Re: pkg/57145: gmake: *** INTERNAL: readdir: Operation not supported. Stop.

[" Taylor R Campbell via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR kern/57145; it has been noted by GNATS.

From: Taylor R Campbell <riastradh%NetBSD.org@localhost>
To: Thomas Klausner <wiz%NetBSD.org@localhost>
Cc: gnats-bugs%NetBSD.org@localhost, netbsd-bugs%NetBSD.org@localhost
Subject: Re: pkg/57145: gmake: *** INTERNAL: readdir: Operation not
	supported. Stop.
Date: Tue, 10 Sep 2024 16:04:58 +0000

 It looks like the EOPNOTSUPP is explained by autofs's lack of lseek.
 
 I'm still unclear on the EINVAL.  My best guess is that:
 
 1. somehow, fp->f_offset gets set to something negative
 2. sys_lseek -> vn_seek -> VOP_SEEK -> genfs_seek fails with EINVAL
 
 I can't find a way for (1) to happen on ufs or tmpfs, but maybe it
 could happen with nfs -- reviewing that code is a lot of work.
 
 One path that crossed my mind is a race in readdir by multiple threads
 or processes on the same file object, which might have been fixed by
 <https://mail-index.netbsd.org/source-changes/2023/04/22/msg144249.html>
 (which has not been pulled up to netbsd-9), but I doubt that's
 happening here.
 
 So I'll go ahead and commit the readdir(3) fix, but I'm still
 suspicious of EINVAL.  It would be worthwhile to try to instrument
 this and reproduce it to find out where it's coming from, in case
 there's another bug lurking.