My personal console is an AlphaStation 200 4/166. I always start xlock before leaving the workstation alone, but, since I upgraded to Redhat 6.2, it's very seldom been there when I return.
I've started to run xlock from xterms, so that I could get any error messages it could print. Here's one I've just got: xlock: caught signal 11 while running sproingies mode (uid 404).
The problem shows up only in modes that depend on MesaGL, so I'm disabling it by now. I'll try to attach the spec patch that accomplishes this.
Created attachment 273 [details] Don't use Mesa on alpha, it crashes within find_last_3f()
Hmm... we'll do that if we can't get this fixed before the next release. Ideally, I'd like to fix the GL libraries...
*** Bug 14275 has been marked as a duplicate of this bug. ***
Reassigning to Mesa, that's where the problem is.
I'm curious if this problem still occurs in Red Hat Linux 7.1 or not, or in any current Alpha builds of Mesa.
I'm afraid the box on which I had seen the problem died a couple of months ago, and the other, much newer alpha boxes I have access to, the problem never occurred, and I've just confirmed that it doesn't occur even on Red Hat Linux 7.0. Perhaps it would be appropriate to change this to WORKSFORME and be done with it? It might well be the case that the machine was already giving signs of dying back then, or it might have been some choice to use instructions not yet available on that old alpha machine.