From: drs AT inxpress DOT net (DRS) Subject: Balter's InCivility Does Not Equal a Bug in ld [was: Re: Stupid stupid remarks 27 Mar 1997 13:09:08 -0800 Sender: daemon AT cygnus DOT com Approved: cygnus DOT gnu-win32 AT cygnus DOT com Distribution: cygnus Message-ID: <3339B9BE.482A.cygnus.gnu-win32@inxpress.net> References: <01BC386B DOT 709FACA0 AT gbird0> <33367A5B DOT 7807 AT netcom DOT com> <333897C6 DOT 777A AT inxpress DOT net> <33389AB5 DOT 4BE4 AT netcom DOT com> Reply-To: drs AT inxpress DOT net X-Mailer: Mozilla 3.01Gold (Win95; I) MIME-Version: 1.0 Original-To: Jim Balter Original-CC: gnu-win32 AT cygnus DOT com Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Original-Sender: owner-gnu-win32 AT cygnus DOT com Jim Balter wrote: > It has been proven that Win NT misbehaves in cases where win95 does not; > ld -s is a well-known example. Yes, as I said, it might be an "NT thing." But what does this have to do with the GNU linker? If the problem is NT related then you should be complaining to microsoft. Maybe the API's in question are not even supported equally across win95/NT. This is a distinct possibility, since one (the WriteProfileSection) is depricated for win32 apps, and the other is a hack no professional would likely need. > > // g++ -o test.exe test.cpp > > This has nothing to do with the bug being discussed. Wrong. You and Phil claimed the linker failed to handle certain API's. I have attempted to verify this and found the opposite. You claim to have proven there is a bug in the linker, but the only code you posted to demo this, turned out not to even compile, much less link. Instead of ad hominum attacks, you might try posting code which proves your point. The code *I* posted works correctly whether you use "g++ -o test.exe -mwindows --subsystem,windows" or not. > > So even though you and Phil claim to have found a linker bug, I have > > yet to be able to reproduce it. > > So what? You haven't run the same code, nor on the same system. Well, I'm using the same linker, which is building the same NE executable, using the same API's, and the code you and phil have provided--after correcting your errors, of course. What more do you want? > > The correct prototype for GetFullPathName looks like this, Jim: > > Well known to me, but irrelevant. If it was known to you, then why did you prototype the function pointer incorrectly? Why did you post code that would not even compile? And why claim "merely referencing the function is enough to cause the problem"? When it plainly is not? > > Regards as always. > > Not from me. Your lack of civility is amusing. But it proves nothing more than your general status as Unix oaf and Windows wanna-be. --DRS - For help on using this list, send a message to "gnu-win32-request AT cygnus DOT com" with one line of text: "help".