Notices


Reply
Thread Tools
Posts: 503 | Thanked: 267 times | Joined on Jul 2006 @ Helsinki
#21
Originally Posted by mlvj
This is the answer:
char somedata [] __attribute__((aligned (4))) = {
.......
};
Hmm, now I'm curious, is the code with different attribute placement wrong (if that's the case, I'll have to look through my code fixing it)? Could you post a testcase showing the problem?
 
Posts: 105 | Thanked: 1 time | Joined on Feb 2006
#22
Hi Serge,

For some reason when I refreshed, I didn't see your answer post!

I located the way I did it using good old google (and lots of it).

Probably both positionings work, if yours works. I stuck mine according to the instructions on the page from where I copied it.

The port still has problems though - that was just the second hurdle. It segfaults on the 770 where it doesn't in scratchbox, and there are some user interface issues - the click of the stylus causes a click in the last position the mouse pointer was, and some things expect a right mouse click.
 
Posts: 503 | Thanked: 267 times | Joined on Jul 2006 @ Helsinki
#23
Originally Posted by mlvj
For some reason when I refreshed, I didn't see your answer post!

I located the way I did it using good old google (and lots of it).

Probably both positionings work, if yours works. I stuck mine according to the instructions on the page from where I copied it.
OK, I understand now, just your previous post sounded like the other positioning is wrong, so I got worried

Well, I actually prefer searching google first and posting questions in forums only if the search did not reveal anything useful, but nevermind

The port still has problems though - that was just the second hurdle. It segfaults on the 770 where it doesn't in scratchbox, and there are some user interface issues - the click of the stylus causes a click in the last position the mouse pointer was, and some things expect a right mouse click.
The segfault is most likely caused by a similar alignment issue. You can try adding lots of guard assert checks to the code verifying pointers alignment and structure packing somewhere near the code that fails.

Anyway, congratulations on your progress so far, now you should be able to eventually hunt down all these bugs. I know that as I was stuck with a similar problem and spent some time figuring out these arm architecture specific tips and tricks. By the way, probably it would be a good idea to make a page at maemo wiki with x86 -> arm porting guide as some people may have the same difficulties but are too shy to ask for help in the forum
 
Posts: 105 | Thanked: 1 time | Joined on Feb 2006
#24
I take your points.

I think I'm getting old!

Spending an hour or so coding each night (after a full day's work) has left me shattered, so I skipped it yesterday.

- good idea about the FAQ on Maemo wiki. When I get a chance I'll put some words in.

- and I've created a maemo garage project for this, so other people will be able to upload stuff.

I got csb working as far as loading, saving, playing whole first level pretty much. Segfaults when you go downstairs.

So my current tasks are:

- sort the segfaults as I find them
- package as an armel debian
- get sound working
- do a wiki entry
 
Posts: 88 | Thanked: 0 times | Joined on Jul 2008
#25
is this game functional?
 
Reply

Thread Tools

 
Forum Jump


All times are GMT. The time now is 21:57.