTIM) == -1)
+			fatal("poll");
+		if ((n = imsg_read(ibuf)) == -1 && errno != EAGAIN)
+			fatal("imsg_read");
+		if (n == 0)
+			fatalx("pipe closed");
+		goto again;
+	}
 
 	ret = imsg.hdr.type;
 	switch (imsg.hdr.type) {