planetmaker wrote:

savegame file - wait till Dec 29.

Attachments

• autosave113.sav (336.61 KiB)

This comment was imported from FlySpray: https://bugs.openttd.org/task/2478#comment5147

SmatZ wrote:

Train 3 is catching its own tail, PBS track following is in endless recursion.

This comment was imported from FlySpray: https://bugs.openttd.org/task/2478#comment5148

Rubidium wrote:

Simply checking whether we reach the begin tile solves the problem, but I fear that that won't work when reserving a path into a figure 9 (or 6) as the begin tile is not within the loop. So we need some kind of loop detection, but what's the upper bound for the length of a reserved path? I don't hope it's in the order of 2048x2048x2 (2 tracks per tile) as that's horribly expensive (CPU wise). A depth of 2048 doesn't adversely affect the behaviour, but there are cases of people building single stretches that long just to tell us that the signals don't work anymore.

This comment was imported from FlySpray: https://bugs.openttd.org/task/2478#comment5167

michi_cc wrote:

This patch should solve the endless looping in all situations I can think of,
but I might have missed some special case. Some more testing wouldn't hurt.

Attachments

• fs2478.diff (1.89 KiB)

This comment was imported from FlySpray: https://bugs.openttd.org/task/2478#comment5199

michi_cc closed the ticket.

Reason for closing: Fixed

In r14849.

This comment was imported from FlySpray: https://bugs.openttd.org/task/2478