Re: [sdcc-devel] longlit.c test within regression test suite fails

[Frieder F. <Fri...@we...>]

Hi Maarten,

"Maarten Brock" <sou...@ds...> schrieb am 07.05.2007 15:40:47:
> Why is it surprising this test fails? It generates a warning about integer
> overflow in expression. If you change 19200 to 19200L it's allright.

Fine. The surprising part (to me) was that apparently some of the
tests were not executed because the white spaces around the function
definition did not match. I noticed with regtrack.c and then looked
for other tests that wouldn't be executed either and found longlit.c. 
I didn't investigate further but I should have noticed the missing L.

> Or did SDCC in the past detect that 12*19200 does not fit in an int and
> automatically upcast the result to a long? I don't know, as I was not
> using SDCC back in 2001 when this test was added. What was this regression
> test exactly supposed to check?

Sorry I don't know why it was added.

Greetings,
Frieder
_______________________________________________________________
SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192