archive.worldofdragon.org

sixxie wrote:Well, patched with the other suggestion in the thread then.

Is this the ssbssa April 21, 2013 at 9:27 PM patch?

Anyway, in the 3rd May version the TUI mode the cursor key problem remains.

However, things improve a little in non-TUI mode so that the up and down arrows move through the command history, but it still has the delete (backspace) key problem and a similar screen refresh problem when using the down arrow to move through the command history. In this respect it is still not as good as the original 29th April non-ncurses version since in that version none of these problems occurred.

Ah well, shame you can't use TUI. No nice user interface for WIndows users (yet) then! I've reverted the build.

Yes, it's a pity. It seems that we will have to wait for a fix in ncurses.

Sorry for necro'ing this, but I thought I'd be better asking my question in a XRoar+GDB thread rather than create a new one.

I was just wondering if its possible to see the raster scan position from Xroar in GDB? I'd normally use MESS's (+ QMC2) debugger for this, but it'd be handy if I could it from GDB on this occasion.

R

Hi Ciaran,
thanks a lot for your last XRoar snapshot for windows.
It works again flawlessly with GDB

I thought I had to tell, so that people that had stopped updating XRoar
and are still using 2015-33312 (the last one that worked fine with GDB)
know that the very last snapshot works alright, too.

cheers
pere

Glad I finally understood where the problem was... I was just testing the wrong thing all along!

That snapshot also has the better simulated NTSC included and selectable through menu options. Release soon, hopefully, once I test all the usual stuff.

Hi Ciaran,
lately I have been forced to debug quite a lot my port of The Hobbit using
DDOS11C, DDOS13A, DDOS41 and DosPlus5.0
As my version works on MAP1, I start my program and then press a key on the GDB window
so that it intercepts XRoar, I can enter a breakpoint and so forth ...
This procedure works 100% of the times with old version 2015-33312 but fails more than
30% of the times with 2016-30041
Here is the result ...
Any idea of a way to overcome that situation?

Hi Ciaran,
now that you have sorted out the error problems when there are no disks in the drives,
I am using the last snapshot 2016-32341 but this one fails to connect to GDB half the times I try.
On the other hand, the old snapshot 2015-33312 connects always at first attempt!
I include here the screenshot of both GDB and XRoar.
Could I find any 'log' file corresponding to that problem?

cheers
pere

You could try running with "-debug-gdb -1" - that should emit all the gdb packets sent & received by XRoar. Of course if it doesn't connect at all, that's not going to be helpful...

Are there particular things you're doing when it fails? Quick testing under Linux shows I can connect & disconnect at will, no fails.

Hi Ciaran,
of course, when it refuses to connect, there is no log file anywhere ...
But, today it has been connecting more than 90% of the times and I have done
quite a lot of debugging!

I don't know if there is a way to tell gdb to wait a bit more time before sending the 'invalid remote reply" message
Will make more tests tomorrow in the morning

cheers
pere