Re: [PATCH] 2.4.19-rc2-ac2 pdc202xx.c update

[Michal Jaegermann]

On Tue, Jul 23, 2002 at 09:19:15AM +0200, Francois Romieu wrote:
> support <[e-mail]> :
> > We think there is no problems,
> 
> After the change, the code is:
> 
> if (speed == XFER_UDMA_2)
>         OUT_BYTE((thold + adj), indexreg);              <- not executed
>         OUT_BYTE((IN_BYTE(datareg) & 0x7f), datareg);   <- executed, damn it !

I have one more question.  Is it really immaterial on the line above in
which order 'datareg' occurences are used?  Regardless of what
'OUT_BYTE' is now or may become in the future and how these macro are
used?  It looks to me that we are trying to read some byte from
'datareg', clear a bit in it and write it back but looks can be
deceptive.  It may turn out that this does or does not work on a
particular compiler whim.

Maybe it is ok now but beeing explicit in a macro definition does
not really cost anything and may save some future grief.

   Michal