Message-ID: <3B38A9FF.5030000@lanet.lv>
Date: Tue, 26 Jun 2001 18:27:59 +0300
From: Andris Pavenis <pavenis AT lanet DOT lv>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.5-ac13 i686; en-US; rv:0.9) Gecko/20010522
X-Accept-Language: en
MIME-Version: 1.0
To: djgpp-workers AT delorie DOT com, snowball3 AT bigfoot DOT com
Subject: Re: gcc 3.0 released
References: <3B376547 DOT 14699 DOT 24583 AT localhost> <3B386A0F DOT 14316 DOT 3F2111 AT localhost>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com

Mark E. wrote:

>>!     if (sym[i].e_scnum == 0)
>>
> 
> This test is big (imo of course). It should also test whether the symbol is 
> actually external before complaining. I suggest the following (not-tested):
> 
> if (sym[i].e_sclass == C_EXT && sym[i].e_scnum == 0)
> 
> Section symbols like '.comment' have an e_sclass of C_STAT and an e_type of 
> T_NULL, so this test should fix the .comment problem at the same time.
> 


Tested under Linux with linux-to-DJGPP cross-compiler (not 3.0 release 
but slightly earlier). This change seems to fix the problem (at least 
for building emu387.dxe)

Andris