linux-kernel - Re: Why use plain numbers and totals rather than predef'd constants for RPC sizes?

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

Message-ID: <c61e6b608f5d5537bb23892be27a52c92e1bd85d.camel@ibm.com>
Date: Thu, 20 Mar 2025 21:48:06 +0000
From: Viacheslav Dubeyko <Slava.Dubeyko@....com>
To: "idryomov@...il.com" <idryomov@...il.com>,
        David Howells
	<dhowells@...hat.com>
CC: "dongsheng.yang@...ystack.cn" <dongsheng.yang@...ystack.cn>,
        "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        "ceph-devel@...r.kernel.org" <ceph-devel@...r.kernel.org>,
        "slava@...eyko.com" <slava@...eyko.com>,
        "linux-kernel@...r.kernel.org"
	<linux-kernel@...r.kernel.org>,
        Alex Markuze <amarkuze@...hat.com>,
        "jlayton@...nel.org" <jlayton@...nel.org>,
        "linux-block@...r.kernel.org"
	<linux-block@...r.kernel.org>
Subject: Re:  Why use plain numbers and totals rather than predef'd constants for
 RPC sizes?

On Thu, 2025-03-20 at 14:59 +0000, David Howells wrote:
> Viacheslav Dubeyko <Slava.Dubeyko@....com> wrote:
> 
> > > -	dbuf = ceph_databuf_reply_alloc(1, 8 + sizeof(struct ceph_timespec), GFP_NOIO);
> > > -	if (!dbuf)
> > > +	request = ceph_databuf_reply_alloc(1, 8 + sizeof(struct ceph_timespec), GFP_NOIO);
> > 
> > Ditto. Why do we have 8 + sizeof(struct ceph_timespec) here?
> 
> Because that's the size of the composite protocol element.
> 
> As to why it's using a total of plain integers and sizeofs rather than
> constant macros, Ilya is the person to ask according to git blame;-).
> 
> I would probably prefer sizeof(__le64) here over 8, but I didn't want to
> change it too far from the existing code.
> 
> If you want macro constants for these sorts of things, someone else who knows
> the protocol better needs to do that.  You could probably write something to
> generate them (akin to rpcgen).
> 

Yes, make sense. I totally agree with you. :)

Thanks,
Slava.