Re: [hatari-devel] Problems with some FPU tests

[Thomas Huth <th.huth@xxxxxxxxx>]

Am Mon, 23 Apr 2018 20:18:56 +0200
schrieb Thorsten Otto <admin@xxxxxxxxxxx>:

> On Montag, 23. April 2018 19:12:51 CEST Thomas Huth wrote:
[...]
> >Hatari should certainly not segfault in this case. Could you attach a
> >gdb and get a backtrace?  
[...]
> With configuration Debug (which still seems to add some optimization
> flags) i get
> 
> #0  0x00000000005dd73e in ScreenConv_BitplaneTo32bppZoomed (coefy=2, 
> coefx=<optimized out>, lowerBorder=0, upperBorder=0, rightBorder=0, 
> leftBorder=0, hscrolloffset=0, nextline=160, vbpp=4, vh=40191600,  
>    vw=21127456, scrheight=80383200, scrwidth=1280, hvram=<optimized
> out>, fvram=0x1406be0 <STRam>)
> out>at /home/sebilla/atari/hatari/src/screenConvert.c:798 

I was finally able to reproduce the crash. Looks like this can also be
reproduced in normal mode when you set bUseSdlRenderer = FALSE in your
config file.

Let me ask to more question to be sure:

- You are using monochrome resolution in TT mode in your config file,
  right?

- Is your host display vertical resolution less than 960 pixels?

My display on got a resolution of 900 lines, and it seems like SDL2 is
only creating a surface with that hight for the window, so the
SDL_GetWindowSurface() in screen.c gets us only a surface that is too
small. Since TT high resolution has a height of 960 pixels, hatari
crashes here as soon as it tries to access anything beyond line 900.
This looks like an SDL2 bug to me.

Anyway, I've now added a patch that sets DisableDisplay = true in dummy
mode, so at least in that case, the problem should now be gone.

 Thomas