Hello Marcus, my first motivation was to remove some compiler warnings when I got conflicts between vararg and stdarg. I think the stdarg interface is "better" because it is a standard (ANSI). So I personally never use the out-of-date vararg interface.
The dependency on glibc headers exists because of the header files involved, but there should not be an object or library dependency because the stdarg interface consists of inline or builtin functions. But to be true: I did not check this. I just tried to use the patched kernel and did not observe any problem which might be related to glibc. Regards, Stefan -----Original Message----- From: Marcus Brinkmann [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 04, 2000 12:09 AM To: OKUJI Yoshinori Cc: Weil, Stefan 3732 EPE-24; [email protected] Subject: Re: gnumach with gcc-2.95.1 or newer On Tue, Feb 29, 2000 at 01:36:30PM +0900, OKUJI Yoshinori wrote: > Although I just glanced your patches, they are basically right, I > think. If the patched kernel can be built cleanly, they should be > applied to the CVS. I am not sure that it is the right thing to remove sys/vararg and replace it with stdarg, as this seems to be creating a dependency on glibc which could cause trouble when bootstrapping the Hurd (gnumach -> hurd-header -> glibc -> hurd). Stefan, what was your motivation to replace these calls? Thanks, Marcus -- `Rhubarb is no Egyptian god.' Debian http://www.debian.org Check Key server Marcus Brinkmann GNU http://www.gnu.org for public PGP Key [EMAIL PROTECTED], [EMAIL PROTECTED] PGP Key ID 36E7CD09 http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/ [EMAIL PROTECTED]
