>
>
> USING THIS CAN CAUSE SEVERE FILESYSTEM CORRUPTION.
Sorry if this happend to you. But I tried to make clear, that this is
really experimental code just to show some basic features of the K6.
>
> In the interests of science I tried out the code that was posted on my
> early k6-166.
>
> The fastcall program segfaulted.
That was the supposed behavior. Since I wasn't able to get the
segments right I forced a segmentation fault to get the information
about register contents.
>
> I tried to run it under gdb and my system instantaneously rebooted,
> leaving me with a badly damaged root filesystem.
Thats the strange thing. I played around with this little program and
it never rebootet my system. Sorry again.
>
>
> So be careful.
I can second that. The code is experimental and just to show some
basic features. It works for me but I didn't had the chance to try it
on other K6 versions. I don't know what will happen, if the version
used is a newer K6 that can specify separate segment selectors for
sysret.
So be careful and try to understand what happens before using the
code. (especially since the small programm is used to measure latency
and the module has to be configured for it (define MEASURE, before
executing fastcall or load a smaller value to ebx)).
Jean
-- I get up each morning, gather my wits. Pick up the paper, read the obits. if I'm not there I know I'm not dead. So I eat a good breakfast and go back to bed. Peete Seeger- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu