This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: anonymous namespaces in GCC source code

From: Alec Teal <a dot teal at warwick dot ac dot uk>
To: Lawrence Crowl <crowl at google dot com>
Cc: Gabriel Dos Reis <gdr at cs dot tamu dot edu>, <jason at gcc dot gnu dot org>, <gcc at gcc dot gnu dot org>, <asutton at cs dot tamu dot edu>
Date: Tue, 19 Mar 2013 11:15:14 +0000
Subject: Re: anonymous namespaces in GCC source code
References: <87k3p4u55x dot fsf at gauss dot cs dot tamu dot edu> <CAGqM8faPay+uo+KYto=0tSCkfno10Qw2530FMmxfeGUR1Y1uOA at mail dot gmail dot com>


On 18/03/13 18:50, Lawrence Crowl wrote:

On 3/18/13, Gabriel Dos Reis <gdr@cs.tamu.edu> wrote:

I have been having discussion with Andrew about uses of anonymous
namespaces in GCC source code.  I seem to remember that they used
to cause troubles when doing binary diff during bootsrap because
we use random names to ensure uniqueness of names; but are we
still doing that?

We could have an option to take the compile time (and probably some
parts of the file path) out of the random number generator.

Why not hash the file name and line-number or something? Or add trueanonymous name spaces? Why do they require some sort of name?

Alec

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]