|
From: | Paul Eggert |
Subject: | bug#60220: 29.0.60; macOS 13.1 crash shortly after starting Emacs |
Date: | Thu, 29 Dec 2022 16:08:37 -0800 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.4.2 |
On 12/29/22 15:59, Aaron Jensen wrote:
On Thu, Dec 29, 2022 at 6:49 PM Paul Eggert <eggert@cs.ucla.edu> wrote:Could it by that the dynamic linker is being invoked recursively? That is, while something is being dynamically linked, a SIGALRM or equivalent arrives, some idle timer code is run, and the dynamic linker is invoked before the outer linker finishes? I imagine that might put some dynamic linkers into a tizzy.I have no idea, would we see that in the trace at all?
Not necessarily from the traces you sent. For example, if the inner dynamic linker function corrupts data but returns, the crash will occur when there's only one dynamic linker activation record on the stack, even though that call wasn't the problem.
It may require some macOS expertise and access to a dynamic debugger to diagnose this further. (I have neither.) Possibly you could deduce it by running Emacs under dtrace or equivalent (sorry, can't help you with that either).
[Prev in Thread] | Current Thread | [Next in Thread] |