Can you apply a patch to DOSBox CVS and compile it for your Mac?

I'll try it but I've got 2 questions :

The official DOSBox source?

This may be a dumb question : Can I apply the patch on the source on a PC and build that source on Mac ?

Thanks for the patch

to this source:
http://dosbox.linuxsecured.net/dosboxcvs.tgz
and yes you can apply it everywhere. however the mac is probably easier as i think it has all utilities needed installed by default

OK thanks 😀
Gonna test in on Mac now.

wd wrote:

Get a debugger, debug it, be happy.

Please wd, read the post will ya 😀. I already found the bug, it's indeed an endian issue. I just don't have the time to actually fix it (luckily HAL did).

JAL

Please wd, read the post will ya . I already found the bug, it's indeed an endian issue. I just don't have the time to actually fix it (luckily HAL did)./

OK I was already planning to do that. I had something else to do first but now I'm working on it, I've applied the patch and now I'm compiling the SDL thing.

Please wd, read the post will ya

I already did that, hence the reply.

luckily HAL did

Nice, thanks.

Sorry for the late reply, got some trouble compiling the SDL.

It works, thanks h-a-l-9000 & jal for the patch 🤣

Uhm don't we just need something like
host_readw(...)
instead of
bitsA = *(Bit16u *)&vga.mem.linear[mouseaddr];
or is there more to it?

I hope my change is more efficient too

Uhm don't we just need something like host_readw(...) instead of bitsA = *(Bit16u *)&vga.mem.linear[mouseaddr]; or is there […]
Show full quote

Uhm don't we just need something like
host_readw(...)
instead of
bitsA = *(Bit16u *)&vga.mem.linear[mouseaddr];
or is there more to it?

I hope my change is more efficient too

If someone makes a .diff I could test it.

I hope my change is more efficient too

It changes some additional stuff that i can't verify, either might be correct.
So i'd prefer to stick to the smaller changes.

At most it would break Win3.11, Win95 and whatvga 😉
(these work for me though)

> If someone makes a .diff I could test it.
The difference is only small, probably.

Bitu mouseaddr = ((Bit32u)vga.s3.hgc.startaddr) * 1024 + moff * 16;

What if moff is negative? moff+=64 doesn't necessarily make it positive,
and mouseaddr is unsigned so might be very large.

I've worked some more on the patch, some of the cursor variables need to be limited in value. I'll probably provide a 8-bit and a 16-bit access version then somebody can test which one is faster. But not today as I'm forced to visit some relatives 😉

Now it doesn't have so much in common with the original any more xD
The cursor values are now more extensively checked for overflows.

I think I'll skip the 16-bit one...

Patch works!

The cursor values are now more extensively checked for overflows.

Does that mean that it's faster/better than the first patch ?

Oh dear. <sarcasm on>Of course, DOSBox will be at least 10x faster, on any system, with that patch applied! This patch alone makes it possible to run DOSBox on anything with a CPU, from a 70's pocket calculator to your grandpa's pacemaker, and play "Blood" in SVGA at full FPS on each device. It's the same thing as with Microsoft security patches, which accelerate the system twofold with each and every single patch. Oh, the wonders of information technology!<sarcasm off>.

Oh dear. <sarcasm on>Of course, DOSBox will be at least 10x faster, on any system, with that patch applied! This patch alone makes it possible to run DOSBox on anything with a CPU, from a 70's pocket calculator to your grandpa's pacemaker, and play "Blood" in SVGA at full FPS on each device. It's the same thing as with Microsoft security patches, which accelerate the system twofold with each and every single patch. Oh, the wonders of
information technology!<sarcasm off>.

I know that I won't make DOSBox any faster with a patch to fix the mouse but I thought maybe the mouse would work better or something like that? Why did he made another patch if it didn't change anything? I'm only asking it to know the difference.

It will be a small bit harder to make dosbox crash in the future 😉

OK, thanks
<Last question> Will patches like these be added in the daily CVS ?