[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] New HSMINST.XDD



>Subject: [MiNT] New HSMINST.XDD
>Message-ID: <19990724133718 A666 -at- tubul dascon de>
>Mime-Version: 1.0
>Content-Type: multipart/mixed; boundary=EVF5PPMfhYS0aIcm
>X-Mailer: Mutt 0.95.3i
>Sender: owner-mint -at- fishpool com
>Precedence: bulk
>X-Loop: mint -at- fishpool com
>
>
>--EVF5PPMfhYS0aIcm
>Content-Type: text/plain; charset=us-ascii
>
>Hi,
>
>I just fixed an error in Jerry Geigers HSMINST.XDD, which caused it to crash
>in certain circumstances (which were random - hsminst.xdd would not
>recognize the end of the RSVF list and access non-initialized memory and
>probably crash). This happened a lot (but not always) when using the last
>Milan TOS (and the version before).

Unfortunately enough it still does not work on my Milan. It has exactly
the same problem as the old version, in detail:
If I boot nothing else but DRVIN.PRG, the two UART_XXX.PRGs, then
RSVX.PRG and then MiNT, booting stops at execution of hsminst.xdd.
hsminst just displays that it has installed SERIAL2 and MODEM2, after
that only the cursor is flashing forever.
However, if I boot e.g. NVDI beforehand, booting continues normally
(which I must say I find rather strange...)
Then under N_AES I switch the MiNT debug level to 2 which causes
thousands of lines passing by saying something like this:
pid xxx [process]: rsvf_ioctl: passing 4601 ...
(Unfortunately I just can´t remember the exact message, as I am at uni
at the moment...)
The process giving that message is mostly MINT, but all the other tasks
also show up with that message. Usually you don´t get many debug
messages at a level of two.

And I have observed another strange effect:
With my own RSVX.PRG - which only differs from the original one in that
way that it doesn´t *copy* the RSVF-Cookie, it just *renames* it - I
hardly get any debug messages apart from the usual ones. So the message
above never shows up. But the price for this is that, for example, the
pppd doesn´t hang up the modem any more.
How can the simple RSVX.PRG have such a strong effect on the whole
HSModem thing?

Has anyone got any ideas (or _solutions_)???

Well, I hope one day I´ll get MiNT-Net running on my Milan...

Cheers, Ingo =;->