> got a look to the issues you brought up:
> 1/ regarding the first one, I rewrote the fix to be consistent with the rest of
> mciSendString (and fixed a bug in returned values - foudn while I was testing
> it -, and a memory leak)
> 2/ regarding the second one, the bug was not in command table storage but in
> the way Wine was linking a loaded specific command table with a driver.
>
> the driver specific command tables shall be returned in the structure passed
> with the DRV_OPEN message. This was ok in case of a 32 bit MCI driver, but
> the message parameters mapping was wrong for 16 drivers, hence not storing a
> relevant specific command table (if the driver did load it, as in your case).
>
> I hope the attached patch should fix your issues (if it does, Alexandre please
> use this patch instead)
It fixes them, thank you :) Alexandre, please use Erics patch :)
Ciao, Marcus
