linux-kernel - Re: [PATCH 1/2] lockref: speculatively spin waiting for the lock to be released

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20240618-spotten-scheren-0c87d248419f@brauner>
Date: Tue, 18 Jun 2024 14:11:44 +0200
From: Christian Brauner <brauner@...nel.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Mateusz Guzik <mjguzik@...il.com>, viro@...iv.linux.org.uk, 
	jack@...e.cz, linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH 1/2] lockref: speculatively spin waiting for the lock to
 be released

On Thu, Jun 13, 2024 at 12:33:59PM GMT, Linus Torvalds wrote:
> On Thu, 13 Jun 2024 at 11:56, Linus Torvalds
> <torvalds@...ux-foundation.org> wrote:
> >
> > I didn't *think* anything in the dentry struct should care about
> > debugging, but clearly that sequence number thing did.
> 
> Looking at the 32-bit build, it looks like out current 'struct dentry'
> is 136 bytes in size, not 128.
> 
> Looks like DNAME_INLINE_LEN should be reduced to 36 on 32-bit.
> 
> And moving d_lockref to after d_fsdata works there too.
> 
> Not that anybody really cares, but let's make sure it's actually
> properly done when this is changed. Christian?

So I verified that indeed on i386 with CONFIG_SMP we need 36 to get to
128 bytes and yes, moving d_lockref after d_fsdata works. So I've put
that patch on top. (Sorry, took a bit. I'm back tomorrow.)