From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from zeniv.linux.org.uk ([195.92.253.2]:37912 "EHLO
        ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1727789AbeK2Bne (ORCPT
        <rfc822;linux-fsdevel@vger.kernel.org>);
        Wed, 28 Nov 2018 20:43:34 -0500
Date: Wed, 28 Nov 2018 14:41:39 +0000
From: Al Viro <viro@zeniv.linux.org.uk>
To: Willy Wolff <willy.mh.wolff.ml@gmail.com>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: oops when ext4 fs is full
Message-ID: <20181128144139.GG2217@ZenIV.linux.org.uk>
References: <20181128085038.iaklny2yuygrixhs@inc028000049.lancs.ac.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20181128085038.iaklny2yuygrixhs@inc028000049.lancs.ac.uk>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

On Wed, Nov 28, 2018 at 08:50:39AM +0000, Willy Wolff wrote:
> Hi,
> I got a Oops when the hard drive was COMPLETELY full using a ext4 fs.
> After it, any command on the directory where the last write should have occurred freezes, while any other directory behave just fine.
> 
> If this email is not relevant for you, please point me what to do.
> 
> Best Regards,
> Willy
> 
> 
> $ cat /proc/version 
> Linux version 4.18.7-dirty (willy@inc028000049) (gcc version 6.3.0 20170516 (Debian 6.3.0-18)) #32 SMP PREEMPT Thu Sep 27 19:31:11 BST 2018
> 
> The kernel is dirty but the modifications are not related to the fs subsystem. The system is running on a Hardkernel Odroid-xu3 but running with the mainline linux at 8ea7eadbef2f7103d5871016f03fc38b3adeaddb
> Also, the fs is on a SD-card. I did a fs check after the oops, no errors are reported.

Disassembly of __d_instantiate in your build would obviously be useful...
_Presumably_ it's one of the spin_lock() inlined there, but it would be
useful to know which one...