Re:Strange behaviour of gcc

2002-12-29 Thread Danny Smith
Hello Randall Yeah, no documentation to speak of. The only reason I know about it because I've been chasing the _alloca bug that Fish reported. It is still present in GCC trunk (3.4). I have submitted a simple patch that fixes, but ... time for a ping in the New Year. Danny, rrschulz wrote:

Re:Strange behaviour of gcc

2002-12-29 Thread Randall R Schulz
Danny, Man! I scanned through the GCC man page for anything that would control this action, and couldn't find anything. I don't see "-mno-stack-arg-probe" listed there at all, nor is any option that includes the word "probe." Google ("GCC mno-stack-arg-probe"

Re:Strange behaviour of gcc

2002-12-29 Thread Danny Smith
> > GCC's __builtin_alloca uses a helper function called _alloca to check the > stack > whenever allocating more that 4000 bytes in one go. > > Danny To disable stack probing, add this switch -mno-stack-arg-probe. Danny http://movies.yahoo.com.au - Yahoo! Movies - What's on at your local ci

Re:Strange behaviour of gcc

2002-12-29 Thread fabrizio_ge-wolit
From: Randall R Schulz To: cygwin at cygwin dot com Date: Tue, 24 Dec 2002 08:16:59 -0800 Subject: Re:Strange behaviour of gcc References: <[EMAIL PROTECTED]> >Stack space is usually far more limited than heap >space, which I assume is what motivates this behavior >in the code

Re:Strange behaviour of gcc

2002-12-24 Thread Randall R Schulz
a performance problem because some heap allocations are involved. Randall Schulz At 07:21 2002-12-24, [EMAIL PROTECTED] wrote: >-- Messaggio Originale -- >Date: Tue, 24 Dec 2002 16:11:34 +1100 (EST) >From: Danny Smith <[EMAIL PROTECTED]> >Subject: Re:Strange behaviour of gcc &

Re:Strange behaviour of gcc

2002-12-24 Thread fabrizio_ge-wolit
>-- Messaggio Originale -- >Date: Tue, 24 Dec 2002 16:11:34 +1100 (EST) >From: Danny Smith <[EMAIL PROTECTED]> >Subject: Re:Strange behaviour of gcc >To: [EMAIL PROTECTED] >Cc: [EMAIL PROTECTED] > > >[EMAIL PROTECTED] wrote: >> Is there a reason why

Re:Strange behaviour of gcc

2002-12-23 Thread Danny Smith
[EMAIL PROTECTED] wrote: > Can somebody explain why gcc (version 3.2 20020927) on Cygwin does > this? Type this simple C program > > void func(void){ > struct {unsigned char data[3985];}var; > } > > and compile with > > gcc -c filename.c > > Then type > > nm filename.o > > The output is > > 0