Re: pgserver SIGFPE crash bug

Thanks for catching this, I just committed a fix to CVS.

I guess this is why the antialiased primitives are still marked
"EXPERIMENTAL" in the config menu ;)


On Mon, Feb 10, 2003 at 12:31:42PM +0100, Daniele Pizzoni wrote:
> The function def_line in dvbl_primitives_aa.c if called with x1 == x2 or
> y1 == y2 (or both) causes a FP exception due to division by zero, as
> obvious looking the first lines of the function itself.
> 
> To trigger the bug run canvastst with a SDL/X11 server and wait. A small
> or stretched window (say 50x300) and a fast machine may help.
> 
> I don't know if the bug is in the caller or in the callee... so gdb
> backtrace follows :)
> 
> #0  0x0808a493 in def_line (dest=0x80d7d44, x1=77, y1=50, x2=77, y2=50, clr=65520, lgop=1)
>     at dvbl_primitives_aa.c:56
> #1  0x08077fdb in gropnode_draw (r=0xbffff53c, n=0xbffff59c) at render.c:786
> #2  0x080768d5 in grop_render (div=0x80f7604, clip=0x0) at render.c:236
> #3  0x08079a56 in divnode_redraw (n=0x80f7604, all=0) at div.c:441
> #4  0x08079b6a in divnode_redraw (n=0x80f75bc, all=0) at div.c:466
> #5  0x08079f86 in update (subtree=0x80f75bc, show=1) at div.c:599
> #6  0x08080bc7 in rqh_updatepart (r=0xbffff6b8) at requests.c:729
> #7  0x0807f53f in request_exec (r=0xbffff6b8) at requests.c:81
> #8  0x080803b2 in rqh_batch (r=0xbffff74c) at requests.c:491
> #9  0x0807f53f in request_exec (r=0xbffff74c) at requests.c:81
> #10 0x0806a69c in readfd (from=10) at request.c:323
> #11 0x0806ae32 in net_iteration () at request.c:598
> #12 0x0806afbd in pgserver_mainloop_iteration () at mainloop.c:91
> #13 0x0806af25 in pgserver_mainloop () at mainloop.c:55
> #14 0x0804b23a in pgserver_main (flags=0, argc=1, argv=0xbffff9f4) at init.c:194
> #15 0x0804ae88 in main (argc=1, argv=0xbffff9f4) at pgmain.c:35
> 
> Bye
> Dan
> -- 
> Daniele Pizzoni --- Ascensit Linux Solutions
> email: <[email protected]> - <[email protected]>
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Pgui-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/pgui-devel

-- 
Only you can prevent creeping featurism!


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com