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]

Re: making naked attr generally available

To: rth at cygnus dot com
Subject: Re: making naked attr generally available
From: "Martin v. Loewis" <martin at mira dot isdn dot cs dot tu-berlin dot de>
Date: Wed, 2 Jun 1999 08:11:47 +0200
CC: bcurrie at tssc dot co dot nz, law at cygnus dot com, egcs at egcs dot cygnus dot com
References: <19990601004855.B20774@cygnus.com> <2944.928227573@upchuck.cygnus.com> <19990601093154.A21044@cygnus.com> <3754BFAA.439166F@tssc.co.nz> <19990601225225.A27332@cygnus.com>

> In the case of (1), we should write the _init and _fini functions to
> process .ctors and .dtors as one unit.

As a side note: Using DT_FINI_ARRAY for .dtors is probably wrong (as
is having .dtors in the first place) - at least for C++. In general,
this technique can't guarantee that destruction is in reverse order
from construction.

Regards,
Martin

References:
- Re: making naked attr generally available
  - From: Richard Henderson
- Re: making naked attr generally available
  - From: Jeffrey A Law
- Re: making naked attr generally available
  - From: Richard Henderson
- Re: making naked attr generally available
  - From: Bill Currie
- Re: making naked attr generally available
  - From: Richard Henderson

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