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: [csl-hpux-branch]: Use targetm structure for manglings


Nathan Sidwell <nathan@codesourcery.com> writes:

>> Also, the hash table initialization
>> would be better broken out into a separate function, called from
>> init_mangle.
> Unfortunately there is a twisty interdependency on initializing the
> mangler. Part needs to be done early and parts late.

I don't see why.  The table's initializer is entirely static data,
isn't it?

> IMO lazy initialization is the simplest manner, and inline with the
> other hash table in mangle.c.

If it must be lazily initialized, at least give the initialization its
own function so that write_builtin_type remains readable.

>> And, since you removed all of the /* Foreign language type mangling
>> section */, you should delete that comment too.
> ok with this one change?

You can check in a patch that has the comment removed and the
initialization of builitin_type_manglings broken out of
write_builtin_type, whether or not that initialization function is
called lazily from write_builtin_type.

zw


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