Mario Becroft (mailing list) wrote:
Steve,
Thanks for your response. I am sorry that we have failed to properly
comprehend the manuals. It is true that I am very busy and may not have
read them as carefully as I could have. Clearly, I was mistaken about
the selection of the parallel port. We will try it again and see if we
can get it to work.
So, what is the procedure for getting changes made to the MSP430 version
of gdbproxy? We still have some problems with it especially when using
newer devices. Are you the only person who has access to the code? Do
you always build new versions and put them on your web site whenever
changes are made? I guess I assumed I would have to get the code from
CVS to get the latest version.
All current devices are supported by the Windows version, as it supports
the USB interface for the Spy-Bi-Wire devices. The latest version for
Linux supports all devices usable with the parallel port FET tool. The
test version for Linux supports the USB tools, so it provides
Spy-Bi-Wire support. That version is not yet very well tested.
On the subject of multiple FET tools. We can support multiple parallel
and USB tools rght now. Most other tools, like IAR, can only support a
single USB FET tool. I regularly use a parallel FET tools and two or
three USB ones.
Examples of some of the problems we are experiencing include:
Programming is very slow on some devices (this may be a device
limitation, but it seems excessively slow).
Yes, that still needs to be properly addressed. The USB tools are much
faster, though.
Debugging with gdb has intermittent problems on some devices. Somehow
the device can get into a state where nothing works properly and it has
to be reset.
This happens with IAR too.
Some newer devices do not appear to work at all--they are not recognised
and we cannot program them.
Use the latest msp430-gdbproxy. Each new device requires an update.
None of these problems happen with the MSP430F1xxx devices that we have
tried.
Maybe we are just doing something stupid and it all works for everyone
else. In which case, I am sorry to have bothered everyone.
I know this is a bit vague, and it is not my intention here to complain
about these specific issues (I can provide more details if this is of
interest). I am just wondering if, in general, there is anything useful
we can do to actually fix problems or improve the software. And if not
(considering that the code is not available to us), is there is a
process whereby changes can be made by you, or someone else who is able
to maintain the software, perhaps if we pay for the work.
Steve
