This is the mail archive of the gcc-patches@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: HPPA constructor merge patch, PR middle-end/45388


On 12/13/2010 5:15 AM, Jan Hubicka wrote:

> Mark, does the C++ change look OK? It changes GLOBAL__I function that will never
> end up being the externally visible static constructor into GLOBAL__sub_I.

This might break targets (if any) that still use "munch".  This is a
utility that runs nm to find static constructors and generates a C
function called __main to call them all; that allows you to implement
C++ static constructors without any linker support for things like .ctors.

I've CC'd Nathan Sidwell and Nathan Froyd here; both of them are
familiar with VxWorks, which (at least the last time I used it) still
depended on munch.  Nathans, does this still apply?

Oh, wait, Jan, is that the whole point of your patch?  That we don't
want utilities like this to recognize these things as static constructors?

Thank you,

-- 
Mark Mitchell
CodeSourcery
mark@codesourcery.com
(650) 331-3385 x713


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